diff --git a/.github/workflows/ci.yml b/.github/workflows/ci.yml index 9bc70ac1..88af9bbd 100644 --- a/.github/workflows/ci.yml +++ b/.github/workflows/ci.yml @@ -80,6 +80,15 @@ jobs: accountId: ${{ secrets.CLOUDFLARE_ACCOUNT_ID }} preCommands: ../db/deploy-${{ matrix.environment }}.sh environment: ${{ matrix.environment }} + - name: Deploy IPFS Retriever and Migrate Database + if: matrix.environment == 'calibration' + uses: cloudflare/wrangler-action@v4 + with: + workingDirectory: ipfs-retriever + apiToken: ${{ secrets.CLOUDFLARE_API_TOKEN }} + accountId: ${{ secrets.CLOUDFLARE_ACCOUNT_ID }} + preCommands: ../db/deploy-${{ matrix.environment }}.sh + environment: ${{ matrix.environment }} - name: Deploy Indexer uses: cloudflare/wrangler-action@v4 with: diff --git a/docs/data-layer.md b/docs/data-layer.md new file mode 100644 index 00000000..5e5c8708 --- /dev/null +++ b/docs/data-layer.md @@ -0,0 +1,236 @@ +# FilBeam Data Layer + +All workers share a **single Cloudflare D1 database** (SQLite). Migrations live in `db/migrations/`. + +- **Tests**: migrations are applied automatically via `wrangler d1 migrations apply test-db --local` (see root `package.json` test script). +- **Deploy**: worker deploys (`npm run deploy:calibration/mainnet`) do **not** apply migrations automatically. Run `db/deploy-calibration.sh` / `db/deploy-mainnet.sh` separately before deploying workers when migrations are pending. + +--- + +## Tables + +| Table | Description | +| ------------------------ | ------------------------------------------------------------------------------------------------------------------------- | +| `service_providers` | SP registry: service URL and deletion status, keyed by on-chain provider ID | +| `data_sets` | CDN deals: links an SP to a payer, tracks CDN/IPFS flags, egress usage, usage reporting watermarks, and termination state | +| `pieces` | Pieces per data set: piece CID, IPFS root CID (from chain metadata), and deletion flag | +| `data_set_egress_quotas` | Remaining byte budgets for CDN delivery and cache-miss charges; only exists for data sets that have been topped up | +| `wallet_details` | Payer addresses with sanction status and last Chainalysis screen timestamp | +| `retrieval_logs` | Per-request audit log: egress bytes, cache hit/miss, performance timings, country code, and bot flag | + +--- + +## ER Diagram + +```mermaid +erDiagram + service_providers { + TEXT id PK + TEXT service_url + INTEGER block_number + BOOLEAN is_deleted + } + data_sets { + TEXT id PK + TEXT service_provider_id + TEXT payer_address + BOOLEAN with_cdn + BOOLEAN with_ipfs_indexing + INTEGER total_egress_bytes_used + TIMESTAMP usage_reported_until + TEXT pending_usage_report_tx_hash + TIMESTAMP cdn_payments_settled_until + TEXT terminate_service_tx_hash + TIMESTAMP lockup_unlocks_at + } + pieces { + TEXT id PK + TEXT data_set_id PK + TEXT cid + TEXT ipfs_root_cid + TEXT x402_price + BOOLEAN is_deleted + } + data_set_egress_quotas { + TEXT data_set_id PK + INTEGER cdn_egress_quota + INTEGER cache_miss_egress_quota + } + wallet_details { + TEXT address PK + BOOLEAN is_sanctioned + TIMESTAMP last_screened_at + } + retrieval_logs { + INTEGER id PK + DATETIME timestamp + TEXT data_set_id + INTEGER response_status + INTEGER egress_bytes + INTEGER cache_miss_egress_bytes + BOOLEAN cache_miss + BOOLEAN cache_miss_response_valid + INTEGER fetch_ttfb + INTEGER fetch_ttlb + INTEGER worker_ttfb + TEXT request_country_code + TEXT bot_name + } + + service_providers ||--o{ data_sets : "hosts" + data_sets ||--o{ pieces : "contains" + data_sets ||--o| data_set_egress_quotas : "has quota" + wallet_details ||--o{ data_sets : "pays for" + data_sets ||--o{ retrieval_logs : "logs" +``` + +> Relationships are logical — no foreign-key constraints are declared in the schema. + +--- + +## How Tables Are Populated + +### `service_providers` + +Written by **indexer** in response to `ServiceProviderRegistry` on-chain events: + +| Event | Handler | Effect | +| ------------------------------------ | -------------------------------------- | --------------------------------------------------------------------------------------------------------- | +| `ProductAdded` / `ProductUpdated` | `handleProductAdded/Updated` | Upserts `id`, `service_url`, `block_number`; skips if stored `block_number` is newer (out-of-order guard) | +| `ProductRemoved` / `ProviderRemoved` | `handleProductRemoved/ProviderRemoved` | Sets `is_deleted = true` | + +### `data_sets` + +Written by **indexer** in response to `FWSS` and `FilBeamOperator` events: + +| Event | Handler | Effect | +| ------------------------- | ----------------------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | +| `DataSetCreated` | `handleFWSSDataSetCreated` | Upserts `id`, `service_provider_id`, `payer_address`, `with_cdn`, `with_ipfs_indexing`; also creates/updates `wallet_details` with sanction screen result **only when `withCDN` is set** | +| `ServiceTerminated` | `handleFWSSServiceTerminated` | Sets `with_cdn = false`, calculates and sets `lockup_unlocks_at` | +| `CDNPaymentRailsToppedUp` | `handleFWSSCDNPaymentRailsToppedUp` | Increments `data_set_egress_quotas` (idempotent via KV event dedup) | +| `CDNPaymentSettled` | `handleCdnPaymentSettled` | Advances `cdn_payments_settled_until` to the block timestamp | + +Written by **usage-reporter** after confirmed on-chain usage report: + +- Sets `usage_reported_until` watermark and clears `pending_usage_report_tx_hash` + +Written by **terminator** after confirmed termination transaction: + +- Sets `terminate_service_tx_hash` + +### `pieces` + +Written by **indexer** in response to `PDPVerifier` on-chain events: + +| Event | Handler | Effect | +| ------------------------- | --------------------- | --------------------------------------------------------------------------------------------------------------- | +| `PieceAdded` / `addPiece` | `insertDataSetPiece` | Upserts `id`, `data_set_id`, `cid`, `ipfs_root_cid`, `x402_price`; `ipfs_root_cid` comes from on-chain metadata | +| `PieceRemoved` | `removeDataSetPieces` | Sets `is_deleted = true` (batch, up to 50 per D1 statement) | + +### `data_set_egress_quotas` + +Written by **indexer** (`handleFWSSCDNPaymentRailsToppedUp`): converts top-up amounts to byte quotas using configured rates and increments both `cdn_egress_quota` and `cache_miss_egress_quota`. + +Decremented by **piece-retriever** / **ipfs-retriever** after each successful retrieval (only when `ENFORCE_EGRESS_QUOTA` is enabled): `cdn_egress_quota` is charged for all egress bytes served to the client; `cache_miss_egress_quota` is charged only on valid cache misses. + +### `wallet_details` + +Created/updated by **indexer** on `DataSetCreated`, but **only when `withCDN` is true** (Chainalysis API call per new payer). + +Re-screened periodically by **indexer** scheduled task (`screenWallets`): re-screens wallets not checked within the configured stale threshold, ordered oldest-first. + +### `retrieval_logs` + +Written by **piece-retriever** and **ipfs-retriever** after every request via `recordRetrieval` in `retrieval/lib/stats.js`. Written inside `ctx.waitUntil` so it does not block the response. + +--- + +## Key Queries by Worker + +### Retrieval candidate lookup (piece-retriever, ipfs-retriever) + +The shared query in `retrieval/lib/access.js` (`buildRetrievalCandidateQuery`) JOINs five tables in one shot: + +```sql +SELECT pieces.data_set_id, data_sets.service_provider_id, data_sets.payer_address, + data_sets.with_cdn, data_set_egress_quotas.cdn_egress_quota, + data_set_egress_quotas.cache_miss_egress_quota, + service_providers.service_url, service_providers.is_deleted AS service_provider_is_deleted, + wallet_details.is_sanctioned +FROM pieces +LEFT OUTER JOIN data_sets ON pieces.data_set_id = data_sets.id +LEFT OUTER JOIN data_set_egress_quotas ON pieces.data_set_id = data_set_egress_quotas.data_set_id +LEFT OUTER JOIN service_providers ON data_sets.service_provider_id = service_providers.id +LEFT OUTER JOIN wallet_details ON data_sets.payer_address = wallet_details.address +WHERE pieces.cid = ? -- piece-retriever +-- or: pieces.ipfs_root_cid = ? -- ipfs-retriever + AND pieces.is_deleted IS FALSE +``` + +The result rows are then filtered by `filterAuthorizedRetrievalCandidates` (authorization cascade) — see `retrieval/lib/access.js`. + +### Slug resolution (ipfs-retriever, slug flow) + +Two sequential queries in `ipfs-retriever/lib/store.js`: + +1. Resolve `(pieceId, dataSetId)` → `ipfs_root_cid` + `payer_address`: + +```sql +SELECT pieces.ipfs_root_cid, data_sets.payer_address +FROM pieces LEFT OUTER JOIN data_sets ON pieces.data_set_id = data_sets.id +WHERE pieces.id = ? AND pieces.data_set_id = ? +``` + +2. Then the retrieval candidate query above, keyed by `ipfs_root_cid`. + +### Usage aggregation (usage-reporter) + +Aggregates `retrieval_logs` per data set between the `usage_reported_until` watermark and a target timestamp, excluding bot traffic and data sets with a pending transaction: + +```sql +SELECT rl.data_set_id, + SUM(rl.egress_bytes) AS cdn_bytes, + SUM(CASE WHEN rl.cache_miss = 1 AND rl.cache_miss_response_valid = 1 + THEN COALESCE(rl.cache_miss_egress_bytes, rl.egress_bytes) ELSE 0 END) AS cache_miss_bytes +FROM retrieval_logs rl +INNER JOIN data_sets ds ON rl.data_set_id = ds.id +WHERE rl.timestamp > datetime(ds.usage_reported_until) + AND rl.timestamp <= datetime(?) + AND rl.egress_bytes IS NOT NULL + AND rl.bot_name IS NULL + AND ds.pending_usage_report_tx_hash IS NULL +GROUP BY rl.data_set_id +HAVING (cdn_bytes > 0 OR cache_miss_bytes > 0) +``` + +### Terminator + +Finds active CDN data sets whose payer is sanctioned and have no pending termination: + +```sql +SELECT DISTINCT data_sets.id +FROM data_sets +LEFT JOIN wallet_details ON data_sets.payer_address = wallet_details.address +WHERE data_sets.with_cdn = 1 + AND wallet_details.is_sanctioned = 1 + AND data_sets.terminate_service_tx_hash IS NULL +``` + +### Payment settler + +Finds data sets that need CDN payment rail settlement (active, recently reporting, not sanctioned): + +```sql +SELECT data_sets.id +FROM data_sets +LEFT JOIN wallet_details ON data_sets.payer_address = wallet_details.address +WHERE (data_sets.with_cdn = 1 OR data_sets.lockup_unlocks_at >= datetime('now')) + AND data_sets.terminate_service_tx_hash IS NULL + AND data_sets.usage_reported_until >= datetime('now', '-30 days') + AND (wallet_details.is_sanctioned IS NULL OR wallet_details.is_sanctioned = 0) +``` + +### Stats API + +- **Per data set** (`stats-api`): reads `data_set_egress_quotas` by `data_set_id` +- **Per payer** (`stats-api`): aggregates `data_set_egress_quotas` and `retrieval_logs` joined through `data_sets`, grouped by `payer_address` diff --git a/docs/ipfs-retriever.md b/docs/ipfs-retriever.md new file mode 100644 index 00000000..61c5e4f5 --- /dev/null +++ b/docs/ipfs-retriever.md @@ -0,0 +1,306 @@ +# ipfs-retriever + +## Overview + +`ipfs-retriever` is a Cloudflare Worker that serves IPFS content stored by Filecoin service providers (SPs). It receives a request identifying a specific dataset and piece, fetches the corresponding CAR archive from an SP, validates every block, and streams the raw file bytes to the client. + +It depends on the shared `@filbeam/retrieval` library for authorization, candidate selection, egress quota tracking, and the fetch lifecycle. + +--- + +## URL formats + +There are two entry points: + +### Slug subdomain (primary) + +``` +https://1-{base32(dataSetId)}-{base32(pieceId)}.ipfs.calibration.filbeam.io/{subpath} +``` + +The slug encodes both the on-chain `dataSetId` and `pieceId` as base32 bigints, prefixed with a version (`1`). Parsed in `ipfs-retriever/lib/request.js`. + +Examples: + +- `https://1-abc123-def456.ipfs.calibration.filbeam.io/` — root of the piece +- `https://1-abc123-def456.ipfs.calibration.filbeam.io/path/to/file.jpg` — specific file + +#### Why the IPFS CID is not in the slug + +The slug does not include the `ipfsRootCid`. The CID is looked up from D1 at request time: when an SP registers a piece on-chain via `addPiece`, it attaches `ipfsRootCID` as a metadata key. The indexer picks this up from the Goldsky webhook event and stores it in the `pieces` table keyed by `(dataSetId, pieceId)`. At retrieval time, the worker decodes the slug to get `dataSetId + pieceId`, queries D1 for the associated `ipfsRootCid`, and uses that CID to both find the right SP and construct the fetch URL. Encoding the CID in the slug itself would push the DNS label past the 63-character limit, and is unnecessary since the indexer already has it. + +ref.: [FIlBeam URL Format Doc](https://space-meridian.github.io/docs/Engineering/fefc0d538c414c8e96a56587e4ca75ce/FilBeam/FilBeam%20URL%20format%2027ecdd5cccdb806caeeaefad80cbf64d.html#27ecdd5c-ccdb-800e-ac46-de8a4a64f0a8) + +#### Why CID + wallet address can't both be in the subdomain + +The initial design proposed combining the IPFS CID and wallet address into a single subdomain component (e.g. `bafk123-0xabc.filbeam.io`). This was ruled out because a single DNS label is limited to 63 characters, too short to fit both a base58 CID and a 42-character Ethereum address together. (see [follow-up comment, issue #297](https://github.com/filbeam/worker/issues/297#issuecomment-3352714474)) + +#### Why the wallet address is not in the slug at all + +The wallet address is not needed in the slug because `dataSetId` alone is sufficient to look it up in D1. Omitting it keeps the subdomain short enough to be valid. The worker resolves the wallet from the dataset record at query time. (see [review comment, PR #312](https://github.com/filbeam/worker/pull/312#issuecomment-4797500253)) + +#### Why identifiers are in the subdomain + +Static websites served over IPFS often load sub-resources at absolute paths (e.g. ``). For these paths to resolve correctly, the dataset/piece identity must live in the subdomain, not the URL path, so that `/style.css` naturally maps to the right SP origin without any path rewriting. (see [original proposal comment, issue #297](https://github.com/filbeam/worker/issues/297#issuecomment-3346046091)) + +### Bare domain redirect (convenience) + +``` +https://ipfs.calibration.filbeam.io/{walletAddress}/{ipfsRootCid}/{subpath} +``` + +Handled by `handleDnsRootRequest` in `ipfs-retriever/bin/ipfs-retriever.js`. Validates that `walletAddress` has an authorized deal for `ipfsRootCid` (running the full authorization cascade: payment rail, CDN flag, sanctions check), looks up the corresponding `dataSetId + pieceId` in D1, builds the slug, and issues a 302 redirect to the slug subdomain URL. If the wallet is not associated with that CID, the request is rejected with 402 before any redirect occurs. This validation is based entirely on indexed on-chain metadata — FilBeam does not verify that the `ipfsRootCid` actually corresponds to the piece data stored by the SP. Useful for constructing shareable links without knowing the on-chain IDs upfront. + +If the request is to the bare domain with no wallet or CID (i.e. `https://ipfs.calibration.filbeam.io/`), redirects to `https://filbeam.com`. + +### Query parameters + +- `?format=car` — serve the raw CAR archive to the client without conversion or block validation. The worker proxies the SP response as-is. Per the [IPFS Trustless Gateway spec](https://specs.ipfs.tech/http-gateways/trustless-gateway/), CAR is a client-validated transport, the caller is responsible for verifying block integrity. +- `?format=raw` — not implemented (returns 400). The trustless gateway `raw` format returns only the single terminal block for a CID, not the full file content, which makes it useful for leaf-node lookups but complex for multi-block files or directories. According to @rvagg we should defer this until a concrete use case arises; tracked in [issue #295](https://github.com/filbeam/worker/issues/295) +- No `?format` — default; converts CAR to raw file bytes. The worker validates every block before streaming. Each block's bytes are hashed and compared to its CID multihash via `validateBlock()`. The client receives only bytes that have passed verification. + +--- + +## Request flow + +There are two entry points — the bare domain redirect and the slug flow — both handled by the same worker. + +```mermaid + sequenceDiagram + participant C as Client + participant W as ipfs-retriever + participant D1 as D1 + participant KV as Bad Bits KV + participant CF as CF Edge Cache + participant SP as Service Provider + + C->>W: GET /wallet/cid or slug.ipfs...filbeam.io/path + + W->>W: handleFetchRequest (lifecycle, bot auth) + + alt Bare domain redirect + W->>D1: getRetrievalCandidatesByWalletAndCid(wallet, cid) + D1-->>W: dataSetId + pieceId + W-->>C: 302 → slug subdomain + else Slug request + W->>W: parseRequest — decode slug → dataSetId + pieceId + subpath + W->>D1: Query 1 — resolve (dataSetId + pieceId) + D1-->>W: ipfsRootCid, payerAddress + W->>D1: Query 2 — find all SPs for ipfsRootCid (auth cascade) + D1-->>W: candidates[] + W->>KV: assertCidNotDenied(ipfsRootCid) + KV-->>W: OK or 410 + + loop Retry across candidates (random order) + W->>CF: fetch {spUrl}/ipfs/{cid}{subpath}?format=car + alt Cache HIT + CF-->>W: CAR stream + else Cache MISS + CF->>SP: GET /ipfs/{cid}{subpath}?format=car + SP-->>CF: CAR stream + CF-->>W: CAR stream + end + end + + alt format=car + W-->>C: CAR stream (no validation) + else default + W->>W: CarBlockIterator + validateBlock + unixfs-exporter + W-->>C: Raw file bytes + end + Note over W,C: Cache-Control: public, max-age=31536000 + end +``` + +### Step-by-step + +**① `handleFetchRequest` (shared lifecycle)** + +Sets up per-request context, registers an abort listener, rejects non-GET/HEAD with 405, redirects legacy `*.filcdn.io` domains to `*.filbeam.io` with 301, and runs `checkBotAuthorization` to validate the `Authorization` header against `BOT_TOKENS`. All errors thrown from here on are caught and converted to HTTP responses via `handleError`. + +**② Route detection** + +If the hostname matches the bare `DNS_ROOT` (e.g. `ipfs.calibration.filbeam.io`), the request is handled by `handleDnsRootRequest`: extracts `walletAddress` and `ipfsRootCid` from the URL path, runs the full authorization cascade to verify the wallet has a deal for that CID, looks up `dataSetId + pieceId` from D1, and issues a **302 redirect** to the slug subdomain. No content is served from this path. + +For slug requests, `parseRequest` strips the `DNS_ROOT` suffix from the hostname, splits the slug into `[version, encodedDataSetId, encodedPieceId]`, decodes each with `base32ToBigInt`, and extracts `ipfsSubpath` from `url.pathname` and `ipfsFormat` from `?format=`. + +**③ `getRetrievalCandidatesByDataSetAndPiece`** + +Two D1 queries: + +1. Resolve `(dataSetId + pieceId)` → `ipfsRootCid + payerAddress`. Throws 404 if the piece doesn't exist, has no payer, or has no `ipfsRootCid`. +2. Query all rows matching `pieces.ipfs_root_cid = ?` joined across `data_sets`, `service_providers`, `data_set_egress_quotas`, and `wallet_details`. Runs the authorization cascade over the results: + +| Check | Error if all rows fail | +| ----------------------------------- | ---------------------------- | +| Any rows at all | 404 — not indexed | +| SP exists and is not deleted | 404 — no SP | +| `payer_address` matches wallet | 402 — no deal for this payer | +| `with_cdn = 1` | 402 — CDN disabled | +| `is_sanctioned` is false | 403 — payer is sanctioned | +| `service_url` is set | 404 — SP not approved | +| `with_ipfs_indexing = 1` | 402 — IPFS indexing disabled | +| `ipfs_root_cid` is set | 404 — no CID on piece | +| (if `enforceEgressQuota`) quota > 0 | 402 — quota exhausted | + +Returns one candidate per authorized SP: `{ serviceUrl, serviceProviderId, dataSetId, pieceId, ipfsRootCid }`. Multiple candidates exist when the same dataset is served by more than one SP — the worker retries across them if one fails. + +**④ `assertCidNotDenied`** + +Checks `ipfsRootCid` against the Bad Bits denylist stored in KV. Returns 410 if blocked. + +**⑤ `selectRetrievalCandidate`** + +Shuffles candidates randomly (no fixed priority) and tries them one by one. A candidate is skipped if its retrieval throws or returns a 5xx. If all candidates fail, logs the failure and returns 502 listing all attempted SPs. + +**⑥ `retrieveIpfsContent`** + +Fetches `{serviceUrl}/ipfs/{ipfsRootCid}{subpath}?format=car` with Cloudflare cache options (`cacheEverything: true`, TTL 86400 for 2xx, 0 for 4xx/5xx). Reads `CF-Cache-Status`: anything other than `HIT` is a cache miss and drives egress quota billing. + +**⑦ `processIpfsResponse`** + +- **`?format=car`**: body passed through as-is, no validation (client's responsibility per the Trustless Gateway spec). +- **Default**: wraps the body in a counting generator (`countingBody`) to track SP egress bytes, then runs the full streaming pipeline — `CarBlockIterator`, per-block multihash check, `validateBlock`, and `ipfs-unixfs-exporter` traversal. A directory entry returns 404. A file or raw entry's `entry.content()` is converted to a `ReadableStream` and returned. + +**⑧ `serveRetrievalOutcome`** + +Pipes the response body through a `TransformStream` that counts `egressBytes` chunk by chunk, preserving backpressure — the SP is pulled only as fast as the client reads. Once the stream ends, `ctx.waitUntil` runs `recordRetrieval` to log the result to D1. Sets `Cache-Control: public, max-age=31536000` and `X-Data-Set-ID` on the response. + +--- + +## CAR streaming pipeline + +The core of the worker is a fully lazy, end-to-end streaming pipeline. Nothing is buffered in memory. The only allocation at any point is one block at a time. + +``` +SP (frisbii/Curio) + └─ CAR stream (HTTP response body, streaming) + └─ countingBody — async generator counting SP egress bytes + └─ CarBlockIterator.fromIterable() — parses CAR header upfront, yields blocks lazily + └─ blockstore.get(cid) — called by the exporter per block + ├─ blocksReader.next() — pulls next block from the CAR stream + ├─ multihash comparison — verifies the block CID matches what the exporter asked for + └─ validateBlock() — hashes the bytes, confirms they match the multihash + └─ ipfs-unixfs-exporter (recursive()) + └─ entry.content() — yields leaf block bytes + └─ ReadableStream → HTTP response to client +``` + +### Why `blockReadConcurrency: 1` + +The exporter is called with `{ blockReadConcurrency: 1 }`. This forces it to request blocks strictly one at a time in DFS traversal order. The blockstore's `get(cid)` does not look blocks up by CID, it calls `blocksReader.next()` and expects that the next block in the CAR is always the one being requested. This works because the SP guarantees DFS-ordered delivery (see [SP integration contract](#sp-integration-contract)). If `blockReadConcurrency` were greater than 1, the exporter would request blocks in parallel and the sequential CAR reader would return the wrong block for each. + +### `CarBlockIterator` vs `CarReader` + +`CarBlockIterator.fromIterable` reads only the CAR header (roots + version) upfront. Block data is pulled lazily as the iterator is consumed. + +### CAR-to-raw conversion + +`processIpfsResponse` uses the `recursive` export from `ipfs-unixfs-exporter` aliased as `exporter`. Despite the name, it is used here only to resolve the first entry (the requested path) and stream its bytes, the loop exits after the first iteration intentionally (`// eslint-disable-next-line no-unreachable-loop`). The actual `exporter()` function would be semantically cleaner for this use case. + +When converting CAR to raw for non-directories, the worker strips the upstream `content-type` and `x-content-type-options` headers and sets `content-disposition: inline`. The browser receives raw bytes and is responsible for inferring the content type. There are two cases: + +- **No subpath** (e.g. `/{ipfsRootCid}/`) — the root CID resolves directly to a file or raw block with no filename. The browser has only magic-byte sniffing to determine the content type. Modern browsers handle this well for common formats (images, video, HTML) but it may fail for less common types. +- **With subpath** (e.g. `/{ipfsRootCid}/path/to/file.jpg`) — The terminal path segment carries a filename and extension. Modern browsers can use the extension, together with magic-byte sniffing, to infer the content type. + Currently, the worker does not set the Content-Type header based on the file extension, leaving that responsibility to the browser. Should we instead set a reasonable Content-Type ourselves? We should verify whether relying on the browser is sufficient for modern websites, including JavaScript, CSS, HTML, images, videos, and other assets. Is there any reason this approach could be risky or lead to compatibility or security issues? + +### Directory entries + +If the resolved path is a UnixFS directory, the worker returns **404 Not Found** (`retrieval.js:195`). Directory listing is not implemented. Since the SP returns a path-scoped CAR with `dag-scope=all`, the directory block and immediate child blocks are present in the CAR. The 404 is a choice, not an architectural limit. This is tracked in [issue #696](https://github.com/filbeam/worker/issues/696). + +**Potential improvement:** @rvagg suggested switching to `dag-scope=entity` on the SP fetch URL. For file paths, `entity` and `all` return identical blocks — the complete file DAG. For directory paths, `entity` returns only the directory block itself rather than all descendants recursively, which avoids fetching child blocks we immediately discard when returning 404. If directory listing is ever implemented (issue #696), `entity` would also be the right scope since listing only needs the directory block, not the full subtree. + +--- + +## SP integration contract + +The pipeline makes specific assumptions about how the SP delivers the CAR. These are satisfied by **Curio**, which implements the [frisbii](https://github.com/ipld/frisbii) trustless HTTP gateway: + +1. **Path-scoped CAR** — the SP is called as `GET /ipfs/{rootCid}{subpath}?format=car`. It returns a CAR containing exactly the blocks needed to walk from `rootCid` to `subpath` and read the file. No more, no less. There are no wasted bytes for single-asset requests regardless of dataset size. + +2. **DFS-ordered blocks** — blocks are written to the CAR in the same order the DAG traversal engine requests them (depth-first). This is enforced in frisbii via `carPipe`, which hooks into the IPLD link system and writes each block to the CAR immediately as it is loaded during traversal. This is what makes the sequential `blocksReader.next()` blockstore safe. + +3. **Streaming-first, silent truncation on missing blocks** — frisbii prioritizes streaming over up-front validation. It is possible to begin receiving a valid CAR and then have it terminate prematurely if a block is missing from SP storage (e.g. the client did not upload the full DAG, or a disk holding part of the DAG was temporarily unavailable). There is no reliable in-band error signal for this — the CAR stream simply ends. The worker detects it by running the same DAG traversal: when `blocksReader.next()` returns `done` before the exporter has finished, it throws a `Block not found` error. + +4. **Single-SP DAG constraint** — the entire DAG must reside on the same SP, but it can span multiple Filecoin pieces on that SP. frisbii fetches blocks from whichever local piece holds them during traversal. Unlike generic IPFS gateways (Rainbow, Boxo) that use the DHT or IPNI to discover blocks scattered across the network, FilBeam has no cross-SP block discovery. A DAG sharded across two SPs cannot be served. + +**If a non-frisbii SP is ever onboarded**, guarantees 1 and 2 (path-scoped CAR and DFS ordering) must be verified before integration (tracked in [issue #692](https://github.com/filbeam/worker/issues/692)). The CID mismatch error (`Unexpected block CID`) is the failure mode if ordering is violated — it is not obvious without this context. + +### Trust model + +FilBeam trusts the `ipfsRootCid` submitted by the SP as on-chain metadata at `addPiece` time. It is indexed as-is without verifying it corresponds to the actual piece data stored on Filecoin. An SP could submit any CID as metadata. + +The only integrity guarantee FilBeam provides is at the block level: `validateBlock()` hashes each block's bytes and confirms they match the block's CID. This proves the CAR is self-consistent. It does not prove the CAR represents the content of the underlying Filecoin piece, since both the CID claim and the CAR bytes originate from the same SP. + +### Alternative: block-by-block retrieval (Discarded) + +An alternative design (used by IPFS Shipyard tooling like Boxo and verified-fetch) drives the exporter with a blockstore whose `get(cid)` makes individual HTTP requests: `GET /ipfs/{blockCid}?format=raw`. This eliminates the ordering dependency and enables per-block Cloudflare caching. + +The trade-offs cut both ways: + +- **CAR:** One round-trip per file, regardless of size. Fully streaming, but without per-block caching. +- **Block-by-block:** Multiple round-trips per file (typically 20–80 for common content), but each block is independently cacheable. Ancestor directory blocks that are shared across many files can be served from cache on subsequent requests. + +For example, imagine a page at `/deep/path/content.html` that references 100 images. With the CAR approach, the ancestor directory blocks are fetched again for each image request. With block-by-block retrieval, those shared directory blocks would be cached after the first request and reused for the remaining images. In practice, however, ancestor directory blocks are typically very small, so the additional overhead of re-fetching them is likely to be minor. + +For FilBeam's narrow case — where the client deliberately stores the entire DAG on one SP and opts into this service — CAR retrieval is the better fit: one round-trip, fully streaming, no ordering complexity from the client side. The block-by-block design exists to solve the distributed-network problem of blocks scattered across unknown peers, which FilBeam does not have. That said, it is not a clear-cut choice. + +[Slack discussion Reference](https://filecoinproject.slack.com/archives/C08TVNKJV7C/p1779163683412569?thread_ts=1778671570.686079&cid=C08TVNKJV7C) + +--- + +## Caching + +There are two independent cache layers. + +### Layer 1 — Cloudflare edge cache (origin fetch) + +Configured via the `cf` option on the SP `fetch` call (`retrieval/lib/origin-cache.js`): + +```js +{ + cacheEverything: true, + cacheTtlByStatus: { '200-299': ORIGIN_CACHE_TTL, 404: 0, '500-599': 0 } +} +``` + +- `cacheEverything: true` — caches the SP response regardless of its `Cache-Control` header +- 2xx responses are cached for `ORIGIN_CACHE_TTL` seconds (currently **86400, 1 day**) +- 404 and 5xx responses are never cached + +**Cache key:** `{spBaseUrl}/ipfs/{ipfsRootCid}{ipfsSubpath}?format=car` + +CIDs are content-addressed and immutable, so in theory the same key always resolves to the same bytes and could be cached indefinitely. In practice two factors push against a very long TTL: + +- **Cache storage pressure** — large CARs cached across ~300 PoPs consume significant edge storage. A very large TTL risks filling PoP caches with content that is rarely re-requested. +- **SP reliability** — a transient SP failure (bad disk, database issue, network blip) can produce a truncated CAR that still arrives with a 200 status. If Cloudflare caches that response, the broken content would be served for the full TTL. A shorter TTL limits the blast radius of such failures. + +The current 1-day value balances both concerns and should be revisited once cache hit rates and storage pressure are measurable. + +**Cache miss detection:** after the fetch, `CF-Cache-Status: HIT` means the edge served a cached copy; anything else is treated as a cache miss. Cache misses are what drive egress quota billing. + +### Layer 2 — Client/browser cache + +Set on every successful response to the client (`retrieval/lib/fetch-handler.js`): + +``` +Cache-Control: public, max-age=31536000 +``` + +1 year. CID-addressed content is immutable, so this is correct in the happy path. + +**Open question:** if the stream terminates mid-way (truncated CAR, block validation error), will the browser reliably detect the response as incomplete and avoid caching a partial file for a year? This depends on HTTP version, whether a `Content-Length` was set, and browser-specific behavior — and needs further investigation before the 1-year TTL can be considered safe. @rvagg suggested asking `@lidel` as a subject-matter expert on this class of HTTP/IPFS gateway interaction. + +### Where cache data is stored + +Both layers use Cloudflare's edge network. Data is cached at the **PoP (Point of Presence)** that handled the request, whichever of Cloudflare's ~300 global data centers is geographically closest to the client. Each PoP maintains its own independent cache. A cache hit in Los Angeles does not warm the Frankfurt PoP. Enabling **Cloudflare Tiered Cache** at the account level would add a regional upper-tier cache, reducing cold-PoP fetches from Storage Providers for popular content if this becomes necessary in the future (it's not needed today). + +### Cache hit cost + +On a cache hit, the SP round-trip is eliminated but the worker still receives the full CAR body and runs the complete pipeline (CAR parsing, block validation, unixfs traversal). The cache saves network bytes from the SP but not CPU work in the worker. + +--- + +## Related + +- `retrieval/` — shared library providing `handleFetchRequest`, `selectRetrievalCandidate`, `assertCidNotDenied`, egress quota tracking, and authorization diff --git a/ipfs-retriever/.dev.vars.template b/ipfs-retriever/.dev.vars.template new file mode 100644 index 00000000..ca1ba659 --- /dev/null +++ b/ipfs-retriever/.dev.vars.template @@ -0,0 +1 @@ +BOT_TOKENS={"secret":"dev"} diff --git a/ipfs-retriever/bin/ipfs-retriever.js b/ipfs-retriever/bin/ipfs-retriever.js new file mode 100644 index 00000000..c5b46eeb --- /dev/null +++ b/ipfs-retriever/bin/ipfs-retriever.js @@ -0,0 +1,200 @@ +import { + isValidEthereumAddress, + httpAssert, + assertCidNotDenied, + handleFetchRequest, + selectRetrievalCandidate, +} from '@filbeam/retrieval' + +import { parseRequest } from '../lib/request.js' +import { + retrieveIpfsContent as defaultRetrieveIpfsContent, + processIpfsResponse, +} from '../lib/retrieval.js' +import { + getRetrievalCandidatesByDataSetAndPiece, + getSlugForWalletAndCid, +} from '../lib/store.js' + +export default { + /** + * @param {Request} request + * @param {Env} env + * @param {ExecutionContext} ctx + * @param {object} options + * @param {typeof defaultRetrieveIpfsContent} [options.retrieveIpfsContent] + * @returns + */ + async fetch(request, env, ctx, options) { + return handleFetchRequest(request, env, ctx, (context) => + this._fetch(request, env, ctx, options, context), + ) + }, + + /** + * @param {Request} request + * @param {Env} env + * @param {ExecutionContext} ctx + * @param {object} options + * @param {typeof defaultRetrieveIpfsContent} [options.retrieveIpfsContent] + * @param {import('@filbeam/retrieval').RequestContext} context + * @returns {Promise} + */ + async _fetch( + request, + env, + ctx, + { retrieveIpfsContent = defaultRetrieveIpfsContent } = {}, + context, + ) { + if ( + URL.parse(request.url)?.hostname === env.DNS_ROOT.slice(1) || + URL.parse(request.url)?.hostname === `link${env.DNS_ROOT}` + ) { + return handleDnsRootRequest(request, env) + } + + const { dataSetId, pieceId, ipfsSubpath, ipfsFormat } = parseRequest( + request, + env, + ) + + return async () => { + // Timestamp to measure file retrieval performance (from cache and from SP) + const fetchStartedAt = performance.now() + + const candidates = await getRetrievalCandidatesByDataSetAndPiece( + env, + dataSetId, + pieceId, + env.ENFORCE_EGRESS_QUOTA, + ) + // Every candidate serves the same content, so they share the root CID. + const ipfsRootCid = candidates[0].ipfsRootCid + + // Now check Bad Bits with the ipfsRootCid we got from the database + await assertCidNotDenied(env, ipfsRootCid) + + const { + failureResponse, + candidate, + result: retrievalResult, + } = await selectRetrievalCandidate( + candidates, + (candidate) => + retrieveIpfsContent( + candidate.serviceUrl, + ipfsRootCid, + ipfsSubpath, + env.ORIGIN_CACHE_TTL, + { signal: request.signal }, + ), + { + env, + ctx, + requestCountryCode: context.requestCountryCode, + timestamp: context.requestTimestamp, + botName: context.botName, + }, + ) + if (failureResponse) return failureResponse + httpAssert(candidate && retrievalResult, 500, 'should never happen') + + const originResponse = retrievalResult.response + const cacheMiss = retrievalResult.cacheMiss + + const { + body: responseBody, + getOriginEgressBytes, + headers: responseHeaders, + } = await processIpfsResponse(originResponse, { + ipfsRootCid, + ipfsSubpath, + ipfsFormat, + signal: request.signal, + }) + + // When converting CAR to raw, the headers already carry the CAR-to-raw + // adjustments. A null body (e.g. a HEAD request) is served as-is. + const response = responseBody + ? new Response(responseBody, { + status: originResponse.status, + statusText: originResponse.statusText, + headers: responseHeaders, + }) + : originResponse + + return { + response, + cacheMiss, + dataSetId: candidate.dataSetId, + fetchStartedAt, + // The client is served the raw bytes. On a cache miss the worker + // fetched a (larger) CAR from the service provider, which the cache-miss + // quota is charged for; for a passed-through CAR (`?format=car`) the two + // are equal. Reaching here means the response streamed successfully (a + // converted CAR is validated during conversion), so charge every cache + // miss. + // The body has fully streamed by the time finalizeCacheMiss runs, so + // the lazily counted CAR byte total is final here. + finalizeCacheMiss: async (egressBytes) => ({ + cacheMissEgressBytes: getOriginEgressBytes() ?? egressBytes, + cacheMissResponseValid: cacheMiss ? true : null, + }), + } + } + }, +} + +/** + * Handles requests to the bare DNS_ROOT domain (e.g., ipfs.filbeam.io). + * + * - If no path is provided, redirects to https://filbeam.com + * - If path is /wallet/cid or /wallet/cid/pathname, generates a slug and + * redirects to the subdomain-based URL + * + * @param {Request} request - The incoming request + * @param {Env} env - Worker environment + * @returns {Promise} Redirect response + */ +async function handleDnsRootRequest(request, env) { + // Parse the URL path to extract wallet, cid, and optional pathname + const parsedUrl = URL.parse(request.url) + const pathname = parsedUrl?.pathname || '/' + + // If no path, redirect to filbeam.com + if (pathname === '/' || pathname === '') { + return Response.redirect('https://filbeam.com/', 302) + } + + // Parse path as /wallet/cid/pathname + const pathParts = pathname.slice(1).split('/') // Remove leading slash and split + + if (pathParts.length < 2) { + httpAssert( + false, + 404, + 'Invalid path format. Expected: /wallet/cid or /wallet/cid/pathname', + ) + } + + const wallet = pathParts[0].toLowerCase() + const cid = pathParts[1] + const subpath = pathParts.slice(2).join('/') + + // Validate wallet address + httpAssert( + isValidEthereumAddress(wallet), + 404, + `Invalid wallet address: ${wallet}. Address must be a valid ethereum address.`, + ) + + // Get slug for the wallet and CID + const slug = await getSlugForWalletAndCid(env, wallet, cid) + + // Build redirect URL + const redirectPath = subpath ? `/${subpath}` : '' + const redirectUrl = `https://${slug}${env.DNS_ROOT}${redirectPath}` + + return Response.redirect(redirectUrl, 302) +} diff --git a/ipfs-retriever/lib/bigint-util.js b/ipfs-retriever/lib/bigint-util.js new file mode 100644 index 00000000..e0e720d3 --- /dev/null +++ b/ipfs-retriever/lib/bigint-util.js @@ -0,0 +1,74 @@ +import { base32 } from 'multiformats/bases/base32' + +/** + * @param {BigInt} value + * @returns {Uint8Array} + */ +export function bigIntToUint8Array(value) { + if (typeof value !== 'bigint') { + throw new TypeError('Expected a BigInt value') + } + if (value < 0n) { + throw new Error('Cannot convert negative bigint to Uint8Array') + } + let hex = value.toString(16) + if (hex.length % 2) hex = '0' + hex + const bytes = hex.match(/.{2}/g)?.map((byte) => parseInt(byte, 16)) ?? [] + return new Uint8Array(bytes) +} + +/** + * @param {Uint8Array} value + * @returns {BigInt} + */ +export function uint8ArrayToBigInt(value) { + if (!(value instanceof Uint8Array)) { + throw new TypeError('Expected a Uint8Array value') + } + if (value.length === 0) { + return 0n + } + const hex = [...value].map((x) => x.toString(16).padStart(2, '0')).join('') + return BigInt('0x' + hex) +} + +/** + * Converts a BigInt to a base32-encoded string + * + * @param {BigInt} value + * @returns {string} + */ +export function bigIntToBase32(value) { + if (typeof value !== 'bigint') { + throw new TypeError('Expected a BigInt value') + } + if (value < 0n) { + throw new Error('Cannot convert negative bigint to base32') + } + // Use "0" for zero value (0 is not a base32 character but is DNS-safe) + if (value === 0n) { + return '0' + } + const bytes = bigIntToUint8Array(value) + // Remove the 'b' prefix that multiformats adds + return base32.encode(bytes).slice(1) +} + +/** + * Converts a base32-encoded string to a BigInt + * + * @param {string} value + * @returns {BigInt} + */ +export function base32ToBigInt(value) { + if (typeof value !== 'string') { + throw new TypeError('Expected a string value') + } + // Handle special case for zero + if (value === '0') { + return 0n + } + // Add back the 'b' prefix that multiformats expects + const bytes = base32.decode('b' + value) + return uint8ArrayToBigInt(bytes) +} diff --git a/ipfs-retriever/lib/request.js b/ipfs-retriever/lib/request.js new file mode 100644 index 00000000..3dc7728f --- /dev/null +++ b/ipfs-retriever/lib/request.js @@ -0,0 +1,77 @@ +import { httpAssert } from '@filbeam/retrieval' +import { base32ToBigInt } from './bigint-util.js' + +/** + * Parse params found in path of the request URL + * + * @param {Request} request + * @param {object} options + * @param {string} options.DNS_ROOT + * @returns {{ + * dataSetId: string + * pieceId: string + * ipfsSubpath: string + * ipfsFormat: string | null + * }} + */ +export function parseRequest(request, { DNS_ROOT }) { + const url = new URL(request.url) + console.log('retrieval request', { DNS_ROOT, url }) + + httpAssert( + url.hostname.endsWith(DNS_ROOT), + 400, + `Invalid hostname: ${url.hostname}. It must end with ${DNS_ROOT}.`, + ) + + const slug = url.hostname.slice(0, -DNS_ROOT.length) + const parts = slug.split('-') + + httpAssert( + parts.length === 3, + 400, + `The hostname must be in the format: 1-{dataSetId}-{pieceId}${DNS_ROOT}`, + ) + + const [version, encodedDataSetId, encodedPieceId] = parts + + httpAssert( + version === '1', + 400, + `Unsupported slug version: ${version}. Expected version 1.`, + ) + + httpAssert( + encodedDataSetId && encodedPieceId, + 400, + `The hostname must be in the format: 1-{dataSetId}-{pieceId}${DNS_ROOT}`, + ) + + let dataSetId + let pieceId + + try { + dataSetId = base32ToBigInt(encodedDataSetId).toString() + } catch (error) { + httpAssert( + false, + 400, + `Invalid dataSetId encoding in slug: ${encodedDataSetId}. ${error instanceof Error ? error.message : String(error)}`, + ) + } + + try { + pieceId = base32ToBigInt(encodedPieceId).toString() + } catch (error) { + httpAssert( + false, + 400, + `Invalid pieceId encoding in slug: ${encodedPieceId}. ${error instanceof Error ? error.message : String(error)}`, + ) + } + + const ipfsSubpath = url.pathname || '/' + const ipfsFormat = url.searchParams.get('format') + + return { dataSetId, pieceId, ipfsSubpath, ipfsFormat } +} diff --git a/ipfs-retriever/lib/retrieval.js b/ipfs-retriever/lib/retrieval.js new file mode 100644 index 00000000..877933fe --- /dev/null +++ b/ipfs-retriever/lib/retrieval.js @@ -0,0 +1,241 @@ +import { CarBlockIterator } from '@ipld/car' +// @ts-ignore - Types exist but package.json exports configuration prevents resolution +import * as carBlockValidator from '@web3-storage/car-block-validator' +import { recursive as exporter } from 'ipfs-unixfs-exporter' +import { httpAssert, originCacheOptions } from '@filbeam/retrieval' + +/** @import {UnixFSBasicEntry} from 'ipfs-unixfs-exporter' */ +/** @typedef {{ cid: import('multiformats').CID; bytes: Uint8Array }} Block */ + +/** @type {(block: Block) => Promise | undefined} */ +const validateBlock = carBlockValidator.validateBlock + +/** + * Retrieves the IPFS content from the SP serving requests at the provided base + * URL. + * + * @param {string} baseUrl - The base URL of service provider. + * @param {string} ipfsRootCid - The IPFS Root CID to retrieve from. + * @param {string} ipfsSubpath - The subpath inside the UnixFS archive to + * retrieve, e.g. `/favicon.ico`. + * @param {number} [cacheTtl=86400] - Cache TTL in seconds (default: 86400). + * Default is `86400` + * @param {object} [options] - Optional parameters. + * @param {AbortSignal} [options.signal] - An optional AbortSignal to cancel the + * fetch request. + * @returns {Promise<{ + * response: Response + * cacheMiss: boolean + * }>} + * + * - The response from the fetch request, the cache miss and the content length. + */ +export async function retrieveIpfsContent( + baseUrl, + ipfsRootCid, + ipfsSubpath, + cacheTtl = 86400, + { signal } = {}, +) { + // TODO: allow the caller to tweak Trustless GW parameters like `dag-scope` when requesting `format=car`. + // See https://specs.ipfs.tech/http-gateways/trustless-gateway/ + // TODO: support `raw` format too, see https://github.com/filbeam/worker/issues/295 + const url = getRetrievalUrl(baseUrl, ipfsRootCid, ipfsSubpath) + '?format=car' + console.log(`Fetching IPFS content from: ${url}`) + const response = await fetch(url, { + cf: originCacheOptions(cacheTtl), + signal, + }) + const cacheStatus = response.headers.get('CF-Cache-Status') + if (!cacheStatus) { + console.log(`CF-Cache-Status was not provided for ${url}`) + } + + const cacheMiss = cacheStatus !== 'HIT' + + return { response, cacheMiss } +} + +/** + * @param {string} serviceUrl + * @param {string} rootCid + * @param {string} subpath + * @returns {string} + */ +export function getRetrievalUrl(serviceUrl, rootCid, subpath) { + if (!serviceUrl.endsWith('/')) { + serviceUrl += '/' + } + let url = `${serviceUrl}ipfs/${rootCid}` + // Curio 404s with trailing slash + if (subpath !== '/') { + url += subpath + } + return url +} + +/** + * @param {Response} response + * @param {object} options + * @param {string} options.ipfsRootCid + * @param {string} options.ipfsSubpath + * @param {string | null} options.ipfsFormat + * @param {AbortSignal} [options.signal] + * @returns {Promise<{ + * body: ReadableStream | null + * getOriginEgressBytes: () => number | null + * headers: Headers + * }>} + * - `body` is the stream to serve to the client: raw bytes when converting from + * CAR, the original body when serving CAR or passing through. + * - `getOriginEgressBytes` returns the number of CAR bytes read from the service + * provider, or `null` when the body is passed through unchanged (in that + * case the bytes served equal the bytes fetched). Because the CAR is + * streamed lazily, the count is only final once `body` has been fully + * consumed, so call this after streaming the response. + * - `headers` are the response headers to serve, with the CAR-to-raw adjustments + * applied when converting. + */ +export async function processIpfsResponse( + response, + { ipfsRootCid, ipfsSubpath, ipfsFormat, signal }, +) { + const body = response.body + if (!response.ok || !body || ipfsFormat === 'car') { + return { + body, + getOriginEgressBytes: () => null, + headers: response.headers, + } + } + + httpAssert( + ipfsFormat === null, + 400, + `Unsupported ?format value: "${ipfsFormat}"`, + ) + + // When converting from CAR to raw, set content-disposition to inline so + // browsers display the content instead of downloading it, and drop the + // upstream content type so the browser sniffs the raw bytes. + const headers = new Headers(response.headers) + headers.set('content-disposition', 'inline') + headers.delete('content-type') + headers.delete('x-content-type-options') + + // Count the CAR bytes fetched from the service provider as they stream + // through. `CarBlockIterator` decodes only the CAR header up front and yields + // blocks lazily, so the whole archive is never held in memory. The byte count + // is therefore only final once the caller has fully consumed the returned + // body, so it is exposed via `getOriginEgressBytes` rather than as a value. + let originEgressBytes = 0 + const countingBody = (async function* () { + for await (const chunk of body) { + originEgressBytes += chunk.length + yield chunk + } + })() + + const blocks = await CarBlockIterator.fromIterable(countingBody) + const blocksReader = blocks[Symbol.asyncIterator]() + + try { + const entries = exporter( + `${ipfsRootCid}${ipfsSubpath}`, + { + async get(blockCid) { + const res = await blocksReader.next() + if (res.done || !res.value) { + throw new Error(`Block ${blockCid} not found in CAR ${ipfsRootCid}`) + } + const block = res.value + + // Compare only the multihashes, so a block stored under an equivalent + // CID with a different codec or CID version still matches. validateBlock + // below verifies the block bytes hash to this multihash. + const actualMultihash = block.cid.multihash.bytes + const expectedMultihash = blockCid.multihash.bytes + if ( + actualMultihash.length !== expectedMultihash.length || + !actualMultihash.every((byte, i) => byte === expectedMultihash[i]) + ) { + throw new Error( + `Unexpected block CID ${block.cid}, expected ${blockCid}`, + ) + } + + try { + await validateBlock(block) + } catch (err) { + throw new Error( + `Invalid block ${blockCid} of root ${ipfsRootCid}`, + { + cause: err, + }, + ) + } + + return block.bytes + }, + }, + { signal, blockReadConcurrency: 1 }, + ) + + // eslint-disable-next-line no-unreachable-loop + for await (const entry of entries) { + signal?.throwIfAborted() + console.log(`Entry: ${entry.path} (${entry.type})`) + + const expectedPath = + ipfsSubpath === '/' ? ipfsRootCid : `${ipfsRootCid}${ipfsSubpath}` + if (entry.path !== expectedPath) { + throw new Error( + `Unexpected entry - wrong path: ${describeEntry(entry)} (expected: ${expectedPath})`, + ) + } + + if (entry.type !== 'file' && entry.type !== 'raw') { + console.log(`Unexpected entry - wrong type: ${describeEntry(entry)}`) + httpAssert(false, 404, 'Not Found') + } + + const entryContent = entry.content() + + // Convert AsyncGenerator to ReadableStream for Response body + const rawDataStream = new ReadableStream({ + async start(controller) { + try { + for await (const chunk of entryContent) { + signal?.throwIfAborted() + controller.enqueue(chunk) + } + controller.close() + } catch (error) { + controller.error(error) + } + }, + }) + + return { + body: rawDataStream, + getOriginEgressBytes: () => originEgressBytes, + headers, + } + } + + httpAssert(false, 404, 'Not Found') + } catch (err) { + // Release the SP connection immediately on error + body.cancel().catch(() => {}) + throw err + } +} + +/** @param {UnixFSBasicEntry} entry */ +export function describeEntry(entry) { + return JSON.stringify( + entry, + (_, v) => (typeof v === 'bigint' ? v.toString() : v), + 2, + ) +} diff --git a/ipfs-retriever/lib/store.js b/ipfs-retriever/lib/store.js new file mode 100644 index 00000000..9a974442 --- /dev/null +++ b/ipfs-retriever/lib/store.js @@ -0,0 +1,229 @@ +import { bigIntToBase32 } from './bigint-util.js' +import { + httpAssert, + filterAuthorizedRetrievalCandidates, + buildRetrievalCandidateQuery, +} from '@filbeam/retrieval' + +const SELECT_CANDIDATES_BY_CID = buildRetrievalCandidateQuery({ + extraColumns: [ + 'pieces.id AS piece_id', + 'pieces.ipfs_root_cid', + 'data_sets.with_ipfs_indexing', + ], + where: 'pieces.ipfs_root_cid = ?', +}) + +/** + * Validates query results and returns every approved retrieval candidate. This + * is a shared helper used by both getRetrievalCandidatesByWalletAndCid and + * getRetrievalCandidatesByDataSetAndPiece. + * + * @param {object} params + * @param {any[]} params.results - The query results to validate + * @param {string} params.payerAddress - The lower-cased address of the client + * paying for the request + * @param {string} params.lookupKey - Descriptive key for error messages (e.g., + * "IPFS Root CID 'bafk...'") + * @param {boolean} [params.enforceEgressQuota] - Whether to require remaining + * egress quota + * @returns {{ + * serviceProviderId: string + * serviceUrl: string + * dataSetId: string + * pieceId: string + * ipfsRootCid: string + * }[]} + */ +function validateQueryResultsAndGetCandidates(params) { + const { + results, + payerAddress, + lookupKey, + enforceEgressQuota = false, + } = params + + const authorizedRetrievalCandidates = filterAuthorizedRetrievalCandidates( + results, + { payerAddress, enforceEgressQuota }, + ) + + const withIpfsIndexing = authorizedRetrievalCandidates.filter( + (row) => row.with_ipfs_indexing === 1, + ) + httpAssert( + withIpfsIndexing.length > 0, + 402, + `The Filecoin Warm Storage Service deal for payer '${payerAddress}' and ${lookupKey} has withIpfsIndexing=false.`, + ) + + const withIpfsRootCid = withIpfsIndexing.filter((row) => row.ipfs_root_cid) + httpAssert( + withIpfsRootCid.length > 0, + 404, + `${lookupKey} exists but has no associated IPFS Root CID.`, + ) + + const candidates = withIpfsRootCid.map((row) => ({ + serviceProviderId: row.service_provider_id, + // We need this cast to suppress a TypeScript error. The compiler cannot + // infer that the filters above keep only rows with service_url defined. + serviceUrl: /** @type {string} */ (row.service_url), + dataSetId: row.data_set_id, + pieceId: row.piece_id, + ipfsRootCid: row.ipfs_root_cid, + })) + + console.log( + `Validated ${candidates.length} retrieval candidate(s) for ${lookupKey} and payer '${payerAddress}'`, + ) + + return candidates +} + +/** + * Retrieves every approved retrieval candidate (one per service provider) for a + * given root CID and payer. + * + * @param {Pick} env - Cloudflare Worker environment with D1 DB + * binding + * @param {string} payerAddress - The lower-cased address of the client paying + * for the request + * @param {string} ipfsRootCid - The IPFS Root CID to look up + * @returns {Promise< + * { + * serviceProviderId: string + * serviceUrl: string + * dataSetId: string + * pieceId: string + * ipfsRootCid: string + * }[] + * >} + */ +export async function getRetrievalCandidatesByWalletAndCid( + env, + payerAddress, + ipfsRootCid, +) { + const results = /** @type {any[]} */ ( + (await env.DB.prepare(SELECT_CANDIDATES_BY_CID).bind(ipfsRootCid).all()) + .results + ) + + return validateQueryResultsAndGetCandidates({ + results, + payerAddress, + lookupKey: `IPFS Root CID '${ipfsRootCid}'`, + }) +} + +/** + * Retrieves every approved retrieval candidate for the content addressed by a + * given data set ID and piece ID. The piece is resolved to its content CID and + * the data set's payer, then every service provider serving that content for + * that payer is returned so the worker can retry across them. + * + * @param {Pick} env - Cloudflare Worker environment with D1 DB + * binding + * @param {string} dataSetId - The data set ID + * @param {string} pieceId - The piece ID + * @param {boolean} [enforceEgressQuota=false] - Whether to require remaining + * egress quota. Default is `false` + * @returns {Promise< + * { + * serviceProviderId: string + * serviceUrl: string + * dataSetId: string + * pieceId: string + * ipfsRootCid: string + * }[] + * >} + */ +export async function getRetrievalCandidatesByDataSetAndPiece( + env, + dataSetId, + pieceId, + enforceEgressQuota = false, +) { + const piece = /** + * @type {{ + * ipfs_root_cid: string | null + * payer_address: string | null + * } | null} + */ ( + await env.DB.prepare( + ` + SELECT pieces.ipfs_root_cid, data_sets.payer_address + FROM pieces + LEFT OUTER JOIN data_sets ON pieces.data_set_id = data_sets.id + WHERE pieces.id = ? AND pieces.data_set_id = ? + `, + ) + .bind(pieceId, dataSetId) + .first() + ) + + httpAssert( + piece, + 404, + `Piece ID '${pieceId}' does not exist in data set ID '${dataSetId}' or may not have been indexed yet.`, + ) + + const ipfsRootCid = piece.ipfs_root_cid + const payerAddress = piece.payer_address + + httpAssert( + payerAddress, + 404, + `Data set ID '${dataSetId}' exists but has no associated payer address.`, + ) + httpAssert( + ipfsRootCid, + 404, + `data set ID '${dataSetId}' and piece ID '${pieceId}' exists but has no associated IPFS Root CID.`, + ) + + const results = /** @type {any[]} */ ( + (await env.DB.prepare(SELECT_CANDIDATES_BY_CID).bind(ipfsRootCid).all()) + .results + ) + + return validateQueryResultsAndGetCandidates({ + results, + payerAddress: payerAddress.toLowerCase(), + lookupKey: `data set ID '${dataSetId}' and piece ID '${pieceId}'`, + enforceEgressQuota, + }) +} + +/** + * Builds a slug from dataSetId and pieceId. + * + * @param {bigint} dataSetId - The data set ID as BigInt + * @param {bigint} pieceId - The piece ID as BigInt + * @returns {string} - The slug in format: + * 1-{base32(dataSetId)}-{base32(pieceId)} + */ +export function buildSlug(dataSetId, pieceId) { + return [ + '1', // version + bigIntToBase32(dataSetId), + bigIntToBase32(pieceId), + ].join('-') +} + +/** + * @param {Pick} env - Cloudflare Worker environment with D1 DB + * binding + * @param {string} payerAddress + * @param {string} ipfsRootCid + */ +export async function getSlugForWalletAndCid(env, payerAddress, ipfsRootCid) { + const [{ dataSetId, pieceId }] = await getRetrievalCandidatesByWalletAndCid( + env, + payerAddress, + ipfsRootCid, + ) + + return buildSlug(BigInt(dataSetId), BigInt(pieceId)) +} diff --git a/ipfs-retriever/package.json b/ipfs-retriever/package.json new file mode 100644 index 00000000..8e3c2158 --- /dev/null +++ b/ipfs-retriever/package.json @@ -0,0 +1,22 @@ +{ + "name": "@filbeam/ipfs-retriever", + "version": "1.0.0", + "private": true, + "description": "FilBeam IPFS Retrieval Worker", + "author": "Space Meridian ", + "type": "module", + "main": "bin/ipfs-retriever.js", + "scripts": { + "build:types": "wrangler types", + "deploy:calibration": "wrangler deploy --env calibration", + "start": "wrangler d1 migrations apply dev-db --local --env dev --cwd ../db && wrangler dev --env dev", + "test": "wrangler d1 migrations apply test-db --local --cwd ../db && vitest run" + }, + "dependencies": { + "@filbeam/retrieval": "^1.0.0", + "@ipld/car": "^5.4.2", + "@web3-storage/car-block-validator": "^1.2.2", + "ipfs-unixfs-exporter": "^13.7.3", + "multiformats": "^13.4.1" + } +} diff --git a/ipfs-retriever/test/apply-migrations.js b/ipfs-retriever/test/apply-migrations.js new file mode 100644 index 00000000..3bad2028 --- /dev/null +++ b/ipfs-retriever/test/apply-migrations.js @@ -0,0 +1,3 @@ +import { applyD1Migrations, env } from 'cloudflare:test' + +await applyD1Migrations(env.DB, env.TEST_MIGRATIONS) diff --git a/ipfs-retriever/test/bigint-util.test.js b/ipfs-retriever/test/bigint-util.test.js new file mode 100644 index 00000000..74177c9b --- /dev/null +++ b/ipfs-retriever/test/bigint-util.test.js @@ -0,0 +1,406 @@ +import { describe, it, expect } from 'vitest' +import { + bigIntToUint8Array, + uint8ArrayToBigInt, + bigIntToBase32, + base32ToBigInt, +} from '../lib/bigint-util.js' + +describe('bigint-util', () => { + describe('bigIntToUint8Array', () => { + it('converts zero correctly', () => { + const result = bigIntToUint8Array(0n) + expect(result).toEqual(new Uint8Array([0])) + }) + + it('converts small positive single-byte values', () => { + expect(bigIntToUint8Array(1n)).toEqual(new Uint8Array([1])) + expect(bigIntToUint8Array(255n)).toEqual(new Uint8Array([255])) + }) + + it('converts two-byte values', () => { + expect(bigIntToUint8Array(256n)).toEqual(new Uint8Array([1, 0])) + expect(bigIntToUint8Array(257n)).toEqual(new Uint8Array([1, 1])) + expect(bigIntToUint8Array(65535n)).toEqual(new Uint8Array([255, 255])) + }) + + it('converts medium values requiring multiple bytes', () => { + // 3 bytes + expect(bigIntToUint8Array(65536n)).toEqual(new Uint8Array([1, 0, 0])) + expect(bigIntToUint8Array(16777215n)).toEqual( + new Uint8Array([255, 255, 255]), + ) + + // 4 bytes + expect(bigIntToUint8Array(16777216n)).toEqual( + new Uint8Array([1, 0, 0, 0]), + ) + }) + + it('converts large values requiring 8 bytes', () => { + const value = 2n ** 64n - 1n // Max 64-bit value + const result = bigIntToUint8Array(value) + expect(result.length).toBe(8) + expect(result).toEqual( + new Uint8Array([255, 255, 255, 255, 255, 255, 255, 255]), + ) + }) + + it('converts very large values requiring 32+ bytes', () => { + const value = 2n ** 256n - 1n + const result = bigIntToUint8Array(value) + expect(result.length).toBe(32) + expect(result.every((byte) => byte === 255)).toBe(true) + }) + + it('converts values requiring 64+ bytes', () => { + const value = 2n ** 512n + const result = bigIntToUint8Array(value) + expect(result.length).toBe(65) // 512 bits = 64 bytes + 1 leading byte + expect(result[0]).toBe(1) + expect(result.slice(1).every((byte) => byte === 0)).toBe(true) + }) + + it('maintains big-endian byte order', () => { + // 0x0102 should be [1, 2], not [2, 1] + expect(bigIntToUint8Array(0x0102n)).toEqual(new Uint8Array([1, 2])) + expect(bigIntToUint8Array(0x123456n)).toEqual( + new Uint8Array([0x12, 0x34, 0x56]), + ) + expect(bigIntToUint8Array(0xabcdefn)).toEqual( + new Uint8Array([0xab, 0xcd, 0xef]), + ) + }) + + it('handles boundary values at byte transitions', () => { + // Test values at byte boundaries + expect(bigIntToUint8Array(254n)).toEqual(new Uint8Array([254])) + expect(bigIntToUint8Array(255n)).toEqual(new Uint8Array([255])) + expect(bigIntToUint8Array(256n)).toEqual(new Uint8Array([1, 0])) + + expect(bigIntToUint8Array(65534n)).toEqual(new Uint8Array([255, 254])) + expect(bigIntToUint8Array(65535n)).toEqual(new Uint8Array([255, 255])) + expect(bigIntToUint8Array(65536n)).toEqual(new Uint8Array([1, 0, 0])) + }) + + it('handles powers of 2 correctly', () => { + expect(bigIntToUint8Array(2n ** 8n)).toEqual(new Uint8Array([1, 0])) + expect(bigIntToUint8Array(2n ** 16n)).toEqual(new Uint8Array([1, 0, 0])) + expect(bigIntToUint8Array(2n ** 24n)).toEqual( + new Uint8Array([1, 0, 0, 0]), + ) + expect(bigIntToUint8Array(2n ** 32n)).toEqual( + new Uint8Array([1, 0, 0, 0, 0]), + ) + }) + + it('throws error for non-bigint input', () => { + expect(() => bigIntToUint8Array(123)).toThrow() + expect(() => bigIntToUint8Array('123')).toThrow() + expect(() => bigIntToUint8Array(null)).toThrow() + expect(() => bigIntToUint8Array(undefined)).toThrow() + expect(() => bigIntToUint8Array({})).toThrow() + expect(() => bigIntToUint8Array([])).toThrow() + }) + }) + + describe('uint8ArrayToBigInt', () => { + it('converts single byte correctly', () => { + expect(uint8ArrayToBigInt(new Uint8Array([0]))).toBe(0n) + expect(uint8ArrayToBigInt(new Uint8Array([1]))).toBe(1n) + expect(uint8ArrayToBigInt(new Uint8Array([255]))).toBe(255n) + }) + + it('converts multiple bytes correctly', () => { + expect(uint8ArrayToBigInt(new Uint8Array([1, 0]))).toBe(256n) + expect(uint8ArrayToBigInt(new Uint8Array([1, 1]))).toBe(257n) + expect(uint8ArrayToBigInt(new Uint8Array([255, 255]))).toBe(65535n) + }) + + it('converts large arrays correctly', () => { + // 4 bytes + expect(uint8ArrayToBigInt(new Uint8Array([1, 0, 0, 0]))).toBe(16777216n) + + // 8 bytes (max 64-bit) + const maxUint64 = new Uint8Array([255, 255, 255, 255, 255, 255, 255, 255]) + expect(uint8ArrayToBigInt(maxUint64)).toBe(2n ** 64n - 1n) + }) + + it('converts very large arrays (32+ bytes)', () => { + const thirtyTwoBytes = new Uint8Array(32).fill(255) + expect(uint8ArrayToBigInt(thirtyTwoBytes)).toBe(2n ** 256n - 1n) + }) + + it('handles empty array', () => { + expect(uint8ArrayToBigInt(new Uint8Array([]))).toBe(0n) + }) + + it('handles arrays with leading zeros', () => { + expect(uint8ArrayToBigInt(new Uint8Array([0, 0, 1]))).toBe(1n) + expect(uint8ArrayToBigInt(new Uint8Array([0, 1, 0]))).toBe(256n) + expect(uint8ArrayToBigInt(new Uint8Array([0, 0, 0, 255]))).toBe(255n) + }) + + it('handles all zeros', () => { + expect(uint8ArrayToBigInt(new Uint8Array([0]))).toBe(0n) + expect(uint8ArrayToBigInt(new Uint8Array([0, 0]))).toBe(0n) + expect(uint8ArrayToBigInt(new Uint8Array([0, 0, 0]))).toBe(0n) + }) + + it('interprets bytes as big-endian', () => { + // [1, 2] should be 0x0102 = 258, not 0x0201 = 513 + expect(uint8ArrayToBigInt(new Uint8Array([1, 2]))).toBe(0x0102n) + expect(uint8ArrayToBigInt(new Uint8Array([0x12, 0x34, 0x56]))).toBe( + 0x123456n, + ) + expect(uint8ArrayToBigInt(new Uint8Array([0xab, 0xcd, 0xef]))).toBe( + 0xabcdefn, + ) + }) + + it('handles boundary values', () => { + expect(uint8ArrayToBigInt(new Uint8Array([254]))).toBe(254n) + expect(uint8ArrayToBigInt(new Uint8Array([255]))).toBe(255n) + expect(uint8ArrayToBigInt(new Uint8Array([255, 254]))).toBe(65534n) + expect(uint8ArrayToBigInt(new Uint8Array([255, 255]))).toBe(65535n) + }) + + it('works with Node.js Buffer', () => { + const buffer = Buffer.from([1, 2, 3]) + expect(uint8ArrayToBigInt(buffer)).toBe(0x010203n) + }) + + it('throws error for non-Uint8Array input', () => { + expect(() => uint8ArrayToBigInt(123)).toThrow() + expect(() => uint8ArrayToBigInt('123')).toThrow() + expect(() => uint8ArrayToBigInt(null)).toThrow() + expect(() => uint8ArrayToBigInt(undefined)).toThrow() + expect(() => uint8ArrayToBigInt({})).toThrow() + }) + + it('throws error for regular arrays', () => { + expect(() => uint8ArrayToBigInt([1, 2, 3])).toThrow() + }) + }) + + describe('Round-Trip Conversion', () => { + it('bigint -> array -> bigint preserves value for small numbers', () => { + const values = [0n, 1n, 127n, 128n, 255n, 256n, 65535n, 65536n] + values.forEach((value) => { + const array = bigIntToUint8Array(value) + const result = uint8ArrayToBigInt(array) + expect(result).toBe(value) + }) + }) + + it('bigint -> array -> bigint preserves value for large numbers', () => { + const values = [2n ** 32n, 2n ** 64n, 2n ** 128n, 2n ** 256n, 2n ** 512n] + values.forEach((value) => { + const array = bigIntToUint8Array(value) + const result = uint8ArrayToBigInt(array) + expect(result).toBe(value) + }) + }) + + it('bigint -> array -> bigint preserves value for powers of 2 minus 1', () => { + const values = [ + 2n ** 8n - 1n, + 2n ** 16n - 1n, + 2n ** 32n - 1n, + 2n ** 64n - 1n, + 2n ** 128n - 1n, + ] + values.forEach((value) => { + const array = bigIntToUint8Array(value) + const result = uint8ArrayToBigInt(array) + expect(result).toBe(value) + }) + }) + + it('array -> bigint -> array preserves array (without leading zeros)', () => { + const arrays = [ + new Uint8Array([0]), + new Uint8Array([1]), + new Uint8Array([255]), + new Uint8Array([1, 0]), + new Uint8Array([255, 255]), + new Uint8Array([1, 2, 3, 4, 5]), + ] + arrays.forEach((array) => { + const bigint = uint8ArrayToBigInt(array) + const result = bigIntToUint8Array(bigint) + expect(result).toEqual(array) + }) + }) + + it('array with leading zeros -> bigint -> array removes leading zeros', () => { + const arrayWithZeros = new Uint8Array([0, 0, 1, 2, 3]) + const bigint = uint8ArrayToBigInt(arrayWithZeros) + const result = bigIntToUint8Array(bigint) + expect(result).toEqual(new Uint8Array([1, 2, 3])) + }) + + it('handles random large values correctly', () => { + // Generate some pseudo-random large bigints + const randomValues = [ + 123456789012345678901234567890n, + 987654321098765432109876543210n, + 111111111111111111111111111111n, + ] + randomValues.forEach((value) => { + const array = bigIntToUint8Array(value) + const result = uint8ArrayToBigInt(array) + expect(result).toBe(value) + }) + }) + }) + + describe('bigIntToBase32', () => { + it('converts zero to the special character "0"', () => { + const result = bigIntToBase32(0n) + expect(result).toBe('0') + }) + + it('converts small positive values', () => { + expect(bigIntToBase32(1n)).toBe('ae') + }) + + it('converts single-byte values', () => { + expect(bigIntToBase32(255n)).toBe('74') + }) + + it('converts two-byte values', () => { + expect(bigIntToBase32(256n)).toBe('aeaa') + expect(bigIntToBase32(65535n)).toBe('777q') + }) + + it('converts large values', () => { + const large = 2n ** 64n - 1n + expect(bigIntToBase32(large)).toBe('7777777777776') + }) + + it('converts very large values (256-bit)', () => { + const veryLarge = 2n ** 256n - 1n + expect(bigIntToBase32(veryLarge)).toBe( + '777777777777777777777777777777777777777777777777777q', + ) + }) + + it('handles powers of 2', () => { + expect(bigIntToBase32(2n ** 8n)).toBe('aeaa') + expect(bigIntToBase32(2n ** 16n)).toBe('aeaaa') + expect(bigIntToBase32(2n ** 32n)).toBe('aeaaaaaa') + expect(bigIntToBase32(2n ** 64n)).toBe('aeaaaaaaaaaaaaa') + }) + + it('throws error for negative values', () => { + expect(() => bigIntToBase32(-1n)).toThrow( + 'Cannot convert negative bigint to base32', + ) + expect(() => bigIntToBase32(-100n)).toThrow() + }) + + it('throws error for non-bigint input', () => { + expect(() => bigIntToBase32(123)).toThrow(TypeError) + expect(() => bigIntToBase32('123')).toThrow(TypeError) + expect(() => bigIntToBase32(null)).toThrow(TypeError) + expect(() => bigIntToBase32(undefined)).toThrow(TypeError) + expect(() => bigIntToBase32({})).toThrow(TypeError) + expect(() => bigIntToBase32([])).toThrow(TypeError) + }) + }) + + describe('base32ToBigInt', () => { + it('converts base32 strings to BigInt', () => { + const base32String = bigIntToBase32(12345n) + const result = base32ToBigInt(base32String) + expect(typeof result).toBe('bigint') + }) + + it('converts small values correctly', () => { + const original = 1n + const base32String = bigIntToBase32(original) + const result = base32ToBigInt(base32String) + expect(result).toBe(original) + }) + + it('converts medium values correctly', () => { + const original = 65535n + const base32String = bigIntToBase32(original) + const result = base32ToBigInt(base32String) + expect(result).toBe(original) + }) + + it('converts large values correctly', () => { + const original = 2n ** 64n - 1n + const base32String = bigIntToBase32(original) + const result = base32ToBigInt(base32String) + expect(result).toBe(original) + }) + + it('converts very large values correctly', () => { + const original = 2n ** 256n - 1n + const base32String = bigIntToBase32(original) + const result = base32ToBigInt(base32String) + expect(result).toBe(original) + }) + + it('handles zero value using special character "0"', () => { + const result = base32ToBigInt('0') + expect(result).toBe(0n) + }) + + it('throws error for non-string input', () => { + expect(() => base32ToBigInt(123)).toThrow(TypeError) + expect(() => base32ToBigInt(123n)).toThrow(TypeError) + expect(() => base32ToBigInt(null)).toThrow(TypeError) + expect(() => base32ToBigInt(undefined)).toThrow(TypeError) + expect(() => base32ToBigInt({})).toThrow(TypeError) + expect(() => base32ToBigInt([])).toThrow(TypeError) + }) + + it('throws error for invalid base32 strings', () => { + expect(() => base32ToBigInt('invalid!@#')).toThrow() + expect(() => base32ToBigInt('not-base32')).toThrow() + expect(() => base32ToBigInt('123')).toThrow() // numbers not in base32 alphabet + }) + }) + + describe('Base32 Round-Trip Conversion', () => { + const TEST_CASES = [ + // special case + 0n, + // small numbers + 1n, + 127n, + 128n, + 255n, + 256n, + 65535n, + 65536n, + // powers of 2 + 2n ** 32n, + 2n ** 64n, + 2n ** 128n, + 2n ** 256n, + // powers of 2 minus 1 + 2n ** 8n - 1n, + 2n ** 16n - 1n, + 2n ** 32n - 1n, + 2n ** 64n - 1n, + 2n ** 128n - 1n, + // random large values + 123456789012345678901234567890n, + 987654321098765432109876543210n, + 111111111111111111111111111111n, + ] + + for (const tc of TEST_CASES) { + it(`preserves ${tc} during the round-trip`, () => { + const base32String = bigIntToBase32(tc) + const result = base32ToBigInt(base32String) + expect(result).toBe(tc) + }) + } + }) +}) diff --git a/ipfs-retriever/test/request.test.js b/ipfs-retriever/test/request.test.js new file mode 100644 index 00000000..ba472c7b --- /dev/null +++ b/ipfs-retriever/test/request.test.js @@ -0,0 +1,286 @@ +import { describe, it, expect } from 'vitest' +import { parseRequest } from '../lib/request.js' +import { bigIntToBase32 } from '../lib/bigint-util.js' + +const DNS_ROOT = '.filbeam.io' +const BOT_TOKENS = JSON.stringify({ secret: 'bot1' }) + +describe('parseRequest', () => { + it('should parse dataSetId and pieceId from a slug URL', () => { + const dataSetId = '12345' + const pieceId = '67890' + const encodedDataSetId = bigIntToBase32(BigInt(dataSetId)) + const encodedPieceId = bigIntToBase32(BigInt(pieceId)) + const slug = `1-${encodedDataSetId}-${encodedPieceId}` + + const request = new Request(`https://${slug}${DNS_ROOT}/`) + const result = parseRequest(request, { DNS_ROOT, BOT_TOKENS }) + + expect(result).toEqual({ + dataSetId, + pieceId, + ipfsSubpath: '/', + ipfsFormat: null, + }) + }) + + it('should parse subpath from URL pathname', () => { + const dataSetId = '100' + const pieceId = '200' + const encodedDataSetId = bigIntToBase32(BigInt(dataSetId)) + const encodedPieceId = bigIntToBase32(BigInt(pieceId)) + const slug = `1-${encodedDataSetId}-${encodedPieceId}` + const subpath = '/path/to/file.txt' + + const request = new Request(`https://${slug}${DNS_ROOT}${subpath}`) + const result = parseRequest(request, { DNS_ROOT, BOT_TOKENS }) + + expect(result).toEqual({ + dataSetId, + pieceId, + ipfsSubpath: subpath, + ipfsFormat: null, + }) + }) + + it('should default to "/" for empty pathname', () => { + const dataSetId = '999' + const pieceId = '888' + const encodedDataSetId = bigIntToBase32(BigInt(dataSetId)) + const encodedPieceId = bigIntToBase32(BigInt(pieceId)) + const slug = `1-${encodedDataSetId}-${encodedPieceId}` + + const request = new Request(`https://${slug}${DNS_ROOT}`) + const result = parseRequest(request, { DNS_ROOT, BOT_TOKENS }) + + expect(result).toEqual({ + dataSetId, + pieceId, + ipfsSubpath: '/', + ipfsFormat: null, + }) + }) + + it('should handle zero values for dataSetId and pieceId', () => { + const slug = '1-0-0' + + const request = new Request(`https://${slug}${DNS_ROOT}/`) + const result = parseRequest(request, { DNS_ROOT, BOT_TOKENS }) + + expect(result).toEqual({ + dataSetId: '0', + pieceId: '0', + ipfsSubpath: '/', + ipfsFormat: null, + }) + }) + + it('should return descriptive error for invalid hostname format - missing parts', () => { + const request = new Request(`https://1-abc${DNS_ROOT}/`) + expect(() => parseRequest(request, { DNS_ROOT, BOT_TOKENS })).toThrowError( + `The hostname must be in the format: 1-{dataSetId}-{pieceId}${DNS_ROOT}`, + ) + }) + + it('should return descriptive error for invalid hostname format - too many parts', () => { + const request = new Request(`https://1-abc-def-ghi${DNS_ROOT}/`) + expect(() => parseRequest(request, { DNS_ROOT, BOT_TOKENS })).toThrowError( + `The hostname must be in the format: 1-{dataSetId}-{pieceId}${DNS_ROOT}`, + ) + }) + + it('should return descriptive error for invalid hostname format - no dashes', () => { + const request = new Request(`https://1abc${DNS_ROOT}/`) + expect(() => parseRequest(request, { DNS_ROOT, BOT_TOKENS })).toThrowError( + `The hostname must be in the format: 1-{dataSetId}-{pieceId}${DNS_ROOT}`, + ) + }) + + it('should return descriptive error for missing dataSetId', () => { + const request = new Request(`https://1--abc${DNS_ROOT}/`) + expect(() => parseRequest(request, { DNS_ROOT, BOT_TOKENS })).toThrowError( + `The hostname must be in the format: 1-{dataSetId}-{pieceId}${DNS_ROOT}`, + ) + }) + + it('should return descriptive error for missing pieceId', () => { + const request = new Request(`https://1-abc-${DNS_ROOT}/`) + expect(() => parseRequest(request, { DNS_ROOT, BOT_TOKENS })).toThrowError( + `The hostname must be in the format: 1-{dataSetId}-{pieceId}${DNS_ROOT}`, + ) + }) + + it('should return descriptive error for unsupported version', () => { + const dataSetId = '12345' + const pieceId = '67890' + const encodedDataSetId = bigIntToBase32(BigInt(dataSetId)) + const encodedPieceId = bigIntToBase32(BigInt(pieceId)) + const slug = `2-${encodedDataSetId}-${encodedPieceId}` + + const request = new Request(`https://${slug}${DNS_ROOT}/`) + expect(() => parseRequest(request, { DNS_ROOT, BOT_TOKENS })).toThrowError( + 'Unsupported slug version: 2. Expected version 1.', + ) + }) + + it('should return descriptive error for invalid base32 dataSetId', () => { + const request = new Request(`https://1-invalid1-aeete${DNS_ROOT}/`) + expect(() => parseRequest(request, { DNS_ROOT, BOT_TOKENS })).toThrowError( + /Invalid dataSetId encoding in slug: invalid1/, + ) + }) + + it('should return descriptive error for invalid base32 pieceId', () => { + const request = new Request(`https://1-ga4q-invalid1${DNS_ROOT}/`) + expect(() => parseRequest(request, { DNS_ROOT, BOT_TOKENS })).toThrowError( + /Invalid pieceId encoding in slug: invalid1/, + ) + }) + + it('should return error for wrong DNS root', () => { + const dataSetId = '12345' + const pieceId = '67890' + const encodedDataSetId = bigIntToBase32(BigInt(dataSetId)) + const encodedPieceId = bigIntToBase32(BigInt(pieceId)) + const slug = `1-${encodedDataSetId}-${encodedPieceId}` + + const request = new Request(`https://${slug}.wrong.io/`) + expect(() => parseRequest(request, { DNS_ROOT, BOT_TOKENS })).toThrowError( + `Invalid hostname: ${slug}.wrong.io. It must end with ${DNS_ROOT}.`, + ) + }) + + it('should ignore query parameters', () => { + const dataSetId = '12345' + const pieceId = '67890' + const encodedDataSetId = bigIntToBase32(BigInt(dataSetId)) + const encodedPieceId = bigIntToBase32(BigInt(pieceId)) + const slug = `1-${encodedDataSetId}-${encodedPieceId}` + const subpath = '/file.txt' + + const request = new Request( + `https://${slug}${DNS_ROOT}${subpath}?foo=bar&baz=qux`, + ) + const result = parseRequest(request, { DNS_ROOT, BOT_TOKENS }) + + expect(result).toEqual({ + dataSetId, + pieceId, + ipfsSubpath: subpath, + ipfsFormat: null, + }) + }) + + it('should parse format=car from URL with subpath', () => { + const dataSetId = '100' + const pieceId = '200' + const encodedDataSetId = bigIntToBase32(BigInt(dataSetId)) + const encodedPieceId = bigIntToBase32(BigInt(pieceId)) + const slug = `1-${encodedDataSetId}-${encodedPieceId}` + const subpath = '/path/to/file.txt' + + const request = new Request( + `https://${slug}${DNS_ROOT}${subpath}?format=car`, + ) + const result = parseRequest(request, { DNS_ROOT, BOT_TOKENS }) + + expect(result).toEqual({ + dataSetId, + pieceId, + ipfsSubpath: subpath, + ipfsFormat: 'car', + }) + }) + + it('should parse any format value from URL', () => { + const dataSetId = '12345' + const pieceId = '67890' + const encodedDataSetId = bigIntToBase32(BigInt(dataSetId)) + const encodedPieceId = bigIntToBase32(BigInt(pieceId)) + const slug = `1-${encodedDataSetId}-${encodedPieceId}` + + const request = new Request(`https://${slug}${DNS_ROOT}/?format=raw`) + const result = parseRequest(request, { DNS_ROOT, BOT_TOKENS }) + + expect(result).toEqual({ + dataSetId, + pieceId, + ipfsSubpath: '/', + ipfsFormat: 'raw', + }) + }) + + it('should return null for ipfsFormat when format parameter is not present', () => { + const dataSetId = '12345' + const pieceId = '67890' + const encodedDataSetId = bigIntToBase32(BigInt(dataSetId)) + const encodedPieceId = bigIntToBase32(BigInt(pieceId)) + const slug = `1-${encodedDataSetId}-${encodedPieceId}` + + const request = new Request(`https://${slug}${DNS_ROOT}/file.txt`) + const result = parseRequest(request, { DNS_ROOT, BOT_TOKENS }) + + expect(result).toEqual({ + dataSetId, + pieceId, + ipfsSubpath: '/file.txt', + ipfsFormat: null, + }) + }) + + it('should preserve trailing slash in subpath', () => { + const dataSetId = '12345' + const pieceId = '67890' + const encodedDataSetId = bigIntToBase32(BigInt(dataSetId)) + const encodedPieceId = bigIntToBase32(BigInt(pieceId)) + const slug = `1-${encodedDataSetId}-${encodedPieceId}` + const subpath = '/directory/' + + const request = new Request(`https://${slug}${DNS_ROOT}${subpath}`) + const result = parseRequest(request, { DNS_ROOT, BOT_TOKENS }) + + expect(result).toEqual({ + dataSetId, + pieceId, + ipfsSubpath: subpath, + ipfsFormat: null, + }) + }) + + it('should handle encoded characters in subpath', () => { + const dataSetId = '12345' + const pieceId = '67890' + const encodedDataSetId = bigIntToBase32(BigInt(dataSetId)) + const encodedPieceId = bigIntToBase32(BigInt(pieceId)) + const slug = `1-${encodedDataSetId}-${encodedPieceId}` + const subpath = '/file%20with%20spaces.txt' + + const request = new Request(`https://${slug}${DNS_ROOT}${subpath}`) + const result = parseRequest(request, { DNS_ROOT, BOT_TOKENS }) + + expect(result).toEqual({ + dataSetId, + pieceId, + ipfsSubpath: subpath, + ipfsFormat: null, + }) + }) + + it('should handle large BigInt values', () => { + const dataSetId = '999999999999999999' + const pieceId = '888888888888888888' + const encodedDataSetId = bigIntToBase32(BigInt(dataSetId)) + const encodedPieceId = bigIntToBase32(BigInt(pieceId)) + const slug = `1-${encodedDataSetId}-${encodedPieceId}` + + const request = new Request(`https://${slug}${DNS_ROOT}/`) + const result = parseRequest(request, { DNS_ROOT, BOT_TOKENS }) + + expect(result).toEqual({ + dataSetId, + pieceId, + ipfsSubpath: '/', + ipfsFormat: null, + }) + }) +}) diff --git a/ipfs-retriever/test/retrieval.test.js b/ipfs-retriever/test/retrieval.test.js new file mode 100644 index 00000000..a47dfedd --- /dev/null +++ b/ipfs-retriever/test/retrieval.test.js @@ -0,0 +1,239 @@ +import { describe, it, expect, vi, beforeEach } from 'vitest' +import { + retrieveIpfsContent, + getRetrievalUrl, + processIpfsResponse, +} from '../lib/retrieval.js' +import { buildRawBlockCar } from './test-data-builders.js' + +describe('retrieveIpfsContent', () => { + const baseUrl = 'https://example.com' + const ipfsRootCid = + 'bafybeigdyrzt5sfp7udm7hu76uh7y26nf3efuylqabf3oclgtqy55fbzdi' + const defaultCacheTtl = 86400 + let fetchMock + + beforeEach(() => { + fetchMock = vi + .fn() + .mockResolvedValue({ ok: true, status: 200, headers: new Headers({}) }) + global.fetch = fetchMock + }) + + it('constructs the correct URL with root path', async () => { + await retrieveIpfsContent(baseUrl, ipfsRootCid, '/') + expect(fetchMock).toHaveBeenCalledWith( + `${baseUrl}/ipfs/${ipfsRootCid}?format=car`, + expect.any(Object), + ) + }) + + it('constructs the correct URL with subpath', async () => { + const subpath = '/path/to/file.txt' + await retrieveIpfsContent(baseUrl, ipfsRootCid, subpath) + expect(fetchMock).toHaveBeenCalledWith( + `${baseUrl}/ipfs/${ipfsRootCid}${subpath}?format=car`, + expect.any(Object), + ) + }) + + it('constructs the correct URL with nested subpath', async () => { + const subpath = '/deep/nested/directory/file.json' + await retrieveIpfsContent(baseUrl, ipfsRootCid, subpath) + expect(fetchMock).toHaveBeenCalledWith( + `${baseUrl}/ipfs/${ipfsRootCid}${subpath}?format=car`, + expect.any(Object), + ) + }) + + it('uses the default cacheTtl if not provided', async () => { + await retrieveIpfsContent(baseUrl, ipfsRootCid, '/') + const options = fetchMock.mock.calls[0][1] + expect(options.cf.cacheTtlByStatus['200-299']).toBe(defaultCacheTtl) + }) + + it('uses the provided cacheTtl', async () => { + await retrieveIpfsContent(baseUrl, ipfsRootCid, '/', 1234) + const options = fetchMock.mock.calls[0][1] + expect(options.cf.cacheTtlByStatus['200-299']).toBe(1234) + }) + + it('sets correct cacheTtlByStatus and cacheEverything', async () => { + await retrieveIpfsContent(baseUrl, ipfsRootCid, '/', 555) + const options = fetchMock.mock.calls[0][1] + expect(options.cf).toEqual({ + cacheTtlByStatus: { + '200-299': 555, + 404: 0, + '500-599': 0, + }, + cacheEverything: true, + }) + }) + + it('passes the signal option correctly', async () => { + const signal = new AbortController().signal + await retrieveIpfsContent(baseUrl, ipfsRootCid, '/', 86400, { signal }) + const options = fetchMock.mock.calls[0][1] + expect(options.signal).toBe(signal) + }) + + it('returns the fetch response and cache miss status', async () => { + const response = { ok: true, status: 200, headers: new Headers({}) } + fetchMock.mockResolvedValueOnce(response) + const result = await retrieveIpfsContent(baseUrl, ipfsRootCid, '/') + expect(result.response).toBe(response) + expect(result.cacheMiss).toBe(true) // No CF-Cache-Status header means cache miss + }) + + it('detects cache hit from CF-Cache-Status header', async () => { + const headers = new Headers({ 'CF-Cache-Status': 'HIT' }) + const response = { ok: true, status: 200, headers } + fetchMock.mockResolvedValueOnce(response) + const result = await retrieveIpfsContent(baseUrl, ipfsRootCid, '/') + expect(result.cacheMiss).toBe(false) + }) + + it('detects cache miss from CF-Cache-Status header', async () => { + const headers = new Headers({ 'CF-Cache-Status': 'MISS' }) + const response = { ok: true, status: 200, headers } + fetchMock.mockResolvedValueOnce(response) + const result = await retrieveIpfsContent(baseUrl, ipfsRootCid, '/') + expect(result.cacheMiss).toBe(true) + }) + + it('always appends format=car query parameter', async () => { + await retrieveIpfsContent(baseUrl, ipfsRootCid, '/file.txt') + expect(fetchMock).toHaveBeenCalledWith( + `${baseUrl}/ipfs/${ipfsRootCid}/file.txt?format=car`, + expect.any(Object), + ) + }) +}) + +describe('getRetrievalUrl', () => { + it('constructs URL with root path', () => { + const url = getRetrievalUrl('https://example.com', 'bafy123abc', '/') + expect(url).toBe('https://example.com/ipfs/bafy123abc') + }) + + it('constructs URL with subpath', () => { + const url = getRetrievalUrl( + 'https://example.com', + 'bafy123abc', + '/file.txt', + ) + expect(url).toBe('https://example.com/ipfs/bafy123abc/file.txt') + }) + + it('constructs URL with nested subpath', () => { + const url = getRetrievalUrl( + 'https://example.com', + 'bafy123abc', + '/path/to/file.json', + ) + expect(url).toBe('https://example.com/ipfs/bafy123abc/path/to/file.json') + }) + + it('avoids double slash in path when the base URL ends with a slash', () => { + const url = getRetrievalUrl( + 'https://example.com/', + 'bafy123abc', + '/file.txt', + ) + expect(url).toBe('https://example.com/ipfs/bafy123abc/file.txt') + }) + + it('handles subpath with trailing slash', () => { + const url = getRetrievalUrl( + 'https://example.com', + 'bafy123abc', + '/directory/', + ) + expect(url).toBe('https://example.com/ipfs/bafy123abc/directory/') + }) + + it('handles empty subpath correctly', () => { + const url = getRetrievalUrl('https://example.com', 'bafy123abc', '') + expect(url).toBe('https://example.com/ipfs/bafy123abc') + }) + + it('preserves special characters in subpath', () => { + const url = getRetrievalUrl( + 'https://example.com', + 'bafy123abc', + '/file%20with%20spaces.txt', + ) + expect(url).toBe( + 'https://example.com/ipfs/bafy123abc/file%20with%20spaces.txt', + ) + }) +}) + +describe('processIpfsResponse', () => { + it('converts CAR to raw, reports the CAR size, and adjusts the headers for raw delivery', async () => { + const fileBytes = new Uint8Array(1000).fill(7) + const { carBytes, rootCid } = await buildRawBlockCar(fileBytes) + expect(carBytes.length).toBeGreaterThan(fileBytes.length) + + const { body, getOriginEgressBytes, headers } = await processIpfsResponse( + new Response(carBytes, { + status: 200, + headers: { + 'content-type': 'application/vnd.ipld.car', + 'x-content-type-options': 'nosniff', + }, + }), + { ipfsRootCid: rootCid, ipfsSubpath: '/', ipfsFormat: null }, + ) + + const served = new Uint8Array(await new Response(body).arrayBuffer()) + expect(served).toEqual(fileBytes) + // The CAR is streamed lazily, so the count is only final once the body has + // been consumed. It reports the full CAR fetched from the SP, not the raw + // bytes. + expect(getOriginEgressBytes()).toBe(carBytes.length) + // The browser should display the raw content and sniff its type. + expect(headers.get('content-disposition')).toBe('inline') + expect(headers.get('content-type')).toBe(null) + expect(headers.get('x-content-type-options')).toBe(null) + }) + + it('passes the body and headers through unchanged for ?format=car with null originEgressBytes', async () => { + const carBytes = new Uint8Array([1, 2, 3, 4]) + const response = new Response(carBytes, { + status: 200, + headers: { 'content-type': 'application/vnd.ipld.car' }, + }) + + const { body, getOriginEgressBytes, headers } = await processIpfsResponse( + response, + { + ipfsRootCid: 'bafyroot', + ipfsSubpath: '/', + ipfsFormat: 'car', + }, + ) + + expect(getOriginEgressBytes()).toBe(null) + expect(new Uint8Array(await new Response(body).arrayBuffer())).toEqual( + carBytes, + ) + // CAR is served as-is, so the upstream content type is preserved. + expect(headers.get('content-type')).toBe('application/vnd.ipld.car') + expect(headers.get('content-disposition')).toBe(null) + }) + + it('passes the body through unchanged for non-ok responses with null originEgressBytes', async () => { + const response = new Response('not found', { status: 404 }) + + const { body, getOriginEgressBytes } = await processIpfsResponse(response, { + ipfsRootCid: 'bafyroot', + ipfsSubpath: '/', + ipfsFormat: null, + }) + + expect(getOriginEgressBytes()).toBe(null) + expect(await new Response(body).text()).toBe('not found') + }) +}) diff --git a/ipfs-retriever/test/retriever.test.js b/ipfs-retriever/test/retriever.test.js new file mode 100644 index 00000000..e20cc596 --- /dev/null +++ b/ipfs-retriever/test/retriever.test.js @@ -0,0 +1,1289 @@ +import { describe, it, expect, vi, beforeAll } from 'vitest' +import worker from '../bin/ipfs-retriever.js' +import { createHash } from 'node:crypto' +import { retrieveIpfsContent } from '../lib/retrieval.js' +import { + env, + createExecutionContext, + waitOnExecutionContext, +} from 'cloudflare:test' +import assert from 'node:assert/strict' +import { + withDataSetPiece, + withApprovedProvider, + withBadBits, + withWalletDetails, + buildRawBlockCar, +} from './test-data-builders.js' +import { CONTENT_STORED_ON_CALIBRATION } from './test-data.js' +import { buildSlug } from '../lib/store.js' + +function sleep(ms) { + return new Promise((resolve) => setTimeout(resolve, ms)) +} + +/** + * Calls the worker and drains the response body so the back-pressured egress + * measurement (which only advances as the client reads) completes before the + * test waits on the execution context. Returns a re-readable response with the + * same status, headers and bytes. A body-less response is returned unchanged. + * + * @param {Request} req + * @param {Env} env + * @param {ExecutionContext} ctx + * @param {object} [options] + */ +async function fetchAndRead(req, env, ctx, options = {}) { + const res = await worker.fetch(req, env, ctx, options) + if (!res.body) return res + return new Response(await res.arrayBuffer(), res) +} + +const DNS_ROOT = '.ipfs.filbeam.io' +env.DNS_ROOT = DNS_ROOT +const botTokens = { secret: 'testbot' } +env.BOT_TOKENS = JSON.stringify(botTokens) + +describe('retriever.fetch', () => { + const defaultPayerAddress = '0x1234567890abcdef1234567890abcdef12345678' + const { + ipfsRootCid: realIpfsRootCid, + dataSetId, + pieceId: realPieceId, + } = CONTENT_STORED_ON_CALIBRATION[0] + const realDataSetId = String(dataSetId) + + beforeAll(async () => { + await env.DB.batch([ + env.DB.prepare('DELETE FROM pieces'), + env.DB.prepare('DELETE FROM data_sets'), + env.DB.prepare('DELETE FROM wallet_details'), + ]) + + let cursor + while (true) { + const list = await env.BAD_BITS_KV.list({ cursor }) + for (const key of list.keys) { + await env.BAD_BITS_KV.delete(key) + } + if (list.list_complete) break + cursor = list.cursor + } + + for (const { + serviceProviderId, + serviceUrl, + pieceCid, + ipfsRootCid, + dataSetId, + pieceId, + } of CONTENT_STORED_ON_CALIBRATION) { + await withDataSetPiece(env, { + serviceProviderId, + pieceCid, + ipfsRootCid, + payerAddress: defaultPayerAddress, + withCDN: true, + withIpfsIndexing: true, + dataSetId: String(dataSetId), + pieceId, + }) + await withApprovedProvider(env, { + id: serviceProviderId, + serviceUrl, + }) + } + }) + + it('redirects to https://filbeam.com when no CID and no wallet address were provided', async () => { + const ctx = createExecutionContext() + const req = new Request(`https://${DNS_ROOT.slice(1)}/`) + const res = await fetchAndRead(req, env, ctx) + await waitOnExecutionContext(ctx) + expect(res.status).toBe(302) + expect(res.headers.get('Location')).toBe('https://filbeam.com/') + }) + + it('returns 404 for invalid path format on DNS_ROOT (missing CID)', async () => { + const ctx = createExecutionContext() + const req = new Request( + `https://${DNS_ROOT.slice(1)}/${defaultPayerAddress}`, + ) + const res = await fetchAndRead(req, env, ctx) + await waitOnExecutionContext(ctx) + expect(res.status).toBe(404) + expect(await res.text()).toContain('Invalid path format') + }) + + it('returns 404 for invalid wallet address on DNS_ROOT path', async () => { + const ctx = createExecutionContext() + const invalidWallet = 'invalid-wallet' + const ipfsRootCid = 'bafk4testslug1' + const req = new Request( + `https://${DNS_ROOT.slice(1)}/${invalidWallet}/${ipfsRootCid}`, + ) + const res = await fetchAndRead(req, env, ctx) + await waitOnExecutionContext(ctx) + expect(res.status).toBe(404) + expect(await res.text()).toContain('Invalid wallet address') + }) + + it('redirects to slug subdomain when valid wallet and CID are provided on DNS_ROOT path', async () => { + // Set up test data with numeric pieceId and dataSetId for slug generation + const testPayerAddress = '0xabcdef1234567890abcdef1234567890abcdef99' + const testIpfsRootCid = 'bafk4testslug2' + const testDataSetId = '12345' + const testPieceId = '67890' + const serviceProviderId = '100' + + await withDataSetPiece(env, { + serviceProviderId, + payerAddress: testPayerAddress, + ipfsRootCid: testIpfsRootCid, + dataSetId: testDataSetId, + pieceId: testPieceId, + withCDN: true, + withIpfsIndexing: true, + }) + await withApprovedProvider(env, { + id: serviceProviderId, + serviceUrl: 'https://test-provider.example.com', + }) + + const ctx = createExecutionContext() + const req = new Request( + `https://${DNS_ROOT.slice(1)}/${testPayerAddress}/${testIpfsRootCid}`, + ) + const res = await fetchAndRead(req, env, ctx) + await waitOnExecutionContext(ctx) + expect(res.status).toBe(302) + const location = res.headers.get('Location') + // Expected slug: 1-ga4q-aeete (version-base32(12345)-base32(67890)) + expect(location).toBe('https://1-ga4q-aeete.ipfs.filbeam.io/') + }) + + it('accepts mixed case addresses', async () => { + const testPayerAddress = '0xabcdef1234567890abcdef1234567890abcdef99' + const testIpfsRootCid = 'bafk4testslug2' + const testDataSetId = '12345' + const testPieceId = '67890' + const serviceProviderId = '100' + + await withDataSetPiece(env, { + serviceProviderId, + payerAddress: testPayerAddress, + ipfsRootCid: testIpfsRootCid, + dataSetId: testDataSetId, + pieceId: testPieceId, + withCDN: true, + withIpfsIndexing: true, + }) + await withApprovedProvider(env, { + id: serviceProviderId, + serviceUrl: 'https://test-provider.example.com', + }) + + const ctx = createExecutionContext() + const req = new Request( + `https://${DNS_ROOT.slice(1)}/${testPayerAddress.toUpperCase()}/${testIpfsRootCid}`, + ) + const res = await fetchAndRead(req, env, ctx) + await waitOnExecutionContext(ctx) + expect(res.status).toBe(302) + const location = res.headers.get('Location') + expect(location).toBe('https://1-ga4q-aeete.ipfs.filbeam.io/') + }) + + it('redirects to slug subdomain with subpath when wallet, CID, and pathname are provided on DNS_ROOT path', async () => { + // Set up test data with numeric pieceId and dataSetId for slug generation + const testPayerAddress = '0xabcdef1234567890abcdef1234567890abcdef98' + const testIpfsRootCid = 'bafk4testslug3' + const testDataSetId = '54321' + const testPieceId = '98765' + const serviceProviderId = '101' + + await withDataSetPiece(env, { + serviceProviderId, + payerAddress: testPayerAddress, + ipfsRootCid: testIpfsRootCid, + dataSetId: testDataSetId, + pieceId: testPieceId, + withCDN: true, + withIpfsIndexing: true, + }) + await withApprovedProvider(env, { + id: serviceProviderId, + serviceUrl: 'https://test-provider2.example.com', + }) + + const ctx = createExecutionContext() + const subpath = 'path/to/file.txt' + const req = new Request( + `https://${DNS_ROOT.slice(1)}/${testPayerAddress}/${testIpfsRootCid}/${subpath}`, + ) + const res = await fetchAndRead(req, env, ctx) + await waitOnExecutionContext(ctx) + expect(res.status).toBe(302) + const location = res.headers.get('Location') + // Expected slug: 1-2qyq-aga42 (version-base32(54321)-base32(98765)) + expect(location).toBe( + 'https://1-2qyq-aga42.ipfs.filbeam.io/path/to/file.txt', + ) + }) + + it('redirects to https://*.filcdn.io/* when old domain was used', async () => { + const ctx = createExecutionContext() + const req = new Request(`https://foo.filcdn.io/bar`) + const res = await fetchAndRead(req, env, ctx) + await waitOnExecutionContext(ctx) + expect(res.status).toBe(301) + expect(res.headers.get('Location')).toBe(`https://foo.filbeam.io/bar`) + }) + + it('returns 405 for unsupported request methods', async () => { + const ctx = createExecutionContext() + const req = withRequest('1', '1', 'POST') + const res = await fetchAndRead(req, env, ctx) + await waitOnExecutionContext(ctx) + expect(res.status).toBe(405) + expect(await res.text()).toBe('Method Not Allowed') + }) + + it('returns 400 if required fields are missing', async () => { + const ctx = createExecutionContext() + const mockRetrieveIpfsContent = vi.fn() + const req = new Request( + `http://${buildSlug(BigInt(realDataSetId), BigInt(realPieceId)).replace(/^(1-)/, '')}.${DNS_ROOT.slice(1)}`, + ) + const res = await fetchAndRead(req, env, ctx, { + retrieveIpfsContent: mockRetrieveIpfsContent, + }) + await waitOnExecutionContext(ctx) + expect(res.status).toBe(400) + }) + + it('returns 400 if slug has invalid base32 encoding', async () => { + const ctx = createExecutionContext() + const mockRetrieveIpfsContent = vi.fn() + const req = new Request( + `http://${buildSlug(BigInt(realDataSetId), BigInt(realPieceId))}1.${DNS_ROOT.slice(1)}`, + ) + const res = await fetchAndRead(req, env, ctx, { + retrieveIpfsContent: mockRetrieveIpfsContent, + }) + await waitOnExecutionContext(ctx) + expect(res.status).toBe(400) + expect(await res.text()).toContain('Invalid pieceId encoding in slug') + }) + + it('returns the response from retrieveIpfsContent', async () => { + const fakeResponse = new Response('hello', { + status: 201, + headers: { 'X-Test': 'yes' }, + }) + const mockRetrieveIpfsContent = vi.fn().mockResolvedValue({ + response: fakeResponse, + cacheMiss: true, + }) + const ctx = createExecutionContext() + const req = withRequest(realDataSetId, realPieceId) + const res = await fetchAndRead(req, env, ctx, { + retrieveIpfsContent: mockRetrieveIpfsContent, + }) + await waitOnExecutionContext(ctx) + expect(await res.text()).toBe('hello') + expect(res.status).toBe(201) + expect(res.headers.get('X-Test')).toBe('yes') + }) + + it('sets Content-Control response header', async () => { + const originResponse = new Response('hello') + const mockRetrieveIpfsContent = vi.fn().mockResolvedValue({ + response: originResponse, + cacheMiss: true, + }) + const ctx = createExecutionContext() + const req = withRequest(realDataSetId, realPieceId) + const res = await fetchAndRead(req, env, ctx, { + retrieveIpfsContent: mockRetrieveIpfsContent, + }) + await waitOnExecutionContext(ctx) + const cacheControlHeaders = res.headers.get('Cache-Control') + expect(cacheControlHeaders).toContain('public') + expect(cacheControlHeaders).toContain(`max-age=${env.CLIENT_CACHE_TTL}`) + }) + + it('sets Content-Control response on empty body', async () => { + const originResponse = new Response(null) + const mockRetrieveIpfsContent = vi.fn().mockResolvedValue({ + response: originResponse, + cacheMiss: false, + }) + const ctx = createExecutionContext() + const req = withRequest(realDataSetId, realPieceId) + const res = await fetchAndRead(req, env, ctx, { + retrieveIpfsContent: mockRetrieveIpfsContent, + }) + await waitOnExecutionContext(ctx) + const cacheControlHeaders = res.headers.get('Cache-Control') + expect(cacheControlHeaders).toContain('public') + expect(cacheControlHeaders).toContain(`max-age=${env.CLIENT_CACHE_TTL}`) + }) + + it('sets Content-Security-Policy response header', async () => { + const originResponse = new Response('hello', { + headers: { + 'Content-Security-Policy': 'report-uri: https://endpoint.example.com', + }, + }) + const mockRetrieveIpfsContent = vi.fn().mockResolvedValue({ + response: originResponse, + cacheMiss: true, + }) + const ctx = createExecutionContext() + const req = withRequest(realDataSetId, realPieceId) + const res = await fetchAndRead(req, env, ctx, { + retrieveIpfsContent: mockRetrieveIpfsContent, + }) + await waitOnExecutionContext(ctx) + const csp = res.headers.get('Content-Security-Policy') + expect(csp).toMatch(/^default-src 'self'/) + expect(csp).toContain('https://*.filbeam.io') + }) + + it('fetches the file from calibration service provider', async () => { + // The default request format is `car`, so the worker serves the CAR file + // unchanged. This is the sha256 of the CAR served for the dataset above. + const expectedHash = + 'd895b1ec0e1fbde5ba2ad3b927e4ea43dcd126e11ddfd9930027a0f594bbe002' + const ctx = createExecutionContext() + const req = withRequest(realDataSetId, realPieceId) + const res = await fetchAndRead(req, env, ctx, { retrieveIpfsContent }) + await waitOnExecutionContext(ctx) + expect(res.status).toBe(200) + // get the sha256 hash of the content + const content = await res.bytes() + const hash = createHash('sha256').update(content).digest('hex') + expect(hash).toEqual(expectedHash) + }) + it('stores retrieval results with cache miss and content length set in D1', async () => { + const body = 'file content' + const expectedEgressBytes = Buffer.byteLength(body, 'utf8') + const fakeResponse = new Response(body, { + status: 200, + headers: { + 'CF-Cache-Status': 'MISS', + }, + }) + const mockRetrieveIpfsContent = vi.fn().mockResolvedValue({ + response: fakeResponse, + cacheMiss: true, + }) + const ctx = createExecutionContext() + const req = withRequest(realDataSetId, realPieceId) + const res = await fetchAndRead(req, env, ctx, { + retrieveIpfsContent: mockRetrieveIpfsContent, + }) + await waitOnExecutionContext(ctx) + assert.strictEqual(res.status, 200) + const readOutput = await env.DB.prepare( + `SELECT id, response_status, egress_bytes, cache_miss + FROM retrieval_logs + WHERE data_set_id = ?`, + ) + .bind(String(realDataSetId)) + .all() + const result = readOutput.results + assert.deepStrictEqual(result, [ + { + id: 1, // Assuming this is the first log entry + response_status: 200, + egress_bytes: expectedEgressBytes, + cache_miss: 1, // 1 for true, 0 for false + }, + ]) + }) + it('stores retrieval results with cache hit and content length set in D1', async () => { + const body = 'file content' + const expectedEgressBytes = Buffer.byteLength(body, 'utf8') + const fakeResponse = new Response(body, { + status: 200, + headers: { + 'CF-Cache-Status': 'HIT', + }, + }) + const mockRetrieveIpfsContent = vi.fn().mockResolvedValue({ + response: fakeResponse, + cacheMiss: false, + }) + const ctx = createExecutionContext() + const req = withRequest(realDataSetId, realPieceId) + const res = await fetchAndRead(req, env, ctx, { + retrieveIpfsContent: mockRetrieveIpfsContent, + }) + await waitOnExecutionContext(ctx) + assert.strictEqual(res.status, 200) + const readOutput = await env.DB.prepare( + `SELECT id, response_status, egress_bytes, cache_miss + FROM retrieval_logs + WHERE data_set_id = ?`, + ) + .bind(String(realDataSetId)) + .all() + const result = readOutput.results + assert.deepStrictEqual(result, [ + { + id: 1, // Assuming this is the first log entry + response_status: 200, + egress_bytes: expectedEgressBytes, + cache_miss: 0, // 1 for true, 0 for false + }, + ]) + }) + it('stores retrieval performance stats in D1', async () => { + const body = 'file content' + const fakeResponse = new Response(body, { + status: 200, + headers: { + 'CF-Cache-Status': 'MISS', + }, + }) + const mockRetrieveIpfsContent = async () => { + await sleep(1) // Simulate a delay + return { + response: fakeResponse, + cacheMiss: true, + } + } + const ctx = createExecutionContext() + const req = withRequest(realDataSetId, realPieceId) + const res = await fetchAndRead(req, env, ctx, { + retrieveIpfsContent: mockRetrieveIpfsContent, + }) + await waitOnExecutionContext(ctx) + assert.strictEqual(res.status, 200) + const readOutput = await env.DB.prepare( + `SELECT + response_status, + fetch_ttfb, + fetch_ttlb, + worker_ttfb + FROM retrieval_logs + WHERE data_set_id = ?`, + ) + .bind(String(realDataSetId)) + .all() + assert.strictEqual(readOutput.results.length, 1) + const result = readOutput.results[0] + + assert.strictEqual(result.response_status, 200) + assert.strictEqual(typeof result.fetch_ttfb, 'number') + assert.strictEqual(typeof result.fetch_ttlb, 'number') + assert.strictEqual(typeof result.worker_ttfb, 'number') + }) + it('stores request country code in D1', async () => { + const body = 'file content' + const mockRetrieveIpfsContent = async () => { + return { + response: new Response(body, { + status: 200, + }), + cacheMiss: true, + } + } + const ctx = createExecutionContext() + const req = withRequest(realDataSetId, realPieceId, 'GET', { + 'CF-IPCountry': 'US', + }) + const res = await fetchAndRead(req, env, ctx, { + retrieveIpfsContent: mockRetrieveIpfsContent, + }) + await waitOnExecutionContext(ctx) + assert.strictEqual(res.status, 200) + const { results } = await env.DB.prepare( + `SELECT request_country_code + FROM retrieval_logs + WHERE data_set_id = ?`, + ) + .bind(String(realDataSetId)) + .all() + assert.deepStrictEqual(results, [ + { + request_country_code: 'US', + }, + ]) + }) + it('logs 0 egress bytes for empty body', async () => { + const fakeResponse = new Response(null, { + status: 200, + headers: { + 'CF-Cache-Status': 'MISS', + }, + }) + const mockRetrieveIpfsContent = vi.fn().mockResolvedValue({ + response: fakeResponse, + cacheMiss: true, + }) + const ctx = createExecutionContext() + const req = withRequest(realDataSetId, realPieceId) + const res = await fetchAndRead(req, env, ctx, { + retrieveIpfsContent: mockRetrieveIpfsContent, + }) + await waitOnExecutionContext(ctx) + assert.strictEqual(res.status, 200) + const readOutput = await env.DB.prepare( + 'SELECT egress_bytes FROM retrieval_logs WHERE data_set_id = ?', + ) + .bind(String(realDataSetId)) + .all() + assert.strictEqual(readOutput.results.length, 1) + assert.strictEqual(readOutput.results[0].egress_bytes, 0) + }) + + it('logs a 900 retrieval result when the response stream errors', async () => { + const erroringBody = new ReadableStream({ + pull(controller) { + controller.enqueue(new Uint8Array([1, 2, 3])) + controller.error(new Error('stream boom')) + }, + }) + const mockRetrieveIpfsContent = vi.fn().mockResolvedValue({ + response: new Response(erroringBody, { status: 200 }), + cacheMiss: true, + }) + + const ctx = createExecutionContext() + // `?format=car` passes the body through unchanged, so the erroring stream + // reaches the egress measurement. + const req = withRequest( + realDataSetId, + realPieceId, + 'GET', + {}, + { + format: 'car', + }, + ) + // Not fetchAndRead: the body errors mid-stream, and the source error (not + // client consumption) drives the 900. Draining it here would throw. + const res = await worker.fetch(req, env, ctx, { + retrieveIpfsContent: mockRetrieveIpfsContent, + }) + expect(res.status).toBe(200) + await waitOnExecutionContext(ctx) + + const log = await env.DB.prepare( + 'SELECT response_status, egress_bytes FROM retrieval_logs WHERE data_set_id = ? AND response_status = 900', + ) + .bind(String(realDataSetId)) + .first() + // The client never reads, so back-pressure stops the chunk being counted + // before the source errors. + expect(log).toEqual({ response_status: 900, egress_bytes: 0 }) + }) + + it( + 'measures egress correctly from real service provider', + { timeout: 10000 }, + async () => { + const tasks = CONTENT_STORED_ON_CALIBRATION.map( + ({ dataSetId, pieceId, serviceProviderId }) => { + return (async () => { + try { + const ctx = createExecutionContext() + const req = withRequest(String(dataSetId), pieceId) + const res = await fetchAndRead(req, env, ctx, { + retrieveIpfsContent, + }) + await waitOnExecutionContext(ctx) + + assert.strictEqual(res.status, 200) + + const content = await res.arrayBuffer() + const actualBytes = content.byteLength + + const log = await env.DB.prepare( + 'SELECT egress_bytes FROM retrieval_logs WHERE data_set_id = ? ORDER BY id DESC LIMIT 1', + ) + .bind(String(dataSetId)) + .first() + + assert.strictEqual(log.egress_bytes, actualBytes) + + return { serviceProviderId, success: true } + } catch (err) { + console.warn( + `⚠️ Warning: Fetch or verification failed for serviceProvider ${serviceProviderId}:`, + err, + ) + throw err + } + })() + }, + ) + + try { + const res = await Promise.allSettled(tasks) + if (!res.some((r) => r.status === 'fulfilled')) { + throw new Error('All tasks failed') + } + } catch (err) { + const serviceProvidersChecked = CONTENT_STORED_ON_CALIBRATION.map( + (o) => o.serviceProviderId, + ) + throw new Error( + `❌ All service providers failed to fetch. Service providers checked: ${serviceProvidersChecked.join(', ')}`, + ) + } + }, + ) + + it('charges bots for egress', async () => { + const botToken = Object.keys(botTokens)[0] + /** @type {string} */ + const botName = env.BOT_TOKENS[botToken] + console.log({ botToken, botName }) + + const mockRetrieveIpfsContent = vi.fn().mockResolvedValue({ + response: new Response('fake'), + cacheMiss: true, + }) + const ctx = createExecutionContext() + const req = withRequest(realDataSetId, realPieceId, 'GET', { + authorization: `Bearer ${botToken}`, + }) + const res = await fetchAndRead(req, env, ctx, { + retrieveIpfsContent: mockRetrieveIpfsContent, + }) + await waitOnExecutionContext(ctx) + expect(res.status).toBe(200) + const readOutput = await env.DB.prepare( + 'SELECT egress_bytes FROM retrieval_logs WHERE data_set_id = ?', + ) + .bind(String(realDataSetId)) + .all() + expect(readOutput.results).toStrictEqual([ + expect.objectContaining({ + egress_bytes: 4, + }), + ]) + }) + + it('logs the CAR size as cache-miss egress when converting CAR to raw', async () => { + const fileBytes = new Uint8Array(1000).fill(42) + const { carBytes, rootCid } = await buildRawBlockCar(fileBytes) + expect(carBytes.length).toBeGreaterThan(fileBytes.length) + + const carDataSetId = '7777' + const carPieceId = '7777' + await withDataSetPiece(env, { + serviceProviderId: 'sp-car', + payerAddress: defaultPayerAddress, + pieceCid: 'bagacartest', + ipfsRootCid: rootCid, + dataSetId: carDataSetId, + pieceId: carPieceId, + }) + await withApprovedProvider(env, { + id: 'sp-car', + serviceUrl: 'https://pdp.example/', + }) + + const mockRetrieveIpfsContent = vi.fn().mockResolvedValue({ + response: new Response(carBytes, { status: 200 }), + cacheMiss: true, + }) + + const ctx = createExecutionContext() + const req = withRequest( + carDataSetId, + carPieceId, + 'GET', + {}, + { format: null }, + ) + const res = await fetchAndRead(req, env, ctx, { + retrieveIpfsContent: mockRetrieveIpfsContent, + }) + await waitOnExecutionContext(ctx) + + expect(res.status).toBe(200) + expect(new Uint8Array(await res.arrayBuffer())).toEqual(fileBytes) + // Converting CAR to raw serves the content inline for browsers to sniff. + expect(res.headers.get('content-disposition')).toBe('inline') + + const readOutput = await env.DB.prepare( + `SELECT egress_bytes, cache_miss_egress_bytes, cache_miss + FROM retrieval_logs + WHERE data_set_id = ?`, + ) + .bind(carDataSetId) + .all() + + // The client is charged the raw bytes served, the cache-miss quota the + // larger CAR fetched from the service provider. + expect(readOutput.results).toStrictEqual([ + { + egress_bytes: fileBytes.length, + cache_miss_egress_bytes: carBytes.length, + cache_miss: 1, + }, + ]) + }) + + it('charges the cache-miss egress quota by the CAR size when enforcing', async () => { + const fileBytes = new Uint8Array(1000).fill(42) + const { carBytes, rootCid } = await buildRawBlockCar(fileBytes) + + const dataSetId = '8888' + const pieceId = '8888' + await withDataSetPiece(env, { + serviceProviderId: 'sp-quota-car', + payerAddress: defaultPayerAddress, + pieceCid: 'bagacarquota', + ipfsRootCid: rootCid, + dataSetId, + pieceId, + }) + await withApprovedProvider(env, { + id: 'sp-quota-car', + serviceUrl: 'https://pdp.example/', + }) + await env.DB.prepare( + 'INSERT INTO data_set_egress_quotas (data_set_id, cdn_egress_quota, cache_miss_egress_quota) VALUES (?, ?, ?)', + ) + .bind(dataSetId, 100000, 100000) + .run() + + const mockRetrieveIpfsContent = vi.fn().mockResolvedValue({ + response: new Response(carBytes, { status: 200 }), + cacheMiss: true, + }) + + const ctx = createExecutionContext() + const req = withRequest(dataSetId, pieceId, 'GET', {}, { format: null }) + const res = await fetchAndRead( + req, + { ...env, ENFORCE_EGRESS_QUOTA: true }, + ctx, + { retrieveIpfsContent: mockRetrieveIpfsContent }, + ) + await waitOnExecutionContext(ctx) + + expect(res.status).toBe(200) + expect(new Uint8Array(await res.arrayBuffer())).toEqual(fileBytes) + + // The CDN quota is charged the raw bytes served, the cache-miss quota the + // larger CAR fetched from the service provider. + const quota = await env.DB.prepare( + 'SELECT cdn_egress_quota, cache_miss_egress_quota FROM data_set_egress_quotas WHERE data_set_id = ?', + ) + .bind(dataSetId) + .first() + + expect(quota).toStrictEqual({ + cdn_egress_quota: 100000 - fileBytes.length, + cache_miss_egress_quota: 100000 - carBytes.length, + }) + }) + + it('charges the cache-miss egress quota for a ?format=car cache miss', async () => { + // `?format=car` passes the CAR through unchanged, so the bytes served equal + // the bytes fetched from the service provider. + const carBytes = new Uint8Array(500).fill(7) + + const dataSetId = '9090' + const pieceId = '9090' + await withDataSetPiece(env, { + serviceProviderId: 'sp-car-passthrough', + payerAddress: defaultPayerAddress, + pieceCid: 'bagacarpassthrough', + ipfsRootCid: 'bafkcarpassthrough', + dataSetId, + pieceId, + }) + await withApprovedProvider(env, { + id: 'sp-car-passthrough', + serviceUrl: 'https://pdp.example/', + }) + await env.DB.prepare( + 'INSERT INTO data_set_egress_quotas (data_set_id, cdn_egress_quota, cache_miss_egress_quota) VALUES (?, ?, ?)', + ) + .bind(dataSetId, 100000, 100000) + .run() + + const mockRetrieveIpfsContent = vi.fn().mockResolvedValue({ + response: new Response(carBytes, { status: 200 }), + cacheMiss: true, + }) + + const ctx = createExecutionContext() + const req = withRequest(dataSetId, pieceId, 'GET', {}, { format: 'car' }) + const res = await fetchAndRead( + req, + { ...env, ENFORCE_EGRESS_QUOTA: true }, + ctx, + { retrieveIpfsContent: mockRetrieveIpfsContent }, + ) + await waitOnExecutionContext(ctx) + + expect(res.status).toBe(200) + expect(new Uint8Array(await res.arrayBuffer())).toEqual(carBytes) + + const quota = await env.DB.prepare( + 'SELECT cdn_egress_quota, cache_miss_egress_quota FROM data_set_egress_quotas WHERE data_set_id = ?', + ) + .bind(dataSetId) + .first() + + expect(quota).toStrictEqual({ + cdn_egress_quota: 100000 - carBytes.length, + cache_miss_egress_quota: 100000 - carBytes.length, + }) + }) + + it('retries another service provider when the first one fails', async () => { + const sharedIpfsRootCid = 'bafkfallbackshared' + const badServiceUrl = 'https://bad-sp.example/' + const goodServiceUrl = 'https://good-sp.example/' + const goodDataSetId = '8801' + + await withDataSetPiece(env, { + serviceProviderId: 'sp-fallback-bad', + payerAddress: defaultPayerAddress, + pieceCid: 'bagafallbackbad', + ipfsRootCid: sharedIpfsRootCid, + dataSetId: '8800', + pieceId: '8800', + }) + await withApprovedProvider(env, { + id: 'sp-fallback-bad', + serviceUrl: badServiceUrl, + }) + await withDataSetPiece(env, { + serviceProviderId: 'sp-fallback-good', + payerAddress: defaultPayerAddress, + pieceCid: 'bagafallbackgood', + ipfsRootCid: sharedIpfsRootCid, + dataSetId: goodDataSetId, + pieceId: '8801', + }) + await withApprovedProvider(env, { + id: 'sp-fallback-good', + serviceUrl: goodServiceUrl, + }) + + const mockRetrieveIpfsContent = vi.fn(async (serviceUrl) => { + if (serviceUrl === goodServiceUrl) { + return { + response: new Response('fake', { + status: 200, + headers: { 'CF-Cache-Status': 'MISS' }, + }), + cacheMiss: true, + } + } + return { + response: new Response('boom', { status: 500 }), + cacheMiss: true, + } + }) + + const ctx = createExecutionContext() + const req = withRequest('8800', '8800', 'GET', {}, { format: 'car' }) + const res = await fetchAndRead(req, env, ctx, { + retrieveIpfsContent: mockRetrieveIpfsContent, + }) + await waitOnExecutionContext(ctx) + + expect(res.status).toBe(200) + expect(await res.text()).toBe('fake') + + // The content is served, and the egress is charged to the data set whose + // service provider succeeded. + const readOutput = await env.DB.prepare( + `SELECT data_set_id, response_status + FROM retrieval_logs + WHERE data_set_id = ?`, + ) + .bind(goodDataSetId) + .all() + expect(readOutput.results).toStrictEqual([ + expect.objectContaining({ + data_set_id: goodDataSetId, + response_status: 200, + }), + ]) + }) + + it('requests payment if withCDN=false', async () => { + const dataSetId = '1004' + const pieceId = '2004' + const pieceCid = + 'baga6ea4seaqaleibb6ud4xeemuzzpsyhl6cxlsymsnfco4cdjka5uzajo2x4ipa' + const ipfsRootCid = 'bafk4test' + const serviceProviderId = 'service-provider' + const payerAddress = '0x1234567890abcdef1234567890abcdef12345678' + + await withApprovedProvider(env, { + id: serviceProviderId, + serviceUrl: 'https://test-provider.xyz', + }) + + await withDataSetPiece(env, { + serviceProviderId, + pieceCid, + ipfsRootCid, + dataSetId, + withCDN: false, + pieceId, + payerAddress, + }) + + const ctx = createExecutionContext() + const req = withRequest(dataSetId, pieceId, 'GET') + const res = await fetchAndRead(req, env, ctx) + await waitOnExecutionContext(ctx) + + assert.strictEqual(res.status, 402) + }) + it('reads the provider URL from the database', async () => { + const serviceProviderId = 'service-provider-id' + const dataSetId = '1001' + const pieceId = '2001' + const payerAddress = '0x1234567890abcdef1234567890abcdef12345608' + const ipfsRootCid = 'bafk4test' + const body = 'file content' + + await withDataSetPiece(env, { + serviceProviderId, + dataSetId, + pieceId, + ipfsRootCid, + payerAddress, + }) + + await withApprovedProvider(env, { + id: serviceProviderId, + serviceUrl: 'https://mock-pdp-url.com', + }) + + const mockRetrieveIpfsContent = async () => { + return { + response: new Response(body, { + status: 200, + }), + cacheMiss: true, + } + } + + const ctx = createExecutionContext() + const req = withRequest(dataSetId, pieceId) + const res = await fetchAndRead(req, env, ctx, { + retrieveIpfsContent: mockRetrieveIpfsContent, + }) + await waitOnExecutionContext(ctx) + + // Check if the URL fetched is from the database + expect(await res.text()).toBe(body) + expect(res.status).toBe(200) + }) + + it('throws an error if the providerAddress is not found in the database', async () => { + const serviceProviderId = 'service-provider-id' + const dataSetId = '1002' + const pieceId = '2002' + const payerAddress = '0x2A06D234246eD18b6C91de8349fF34C22C7268e8' + const ipfsRootCid = 'bafk4test' + + await withDataSetPiece(env, { + serviceProviderId, + dataSetId, + pieceId, + ipfsRootCid, + payerAddress, + }) + + const ctx = createExecutionContext() + const req = withRequest(dataSetId, pieceId) + const res = await fetchAndRead(req, env, ctx) + await waitOnExecutionContext(ctx) + + // Expect an error because no URL was found + expect(res.status).toBe(404) + expect(await res.text()).toBe( + `No approved service provider found for payer '0x2a06d234246ed18b6c91de8349ff34c22c7268e8' and the requested content.`, + ) + }) + + it('returns data set ID in the X-Data-Set-ID response header', async () => { + const { dataSetId } = CONTENT_STORED_ON_CALIBRATION[0] + const mockRetrieveIpfsContent = vi.fn().mockResolvedValue({ + response: new Response('hello'), + cacheMiss: true, + }) + const ctx = createExecutionContext() + const req = withRequest(realDataSetId, realPieceId) + const res = await fetchAndRead(req, env, ctx, { + retrieveIpfsContent: mockRetrieveIpfsContent, + }) + await waitOnExecutionContext(ctx) + expect(await res.text()).toBe('hello') + expect(res.headers.get('X-Data-Set-ID')).toBe(String(dataSetId)) + }) + + it('stores data set ID in retrieval logs', async () => { + const { dataSetId } = CONTENT_STORED_ON_CALIBRATION[0] + const mockRetrieveIpfsContent = vi.fn().mockResolvedValue({ + response: new Response('hello'), + cacheMiss: true, + }) + const ctx = createExecutionContext() + const req = withRequest(realDataSetId, realPieceId) + const res = await fetchAndRead(req, env, ctx, { + retrieveIpfsContent: mockRetrieveIpfsContent, + }) + await waitOnExecutionContext(ctx) + expect(await res.text()).toBe('hello') + + assert.strictEqual(res.status, 200) + const { results } = await env.DB.prepare( + `SELECT id, response_status, cache_miss + FROM retrieval_logs + WHERE data_set_id = ?`, + ) + .bind(String(dataSetId)) + .all() + assert.deepStrictEqual(results, [ + { + id: 1, // Assuming this is the first log entry + response_status: 200, + cache_miss: 1, // 1 for true, 0 for false + }, + ]) + }) + + it('returns data set ID in the X-Data-Set-ID response header when the response body is empty', async () => { + const { dataSetId } = CONTENT_STORED_ON_CALIBRATION[0] + const mockRetrieveIpfsContent = vi.fn().mockResolvedValue({ + response: new Response(null, { status: 404 }), + cacheMiss: true, + }) + const ctx = createExecutionContext() + const req = withRequest(realDataSetId, realPieceId) + const res = await fetchAndRead(req, env, ctx, { + retrieveIpfsContent: mockRetrieveIpfsContent, + }) + await waitOnExecutionContext(ctx) + expect(res.body).toBeNull() + expect(res.headers.get('X-Data-Set-ID')).toBe(String(dataSetId)) + }) + + it('supports HEAD requests', async () => { + const fakeResponse = new Response('file content', { + status: 200, + }) + const mockRetrieveIpfsContent = vi.fn().mockResolvedValue({ + response: fakeResponse, + cacheMiss: true, + }) + const ctx = createExecutionContext() + const req = withRequest(realDataSetId, realPieceId, 'HEAD') + const res = await fetchAndRead(req, env, ctx, { + retrieveIpfsContent: mockRetrieveIpfsContent, + }) + await waitOnExecutionContext(ctx) + expect(res.status).toBe(200) + }) + + it('rejects retrieval requests for CIDs found in the Bad Bits denylist', async () => { + await withBadBits(env, realIpfsRootCid) + + const fakeResponse = new Response('hello') + const mockRetrieveIpfsContent = vi.fn().mockResolvedValue({ + response: fakeResponse, + cacheMiss: true, + }) + + const ctx = createExecutionContext() + const req = withRequest(realDataSetId, realPieceId) + const res = await fetchAndRead(req, env, ctx, { + retrieveIpfsContent: mockRetrieveIpfsContent, + }) + await waitOnExecutionContext(ctx) + expect(res.status).toBe(404) + expect(await res.text()).toBe( + 'The requested CID was flagged by the Bad Bits Denylist at https://badbits.dwebops.pub', + ) + }) + + it('reject retrieval request if payer is sanctioned', async () => { + const dataSetId = '1003' + const pieceId = '2003' + const pieceCid = + 'baga6ea4seaqaleibb6ud4xeemuzzpsyhl6cxlsymsnfco4cdjka5uzajo2x4ipa' + const ipfsRootCid = 'bafk4test' + const serviceProviderId = 'service-provider-id' + const payerAddress = '0x999999cf1046e68e36E1aA2E0E07105eDDD1f08E' + + await withApprovedProvider(env, { + id: serviceProviderId, + serviceUrl: 'https://test-provider.xyz', + }) + + await withDataSetPiece(env, { + serviceProviderId, + payerAddress, + pieceCid, + ipfsRootCid, + dataSetId, + withCDN: true, + withIpfsIndexing: true, + pieceId, + }) + + await withWalletDetails( + env, + payerAddress, + true, // Sanctioned + ) + const ctx = createExecutionContext() + const req = withRequest(dataSetId, pieceId) + const res = await fetchAndRead(req, env, ctx) + await waitOnExecutionContext(ctx) + + assert.strictEqual(res.status, 403) + }) + it('does not log to retrieval_logs on method not allowed (405)', async () => { + const ctx = createExecutionContext() + const req = withRequest(realDataSetId, realPieceId, 'POST') + const res = await fetchAndRead(req, env, ctx) + await waitOnExecutionContext(ctx) + + expect(res.status).toBe(405) + expect(await res.text()).toBe('Method Not Allowed') + + const result = await env.DB.prepare( + `SELECT response_status FROM retrieval_logs WHERE data_set_id = ? ORDER BY id DESC LIMIT 1`, + ) + .bind(realDataSetId) + .first() + expect(result).toBeNull() + }) + + it('logs to retrieval_logs on unsupported service provider (404)', async () => { + const invalidPieceCid = 'bafiknvalidpieceid' + const pieceId = '9' + const invalidIpfsRootCid = 'bafkinvalidrootcid' + const dataSetId = '13' + const unsupportedServiceProviderId = 0 + + await env.DB.batch([ + env.DB.prepare( + 'INSERT INTO data_sets (id, service_provider_id, payer_address, with_cdn, with_ipfs_indexing) VALUES (?, ?, ?, ?, ?)', + ).bind( + dataSetId, + unsupportedServiceProviderId, + defaultPayerAddress, + true, + true, + ), + env.DB.prepare( + 'INSERT INTO pieces (id, data_set_id, cid, ipfs_root_cid) VALUES (?, ?, ?, ?)', + ).bind(pieceId, dataSetId, invalidPieceCid, invalidIpfsRootCid), + ]) + + const ctx = createExecutionContext() + const req = withRequest(dataSetId, pieceId) + const res = await fetchAndRead(req, env, ctx) + await waitOnExecutionContext(ctx) + + expect(res.status).toBe(404) + expect(await res.text()).toContain('No approved service provider found') + + const result = await env.DB.prepare( + 'SELECT * FROM retrieval_logs WHERE data_set_id IS NULL AND response_status = 404 and CACHE_MISS IS NULL and egress_bytes IS NULL', + ).first() + expect(result).toBeTruthy() + }) + it('does not log to retrieval_logs when slug encoding is invalid (400)', async () => { + const ctx = createExecutionContext() + const { count: countBefore } = await env.DB.prepare( + 'SELECT COUNT(*) AS count FROM retrieval_logs', + ).first() + + const req = new Request( + `http://${buildSlug(BigInt(realDataSetId), BigInt(realPieceId))}1.${DNS_ROOT.slice(1)}`, + ) + const res = await fetchAndRead(req, env, ctx) + await waitOnExecutionContext(ctx) + + expect(res.status).toBe(400) + expect(await res.text()).toContain('Invalid pieceId encoding in slug') + + const { count: countAfter } = await env.DB.prepare( + 'SELECT COUNT(*) AS count FROM retrieval_logs', + ).first() + expect(countAfter).toBe(countBefore) + }) + + it('converts CAR to RAW by default (no format parameter)', async () => { + // CONTENT_STORED_ON_CALIBRATION[1] is a single raw block holding a PNG. + const { dataSetId, pieceId } = CONTENT_STORED_ON_CALIBRATION[1] + const ctx = createExecutionContext() + + const req = withRequest( + String(dataSetId), + pieceId, + 'GET', + {}, + { format: null }, + ) + + const res = await fetchAndRead(req, env, ctx, { retrieveIpfsContent }) + await waitOnExecutionContext(ctx) + + expect(res.status).toBe(200) + + // Verify content-disposition is set to inline (not attachment) + expect(res.headers.get('content-disposition')).toBe('inline') + + // Verify we got RAW PNG data, not a CAR file + const content = await res.bytes() + expect(content.length).toBeGreaterThan(0) + + // PNG files start with the magic bytes: 89 50 4E 47 0D 0A 1A 0A + expect(content.slice(0, 4)).toEqual( + new Uint8Array([ + 0x89, + 0x50, // 'P' + 0x4e, // 'N' + 0x47, // 'G' + ]), + ) + }) +}) + +/** + * @param {string} dataSetId + * @param {string} pieceId + * @param {string} method + * @param {Object} headers + * @param {Object} options + * @param {string} options.subpath + * @returns {Request} + */ +function withRequest( + dataSetId, + pieceId, + method = 'GET', + headers = {}, + { subpath = '', format = 'car' } = {}, +) { + let url = `http://${buildSlug(BigInt(dataSetId), BigInt(pieceId))}.` + url += DNS_ROOT.slice(1) // remove the leading '.' + if (subpath) url += `${subpath}` + if (format) url += `?format=${format}` + + return new Request(url, { method, headers }) +} diff --git a/ipfs-retriever/test/store.test.js b/ipfs-retriever/test/store.test.js new file mode 100644 index 00000000..ffb885c1 --- /dev/null +++ b/ipfs-retriever/test/store.test.js @@ -0,0 +1,725 @@ +import { describe, it, beforeAll } from 'vitest' +import assert from 'node:assert/strict' +import { + getRetrievalCandidatesByWalletAndCid, + getRetrievalCandidatesByDataSetAndPiece, + getSlugForWalletAndCid, +} from '../lib/store.js' +import { env } from 'cloudflare:test' +import { withDataSetPiece, withApprovedProvider } from './test-data-builders.js' + +describe('getRetrievalCandidatesByWalletAndCid', () => { + const APPROVED_SERVICE_PROVIDER_ID = '20' + beforeAll(async () => { + await withApprovedProvider(env, { + id: APPROVED_SERVICE_PROVIDER_ID, + serviceUrl: 'https://approved-provider.xyz', + }) + }) + + it('returns service provider for valid ipfsRootCid', async () => { + const dataSetId = 'test-set-1' + const ipfsRootCid = 'bafk4test' + const payerAddress = '0x1234567890abcdef1234567890abcdef12345678' + + await env.DB.prepare( + 'INSERT INTO data_sets (id, service_provider_id, payer_address, with_cdn, with_ipfs_indexing) VALUES (?, ?, ?, ?, ?)', + ) + .bind(dataSetId, APPROVED_SERVICE_PROVIDER_ID, payerAddress, true, true) + .run() + await env.DB.prepare( + 'INSERT INTO pieces (id, data_set_id, cid, ipfs_root_cid) VALUES (?, ?, ?, ?)', + ) + .bind('piece-1', dataSetId, 'baga4piece', ipfsRootCid) + .run() + + const result = await getRetrievalCandidatesByWalletAndCid( + env, + payerAddress, + ipfsRootCid, + ) + assert.strictEqual( + result[0].serviceProviderId, + APPROVED_SERVICE_PROVIDER_ID, + ) + }) + + it('throws error if ipfsRootCid not found', async () => { + const payerAddress = '0x1234567890abcdef1234567890abcdef12345678' + await assert.rejects( + async () => + await getRetrievalCandidatesByWalletAndCid( + env, + payerAddress, + 'nonexistent-cid', + ), + /does not exist/, + ) + }) + + it('excludes deleted pieces', async () => { + const dataSetId = 'test-set-deleted' + const ipfsRootCid = 'bafk4deleted' + const payerAddress = '0x1234567890abcdef1234567890abcdef12345678' + + await env.DB.prepare( + 'INSERT INTO data_sets (id, service_provider_id, payer_address, with_cdn, with_ipfs_indexing) VALUES (?, ?, ?, ?, ?)', + ) + .bind(dataSetId, APPROVED_SERVICE_PROVIDER_ID, payerAddress, true, true) + .run() + await env.DB.prepare( + 'INSERT INTO pieces (id, data_set_id, cid, ipfs_root_cid, is_deleted) VALUES (?, ?, ?, ?, ?)', + ) + .bind('piece-deleted', dataSetId, 'baga4deleted', ipfsRootCid, true) + .run() + + await assert.rejects( + async () => + await getRetrievalCandidatesByWalletAndCid( + env, + payerAddress, + ipfsRootCid, + ), + /does not exist/, + ) + }) + + it('throws error if data_set_id exists but has no associated service provider', async () => { + const cid = 'cid-no-owner' + const dataSetId = 'data-set-no-owner' + const payerAddress = '0x1234567890abcdef1234567890abcdef12345678' + + await env.DB.prepare( + ` + INSERT INTO pieces (id, data_set_id, cid, ipfs_root_cid) + VALUES (?, ?, ?, ?) + `, + ) + .bind('piece-1', dataSetId, `bagatestpiece`, cid) + .run() + + await assert.rejects( + async () => + await getRetrievalCandidatesByWalletAndCid(env, payerAddress, cid), + /no associated service provider/, + ) + }) + + it('returns error if no payment rail', async () => { + const cid = 'cid-unapproved' + const dataSetId = 'data-set-unapproved' + const serviceProviderId = APPROVED_SERVICE_PROVIDER_ID + const payerAddress = '0xabcdef1234567890abcdef1234567890abcdef12' + + await env.DB.batch([ + env.DB.prepare( + 'INSERT INTO data_sets (id, service_provider_id, payer_address, with_cdn) VALUES (?, ?, ?, ?)', + ).bind( + dataSetId, + serviceProviderId, + payerAddress.replace('a', 'b'), + true, + ), + env.DB.prepare( + 'INSERT INTO pieces (id, data_set_id, cid, ipfs_root_cid) VALUES (?, ?, ?, ?)', + ).bind('piece-2', dataSetId, 'bagatest', cid), + ]) + + await assert.rejects( + async () => + await getRetrievalCandidatesByWalletAndCid(env, payerAddress, cid), + /There is no Filecoin Warm Storage Service deal for payer/, + ) + }) + + it('returns error if withCDN=false', async () => { + const cid = 'cid-unapproved' + const dataSetId = 'data-set-unapproved' + const serviceProviderId = APPROVED_SERVICE_PROVIDER_ID + const payerAddress = '0xabcdef1234567890abcdef1234567890abcdef12' + + await env.DB.batch([ + env.DB.prepare( + 'INSERT INTO data_sets (id, service_provider_id, payer_address, with_cdn) VALUES (?, ?, ?, ?)', + ).bind(dataSetId, serviceProviderId, payerAddress, false), + env.DB.prepare( + 'INSERT INTO pieces (id, data_set_id, cid, ipfs_root_cid) VALUES (?, ?, ?, ?)', + ).bind('piece-2', dataSetId, 'bagatest', cid), + ]) + + await assert.rejects( + async () => + await getRetrievalCandidatesByWalletAndCid(env, payerAddress, cid), + /withCDN=false/, + ) + }) + + it('returns serviceProviderId for approved service provider', async () => { + const cid = 'cid-approved' + const dataSetId = 'data-set-approved' + const payerAddress = '0xabcdef1234567890abcdef1234567890abcdef12' + + await env.DB.batch([ + env.DB.prepare( + 'INSERT INTO data_sets (id, service_provider_id, payer_address, with_cdn, with_ipfs_indexing) VALUES (?, ?, ?, ?, ?)', + ).bind(dataSetId, APPROVED_SERVICE_PROVIDER_ID, payerAddress, true, true), + env.DB.prepare( + 'INSERT INTO pieces (id, data_set_id, cid, ipfs_root_cid) VALUES (?, ?, ?, ?)', + ).bind('piece-3', dataSetId, 'bagatest', cid), + ]) + + const result = await getRetrievalCandidatesByWalletAndCid( + env, + payerAddress, + cid, + ) + + assert.strictEqual( + result[0].serviceProviderId, + APPROVED_SERVICE_PROVIDER_ID, + ) + }) + it('returns the service provider first in the ordering when multiple service providers share the same ipfsRootCid', async () => { + const dataSetId1 = 'data-set-a' + const dataSetId2 = 'data-set-b' + const ipfsRootCid = 'shared-ipfs-cid' + const payerAddress = '0x1234567890abcdef1234567890abcdef12345678' + const serviceProviderId1 = 'service-provider-a' + const serviceProviderId2 = 'service-provicer-b' + + await withApprovedProvider(env, { + id: serviceProviderId1, + }) + await withApprovedProvider(env, { + id: serviceProviderId2, + }) + + // Insert both owners into separate sets with the same ipfsRootCid + await env.DB.prepare( + 'INSERT INTO data_sets (id, service_provider_id, payer_address, with_cdn, with_ipfs_indexing) VALUES (?, ?, ?, ?, ?)', + ) + .bind(dataSetId1, serviceProviderId1, payerAddress, true, true) + .run() + + await env.DB.prepare( + 'INSERT INTO data_sets (id, service_provider_id, payer_address, with_cdn) VALUES (?, ?, ?, ?)', + ) + .bind(dataSetId2, serviceProviderId2, payerAddress, true) + .run() + + // Insert same ipfsRootCid for both sets + await env.DB.prepare( + 'INSERT INTO pieces (id, data_set_id, cid, ipfs_root_cid) VALUES (?, ?, ?, ?)', + ) + .bind('piece-a', dataSetId1, 'bagatest', ipfsRootCid) + .run() + + await env.DB.prepare( + 'INSERT INTO pieces (id, data_set_id, cid, ipfs_root_cid) VALUES (?, ?, ?, ?)', + ) + .bind('piece-b', dataSetId2, 'bagatest', ipfsRootCid) + .run() + + // Should return only the serviceProviderId1 which is the first in the ordering + const result = await getRetrievalCandidatesByWalletAndCid( + env, + payerAddress, + ipfsRootCid, + ) + assert.strictEqual(result[0].serviceProviderId, serviceProviderId1) + }) + + it('ignores owners that are not approved by Filecoin Warm Storage Service', async () => { + const dataSetId1 = '0' + const dataSetId2 = '1' + const ipfsRootCid = 'shared-piece-cid' + const payerAddress = '0x1234567890abcdef1234567890abcdef12345678' + const serviceProviderId1 = '0' + const serviceProviderId2 = '1' + + await withApprovedProvider(env, { + id: serviceProviderId1, + serviceUrl: 'https://pdp-provider-1.xyz', + }) + + // NOTE: the second provider is not registered as an approved provider + + // Important: we must insert the unapproved provider first! + await withDataSetPiece(env, { + payerAddress, + serviceProviderId: serviceProviderId2, + dataSetId: dataSetId2, + withCDN: true, + ipfsRootCid, + }) + + await withDataSetPiece(env, { + payerAddress, + serviceProviderId: serviceProviderId1, + dataSetId: dataSetId1, + withCDN: true, + ipfsRootCid, + }) + + // Should return service provider 1 because service provider 2 is not approved + const result = await getRetrievalCandidatesByWalletAndCid( + env, + payerAddress, + ipfsRootCid, + ) + assert.deepStrictEqual(result[0], { + dataSetId: dataSetId1, + pieceId: '0', + serviceProviderId: serviceProviderId1.toLowerCase(), + serviceUrl: 'https://pdp-provider-1.xyz', + ipfsRootCid, + }) + }) +}) + +describe('getRetrievalCandidatesByDataSetAndPiece', () => { + const APPROVED_SERVICE_PROVIDER_ID = '25' + beforeAll(async () => { + await withApprovedProvider(env, { + id: APPROVED_SERVICE_PROVIDER_ID, + serviceUrl: 'https://approved-provider-byids.xyz', + }) + }) + + it('returns service provider for valid dataSetId and pieceId', async () => { + const dataSetId = 'test-set-byids-1' + const pieceId = 'piece-byids-1' + const payerAddress = '0xabc123def456abc123def456abc123def456abc1' + + await withDataSetPiece(env, { + payerAddress, + serviceProviderId: APPROVED_SERVICE_PROVIDER_ID, + dataSetId, + pieceId, + withCDN: true, + withIpfsIndexing: true, + ipfsRootCid: 'bafkbyids1', + }) + + const result = await getRetrievalCandidatesByDataSetAndPiece( + env, + dataSetId, + pieceId, + ) + + assert.strictEqual( + result[0].serviceProviderId, + APPROVED_SERVICE_PROVIDER_ID, + ) + assert.strictEqual( + result[0].serviceUrl, + 'https://approved-provider-byids.xyz', + ) + assert.strictEqual(result[0].dataSetId, dataSetId) + assert.strictEqual(result[0].pieceId, pieceId) + }) + + it('throws error if pieceId does not exist in the data set', async () => { + const dataSetId = 'test-set-byids-2' + const pieceId = 'nonexistent-piece' + + await withDataSetPiece(env, { + payerAddress: '0xabc123def456abc123def456abc123def456abc2', + serviceProviderId: APPROVED_SERVICE_PROVIDER_ID, + dataSetId, + pieceId: 'existing-piece', + withCDN: true, + withIpfsIndexing: true, + ipfsRootCid: 'bafkbyids2', + }) + + await assert.rejects( + async () => + await getRetrievalCandidatesByDataSetAndPiece(env, dataSetId, pieceId), + /does not exist in data set/, + ) + }) + + it('throws error if pieceId exists but in different dataSetId', async () => { + const dataSetId1 = 'test-set-byids-3a' + const dataSetId2 = 'test-set-byids-3b' + const pieceId = 'piece-byids-3' + const payerAddress = '0xabc123def456abc123def456abc123def456abc3' + + await withDataSetPiece(env, { + payerAddress, + serviceProviderId: APPROVED_SERVICE_PROVIDER_ID, + dataSetId: dataSetId1, + pieceId, + withCDN: true, + withIpfsIndexing: true, + ipfsRootCid: 'bafkbyids3a', + }) + + await withDataSetPiece(env, { + payerAddress, + serviceProviderId: APPROVED_SERVICE_PROVIDER_ID, + dataSetId: dataSetId2, + pieceId: 'different-piece', + withCDN: true, + withIpfsIndexing: true, + ipfsRootCid: 'bafkbyids3b', + }) + + await assert.rejects( + async () => + await getRetrievalCandidatesByDataSetAndPiece(env, dataSetId2, pieceId), + /does not exist in data set/, + ) + }) + + it('throws error if withCDN=false', async () => { + const dataSetId = 'test-set-byids-4' + const pieceId = 'piece-byids-4' + + await withDataSetPiece(env, { + payerAddress: '0xabc123def456abc123def456abc123def456abc4', + serviceProviderId: APPROVED_SERVICE_PROVIDER_ID, + dataSetId, + pieceId, + withCDN: false, + withIpfsIndexing: true, + ipfsRootCid: 'bafkbyids4', + }) + + await assert.rejects( + async () => + await getRetrievalCandidatesByDataSetAndPiece(env, dataSetId, pieceId), + /withCDN=false/, + ) + }) + + it('throws error if withIpfsIndexing=false', async () => { + const dataSetId = 'test-set-byids-5' + const pieceId = 'piece-byids-5' + + await withDataSetPiece(env, { + payerAddress: '0xabc123def456abc123def456abc123def456abc5', + serviceProviderId: APPROVED_SERVICE_PROVIDER_ID, + dataSetId, + pieceId, + withCDN: true, + withIpfsIndexing: false, + ipfsRootCid: 'bafkbyids5', + }) + + await assert.rejects( + async () => + await getRetrievalCandidatesByDataSetAndPiece(env, dataSetId, pieceId), + /withIpfsIndexing=false/, + ) + }) + + it('throws error if payer is sanctioned', async () => { + const dataSetId = 'test-set-byids-6' + const pieceId = 'piece-byids-6' + const payerAddress = '0xabc123def456abc123def456abc123def456abc6' + + await env.DB.prepare( + 'INSERT INTO wallet_details (address, is_sanctioned) VALUES (?, ?)', + ) + .bind(payerAddress, true) + .run() + + await withDataSetPiece(env, { + payerAddress, + serviceProviderId: APPROVED_SERVICE_PROVIDER_ID, + dataSetId, + pieceId, + withCDN: true, + withIpfsIndexing: true, + ipfsRootCid: 'bafkbyids6', + }) + + await assert.rejects( + async () => + await getRetrievalCandidatesByDataSetAndPiece(env, dataSetId, pieceId), + /is sanctioned/, + ) + }) + + it('handles zero values for dataSetId and pieceId', async () => { + const dataSetId = '0' + const pieceId = '0' + + await withDataSetPiece(env, { + payerAddress: '0xabc123def456abc123def456abc123def456abc7', + serviceProviderId: APPROVED_SERVICE_PROVIDER_ID, + dataSetId, + pieceId, + withCDN: true, + withIpfsIndexing: true, + ipfsRootCid: 'bafkbyids7', + }) + + const result = await getRetrievalCandidatesByDataSetAndPiece( + env, + dataSetId, + pieceId, + ) + + assert.strictEqual(result[0].dataSetId, '0') + assert.strictEqual(result[0].pieceId, '0') + assert.strictEqual( + result[0].serviceProviderId, + APPROVED_SERVICE_PROVIDER_ID, + ) + }) + + it('returns every service provider serving the same content for the payer', async () => { + const payerAddress = '0xabc123def456abc123def456abc123def456abc8' + const ipfsRootCid = 'bafkbyids8shared' + const serviceProviderId1 = 'sp-byids-8a' + const serviceProviderId2 = 'sp-byids-8b' + + await withApprovedProvider(env, { + id: serviceProviderId1, + serviceUrl: 'https://sp8a.xyz', + }) + await withApprovedProvider(env, { + id: serviceProviderId2, + serviceUrl: 'https://sp8b.xyz', + }) + await withDataSetPiece(env, { + payerAddress, + serviceProviderId: serviceProviderId1, + dataSetId: 'ds-8a', + pieceId: 'piece-8a', + withCDN: true, + withIpfsIndexing: true, + ipfsRootCid, + }) + await withDataSetPiece(env, { + payerAddress, + serviceProviderId: serviceProviderId2, + dataSetId: 'ds-8b', + pieceId: 'piece-8b', + withCDN: true, + withIpfsIndexing: true, + ipfsRootCid, + }) + + // Looking up by one (data set, piece) returns the candidates for every + // service provider serving the same content for the payer. + const result = await getRetrievalCandidatesByDataSetAndPiece( + env, + 'ds-8a', + 'piece-8a', + ) + + assert.strictEqual(result.length, 2) + assert.deepStrictEqual( + result.map((c) => c.serviceProviderId).sort(), + [serviceProviderId1, serviceProviderId2].sort(), + ) + }) + + it('excludes soft-deleted service providers', async () => { + const payerAddress = '0xabc123def456abc123def456abc123def456abc9' + const ipfsRootCid = 'bafkbyids9deleted' + const serviceProviderId = 'sp-byids-9-deleted' + + await withApprovedProvider(env, { + id: serviceProviderId, + serviceUrl: 'https://sp9.xyz', + }) + await env.DB.prepare( + 'UPDATE service_providers SET is_deleted = TRUE WHERE id = ?', + ) + .bind(serviceProviderId) + .run() + await withDataSetPiece(env, { + payerAddress, + serviceProviderId, + dataSetId: 'ds-9', + pieceId: 'piece-9', + withCDN: true, + withIpfsIndexing: true, + ipfsRootCid, + }) + + await assert.rejects( + async () => + await getRetrievalCandidatesByDataSetAndPiece(env, 'ds-9', 'piece-9'), + /has no associated service provider/, + ) + }) + + it('throws 402 when the CDN egress quota is exhausted and enforcement is on', async () => { + await withApprovedProvider(env, { + id: 'sp-quota-cdn', + serviceUrl: 'https://qcdn.xyz', + }) + await withDataSetPiece(env, { + payerAddress: '0xabc123def456abc123def456abc123def456abca', + serviceProviderId: 'sp-quota-cdn', + dataSetId: 'ds-quota-cdn', + pieceId: 'piece-quota-cdn', + withCDN: true, + withIpfsIndexing: true, + ipfsRootCid: 'bafkbyidsquotacdn', + }) + await env.DB.prepare( + 'INSERT INTO data_set_egress_quotas (data_set_id, cdn_egress_quota, cache_miss_egress_quota) VALUES (?, ?, ?)', + ) + .bind('ds-quota-cdn', 0, 100) + .run() + + await assert.rejects( + async () => + await getRetrievalCandidatesByDataSetAndPiece( + env, + 'ds-quota-cdn', + 'piece-quota-cdn', + true, + ), + /CDN egress quota exhausted/, + ) + }) + + it('throws 402 when the cache-miss egress quota is exhausted and enforcement is on', async () => { + await withApprovedProvider(env, { + id: 'sp-quota-cm', + serviceUrl: 'https://qcm.xyz', + }) + await withDataSetPiece(env, { + payerAddress: '0xabc123def456abc123def456abc123def456abcb', + serviceProviderId: 'sp-quota-cm', + dataSetId: 'ds-quota-cm', + pieceId: 'piece-quota-cm', + withCDN: true, + withIpfsIndexing: true, + ipfsRootCid: 'bafkbyidsquotacm', + }) + await env.DB.prepare( + 'INSERT INTO data_set_egress_quotas (data_set_id, cdn_egress_quota, cache_miss_egress_quota) VALUES (?, ?, ?)', + ) + .bind('ds-quota-cm', 100, 0) + .run() + + await assert.rejects( + async () => + await getRetrievalCandidatesByDataSetAndPiece( + env, + 'ds-quota-cm', + 'piece-quota-cm', + true, + ), + /Cache miss egress quota exhausted/, + ) + }) + + it('returns candidates when enforcement is on and quota is sufficient', async () => { + await withApprovedProvider(env, { + id: 'sp-quota-ok', + serviceUrl: 'https://qok.xyz', + }) + await withDataSetPiece(env, { + payerAddress: '0xabc123def456abc123def456abc123def456abcc', + serviceProviderId: 'sp-quota-ok', + dataSetId: 'ds-quota-ok', + pieceId: 'piece-quota-ok', + withCDN: true, + withIpfsIndexing: true, + ipfsRootCid: 'bafkbyidsquotaok', + }) + await env.DB.prepare( + 'INSERT INTO data_set_egress_quotas (data_set_id, cdn_egress_quota, cache_miss_egress_quota) VALUES (?, ?, ?)', + ) + .bind('ds-quota-ok', 100, 100) + .run() + + const result = await getRetrievalCandidatesByDataSetAndPiece( + env, + 'ds-quota-ok', + 'piece-quota-ok', + true, + ) + + assert.strictEqual(result.length, 1) + assert.strictEqual(result[0].serviceProviderId, 'sp-quota-ok') + }) +}) + +describe('getSlugForWalletAndCid', () => { + const APPROVED_SERVICE_PROVIDER_ID = '30' + beforeAll(async () => { + await withApprovedProvider(env, { + id: APPROVED_SERVICE_PROVIDER_ID, + serviceUrl: 'https://approved-provider-slug.xyz', + }) + }) + + it('returns slug with version, dataSetId and pieceId encoded in base32', async () => { + const dataSetId = '12345' + const pieceId = '67890' + const ipfsRootCid = 'bafk4slugtest1' + const payerAddress = '0xabcdef1234567890abcdef1234567890abcdef34' + + await withDataSetPiece(env, { + payerAddress, + serviceProviderId: APPROVED_SERVICE_PROVIDER_ID, + dataSetId, + pieceId, + withCDN: true, + withIpfsIndexing: true, + ipfsRootCid, + }) + + const result = await getSlugForWalletAndCid(env, payerAddress, ipfsRootCid) + + // Slug format: version-base32(dataSetId)-base32(pieceId) + assert.strictEqual(result, '1-ga4q-aeete') + }) + + it('returns slug with zero-encoded values for dataSetId=0 and pieceId=0', async () => { + const dataSetId = '0' + const pieceId = '0' + const ipfsRootCid = 'bafk4slugtest2' + const payerAddress = '0xabcdef1234567890abcdef1234567890abcdef35' + + await withDataSetPiece(env, { + payerAddress, + serviceProviderId: APPROVED_SERVICE_PROVIDER_ID, + dataSetId, + pieceId, + withCDN: true, + withIpfsIndexing: true, + ipfsRootCid, + }) + + const result = await getSlugForWalletAndCid(env, payerAddress, ipfsRootCid) + + // For dataSetId=0 and pieceId=0, bigIntToBase32 returns '0' + assert.strictEqual(result, '1-0-0') + }) + + it('throws error for invalid payer address', async () => { + const dataSetId = '99999' + const pieceId = '88888' + const ipfsRootCid = 'bafk4slugtest3' + const validPayerAddress = '0xabcdef1234567890abcdef1234567890abcdef36' + const invalidPayerAddress = '0x0000000000000000000000000000000000000000' + + await withDataSetPiece(env, { + payerAddress: validPayerAddress, + serviceProviderId: APPROVED_SERVICE_PROVIDER_ID, + dataSetId, + pieceId, + withCDN: true, + withIpfsIndexing: true, + ipfsRootCid, + }) + + await assert.rejects( + async () => + await getSlugForWalletAndCid(env, invalidPayerAddress, ipfsRootCid), + /There is no Filecoin Warm Storage Service deal for payer/, + ) + }) +}) diff --git a/ipfs-retriever/test/test-data-builders.js b/ipfs-retriever/test/test-data-builders.js new file mode 100644 index 00000000..51eab806 --- /dev/null +++ b/ipfs-retriever/test/test-data-builders.js @@ -0,0 +1,133 @@ +import { getBadBitsEntry } from '@filbeam/retrieval' +import { CarWriter } from '@ipld/car' +import * as raw from 'multiformats/codecs/raw' +import { sha256 } from 'multiformats/hashes/sha2' +import { CID } from 'multiformats/cid' + +/** + * Builds an in-memory CAR holding a single raw block, so tests can exercise the + * CAR-to-raw conversion without a live service provider. The CAR is larger than + * the raw block it wraps (header + block framing). + * + * @param {Uint8Array} fileBytes - The raw content to wrap. + * @returns {Promise<{ carBytes: Uint8Array; rootCid: string }>} + */ +export async function buildRawBlockCar(fileBytes) { + const cid = CID.create(1, raw.code, await sha256.digest(fileBytes)) + const { writer, out } = CarWriter.create([cid]) + + /** @type {Uint8Array[]} */ + const chunks = [] + const collecting = (async () => { + for await (const chunk of out) chunks.push(chunk) + })() + await writer.put({ cid, bytes: fileBytes }) + await writer.close() + await collecting + + const carBytes = new Uint8Array(chunks.reduce((sum, c) => sum + c.length, 0)) + let offset = 0 + for (const chunk of chunks) { + carBytes.set(chunk, offset) + offset += chunk.length + } + + return { carBytes, rootCid: cid.toString() } +} + +/** + * @param {Env} env + * @param {Object} options + * @param {number} options.serviceProviderId + * @param {string} options.pieceCid + * @param {number} options.dataSetId + * @param {boolean} options.withCDN + * @param {string} options.payerAddress + * @param {string} options.pieceId + */ +export async function withDataSetPiece( + env, + { + serviceProviderId = 0, + payerAddress = '0x1234567890abcdef1234567890abcdef12345608', + pieceCid = 'bagaTEST', + ipfsRootCid = 'bafk4test', + dataSetId = 0, + withCDN = true, + withIpfsIndexing = true, + pieceId = 0, + } = {}, +) { + await env.DB.batch([ + env.DB.prepare( + ` + INSERT INTO data_sets (id, service_provider_id, payer_address, with_cdn, with_ipfs_indexing) + VALUES (?, ?, ?, ?, ?) + `, + ).bind( + String(dataSetId), + String(serviceProviderId), + payerAddress.toLowerCase(), + withCDN, + withIpfsIndexing, + ), + + env.DB.prepare( + ` + INSERT INTO pieces (id, data_set_id, cid, ipfs_root_cid) + VALUES (?, ?, ?, ?) + `, + ).bind(String(pieceId), String(dataSetId), pieceCid, ipfsRootCid ?? null), + ]) +} + +/** + * @param {Env} env + * @param {Object} options + * @param {number} id + * @param {string} [options.serviceUrl] + */ +export async function withApprovedProvider( + env, + { id, serviceUrl = 'https://pdp.xyz/' } = {}, +) { + await env.DB.prepare( + ` + INSERT INTO service_providers (id, service_url) + VALUES (?, ?) + `, + ) + .bind(String(id), serviceUrl) + .run() +} + +/** + * @param {Env} env + * @param {...string} cids + */ +export async function withBadBits(env, ...cids) { + await Promise.all( + cids.map(async (cid) => + env.BAD_BITS_KV.put(`bad-bits:${await getBadBitsEntry(cid)}`, 'true'), + ), + ) +} + +/** + * Inserts an address into the database with an optional sanctioned flag. + * + * @param {Env} env + * @param {string} address + * @param {boolean} [isSanctioned=false] Default is `false` + * @returns {Promise} + */ +export async function withWalletDetails(env, address, isSanctioned = false) { + await env.DB.prepare( + ` + INSERT INTO wallet_details (address, is_sanctioned) + VALUES (?, ?) + `, + ) + .bind(address.toLowerCase(), isSanctioned ? 1 : 0) + .run() +} diff --git a/ipfs-retriever/test/test-data.js b/ipfs-retriever/test/test-data.js new file mode 100644 index 00000000..548fb57f --- /dev/null +++ b/ipfs-retriever/test/test-data.js @@ -0,0 +1,31 @@ +/** + * @type {{ + * serviceProviderId: string + * serviceUrl: string + * pieceCid: string + * ipfsRootCid: string + * dataSetId: number + * pieceId: string + * }[]} + */ +export const CONTENT_STORED_ON_CALIBRATION = [ + { + // This Piece must have IPFS RootCID set and IPFS Indexing enabled at the dataset level + serviceProviderId: '2', + serviceUrl: 'https://calib2.ezpdpz.net/', + pieceCid: + 'bafkzcibdzabqtx4ovk72zspicej5vmbjse2237cfzduljnevpmd4kfvccb5h44y4', + ipfsRootCid: 'bafkreiheygfzn22dfeos3xoay5cxnfb464znd2rszieyzcinlsgu2z7kau', + dataSetId: 14578, + pieceId: '0', + }, + { + serviceProviderId: '4', + serviceUrl: 'https://caliberation-pdp.infrafolio.com/', + pieceCid: + 'bafkzcibd7r7avok5z3tdn4uq6shuqghxm75e5jgirvdzsdmrkly2u5dldodic4jb', + ipfsRootCid: 'bafkreigo55ody3xm4g6mbkitgdytcshanhluzywpi25f3qalfypc7bpna4', + dataSetId: 14577, + pieceId: '1', + }, +] diff --git a/ipfs-retriever/tsconfig.json b/ipfs-retriever/tsconfig.json new file mode 100644 index 00000000..e9257aa3 --- /dev/null +++ b/ipfs-retriever/tsconfig.json @@ -0,0 +1,15 @@ +{ + "$schema": "https://json.schemastore.org/tsconfig", + "extends": "../tsconfig.base.json", + "compilerOptions": { + "composite": true, + "outDir": "dist" + }, + "include": ["**/*.ts", "**/*.js", "src/**/*.json"], + "exclude": ["dist", "test"], + "references": [ + { + "path": "../retrieval/tsconfig.json" + } + ] +} diff --git a/ipfs-retriever/vitest.config.js b/ipfs-retriever/vitest.config.js new file mode 100644 index 00000000..4b7a065e --- /dev/null +++ b/ipfs-retriever/vitest.config.js @@ -0,0 +1,30 @@ +import path from 'node:path' +import { + defineWorkersProject, + readD1Migrations, +} from '@cloudflare/vitest-pool-workers/config' + +export default defineWorkersProject(async () => { + // Read all migrations in the `migrations` directory + const migrationsPath = path.join(__dirname, '../db/migrations') + const migrations = await readD1Migrations(migrationsPath) + return { + test: { + setupFiles: ['./test/apply-migrations.js'], + poolOptions: { + workers: { + singleWorker: true, + wrangler: { + configPath: './wrangler.toml', + environment: 'dev', + }, + miniflare: { + // Add a test-only binding for migrations, so we can apply them in a + // setup file + bindings: { TEST_MIGRATIONS: migrations }, + }, + }, + }, + }, + } +}) diff --git a/ipfs-retriever/worker-configuration.d.ts b/ipfs-retriever/worker-configuration.d.ts new file mode 100644 index 00000000..fc0cfcc4 --- /dev/null +++ b/ipfs-retriever/worker-configuration.d.ts @@ -0,0 +1,10905 @@ +/* eslint-disable */ +// Generated by Wrangler by running `wrangler types` (hash: 4d117a9bf0aadb02f8235710f1b9e1a0) +// Runtime types generated with workerd@1.20260124.0 2024-12-05 enable_request_signal,nodejs_compat +declare namespace Cloudflare { + interface GlobalProps { + mainModule: typeof import("./bin/ipfs-retriever"); + } + interface DevEnv { + BAD_BITS_KV: KVNamespace; + DB: D1Database; + ENVIRONMENT: "dev"; + ORIGIN_CACHE_TTL: 86400; + CLIENT_CACHE_TTL: 31536000; + DNS_ROOT: ".localhost"; + ENFORCE_EGRESS_QUOTA: false; + BOT_TOKENS: string; + } + interface CalibrationEnv { + BAD_BITS_KV: KVNamespace; + DB: D1Database; + ENVIRONMENT: "calibration"; + ORIGIN_CACHE_TTL: 86400; + CLIENT_CACHE_TTL: 31536000; + DNS_ROOT: ".ipfs.calibration.filbeam.io"; + ENFORCE_EGRESS_QUOTA: true; + BOT_TOKENS: string; + } + interface MainnetEnv { + BAD_BITS_KV: KVNamespace; + DB: D1Database; + ENVIRONMENT: "mainnet"; + ORIGIN_CACHE_TTL: 86400; + CLIENT_CACHE_TTL: 31536000; + DNS_ROOT: ".ipfs.filbeam.io"; + ENFORCE_EGRESS_QUOTA: true; + BOT_TOKENS: string; + } + interface Env { + BOT_TOKENS: string; + BAD_BITS_KV: KVNamespace; + DB: D1Database; + ENVIRONMENT?: "dev" | "calibration" | "mainnet"; + ORIGIN_CACHE_TTL: 86400; + CLIENT_CACHE_TTL: 31536000; + DNS_ROOT: ".localhost" | ".ipfs.calibration.filbeam.io" | ".ipfs.filbeam.io"; + ENFORCE_EGRESS_QUOTA: false | true; + } +} +interface Env extends Cloudflare.Env {} + +// Begin runtime types +/*! ***************************************************************************** +Copyright (c) Cloudflare. All rights reserved. +Copyright (c) Microsoft Corporation. All rights reserved. + +Licensed under the Apache License, Version 2.0 (the "License"); you may not use +this file except in compliance with the License. You may obtain a copy of the +License at http://www.apache.org/licenses/LICENSE-2.0 +THIS CODE IS PROVIDED ON AN *AS IS* BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY +KIND, EITHER EXPRESS OR IMPLIED, INCLUDING WITHOUT LIMITATION ANY IMPLIED +WARRANTIES OR CONDITIONS OF TITLE, FITNESS FOR A PARTICULAR PURPOSE, +MERCHANTABLITY OR NON-INFRINGEMENT. +See the Apache Version 2.0 License for specific language governing permissions +and limitations under the License. +***************************************************************************** */ +/* eslint-disable */ +// noinspection JSUnusedGlobalSymbols +declare var onmessage: never; +/** + * The **`DOMException`** interface represents an abnormal event (called an **exception**) that occurs as a result of calling a method or accessing a property of a web API. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/DOMException) + */ +declare class DOMException extends Error { + constructor(message?: string, name?: string); + /** + * The **`message`** read-only property of the a message or description associated with the given error name. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/DOMException/message) + */ + readonly message: string; + /** + * The **`name`** read-only property of the one of the strings associated with an error name. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/DOMException/name) + */ + readonly name: string; + /** + * The **`code`** read-only property of the DOMException interface returns one of the legacy error code constants, or `0` if none match. + * @deprecated + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/DOMException/code) + */ + readonly code: number; + static readonly INDEX_SIZE_ERR: number; + static readonly DOMSTRING_SIZE_ERR: number; + static readonly HIERARCHY_REQUEST_ERR: number; + static readonly WRONG_DOCUMENT_ERR: number; + static readonly INVALID_CHARACTER_ERR: number; + static readonly NO_DATA_ALLOWED_ERR: number; + static readonly NO_MODIFICATION_ALLOWED_ERR: number; + static readonly NOT_FOUND_ERR: number; + static readonly NOT_SUPPORTED_ERR: number; + static readonly INUSE_ATTRIBUTE_ERR: number; + static readonly INVALID_STATE_ERR: number; + static readonly SYNTAX_ERR: number; + static readonly INVALID_MODIFICATION_ERR: number; + static readonly NAMESPACE_ERR: number; + static readonly INVALID_ACCESS_ERR: number; + static readonly VALIDATION_ERR: number; + static readonly TYPE_MISMATCH_ERR: number; + static readonly SECURITY_ERR: number; + static readonly NETWORK_ERR: number; + static readonly ABORT_ERR: number; + static readonly URL_MISMATCH_ERR: number; + static readonly QUOTA_EXCEEDED_ERR: number; + static readonly TIMEOUT_ERR: number; + static readonly INVALID_NODE_TYPE_ERR: number; + static readonly DATA_CLONE_ERR: number; + get stack(): any; + set stack(value: any); +} +type WorkerGlobalScopeEventMap = { + fetch: FetchEvent; + scheduled: ScheduledEvent; + queue: QueueEvent; + unhandledrejection: PromiseRejectionEvent; + rejectionhandled: PromiseRejectionEvent; +}; +declare abstract class WorkerGlobalScope extends EventTarget { + EventTarget: typeof EventTarget; +} +/* The **`console`** object provides access to the debugging console (e.g., the Web console in Firefox). * + * The **`console`** object provides access to the debugging console (e.g., the Web console in Firefox). + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/console) + */ +interface Console { + "assert"(condition?: boolean, ...data: any[]): void; + /** + * The **`console.clear()`** static method clears the console if possible. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/console/clear_static) + */ + clear(): void; + /** + * The **`console.count()`** static method logs the number of times that this particular call to `count()` has been called. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/console/count_static) + */ + count(label?: string): void; + /** + * The **`console.countReset()`** static method resets counter used with console/count_static. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/console/countReset_static) + */ + countReset(label?: string): void; + /** + * The **`console.debug()`** static method outputs a message to the console at the 'debug' log level. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/console/debug_static) + */ + debug(...data: any[]): void; + /** + * The **`console.dir()`** static method displays a list of the properties of the specified JavaScript object. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/console/dir_static) + */ + dir(item?: any, options?: any): void; + /** + * The **`console.dirxml()`** static method displays an interactive tree of the descendant elements of the specified XML/HTML element. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/console/dirxml_static) + */ + dirxml(...data: any[]): void; + /** + * The **`console.error()`** static method outputs a message to the console at the 'error' log level. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/console/error_static) + */ + error(...data: any[]): void; + /** + * The **`console.group()`** static method creates a new inline group in the Web console log, causing any subsequent console messages to be indented by an additional level, until console/groupEnd_static is called. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/console/group_static) + */ + group(...data: any[]): void; + /** + * The **`console.groupCollapsed()`** static method creates a new inline group in the console. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/console/groupCollapsed_static) + */ + groupCollapsed(...data: any[]): void; + /** + * The **`console.groupEnd()`** static method exits the current inline group in the console. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/console/groupEnd_static) + */ + groupEnd(): void; + /** + * The **`console.info()`** static method outputs a message to the console at the 'info' log level. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/console/info_static) + */ + info(...data: any[]): void; + /** + * The **`console.log()`** static method outputs a message to the console. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/console/log_static) + */ + log(...data: any[]): void; + /** + * The **`console.table()`** static method displays tabular data as a table. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/console/table_static) + */ + table(tabularData?: any, properties?: string[]): void; + /** + * The **`console.time()`** static method starts a timer you can use to track how long an operation takes. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/console/time_static) + */ + time(label?: string): void; + /** + * The **`console.timeEnd()`** static method stops a timer that was previously started by calling console/time_static. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/console/timeEnd_static) + */ + timeEnd(label?: string): void; + /** + * The **`console.timeLog()`** static method logs the current value of a timer that was previously started by calling console/time_static. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/console/timeLog_static) + */ + timeLog(label?: string, ...data: any[]): void; + timeStamp(label?: string): void; + /** + * The **`console.trace()`** static method outputs a stack trace to the console. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/console/trace_static) + */ + trace(...data: any[]): void; + /** + * The **`console.warn()`** static method outputs a warning message to the console at the 'warning' log level. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/console/warn_static) + */ + warn(...data: any[]): void; +} +declare const console: Console; +type BufferSource = ArrayBufferView | ArrayBuffer; +type TypedArray = Int8Array | Uint8Array | Uint8ClampedArray | Int16Array | Uint16Array | Int32Array | Uint32Array | Float32Array | Float64Array | BigInt64Array | BigUint64Array; +declare namespace WebAssembly { + class CompileError extends Error { + constructor(message?: string); + } + class RuntimeError extends Error { + constructor(message?: string); + } + type ValueType = "anyfunc" | "externref" | "f32" | "f64" | "i32" | "i64" | "v128"; + interface GlobalDescriptor { + value: ValueType; + mutable?: boolean; + } + class Global { + constructor(descriptor: GlobalDescriptor, value?: any); + value: any; + valueOf(): any; + } + type ImportValue = ExportValue | number; + type ModuleImports = Record; + type Imports = Record; + type ExportValue = Function | Global | Memory | Table; + type Exports = Record; + class Instance { + constructor(module: Module, imports?: Imports); + readonly exports: Exports; + } + interface MemoryDescriptor { + initial: number; + maximum?: number; + shared?: boolean; + } + class Memory { + constructor(descriptor: MemoryDescriptor); + readonly buffer: ArrayBuffer; + grow(delta: number): number; + } + type ImportExportKind = "function" | "global" | "memory" | "table"; + interface ModuleExportDescriptor { + kind: ImportExportKind; + name: string; + } + interface ModuleImportDescriptor { + kind: ImportExportKind; + module: string; + name: string; + } + abstract class Module { + static customSections(module: Module, sectionName: string): ArrayBuffer[]; + static exports(module: Module): ModuleExportDescriptor[]; + static imports(module: Module): ModuleImportDescriptor[]; + } + type TableKind = "anyfunc" | "externref"; + interface TableDescriptor { + element: TableKind; + initial: number; + maximum?: number; + } + class Table { + constructor(descriptor: TableDescriptor, value?: any); + readonly length: number; + get(index: number): any; + grow(delta: number, value?: any): number; + set(index: number, value?: any): void; + } + function instantiate(module: Module, imports?: Imports): Promise; + function validate(bytes: BufferSource): boolean; +} +/** + * The **`ServiceWorkerGlobalScope`** interface of the Service Worker API represents the global execution context of a service worker. + * Available only in secure contexts. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/ServiceWorkerGlobalScope) + */ +interface ServiceWorkerGlobalScope extends WorkerGlobalScope { + DOMException: typeof DOMException; + WorkerGlobalScope: typeof WorkerGlobalScope; + btoa(data: string): string; + atob(data: string): string; + setTimeout(callback: (...args: any[]) => void, msDelay?: number): number; + setTimeout(callback: (...args: Args) => void, msDelay?: number, ...args: Args): number; + clearTimeout(timeoutId: number | null): void; + setInterval(callback: (...args: any[]) => void, msDelay?: number): number; + setInterval(callback: (...args: Args) => void, msDelay?: number, ...args: Args): number; + clearInterval(timeoutId: number | null): void; + queueMicrotask(task: Function): void; + structuredClone(value: T, options?: StructuredSerializeOptions): T; + reportError(error: any): void; + fetch(input: RequestInfo | URL, init?: RequestInit): Promise; + self: ServiceWorkerGlobalScope; + crypto: Crypto; + caches: CacheStorage; + scheduler: Scheduler; + performance: Performance; + Cloudflare: Cloudflare; + readonly origin: string; + Event: typeof Event; + ExtendableEvent: typeof ExtendableEvent; + CustomEvent: typeof CustomEvent; + PromiseRejectionEvent: typeof PromiseRejectionEvent; + FetchEvent: typeof FetchEvent; + TailEvent: typeof TailEvent; + TraceEvent: typeof TailEvent; + ScheduledEvent: typeof ScheduledEvent; + MessageEvent: typeof MessageEvent; + CloseEvent: typeof CloseEvent; + ReadableStreamDefaultReader: typeof ReadableStreamDefaultReader; + ReadableStreamBYOBReader: typeof ReadableStreamBYOBReader; + ReadableStream: typeof ReadableStream; + WritableStream: typeof WritableStream; + WritableStreamDefaultWriter: typeof WritableStreamDefaultWriter; + TransformStream: typeof TransformStream; + ByteLengthQueuingStrategy: typeof ByteLengthQueuingStrategy; + CountQueuingStrategy: typeof CountQueuingStrategy; + ErrorEvent: typeof ErrorEvent; + EventSource: typeof EventSource; + ReadableStreamBYOBRequest: typeof ReadableStreamBYOBRequest; + ReadableStreamDefaultController: typeof ReadableStreamDefaultController; + ReadableByteStreamController: typeof ReadableByteStreamController; + WritableStreamDefaultController: typeof WritableStreamDefaultController; + TransformStreamDefaultController: typeof TransformStreamDefaultController; + CompressionStream: typeof CompressionStream; + DecompressionStream: typeof DecompressionStream; + TextEncoderStream: typeof TextEncoderStream; + TextDecoderStream: typeof TextDecoderStream; + Headers: typeof Headers; + Body: typeof Body; + Request: typeof Request; + Response: typeof Response; + WebSocket: typeof WebSocket; + WebSocketPair: typeof WebSocketPair; + WebSocketRequestResponsePair: typeof WebSocketRequestResponsePair; + AbortController: typeof AbortController; + AbortSignal: typeof AbortSignal; + TextDecoder: typeof TextDecoder; + TextEncoder: typeof TextEncoder; + navigator: Navigator; + Navigator: typeof Navigator; + URL: typeof URL; + URLSearchParams: typeof URLSearchParams; + URLPattern: typeof URLPattern; + Blob: typeof Blob; + File: typeof File; + FormData: typeof FormData; + Crypto: typeof Crypto; + SubtleCrypto: typeof SubtleCrypto; + CryptoKey: typeof CryptoKey; + CacheStorage: typeof CacheStorage; + Cache: typeof Cache; + FixedLengthStream: typeof FixedLengthStream; + IdentityTransformStream: typeof IdentityTransformStream; + HTMLRewriter: typeof HTMLRewriter; +} +declare function addEventListener(type: Type, handler: EventListenerOrEventListenerObject, options?: EventTargetAddEventListenerOptions | boolean): void; +declare function removeEventListener(type: Type, handler: EventListenerOrEventListenerObject, options?: EventTargetEventListenerOptions | boolean): void; +/** + * The **`dispatchEvent()`** method of the EventTarget sends an Event to the object, (synchronously) invoking the affected event listeners in the appropriate order. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/EventTarget/dispatchEvent) + */ +declare function dispatchEvent(event: WorkerGlobalScopeEventMap[keyof WorkerGlobalScopeEventMap]): boolean; +/* [MDN Reference](https://developer.mozilla.org/docs/Web/API/Window/btoa) */ +declare function btoa(data: string): string; +/* [MDN Reference](https://developer.mozilla.org/docs/Web/API/Window/atob) */ +declare function atob(data: string): string; +/* [MDN Reference](https://developer.mozilla.org/docs/Web/API/Window/setTimeout) */ +declare function setTimeout(callback: (...args: any[]) => void, msDelay?: number): number; +/* [MDN Reference](https://developer.mozilla.org/docs/Web/API/Window/setTimeout) */ +declare function setTimeout(callback: (...args: Args) => void, msDelay?: number, ...args: Args): number; +/* [MDN Reference](https://developer.mozilla.org/docs/Web/API/Window/clearTimeout) */ +declare function clearTimeout(timeoutId: number | null): void; +/* [MDN Reference](https://developer.mozilla.org/docs/Web/API/Window/setInterval) */ +declare function setInterval(callback: (...args: any[]) => void, msDelay?: number): number; +/* [MDN Reference](https://developer.mozilla.org/docs/Web/API/Window/setInterval) */ +declare function setInterval(callback: (...args: Args) => void, msDelay?: number, ...args: Args): number; +/* [MDN Reference](https://developer.mozilla.org/docs/Web/API/Window/clearInterval) */ +declare function clearInterval(timeoutId: number | null): void; +/* [MDN Reference](https://developer.mozilla.org/docs/Web/API/Window/queueMicrotask) */ +declare function queueMicrotask(task: Function): void; +/* [MDN Reference](https://developer.mozilla.org/docs/Web/API/Window/structuredClone) */ +declare function structuredClone(value: T, options?: StructuredSerializeOptions): T; +/* [MDN Reference](https://developer.mozilla.org/docs/Web/API/Window/reportError) */ +declare function reportError(error: any): void; +/* [MDN Reference](https://developer.mozilla.org/docs/Web/API/Window/fetch) */ +declare function fetch(input: RequestInfo | URL, init?: RequestInit): Promise; +declare const self: ServiceWorkerGlobalScope; +/** +* The Web Crypto API provides a set of low-level functions for common cryptographic tasks. +* The Workers runtime implements the full surface of this API, but with some differences in +* the [supported algorithms](https://developers.cloudflare.com/workers/runtime-apis/web-crypto/#supported-algorithms) +* compared to those implemented in most browsers. +* +* [Cloudflare Docs Reference](https://developers.cloudflare.com/workers/runtime-apis/web-crypto/) +*/ +declare const crypto: Crypto; +/** +* The Cache API allows fine grained control of reading and writing from the Cloudflare global network cache. +* +* [Cloudflare Docs Reference](https://developers.cloudflare.com/workers/runtime-apis/cache/) +*/ +declare const caches: CacheStorage; +declare const scheduler: Scheduler; +/** +* The Workers runtime supports a subset of the Performance API, used to measure timing and performance, +* as well as timing of subrequests and other operations. +* +* [Cloudflare Docs Reference](https://developers.cloudflare.com/workers/runtime-apis/performance/) +*/ +declare const performance: Performance; +declare const Cloudflare: Cloudflare; +declare const origin: string; +declare const navigator: Navigator; +interface TestController { +} +interface ExecutionContext { + waitUntil(promise: Promise): void; + passThroughOnException(): void; + readonly props: Props; +} +type ExportedHandlerFetchHandler = (request: Request>, env: Env, ctx: ExecutionContext) => Response | Promise; +type ExportedHandlerTailHandler = (events: TraceItem[], env: Env, ctx: ExecutionContext) => void | Promise; +type ExportedHandlerTraceHandler = (traces: TraceItem[], env: Env, ctx: ExecutionContext) => void | Promise; +type ExportedHandlerTailStreamHandler = (event: TailStream.TailEvent, env: Env, ctx: ExecutionContext) => TailStream.TailEventHandlerType | Promise; +type ExportedHandlerScheduledHandler = (controller: ScheduledController, env: Env, ctx: ExecutionContext) => void | Promise; +type ExportedHandlerQueueHandler = (batch: MessageBatch, env: Env, ctx: ExecutionContext) => void | Promise; +type ExportedHandlerTestHandler = (controller: TestController, env: Env, ctx: ExecutionContext) => void | Promise; +interface ExportedHandler { + fetch?: ExportedHandlerFetchHandler; + tail?: ExportedHandlerTailHandler; + trace?: ExportedHandlerTraceHandler; + tailStream?: ExportedHandlerTailStreamHandler; + scheduled?: ExportedHandlerScheduledHandler; + test?: ExportedHandlerTestHandler; + email?: EmailExportedHandler; + queue?: ExportedHandlerQueueHandler; +} +interface StructuredSerializeOptions { + transfer?: any[]; +} +declare abstract class Navigator { + sendBeacon(url: string, body?: BodyInit): boolean; + readonly userAgent: string; + readonly hardwareConcurrency: number; +} +interface AlarmInvocationInfo { + readonly isRetry: boolean; + readonly retryCount: number; +} +interface Cloudflare { + readonly compatibilityFlags: Record; +} +interface DurableObject { + fetch(request: Request): Response | Promise; + alarm?(alarmInfo?: AlarmInvocationInfo): void | Promise; + webSocketMessage?(ws: WebSocket, message: string | ArrayBuffer): void | Promise; + webSocketClose?(ws: WebSocket, code: number, reason: string, wasClean: boolean): void | Promise; + webSocketError?(ws: WebSocket, error: unknown): void | Promise; +} +type DurableObjectStub = Fetcher & { + readonly id: DurableObjectId; + readonly name?: string; +}; +interface DurableObjectId { + toString(): string; + equals(other: DurableObjectId): boolean; + readonly name?: string; +} +declare abstract class DurableObjectNamespace { + newUniqueId(options?: DurableObjectNamespaceNewUniqueIdOptions): DurableObjectId; + idFromName(name: string): DurableObjectId; + idFromString(id: string): DurableObjectId; + get(id: DurableObjectId, options?: DurableObjectNamespaceGetDurableObjectOptions): DurableObjectStub; + getByName(name: string, options?: DurableObjectNamespaceGetDurableObjectOptions): DurableObjectStub; + jurisdiction(jurisdiction: DurableObjectJurisdiction): DurableObjectNamespace; +} +type DurableObjectJurisdiction = "eu" | "fedramp" | "fedramp-high"; +interface DurableObjectNamespaceNewUniqueIdOptions { + jurisdiction?: DurableObjectJurisdiction; +} +type DurableObjectLocationHint = "wnam" | "enam" | "sam" | "weur" | "eeur" | "apac" | "oc" | "afr" | "me"; +type DurableObjectRoutingMode = "primary-only"; +interface DurableObjectNamespaceGetDurableObjectOptions { + locationHint?: DurableObjectLocationHint; + routingMode?: DurableObjectRoutingMode; +} +interface DurableObjectClass<_T extends Rpc.DurableObjectBranded | undefined = undefined> { +} +interface DurableObjectState { + waitUntil(promise: Promise): void; + readonly props: Props; + readonly id: DurableObjectId; + readonly storage: DurableObjectStorage; + container?: Container; + blockConcurrencyWhile(callback: () => Promise): Promise; + acceptWebSocket(ws: WebSocket, tags?: string[]): void; + getWebSockets(tag?: string): WebSocket[]; + setWebSocketAutoResponse(maybeReqResp?: WebSocketRequestResponsePair): void; + getWebSocketAutoResponse(): WebSocketRequestResponsePair | null; + getWebSocketAutoResponseTimestamp(ws: WebSocket): Date | null; + setHibernatableWebSocketEventTimeout(timeoutMs?: number): void; + getHibernatableWebSocketEventTimeout(): number | null; + getTags(ws: WebSocket): string[]; + abort(reason?: string): void; +} +interface DurableObjectTransaction { + get(key: string, options?: DurableObjectGetOptions): Promise; + get(keys: string[], options?: DurableObjectGetOptions): Promise>; + list(options?: DurableObjectListOptions): Promise>; + put(key: string, value: T, options?: DurableObjectPutOptions): Promise; + put(entries: Record, options?: DurableObjectPutOptions): Promise; + delete(key: string, options?: DurableObjectPutOptions): Promise; + delete(keys: string[], options?: DurableObjectPutOptions): Promise; + rollback(): void; + getAlarm(options?: DurableObjectGetAlarmOptions): Promise; + setAlarm(scheduledTime: number | Date, options?: DurableObjectSetAlarmOptions): Promise; + deleteAlarm(options?: DurableObjectSetAlarmOptions): Promise; +} +interface DurableObjectStorage { + get(key: string, options?: DurableObjectGetOptions): Promise; + get(keys: string[], options?: DurableObjectGetOptions): Promise>; + list(options?: DurableObjectListOptions): Promise>; + put(key: string, value: T, options?: DurableObjectPutOptions): Promise; + put(entries: Record, options?: DurableObjectPutOptions): Promise; + delete(key: string, options?: DurableObjectPutOptions): Promise; + delete(keys: string[], options?: DurableObjectPutOptions): Promise; + deleteAll(options?: DurableObjectPutOptions): Promise; + transaction(closure: (txn: DurableObjectTransaction) => Promise): Promise; + getAlarm(options?: DurableObjectGetAlarmOptions): Promise; + setAlarm(scheduledTime: number | Date, options?: DurableObjectSetAlarmOptions): Promise; + deleteAlarm(options?: DurableObjectSetAlarmOptions): Promise; + sync(): Promise; + sql: SqlStorage; + kv: SyncKvStorage; + transactionSync(closure: () => T): T; + getCurrentBookmark(): Promise; + getBookmarkForTime(timestamp: number | Date): Promise; + onNextSessionRestoreBookmark(bookmark: string): Promise; +} +interface DurableObjectListOptions { + start?: string; + startAfter?: string; + end?: string; + prefix?: string; + reverse?: boolean; + limit?: number; + allowConcurrency?: boolean; + noCache?: boolean; +} +interface DurableObjectGetOptions { + allowConcurrency?: boolean; + noCache?: boolean; +} +interface DurableObjectGetAlarmOptions { + allowConcurrency?: boolean; +} +interface DurableObjectPutOptions { + allowConcurrency?: boolean; + allowUnconfirmed?: boolean; + noCache?: boolean; +} +interface DurableObjectSetAlarmOptions { + allowConcurrency?: boolean; + allowUnconfirmed?: boolean; +} +declare class WebSocketRequestResponsePair { + constructor(request: string, response: string); + get request(): string; + get response(): string; +} +interface AnalyticsEngineDataset { + writeDataPoint(event?: AnalyticsEngineDataPoint): void; +} +interface AnalyticsEngineDataPoint { + indexes?: ((ArrayBuffer | string) | null)[]; + doubles?: number[]; + blobs?: ((ArrayBuffer | string) | null)[]; +} +/** + * The **`Event`** interface represents an event which takes place on an `EventTarget`. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Event) + */ +declare class Event { + constructor(type: string, init?: EventInit); + /** + * The **`type`** read-only property of the Event interface returns a string containing the event's type. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Event/type) + */ + get type(): string; + /** + * The **`eventPhase`** read-only property of the being evaluated. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Event/eventPhase) + */ + get eventPhase(): number; + /** + * The read-only **`composed`** property of the or not the event will propagate across the shadow DOM boundary into the standard DOM. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Event/composed) + */ + get composed(): boolean; + /** + * The **`bubbles`** read-only property of the Event interface indicates whether the event bubbles up through the DOM tree or not. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Event/bubbles) + */ + get bubbles(): boolean; + /** + * The **`cancelable`** read-only property of the Event interface indicates whether the event can be canceled, and therefore prevented as if the event never happened. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Event/cancelable) + */ + get cancelable(): boolean; + /** + * The **`defaultPrevented`** read-only property of the Event interface returns a boolean value indicating whether or not the call to Event.preventDefault() canceled the event. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Event/defaultPrevented) + */ + get defaultPrevented(): boolean; + /** + * The Event property **`returnValue`** indicates whether the default action for this event has been prevented or not. + * @deprecated + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Event/returnValue) + */ + get returnValue(): boolean; + /** + * The **`currentTarget`** read-only property of the Event interface identifies the element to which the event handler has been attached. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Event/currentTarget) + */ + get currentTarget(): EventTarget | undefined; + /** + * The read-only **`target`** property of the dispatched. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Event/target) + */ + get target(): EventTarget | undefined; + /** + * The deprecated **`Event.srcElement`** is an alias for the Event.target property. + * @deprecated + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Event/srcElement) + */ + get srcElement(): EventTarget | undefined; + /** + * The **`timeStamp`** read-only property of the Event interface returns the time (in milliseconds) at which the event was created. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Event/timeStamp) + */ + get timeStamp(): number; + /** + * The **`isTrusted`** read-only property of the when the event was generated by the user agent (including via user actions and programmatic methods such as HTMLElement.focus()), and `false` when the event was dispatched via The only exception is the `click` event, which initializes the `isTrusted` property to `false` in user agents. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Event/isTrusted) + */ + get isTrusted(): boolean; + /** + * The **`cancelBubble`** property of the Event interface is deprecated. + * @deprecated + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Event/cancelBubble) + */ + get cancelBubble(): boolean; + /** + * The **`cancelBubble`** property of the Event interface is deprecated. + * @deprecated + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Event/cancelBubble) + */ + set cancelBubble(value: boolean); + /** + * The **`stopImmediatePropagation()`** method of the If several listeners are attached to the same element for the same event type, they are called in the order in which they were added. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Event/stopImmediatePropagation) + */ + stopImmediatePropagation(): void; + /** + * The **`preventDefault()`** method of the Event interface tells the user agent that if the event does not get explicitly handled, its default action should not be taken as it normally would be. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Event/preventDefault) + */ + preventDefault(): void; + /** + * The **`stopPropagation()`** method of the Event interface prevents further propagation of the current event in the capturing and bubbling phases. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Event/stopPropagation) + */ + stopPropagation(): void; + /** + * The **`composedPath()`** method of the Event interface returns the event's path which is an array of the objects on which listeners will be invoked. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Event/composedPath) + */ + composedPath(): EventTarget[]; + static readonly NONE: number; + static readonly CAPTURING_PHASE: number; + static readonly AT_TARGET: number; + static readonly BUBBLING_PHASE: number; +} +interface EventInit { + bubbles?: boolean; + cancelable?: boolean; + composed?: boolean; +} +type EventListener = (event: EventType) => void; +interface EventListenerObject { + handleEvent(event: EventType): void; +} +type EventListenerOrEventListenerObject = EventListener | EventListenerObject; +/** + * The **`EventTarget`** interface is implemented by objects that can receive events and may have listeners for them. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/EventTarget) + */ +declare class EventTarget = Record> { + constructor(); + /** + * The **`addEventListener()`** method of the EventTarget interface sets up a function that will be called whenever the specified event is delivered to the target. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/EventTarget/addEventListener) + */ + addEventListener(type: Type, handler: EventListenerOrEventListenerObject, options?: EventTargetAddEventListenerOptions | boolean): void; + /** + * The **`removeEventListener()`** method of the EventTarget interface removes an event listener previously registered with EventTarget.addEventListener() from the target. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/EventTarget/removeEventListener) + */ + removeEventListener(type: Type, handler: EventListenerOrEventListenerObject, options?: EventTargetEventListenerOptions | boolean): void; + /** + * The **`dispatchEvent()`** method of the EventTarget sends an Event to the object, (synchronously) invoking the affected event listeners in the appropriate order. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/EventTarget/dispatchEvent) + */ + dispatchEvent(event: EventMap[keyof EventMap]): boolean; +} +interface EventTargetEventListenerOptions { + capture?: boolean; +} +interface EventTargetAddEventListenerOptions { + capture?: boolean; + passive?: boolean; + once?: boolean; + signal?: AbortSignal; +} +interface EventTargetHandlerObject { + handleEvent: (event: Event) => any | undefined; +} +/** + * The **`AbortController`** interface represents a controller object that allows you to abort one or more Web requests as and when desired. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/AbortController) + */ +declare class AbortController { + constructor(); + /** + * The **`signal`** read-only property of the AbortController interface returns an AbortSignal object instance, which can be used to communicate with/abort an asynchronous operation as desired. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/AbortController/signal) + */ + get signal(): AbortSignal; + /** + * The **`abort()`** method of the AbortController interface aborts an asynchronous operation before it has completed. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/AbortController/abort) + */ + abort(reason?: any): void; +} +/** + * The **`AbortSignal`** interface represents a signal object that allows you to communicate with an asynchronous operation (such as a fetch request) and abort it if required via an AbortController object. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/AbortSignal) + */ +declare abstract class AbortSignal extends EventTarget { + /** + * The **`AbortSignal.abort()`** static method returns an AbortSignal that is already set as aborted (and which does not trigger an AbortSignal/abort_event event). + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/AbortSignal/abort_static) + */ + static abort(reason?: any): AbortSignal; + /** + * The **`AbortSignal.timeout()`** static method returns an AbortSignal that will automatically abort after a specified time. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/AbortSignal/timeout_static) + */ + static timeout(delay: number): AbortSignal; + /** + * The **`AbortSignal.any()`** static method takes an iterable of abort signals and returns an AbortSignal. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/AbortSignal/any_static) + */ + static any(signals: AbortSignal[]): AbortSignal; + /** + * The **`aborted`** read-only property returns a value that indicates whether the asynchronous operations the signal is communicating with are aborted (`true`) or not (`false`). + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/AbortSignal/aborted) + */ + get aborted(): boolean; + /** + * The **`reason`** read-only property returns a JavaScript value that indicates the abort reason. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/AbortSignal/reason) + */ + get reason(): any; + /* [MDN Reference](https://developer.mozilla.org/docs/Web/API/AbortSignal/abort_event) */ + get onabort(): any | null; + /* [MDN Reference](https://developer.mozilla.org/docs/Web/API/AbortSignal/abort_event) */ + set onabort(value: any | null); + /** + * The **`throwIfAborted()`** method throws the signal's abort AbortSignal.reason if the signal has been aborted; otherwise it does nothing. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/AbortSignal/throwIfAborted) + */ + throwIfAborted(): void; +} +interface Scheduler { + wait(delay: number, maybeOptions?: SchedulerWaitOptions): Promise; +} +interface SchedulerWaitOptions { + signal?: AbortSignal; +} +/** + * The **`ExtendableEvent`** interface extends the lifetime of the `install` and `activate` events dispatched on the global scope as part of the service worker lifecycle. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/ExtendableEvent) + */ +declare abstract class ExtendableEvent extends Event { + /** + * The **`ExtendableEvent.waitUntil()`** method tells the event dispatcher that work is ongoing. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/ExtendableEvent/waitUntil) + */ + waitUntil(promise: Promise): void; +} +/** + * The **`CustomEvent`** interface represents events initialized by an application for any purpose. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/CustomEvent) + */ +declare class CustomEvent extends Event { + constructor(type: string, init?: CustomEventCustomEventInit); + /** + * The read-only **`detail`** property of the CustomEvent interface returns any data passed when initializing the event. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/CustomEvent/detail) + */ + get detail(): T; +} +interface CustomEventCustomEventInit { + bubbles?: boolean; + cancelable?: boolean; + composed?: boolean; + detail?: any; +} +/** + * The **`Blob`** interface represents a blob, which is a file-like object of immutable, raw data; they can be read as text or binary data, or converted into a ReadableStream so its methods can be used for processing the data. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Blob) + */ +declare class Blob { + constructor(type?: ((ArrayBuffer | ArrayBufferView) | string | Blob)[], options?: BlobOptions); + /** + * The **`size`** read-only property of the Blob interface returns the size of the Blob or File in bytes. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Blob/size) + */ + get size(): number; + /** + * The **`type`** read-only property of the Blob interface returns the MIME type of the file. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Blob/type) + */ + get type(): string; + /** + * The **`slice()`** method of the Blob interface creates and returns a new `Blob` object which contains data from a subset of the blob on which it's called. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Blob/slice) + */ + slice(start?: number, end?: number, type?: string): Blob; + /** + * The **`arrayBuffer()`** method of the Blob interface returns a Promise that resolves with the contents of the blob as binary data contained in an ArrayBuffer. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Blob/arrayBuffer) + */ + arrayBuffer(): Promise; + /** + * The **`bytes()`** method of the Blob interface returns a Promise that resolves with a Uint8Array containing the contents of the blob as an array of bytes. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Blob/bytes) + */ + bytes(): Promise; + /** + * The **`text()`** method of the string containing the contents of the blob, interpreted as UTF-8. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Blob/text) + */ + text(): Promise; + /** + * The **`stream()`** method of the Blob interface returns a ReadableStream which upon reading returns the data contained within the `Blob`. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Blob/stream) + */ + stream(): ReadableStream; +} +interface BlobOptions { + type?: string; +} +/** + * The **`File`** interface provides information about files and allows JavaScript in a web page to access their content. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/File) + */ +declare class File extends Blob { + constructor(bits: ((ArrayBuffer | ArrayBufferView) | string | Blob)[] | undefined, name: string, options?: FileOptions); + /** + * The **`name`** read-only property of the File interface returns the name of the file represented by a File object. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/File/name) + */ + get name(): string; + /** + * The **`lastModified`** read-only property of the File interface provides the last modified date of the file as the number of milliseconds since the Unix epoch (January 1, 1970 at midnight). + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/File/lastModified) + */ + get lastModified(): number; +} +interface FileOptions { + type?: string; + lastModified?: number; +} +/** +* The Cache API allows fine grained control of reading and writing from the Cloudflare global network cache. +* +* [Cloudflare Docs Reference](https://developers.cloudflare.com/workers/runtime-apis/cache/) +*/ +declare abstract class CacheStorage { + /** + * The **`open()`** method of the the Cache object matching the `cacheName`. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/CacheStorage/open) + */ + open(cacheName: string): Promise; + readonly default: Cache; +} +/** +* The Cache API allows fine grained control of reading and writing from the Cloudflare global network cache. +* +* [Cloudflare Docs Reference](https://developers.cloudflare.com/workers/runtime-apis/cache/) +*/ +declare abstract class Cache { + /* [Cloudflare Docs Reference](https://developers.cloudflare.com/workers/runtime-apis/cache/#delete) */ + delete(request: RequestInfo | URL, options?: CacheQueryOptions): Promise; + /* [Cloudflare Docs Reference](https://developers.cloudflare.com/workers/runtime-apis/cache/#match) */ + match(request: RequestInfo | URL, options?: CacheQueryOptions): Promise; + /* [Cloudflare Docs Reference](https://developers.cloudflare.com/workers/runtime-apis/cache/#put) */ + put(request: RequestInfo | URL, response: Response): Promise; +} +interface CacheQueryOptions { + ignoreMethod?: boolean; +} +/** +* The Web Crypto API provides a set of low-level functions for common cryptographic tasks. +* The Workers runtime implements the full surface of this API, but with some differences in +* the [supported algorithms](https://developers.cloudflare.com/workers/runtime-apis/web-crypto/#supported-algorithms) +* compared to those implemented in most browsers. +* +* [Cloudflare Docs Reference](https://developers.cloudflare.com/workers/runtime-apis/web-crypto/) +*/ +declare abstract class Crypto { + /** + * The **`Crypto.subtle`** read-only property returns a cryptographic operations. + * Available only in secure contexts. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Crypto/subtle) + */ + get subtle(): SubtleCrypto; + /** + * The **`Crypto.getRandomValues()`** method lets you get cryptographically strong random values. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Crypto/getRandomValues) + */ + getRandomValues(buffer: T): T; + /** + * The **`randomUUID()`** method of the Crypto interface is used to generate a v4 UUID using a cryptographically secure random number generator. + * Available only in secure contexts. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Crypto/randomUUID) + */ + randomUUID(): string; + DigestStream: typeof DigestStream; +} +/** + * The **`SubtleCrypto`** interface of the Web Crypto API provides a number of low-level cryptographic functions. + * Available only in secure contexts. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/SubtleCrypto) + */ +declare abstract class SubtleCrypto { + /** + * The **`encrypt()`** method of the SubtleCrypto interface encrypts data. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/SubtleCrypto/encrypt) + */ + encrypt(algorithm: string | SubtleCryptoEncryptAlgorithm, key: CryptoKey, plainText: ArrayBuffer | ArrayBufferView): Promise; + /** + * The **`decrypt()`** method of the SubtleCrypto interface decrypts some encrypted data. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/SubtleCrypto/decrypt) + */ + decrypt(algorithm: string | SubtleCryptoEncryptAlgorithm, key: CryptoKey, cipherText: ArrayBuffer | ArrayBufferView): Promise; + /** + * The **`sign()`** method of the SubtleCrypto interface generates a digital signature. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/SubtleCrypto/sign) + */ + sign(algorithm: string | SubtleCryptoSignAlgorithm, key: CryptoKey, data: ArrayBuffer | ArrayBufferView): Promise; + /** + * The **`verify()`** method of the SubtleCrypto interface verifies a digital signature. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/SubtleCrypto/verify) + */ + verify(algorithm: string | SubtleCryptoSignAlgorithm, key: CryptoKey, signature: ArrayBuffer | ArrayBufferView, data: ArrayBuffer | ArrayBufferView): Promise; + /** + * The **`digest()`** method of the SubtleCrypto interface generates a _digest_ of the given data, using the specified hash function. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/SubtleCrypto/digest) + */ + digest(algorithm: string | SubtleCryptoHashAlgorithm, data: ArrayBuffer | ArrayBufferView): Promise; + /** + * The **`generateKey()`** method of the SubtleCrypto interface is used to generate a new key (for symmetric algorithms) or key pair (for public-key algorithms). + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/SubtleCrypto/generateKey) + */ + generateKey(algorithm: string | SubtleCryptoGenerateKeyAlgorithm, extractable: boolean, keyUsages: string[]): Promise; + /** + * The **`deriveKey()`** method of the SubtleCrypto interface can be used to derive a secret key from a master key. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/SubtleCrypto/deriveKey) + */ + deriveKey(algorithm: string | SubtleCryptoDeriveKeyAlgorithm, baseKey: CryptoKey, derivedKeyAlgorithm: string | SubtleCryptoImportKeyAlgorithm, extractable: boolean, keyUsages: string[]): Promise; + /** + * The **`deriveBits()`** method of the key. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/SubtleCrypto/deriveBits) + */ + deriveBits(algorithm: string | SubtleCryptoDeriveKeyAlgorithm, baseKey: CryptoKey, length?: number | null): Promise; + /** + * The **`importKey()`** method of the SubtleCrypto interface imports a key: that is, it takes as input a key in an external, portable format and gives you a CryptoKey object that you can use in the Web Crypto API. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/SubtleCrypto/importKey) + */ + importKey(format: string, keyData: (ArrayBuffer | ArrayBufferView) | JsonWebKey, algorithm: string | SubtleCryptoImportKeyAlgorithm, extractable: boolean, keyUsages: string[]): Promise; + /** + * The **`exportKey()`** method of the SubtleCrypto interface exports a key: that is, it takes as input a CryptoKey object and gives you the key in an external, portable format. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/SubtleCrypto/exportKey) + */ + exportKey(format: string, key: CryptoKey): Promise; + /** + * The **`wrapKey()`** method of the SubtleCrypto interface 'wraps' a key. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/SubtleCrypto/wrapKey) + */ + wrapKey(format: string, key: CryptoKey, wrappingKey: CryptoKey, wrapAlgorithm: string | SubtleCryptoEncryptAlgorithm): Promise; + /** + * The **`unwrapKey()`** method of the SubtleCrypto interface 'unwraps' a key. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/SubtleCrypto/unwrapKey) + */ + unwrapKey(format: string, wrappedKey: ArrayBuffer | ArrayBufferView, unwrappingKey: CryptoKey, unwrapAlgorithm: string | SubtleCryptoEncryptAlgorithm, unwrappedKeyAlgorithm: string | SubtleCryptoImportKeyAlgorithm, extractable: boolean, keyUsages: string[]): Promise; + timingSafeEqual(a: ArrayBuffer | ArrayBufferView, b: ArrayBuffer | ArrayBufferView): boolean; +} +/** + * The **`CryptoKey`** interface of the Web Crypto API represents a cryptographic key obtained from one of the SubtleCrypto methods SubtleCrypto.generateKey, SubtleCrypto.deriveKey, SubtleCrypto.importKey, or SubtleCrypto.unwrapKey. + * Available only in secure contexts. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/CryptoKey) + */ +declare abstract class CryptoKey { + /** + * The read-only **`type`** property of the CryptoKey interface indicates which kind of key is represented by the object. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/CryptoKey/type) + */ + readonly type: string; + /** + * The read-only **`extractable`** property of the CryptoKey interface indicates whether or not the key may be extracted using `SubtleCrypto.exportKey()` or `SubtleCrypto.wrapKey()`. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/CryptoKey/extractable) + */ + readonly extractable: boolean; + /** + * The read-only **`algorithm`** property of the CryptoKey interface returns an object describing the algorithm for which this key can be used, and any associated extra parameters. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/CryptoKey/algorithm) + */ + readonly algorithm: CryptoKeyKeyAlgorithm | CryptoKeyAesKeyAlgorithm | CryptoKeyHmacKeyAlgorithm | CryptoKeyRsaKeyAlgorithm | CryptoKeyEllipticKeyAlgorithm | CryptoKeyArbitraryKeyAlgorithm; + /** + * The read-only **`usages`** property of the CryptoKey interface indicates what can be done with the key. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/CryptoKey/usages) + */ + readonly usages: string[]; +} +interface CryptoKeyPair { + publicKey: CryptoKey; + privateKey: CryptoKey; +} +interface JsonWebKey { + kty: string; + use?: string; + key_ops?: string[]; + alg?: string; + ext?: boolean; + crv?: string; + x?: string; + y?: string; + d?: string; + n?: string; + e?: string; + p?: string; + q?: string; + dp?: string; + dq?: string; + qi?: string; + oth?: RsaOtherPrimesInfo[]; + k?: string; +} +interface RsaOtherPrimesInfo { + r?: string; + d?: string; + t?: string; +} +interface SubtleCryptoDeriveKeyAlgorithm { + name: string; + salt?: (ArrayBuffer | ArrayBufferView); + iterations?: number; + hash?: (string | SubtleCryptoHashAlgorithm); + $public?: CryptoKey; + info?: (ArrayBuffer | ArrayBufferView); +} +interface SubtleCryptoEncryptAlgorithm { + name: string; + iv?: (ArrayBuffer | ArrayBufferView); + additionalData?: (ArrayBuffer | ArrayBufferView); + tagLength?: number; + counter?: (ArrayBuffer | ArrayBufferView); + length?: number; + label?: (ArrayBuffer | ArrayBufferView); +} +interface SubtleCryptoGenerateKeyAlgorithm { + name: string; + hash?: (string | SubtleCryptoHashAlgorithm); + modulusLength?: number; + publicExponent?: (ArrayBuffer | ArrayBufferView); + length?: number; + namedCurve?: string; +} +interface SubtleCryptoHashAlgorithm { + name: string; +} +interface SubtleCryptoImportKeyAlgorithm { + name: string; + hash?: (string | SubtleCryptoHashAlgorithm); + length?: number; + namedCurve?: string; + compressed?: boolean; +} +interface SubtleCryptoSignAlgorithm { + name: string; + hash?: (string | SubtleCryptoHashAlgorithm); + dataLength?: number; + saltLength?: number; +} +interface CryptoKeyKeyAlgorithm { + name: string; +} +interface CryptoKeyAesKeyAlgorithm { + name: string; + length: number; +} +interface CryptoKeyHmacKeyAlgorithm { + name: string; + hash: CryptoKeyKeyAlgorithm; + length: number; +} +interface CryptoKeyRsaKeyAlgorithm { + name: string; + modulusLength: number; + publicExponent: ArrayBuffer | ArrayBufferView; + hash?: CryptoKeyKeyAlgorithm; +} +interface CryptoKeyEllipticKeyAlgorithm { + name: string; + namedCurve: string; +} +interface CryptoKeyArbitraryKeyAlgorithm { + name: string; + hash?: CryptoKeyKeyAlgorithm; + namedCurve?: string; + length?: number; +} +declare class DigestStream extends WritableStream { + constructor(algorithm: string | SubtleCryptoHashAlgorithm); + readonly digest: Promise; + get bytesWritten(): number | bigint; +} +/** + * The **`TextDecoder`** interface represents a decoder for a specific text encoding, such as `UTF-8`, `ISO-8859-2`, `KOI8-R`, `GBK`, etc. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/TextDecoder) + */ +declare class TextDecoder { + constructor(label?: string, options?: TextDecoderConstructorOptions); + /** + * The **`TextDecoder.decode()`** method returns a string containing text decoded from the buffer passed as a parameter. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/TextDecoder/decode) + */ + decode(input?: (ArrayBuffer | ArrayBufferView), options?: TextDecoderDecodeOptions): string; + get encoding(): string; + get fatal(): boolean; + get ignoreBOM(): boolean; +} +/** + * The **`TextEncoder`** interface takes a stream of code points as input and emits a stream of UTF-8 bytes. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/TextEncoder) + */ +declare class TextEncoder { + constructor(); + /** + * The **`TextEncoder.encode()`** method takes a string as input, and returns a Global_Objects/Uint8Array containing the text given in parameters encoded with the specific method for that TextEncoder object. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/TextEncoder/encode) + */ + encode(input?: string): Uint8Array; + /** + * The **`TextEncoder.encodeInto()`** method takes a string to encode and a destination Uint8Array to put resulting UTF-8 encoded text into, and returns a dictionary object indicating the progress of the encoding. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/TextEncoder/encodeInto) + */ + encodeInto(input: string, buffer: Uint8Array): TextEncoderEncodeIntoResult; + get encoding(): string; +} +interface TextDecoderConstructorOptions { + fatal: boolean; + ignoreBOM: boolean; +} +interface TextDecoderDecodeOptions { + stream: boolean; +} +interface TextEncoderEncodeIntoResult { + read: number; + written: number; +} +/** + * The **`ErrorEvent`** interface represents events providing information related to errors in scripts or in files. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/ErrorEvent) + */ +declare class ErrorEvent extends Event { + constructor(type: string, init?: ErrorEventErrorEventInit); + /** + * The **`filename`** read-only property of the ErrorEvent interface returns a string containing the name of the script file in which the error occurred. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/ErrorEvent/filename) + */ + get filename(): string; + /** + * The **`message`** read-only property of the ErrorEvent interface returns a string containing a human-readable error message describing the problem. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/ErrorEvent/message) + */ + get message(): string; + /** + * The **`lineno`** read-only property of the ErrorEvent interface returns an integer containing the line number of the script file on which the error occurred. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/ErrorEvent/lineno) + */ + get lineno(): number; + /** + * The **`colno`** read-only property of the ErrorEvent interface returns an integer containing the column number of the script file on which the error occurred. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/ErrorEvent/colno) + */ + get colno(): number; + /** + * The **`error`** read-only property of the ErrorEvent interface returns a JavaScript value, such as an Error or DOMException, representing the error associated with this event. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/ErrorEvent/error) + */ + get error(): any; +} +interface ErrorEventErrorEventInit { + message?: string; + filename?: string; + lineno?: number; + colno?: number; + error?: any; +} +/** + * The **`MessageEvent`** interface represents a message received by a target object. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/MessageEvent) + */ +declare class MessageEvent extends Event { + constructor(type: string, initializer: MessageEventInit); + /** + * The **`data`** read-only property of the The data sent by the message emitter; this can be any data type, depending on what originated this event. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/MessageEvent/data) + */ + readonly data: any; + /** + * The **`origin`** read-only property of the origin of the message emitter. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/MessageEvent/origin) + */ + readonly origin: string | null; + /** + * The **`lastEventId`** read-only property of the unique ID for the event. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/MessageEvent/lastEventId) + */ + readonly lastEventId: string; + /** + * The **`source`** read-only property of the a WindowProxy, MessagePort, or a `MessageEventSource` (which can be a WindowProxy, message emitter. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/MessageEvent/source) + */ + readonly source: MessagePort | null; + /** + * The **`ports`** read-only property of the containing all MessagePort objects sent with the message, in order. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/MessageEvent/ports) + */ + readonly ports: MessagePort[]; +} +interface MessageEventInit { + data: ArrayBuffer | string; +} +/** + * The **`PromiseRejectionEvent`** interface represents events which are sent to the global script context when JavaScript Promises are rejected. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/PromiseRejectionEvent) + */ +declare abstract class PromiseRejectionEvent extends Event { + /** + * The PromiseRejectionEvent interface's **`promise`** read-only property indicates the JavaScript rejected. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/PromiseRejectionEvent/promise) + */ + readonly promise: Promise; + /** + * The PromiseRejectionEvent **`reason`** read-only property is any JavaScript value or Object which provides the reason passed into Promise.reject(). + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/PromiseRejectionEvent/reason) + */ + readonly reason: any; +} +/** + * The **`FormData`** interface provides a way to construct a set of key/value pairs representing form fields and their values, which can be sent using the Window/fetch, XMLHttpRequest.send() or navigator.sendBeacon() methods. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/FormData) + */ +declare class FormData { + constructor(); + /** + * The **`append()`** method of the FormData interface appends a new value onto an existing key inside a `FormData` object, or adds the key if it does not already exist. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/FormData/append) + */ + append(name: string, value: string): void; + /** + * The **`append()`** method of the FormData interface appends a new value onto an existing key inside a `FormData` object, or adds the key if it does not already exist. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/FormData/append) + */ + append(name: string, value: Blob, filename?: string): void; + /** + * The **`delete()`** method of the FormData interface deletes a key and its value(s) from a `FormData` object. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/FormData/delete) + */ + delete(name: string): void; + /** + * The **`get()`** method of the FormData interface returns the first value associated with a given key from within a `FormData` object. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/FormData/get) + */ + get(name: string): (File | string) | null; + /** + * The **`getAll()`** method of the FormData interface returns all the values associated with a given key from within a `FormData` object. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/FormData/getAll) + */ + getAll(name: string): (File | string)[]; + /** + * The **`has()`** method of the FormData interface returns whether a `FormData` object contains a certain key. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/FormData/has) + */ + has(name: string): boolean; + /** + * The **`set()`** method of the FormData interface sets a new value for an existing key inside a `FormData` object, or adds the key/value if it does not already exist. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/FormData/set) + */ + set(name: string, value: string): void; + /** + * The **`set()`** method of the FormData interface sets a new value for an existing key inside a `FormData` object, or adds the key/value if it does not already exist. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/FormData/set) + */ + set(name: string, value: Blob, filename?: string): void; + /* Returns an array of key, value pairs for every entry in the list. */ + entries(): IterableIterator<[ + key: string, + value: File | string + ]>; + /* Returns a list of keys in the list. */ + keys(): IterableIterator; + /* Returns a list of values in the list. */ + values(): IterableIterator<(File | string)>; + forEach(callback: (this: This, value: File | string, key: string, parent: FormData) => void, thisArg?: This): void; + [Symbol.iterator](): IterableIterator<[ + key: string, + value: File | string + ]>; +} +interface ContentOptions { + html?: boolean; +} +declare class HTMLRewriter { + constructor(); + on(selector: string, handlers: HTMLRewriterElementContentHandlers): HTMLRewriter; + onDocument(handlers: HTMLRewriterDocumentContentHandlers): HTMLRewriter; + transform(response: Response): Response; +} +interface HTMLRewriterElementContentHandlers { + element?(element: Element): void | Promise; + comments?(comment: Comment): void | Promise; + text?(element: Text): void | Promise; +} +interface HTMLRewriterDocumentContentHandlers { + doctype?(doctype: Doctype): void | Promise; + comments?(comment: Comment): void | Promise; + text?(text: Text): void | Promise; + end?(end: DocumentEnd): void | Promise; +} +interface Doctype { + readonly name: string | null; + readonly publicId: string | null; + readonly systemId: string | null; +} +interface Element { + tagName: string; + readonly attributes: IterableIterator; + readonly removed: boolean; + readonly namespaceURI: string; + getAttribute(name: string): string | null; + hasAttribute(name: string): boolean; + setAttribute(name: string, value: string): Element; + removeAttribute(name: string): Element; + before(content: string | ReadableStream | Response, options?: ContentOptions): Element; + after(content: string | ReadableStream | Response, options?: ContentOptions): Element; + prepend(content: string | ReadableStream | Response, options?: ContentOptions): Element; + append(content: string | ReadableStream | Response, options?: ContentOptions): Element; + replace(content: string | ReadableStream | Response, options?: ContentOptions): Element; + remove(): Element; + removeAndKeepContent(): Element; + setInnerContent(content: string | ReadableStream | Response, options?: ContentOptions): Element; + onEndTag(handler: (tag: EndTag) => void | Promise): void; +} +interface EndTag { + name: string; + before(content: string | ReadableStream | Response, options?: ContentOptions): EndTag; + after(content: string | ReadableStream | Response, options?: ContentOptions): EndTag; + remove(): EndTag; +} +interface Comment { + text: string; + readonly removed: boolean; + before(content: string, options?: ContentOptions): Comment; + after(content: string, options?: ContentOptions): Comment; + replace(content: string, options?: ContentOptions): Comment; + remove(): Comment; +} +interface Text { + readonly text: string; + readonly lastInTextNode: boolean; + readonly removed: boolean; + before(content: string | ReadableStream | Response, options?: ContentOptions): Text; + after(content: string | ReadableStream | Response, options?: ContentOptions): Text; + replace(content: string | ReadableStream | Response, options?: ContentOptions): Text; + remove(): Text; +} +interface DocumentEnd { + append(content: string, options?: ContentOptions): DocumentEnd; +} +/** + * This is the event type for `fetch` events dispatched on the ServiceWorkerGlobalScope. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/FetchEvent) + */ +declare abstract class FetchEvent extends ExtendableEvent { + /** + * The **`request`** read-only property of the the event handler. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/FetchEvent/request) + */ + readonly request: Request; + /** + * The **`respondWith()`** method of allows you to provide a promise for a Response yourself. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/FetchEvent/respondWith) + */ + respondWith(promise: Response | Promise): void; + passThroughOnException(): void; +} +type HeadersInit = Headers | Iterable> | Record; +/** + * The **`Headers`** interface of the Fetch API allows you to perform various actions on HTTP request and response headers. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Headers) + */ +declare class Headers { + constructor(init?: HeadersInit); + /** + * The **`get()`** method of the Headers interface returns a byte string of all the values of a header within a `Headers` object with a given name. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Headers/get) + */ + get(name: string): string | null; + getAll(name: string): string[]; + /** + * The **`getSetCookie()`** method of the Headers interface returns an array containing the values of all Set-Cookie headers associated with a response. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Headers/getSetCookie) + */ + getSetCookie(): string[]; + /** + * The **`has()`** method of the Headers interface returns a boolean stating whether a `Headers` object contains a certain header. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Headers/has) + */ + has(name: string): boolean; + /** + * The **`set()`** method of the Headers interface sets a new value for an existing header inside a `Headers` object, or adds the header if it does not already exist. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Headers/set) + */ + set(name: string, value: string): void; + /** + * The **`append()`** method of the Headers interface appends a new value onto an existing header inside a `Headers` object, or adds the header if it does not already exist. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Headers/append) + */ + append(name: string, value: string): void; + /** + * The **`delete()`** method of the Headers interface deletes a header from the current `Headers` object. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Headers/delete) + */ + delete(name: string): void; + forEach(callback: (this: This, value: string, key: string, parent: Headers) => void, thisArg?: This): void; + /* Returns an iterator allowing to go through all key/value pairs contained in this object. */ + entries(): IterableIterator<[ + key: string, + value: string + ]>; + /* Returns an iterator allowing to go through all keys of the key/value pairs contained in this object. */ + keys(): IterableIterator; + /* Returns an iterator allowing to go through all values of the key/value pairs contained in this object. */ + values(): IterableIterator; + [Symbol.iterator](): IterableIterator<[ + key: string, + value: string + ]>; +} +type BodyInit = ReadableStream | string | ArrayBuffer | ArrayBufferView | Blob | URLSearchParams | FormData; +declare abstract class Body { + /* [MDN Reference](https://developer.mozilla.org/docs/Web/API/Request/body) */ + get body(): ReadableStream | null; + /* [MDN Reference](https://developer.mozilla.org/docs/Web/API/Request/bodyUsed) */ + get bodyUsed(): boolean; + /* [MDN Reference](https://developer.mozilla.org/docs/Web/API/Request/arrayBuffer) */ + arrayBuffer(): Promise; + /* [MDN Reference](https://developer.mozilla.org/docs/Web/API/Request/bytes) */ + bytes(): Promise; + /* [MDN Reference](https://developer.mozilla.org/docs/Web/API/Request/text) */ + text(): Promise; + /* [MDN Reference](https://developer.mozilla.org/docs/Web/API/Request/json) */ + json(): Promise; + /* [MDN Reference](https://developer.mozilla.org/docs/Web/API/Request/formData) */ + formData(): Promise; + /* [MDN Reference](https://developer.mozilla.org/docs/Web/API/Request/blob) */ + blob(): Promise; +} +/** + * The **`Response`** interface of the Fetch API represents the response to a request. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Response) + */ +declare var Response: { + prototype: Response; + new (body?: BodyInit | null, init?: ResponseInit): Response; + error(): Response; + redirect(url: string, status?: number): Response; + json(any: any, maybeInit?: (ResponseInit | Response)): Response; +}; +/** + * The **`Response`** interface of the Fetch API represents the response to a request. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Response) + */ +interface Response extends Body { + /** + * The **`clone()`** method of the Response interface creates a clone of a response object, identical in every way, but stored in a different variable. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Response/clone) + */ + clone(): Response; + /** + * The **`status`** read-only property of the Response interface contains the HTTP status codes of the response. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Response/status) + */ + status: number; + /** + * The **`statusText`** read-only property of the Response interface contains the status message corresponding to the HTTP status code in Response.status. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Response/statusText) + */ + statusText: string; + /** + * The **`headers`** read-only property of the with the response. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Response/headers) + */ + headers: Headers; + /** + * The **`ok`** read-only property of the Response interface contains a Boolean stating whether the response was successful (status in the range 200-299) or not. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Response/ok) + */ + ok: boolean; + /** + * The **`redirected`** read-only property of the Response interface indicates whether or not the response is the result of a request you made which was redirected. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Response/redirected) + */ + redirected: boolean; + /** + * The **`url`** read-only property of the Response interface contains the URL of the response. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Response/url) + */ + url: string; + webSocket: WebSocket | null; + cf: any | undefined; + /** + * The **`type`** read-only property of the Response interface contains the type of the response. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Response/type) + */ + type: "default" | "error"; +} +interface ResponseInit { + status?: number; + statusText?: string; + headers?: HeadersInit; + cf?: any; + webSocket?: (WebSocket | null); + encodeBody?: "automatic" | "manual"; +} +type RequestInfo> = Request | string; +/** + * The **`Request`** interface of the Fetch API represents a resource request. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Request) + */ +declare var Request: { + prototype: Request; + new >(input: RequestInfo | URL, init?: RequestInit): Request; +}; +/** + * The **`Request`** interface of the Fetch API represents a resource request. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Request) + */ +interface Request> extends Body { + /** + * The **`clone()`** method of the Request interface creates a copy of the current `Request` object. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Request/clone) + */ + clone(): Request; + /** + * The **`method`** read-only property of the `POST`, etc.) A String indicating the method of the request. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Request/method) + */ + method: string; + /** + * The **`url`** read-only property of the Request interface contains the URL of the request. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Request/url) + */ + url: string; + /** + * The **`headers`** read-only property of the with the request. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Request/headers) + */ + headers: Headers; + /** + * The **`redirect`** read-only property of the Request interface contains the mode for how redirects are handled. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Request/redirect) + */ + redirect: string; + fetcher: Fetcher | null; + /** + * The read-only **`signal`** property of the Request interface returns the AbortSignal associated with the request. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Request/signal) + */ + signal: AbortSignal; + cf: Cf | undefined; + /** + * The **`integrity`** read-only property of the Request interface contains the subresource integrity value of the request. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Request/integrity) + */ + integrity: string; + /** + * The **`keepalive`** read-only property of the Request interface contains the request's `keepalive` setting (`true` or `false`), which indicates whether the browser will keep the associated request alive if the page that initiated it is unloaded before the request is complete. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Request/keepalive) + */ + keepalive: boolean; + /** + * The **`cache`** read-only property of the Request interface contains the cache mode of the request. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/Request/cache) + */ + cache?: "no-store"; +} +interface RequestInit { + /* A string to set request's method. */ + method?: string; + /* A Headers object, an object literal, or an array of two-item arrays to set request's headers. */ + headers?: HeadersInit; + /* A BodyInit object or null to set request's body. */ + body?: BodyInit | null; + /* A string indicating whether request follows redirects, results in an error upon encountering a redirect, or returns the redirect (in an opaque fashion). Sets request's redirect. */ + redirect?: string; + fetcher?: (Fetcher | null); + cf?: Cf; + /* A string indicating how the request will interact with the browser's cache to set request's cache. */ + cache?: "no-store"; + /* A cryptographic hash of the resource to be fetched by request. Sets request's integrity. */ + integrity?: string; + /* An AbortSignal to set request's signal. */ + signal?: (AbortSignal | null); + encodeResponseBody?: "automatic" | "manual"; +} +type Service Rpc.WorkerEntrypointBranded) | Rpc.WorkerEntrypointBranded | ExportedHandler | undefined = undefined> = T extends new (...args: any[]) => Rpc.WorkerEntrypointBranded ? Fetcher> : T extends Rpc.WorkerEntrypointBranded ? Fetcher : T extends Exclude ? never : Fetcher; +type Fetcher = (T extends Rpc.EntrypointBranded ? Rpc.Provider : unknown) & { + fetch(input: RequestInfo | URL, init?: RequestInit): Promise; + connect(address: SocketAddress | string, options?: SocketOptions): Socket; +}; +interface KVNamespaceListKey { + name: Key; + expiration?: number; + metadata?: Metadata; +} +type KVNamespaceListResult = { + list_complete: false; + keys: KVNamespaceListKey[]; + cursor: string; + cacheStatus: string | null; +} | { + list_complete: true; + keys: KVNamespaceListKey[]; + cacheStatus: string | null; +}; +interface KVNamespace { + get(key: Key, options?: Partial>): Promise; + get(key: Key, type: "text"): Promise; + get(key: Key, type: "json"): Promise; + get(key: Key, type: "arrayBuffer"): Promise; + get(key: Key, type: "stream"): Promise; + get(key: Key, options?: KVNamespaceGetOptions<"text">): Promise; + get(key: Key, options?: KVNamespaceGetOptions<"json">): Promise; + get(key: Key, options?: KVNamespaceGetOptions<"arrayBuffer">): Promise; + get(key: Key, options?: KVNamespaceGetOptions<"stream">): Promise; + get(key: Array, type: "text"): Promise>; + get(key: Array, type: "json"): Promise>; + get(key: Array, options?: Partial>): Promise>; + get(key: Array, options?: KVNamespaceGetOptions<"text">): Promise>; + get(key: Array, options?: KVNamespaceGetOptions<"json">): Promise>; + list(options?: KVNamespaceListOptions): Promise>; + put(key: Key, value: string | ArrayBuffer | ArrayBufferView | ReadableStream, options?: KVNamespacePutOptions): Promise; + getWithMetadata(key: Key, options?: Partial>): Promise>; + getWithMetadata(key: Key, type: "text"): Promise>; + getWithMetadata(key: Key, type: "json"): Promise>; + getWithMetadata(key: Key, type: "arrayBuffer"): Promise>; + getWithMetadata(key: Key, type: "stream"): Promise>; + getWithMetadata(key: Key, options: KVNamespaceGetOptions<"text">): Promise>; + getWithMetadata(key: Key, options: KVNamespaceGetOptions<"json">): Promise>; + getWithMetadata(key: Key, options: KVNamespaceGetOptions<"arrayBuffer">): Promise>; + getWithMetadata(key: Key, options: KVNamespaceGetOptions<"stream">): Promise>; + getWithMetadata(key: Array, type: "text"): Promise>>; + getWithMetadata(key: Array, type: "json"): Promise>>; + getWithMetadata(key: Array, options?: Partial>): Promise>>; + getWithMetadata(key: Array, options?: KVNamespaceGetOptions<"text">): Promise>>; + getWithMetadata(key: Array, options?: KVNamespaceGetOptions<"json">): Promise>>; + delete(key: Key): Promise; +} +interface KVNamespaceListOptions { + limit?: number; + prefix?: (string | null); + cursor?: (string | null); +} +interface KVNamespaceGetOptions { + type: Type; + cacheTtl?: number; +} +interface KVNamespacePutOptions { + expiration?: number; + expirationTtl?: number; + metadata?: (any | null); +} +interface KVNamespaceGetWithMetadataResult { + value: Value | null; + metadata: Metadata | null; + cacheStatus: string | null; +} +type QueueContentType = "text" | "bytes" | "json" | "v8"; +interface Queue { + send(message: Body, options?: QueueSendOptions): Promise; + sendBatch(messages: Iterable>, options?: QueueSendBatchOptions): Promise; +} +interface QueueSendOptions { + contentType?: QueueContentType; + delaySeconds?: number; +} +interface QueueSendBatchOptions { + delaySeconds?: number; +} +interface MessageSendRequest { + body: Body; + contentType?: QueueContentType; + delaySeconds?: number; +} +interface QueueRetryOptions { + delaySeconds?: number; +} +interface Message { + readonly id: string; + readonly timestamp: Date; + readonly body: Body; + readonly attempts: number; + retry(options?: QueueRetryOptions): void; + ack(): void; +} +interface QueueEvent extends ExtendableEvent { + readonly messages: readonly Message[]; + readonly queue: string; + retryAll(options?: QueueRetryOptions): void; + ackAll(): void; +} +interface MessageBatch { + readonly messages: readonly Message[]; + readonly queue: string; + retryAll(options?: QueueRetryOptions): void; + ackAll(): void; +} +interface R2Error extends Error { + readonly name: string; + readonly code: number; + readonly message: string; + readonly action: string; + readonly stack: any; +} +interface R2ListOptions { + limit?: number; + prefix?: string; + cursor?: string; + delimiter?: string; + startAfter?: string; + include?: ("httpMetadata" | "customMetadata")[]; +} +declare abstract class R2Bucket { + head(key: string): Promise; + get(key: string, options: R2GetOptions & { + onlyIf: R2Conditional | Headers; + }): Promise; + get(key: string, options?: R2GetOptions): Promise; + put(key: string, value: ReadableStream | ArrayBuffer | ArrayBufferView | string | null | Blob, options?: R2PutOptions & { + onlyIf: R2Conditional | Headers; + }): Promise; + put(key: string, value: ReadableStream | ArrayBuffer | ArrayBufferView | string | null | Blob, options?: R2PutOptions): Promise; + createMultipartUpload(key: string, options?: R2MultipartOptions): Promise; + resumeMultipartUpload(key: string, uploadId: string): R2MultipartUpload; + delete(keys: string | string[]): Promise; + list(options?: R2ListOptions): Promise; +} +interface R2MultipartUpload { + readonly key: string; + readonly uploadId: string; + uploadPart(partNumber: number, value: ReadableStream | (ArrayBuffer | ArrayBufferView) | string | Blob, options?: R2UploadPartOptions): Promise; + abort(): Promise; + complete(uploadedParts: R2UploadedPart[]): Promise; +} +interface R2UploadedPart { + partNumber: number; + etag: string; +} +declare abstract class R2Object { + readonly key: string; + readonly version: string; + readonly size: number; + readonly etag: string; + readonly httpEtag: string; + readonly checksums: R2Checksums; + readonly uploaded: Date; + readonly httpMetadata?: R2HTTPMetadata; + readonly customMetadata?: Record; + readonly range?: R2Range; + readonly storageClass: string; + readonly ssecKeyMd5?: string; + writeHttpMetadata(headers: Headers): void; +} +interface R2ObjectBody extends R2Object { + get body(): ReadableStream; + get bodyUsed(): boolean; + arrayBuffer(): Promise; + bytes(): Promise; + text(): Promise; + json(): Promise; + blob(): Promise; +} +type R2Range = { + offset: number; + length?: number; +} | { + offset?: number; + length: number; +} | { + suffix: number; +}; +interface R2Conditional { + etagMatches?: string; + etagDoesNotMatch?: string; + uploadedBefore?: Date; + uploadedAfter?: Date; + secondsGranularity?: boolean; +} +interface R2GetOptions { + onlyIf?: (R2Conditional | Headers); + range?: (R2Range | Headers); + ssecKey?: (ArrayBuffer | string); +} +interface R2PutOptions { + onlyIf?: (R2Conditional | Headers); + httpMetadata?: (R2HTTPMetadata | Headers); + customMetadata?: Record; + md5?: ((ArrayBuffer | ArrayBufferView) | string); + sha1?: ((ArrayBuffer | ArrayBufferView) | string); + sha256?: ((ArrayBuffer | ArrayBufferView) | string); + sha384?: ((ArrayBuffer | ArrayBufferView) | string); + sha512?: ((ArrayBuffer | ArrayBufferView) | string); + storageClass?: string; + ssecKey?: (ArrayBuffer | string); +} +interface R2MultipartOptions { + httpMetadata?: (R2HTTPMetadata | Headers); + customMetadata?: Record; + storageClass?: string; + ssecKey?: (ArrayBuffer | string); +} +interface R2Checksums { + readonly md5?: ArrayBuffer; + readonly sha1?: ArrayBuffer; + readonly sha256?: ArrayBuffer; + readonly sha384?: ArrayBuffer; + readonly sha512?: ArrayBuffer; + toJSON(): R2StringChecksums; +} +interface R2StringChecksums { + md5?: string; + sha1?: string; + sha256?: string; + sha384?: string; + sha512?: string; +} +interface R2HTTPMetadata { + contentType?: string; + contentLanguage?: string; + contentDisposition?: string; + contentEncoding?: string; + cacheControl?: string; + cacheExpiry?: Date; +} +type R2Objects = { + objects: R2Object[]; + delimitedPrefixes: string[]; +} & ({ + truncated: true; + cursor: string; +} | { + truncated: false; +}); +interface R2UploadPartOptions { + ssecKey?: (ArrayBuffer | string); +} +declare abstract class ScheduledEvent extends ExtendableEvent { + readonly scheduledTime: number; + readonly cron: string; + noRetry(): void; +} +interface ScheduledController { + readonly scheduledTime: number; + readonly cron: string; + noRetry(): void; +} +interface QueuingStrategy { + highWaterMark?: (number | bigint); + size?: (chunk: T) => number | bigint; +} +interface UnderlyingSink { + type?: string; + start?: (controller: WritableStreamDefaultController) => void | Promise; + write?: (chunk: W, controller: WritableStreamDefaultController) => void | Promise; + abort?: (reason: any) => void | Promise; + close?: () => void | Promise; +} +interface UnderlyingByteSource { + type: "bytes"; + autoAllocateChunkSize?: number; + start?: (controller: ReadableByteStreamController) => void | Promise; + pull?: (controller: ReadableByteStreamController) => void | Promise; + cancel?: (reason: any) => void | Promise; +} +interface UnderlyingSource { + type?: "" | undefined; + start?: (controller: ReadableStreamDefaultController) => void | Promise; + pull?: (controller: ReadableStreamDefaultController) => void | Promise; + cancel?: (reason: any) => void | Promise; + expectedLength?: (number | bigint); +} +interface Transformer { + readableType?: string; + writableType?: string; + start?: (controller: TransformStreamDefaultController) => void | Promise; + transform?: (chunk: I, controller: TransformStreamDefaultController) => void | Promise; + flush?: (controller: TransformStreamDefaultController) => void | Promise; + cancel?: (reason: any) => void | Promise; + expectedLength?: number; +} +interface StreamPipeOptions { + preventAbort?: boolean; + preventCancel?: boolean; + /** + * Pipes this readable stream to a given writable stream destination. The way in which the piping process behaves under various error conditions can be customized with a number of passed options. It returns a promise that fulfills when the piping process completes successfully, or rejects if any errors were encountered. + * + * Piping a stream will lock it for the duration of the pipe, preventing any other consumer from acquiring a reader. + * + * Errors and closures of the source and destination streams propagate as follows: + * + * An error in this source readable stream will abort destination, unless preventAbort is truthy. The returned promise will be rejected with the source's error, or with any error that occurs during aborting the destination. + * + * An error in destination will cancel this source readable stream, unless preventCancel is truthy. The returned promise will be rejected with the destination's error, or with any error that occurs during canceling the source. + * + * When this source readable stream closes, destination will be closed, unless preventClose is truthy. The returned promise will be fulfilled once this process completes, unless an error is encountered while closing the destination, in which case it will be rejected with that error. + * + * If destination starts out closed or closing, this source readable stream will be canceled, unless preventCancel is true. The returned promise will be rejected with an error indicating piping to a closed stream failed, or with any error that occurs during canceling the source. + * + * The signal option can be set to an AbortSignal to allow aborting an ongoing pipe operation via the corresponding AbortController. In this case, this source readable stream will be canceled, and destination aborted, unless the respective options preventCancel or preventAbort are set. + */ + preventClose?: boolean; + signal?: AbortSignal; +} +type ReadableStreamReadResult = { + done: false; + value: R; +} | { + done: true; + value?: undefined; +}; +/** + * The `ReadableStream` interface of the Streams API represents a readable stream of byte data. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/ReadableStream) + */ +interface ReadableStream { + /** + * The **`locked`** read-only property of the ReadableStream interface returns whether or not the readable stream is locked to a reader. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/ReadableStream/locked) + */ + get locked(): boolean; + /** + * The **`cancel()`** method of the ReadableStream interface returns a Promise that resolves when the stream is canceled. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/ReadableStream/cancel) + */ + cancel(reason?: any): Promise; + /** + * The **`getReader()`** method of the ReadableStream interface creates a reader and locks the stream to it. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/ReadableStream/getReader) + */ + getReader(): ReadableStreamDefaultReader; + /** + * The **`getReader()`** method of the ReadableStream interface creates a reader and locks the stream to it. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/ReadableStream/getReader) + */ + getReader(options: ReadableStreamGetReaderOptions): ReadableStreamBYOBReader; + /** + * The **`pipeThrough()`** method of the ReadableStream interface provides a chainable way of piping the current stream through a transform stream or any other writable/readable pair. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/ReadableStream/pipeThrough) + */ + pipeThrough(transform: ReadableWritablePair, options?: StreamPipeOptions): ReadableStream; + /** + * The **`pipeTo()`** method of the ReadableStream interface pipes the current `ReadableStream` to a given WritableStream and returns a Promise that fulfills when the piping process completes successfully, or rejects if any errors were encountered. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/ReadableStream/pipeTo) + */ + pipeTo(destination: WritableStream, options?: StreamPipeOptions): Promise; + /** + * The **`tee()`** method of the two-element array containing the two resulting branches as new ReadableStream instances. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/ReadableStream/tee) + */ + tee(): [ + ReadableStream, + ReadableStream + ]; + values(options?: ReadableStreamValuesOptions): AsyncIterableIterator; + [Symbol.asyncIterator](options?: ReadableStreamValuesOptions): AsyncIterableIterator; +} +/** + * The `ReadableStream` interface of the Streams API represents a readable stream of byte data. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/ReadableStream) + */ +declare const ReadableStream: { + prototype: ReadableStream; + new (underlyingSource: UnderlyingByteSource, strategy?: QueuingStrategy): ReadableStream; + new (underlyingSource?: UnderlyingSource, strategy?: QueuingStrategy): ReadableStream; +}; +/** + * The **`ReadableStreamDefaultReader`** interface of the Streams API represents a default reader that can be used to read stream data supplied from a network (such as a fetch request). + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/ReadableStreamDefaultReader) + */ +declare class ReadableStreamDefaultReader { + constructor(stream: ReadableStream); + get closed(): Promise; + cancel(reason?: any): Promise; + /** + * The **`read()`** method of the ReadableStreamDefaultReader interface returns a Promise providing access to the next chunk in the stream's internal queue. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/ReadableStreamDefaultReader/read) + */ + read(): Promise>; + /** + * The **`releaseLock()`** method of the ReadableStreamDefaultReader interface releases the reader's lock on the stream. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/ReadableStreamDefaultReader/releaseLock) + */ + releaseLock(): void; +} +/** + * The `ReadableStreamBYOBReader` interface of the Streams API defines a reader for a ReadableStream that supports zero-copy reading from an underlying byte source. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/ReadableStreamBYOBReader) + */ +declare class ReadableStreamBYOBReader { + constructor(stream: ReadableStream); + get closed(): Promise; + cancel(reason?: any): Promise; + /** + * The **`read()`** method of the ReadableStreamBYOBReader interface is used to read data into a view on a user-supplied buffer from an associated readable byte stream. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/ReadableStreamBYOBReader/read) + */ + read(view: T): Promise>; + /** + * The **`releaseLock()`** method of the ReadableStreamBYOBReader interface releases the reader's lock on the stream. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/ReadableStreamBYOBReader/releaseLock) + */ + releaseLock(): void; + readAtLeast(minElements: number, view: T): Promise>; +} +interface ReadableStreamBYOBReaderReadableStreamBYOBReaderReadOptions { + min?: number; +} +interface ReadableStreamGetReaderOptions { + /** + * Creates a ReadableStreamBYOBReader and locks the stream to the new reader. + * + * This call behaves the same way as the no-argument variant, except that it only works on readable byte streams, i.e. streams which were constructed specifically with the ability to handle "bring your own buffer" reading. The returned BYOB reader provides the ability to directly read individual chunks from the stream via its read() method, into developer-supplied buffers, allowing more precise control over allocation. + */ + mode: "byob"; +} +/** + * The **`ReadableStreamBYOBRequest`** interface of the Streams API represents a 'pull request' for data from an underlying source that will made as a zero-copy transfer to a consumer (bypassing the stream's internal queues). + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/ReadableStreamBYOBRequest) + */ +declare abstract class ReadableStreamBYOBRequest { + /** + * The **`view`** getter property of the ReadableStreamBYOBRequest interface returns the current view. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/ReadableStreamBYOBRequest/view) + */ + get view(): Uint8Array | null; + /** + * The **`respond()`** method of the ReadableStreamBYOBRequest interface is used to signal to the associated readable byte stream that the specified number of bytes were written into the ReadableStreamBYOBRequest.view. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/ReadableStreamBYOBRequest/respond) + */ + respond(bytesWritten: number): void; + /** + * The **`respondWithNewView()`** method of the ReadableStreamBYOBRequest interface specifies a new view that the consumer of the associated readable byte stream should write to instead of ReadableStreamBYOBRequest.view. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/ReadableStreamBYOBRequest/respondWithNewView) + */ + respondWithNewView(view: ArrayBuffer | ArrayBufferView): void; + get atLeast(): number | null; +} +/** + * The **`ReadableStreamDefaultController`** interface of the Streams API represents a controller allowing control of a ReadableStream's state and internal queue. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/ReadableStreamDefaultController) + */ +declare abstract class ReadableStreamDefaultController { + /** + * The **`desiredSize`** read-only property of the required to fill the stream's internal queue. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/ReadableStreamDefaultController/desiredSize) + */ + get desiredSize(): number | null; + /** + * The **`close()`** method of the ReadableStreamDefaultController interface closes the associated stream. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/ReadableStreamDefaultController/close) + */ + close(): void; + /** + * The **`enqueue()`** method of the ```js-nolint enqueue(chunk) ``` - `chunk` - : The chunk to enqueue. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/ReadableStreamDefaultController/enqueue) + */ + enqueue(chunk?: R): void; + /** + * The **`error()`** method of the with the associated stream to error. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/ReadableStreamDefaultController/error) + */ + error(reason: any): void; +} +/** + * The **`ReadableByteStreamController`** interface of the Streams API represents a controller for a readable byte stream. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/ReadableByteStreamController) + */ +declare abstract class ReadableByteStreamController { + /** + * The **`byobRequest`** read-only property of the ReadableByteStreamController interface returns the current BYOB request, or `null` if there are no pending requests. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/ReadableByteStreamController/byobRequest) + */ + get byobRequest(): ReadableStreamBYOBRequest | null; + /** + * The **`desiredSize`** read-only property of the ReadableByteStreamController interface returns the number of bytes required to fill the stream's internal queue to its 'desired size'. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/ReadableByteStreamController/desiredSize) + */ + get desiredSize(): number | null; + /** + * The **`close()`** method of the ReadableByteStreamController interface closes the associated stream. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/ReadableByteStreamController/close) + */ + close(): void; + /** + * The **`enqueue()`** method of the ReadableByteStreamController interface enqueues a given chunk on the associated readable byte stream (the chunk is copied into the stream's internal queues). + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/ReadableByteStreamController/enqueue) + */ + enqueue(chunk: ArrayBuffer | ArrayBufferView): void; + /** + * The **`error()`** method of the ReadableByteStreamController interface causes any future interactions with the associated stream to error with the specified reason. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/ReadableByteStreamController/error) + */ + error(reason: any): void; +} +/** + * The **`WritableStreamDefaultController`** interface of the Streams API represents a controller allowing control of a WritableStream's state. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/WritableStreamDefaultController) + */ +declare abstract class WritableStreamDefaultController { + /** + * The read-only **`signal`** property of the WritableStreamDefaultController interface returns the AbortSignal associated with the controller. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/WritableStreamDefaultController/signal) + */ + get signal(): AbortSignal; + /** + * The **`error()`** method of the with the associated stream to error. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/WritableStreamDefaultController/error) + */ + error(reason?: any): void; +} +/** + * The **`TransformStreamDefaultController`** interface of the Streams API provides methods to manipulate the associated ReadableStream and WritableStream. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/TransformStreamDefaultController) + */ +declare abstract class TransformStreamDefaultController { + /** + * The **`desiredSize`** read-only property of the TransformStreamDefaultController interface returns the desired size to fill the queue of the associated ReadableStream. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/TransformStreamDefaultController/desiredSize) + */ + get desiredSize(): number | null; + /** + * The **`enqueue()`** method of the TransformStreamDefaultController interface enqueues the given chunk in the readable side of the stream. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/TransformStreamDefaultController/enqueue) + */ + enqueue(chunk?: O): void; + /** + * The **`error()`** method of the TransformStreamDefaultController interface errors both sides of the stream. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/TransformStreamDefaultController/error) + */ + error(reason: any): void; + /** + * The **`terminate()`** method of the TransformStreamDefaultController interface closes the readable side and errors the writable side of the stream. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/TransformStreamDefaultController/terminate) + */ + terminate(): void; +} +interface ReadableWritablePair { + readable: ReadableStream; + /** + * Provides a convenient, chainable way of piping this readable stream through a transform stream (or any other { writable, readable } pair). It simply pipes the stream into the writable side of the supplied pair, and returns the readable side for further use. + * + * Piping a stream will lock it for the duration of the pipe, preventing any other consumer from acquiring a reader. + */ + writable: WritableStream; +} +/** + * The **`WritableStream`** interface of the Streams API provides a standard abstraction for writing streaming data to a destination, known as a sink. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/WritableStream) + */ +declare class WritableStream { + constructor(underlyingSink?: UnderlyingSink, queuingStrategy?: QueuingStrategy); + /** + * The **`locked`** read-only property of the WritableStream interface returns a boolean indicating whether the `WritableStream` is locked to a writer. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/WritableStream/locked) + */ + get locked(): boolean; + /** + * The **`abort()`** method of the WritableStream interface aborts the stream, signaling that the producer can no longer successfully write to the stream and it is to be immediately moved to an error state, with any queued writes discarded. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/WritableStream/abort) + */ + abort(reason?: any): Promise; + /** + * The **`close()`** method of the WritableStream interface closes the associated stream. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/WritableStream/close) + */ + close(): Promise; + /** + * The **`getWriter()`** method of the WritableStream interface returns a new instance of WritableStreamDefaultWriter and locks the stream to that instance. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/WritableStream/getWriter) + */ + getWriter(): WritableStreamDefaultWriter; +} +/** + * The **`WritableStreamDefaultWriter`** interface of the Streams API is the object returned by WritableStream.getWriter() and once created locks the writer to the `WritableStream` ensuring that no other streams can write to the underlying sink. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/WritableStreamDefaultWriter) + */ +declare class WritableStreamDefaultWriter { + constructor(stream: WritableStream); + /** + * The **`closed`** read-only property of the the stream errors or the writer's lock is released. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/WritableStreamDefaultWriter/closed) + */ + get closed(): Promise; + /** + * The **`ready`** read-only property of the that resolves when the desired size of the stream's internal queue transitions from non-positive to positive, signaling that it is no longer applying backpressure. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/WritableStreamDefaultWriter/ready) + */ + get ready(): Promise; + /** + * The **`desiredSize`** read-only property of the to fill the stream's internal queue. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/WritableStreamDefaultWriter/desiredSize) + */ + get desiredSize(): number | null; + /** + * The **`abort()`** method of the the producer can no longer successfully write to the stream and it is to be immediately moved to an error state, with any queued writes discarded. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/WritableStreamDefaultWriter/abort) + */ + abort(reason?: any): Promise; + /** + * The **`close()`** method of the stream. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/WritableStreamDefaultWriter/close) + */ + close(): Promise; + /** + * The **`write()`** method of the operation. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/WritableStreamDefaultWriter/write) + */ + write(chunk?: W): Promise; + /** + * The **`releaseLock()`** method of the corresponding stream. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/WritableStreamDefaultWriter/releaseLock) + */ + releaseLock(): void; +} +/** + * The **`TransformStream`** interface of the Streams API represents a concrete implementation of the pipe chain _transform stream_ concept. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/TransformStream) + */ +declare class TransformStream { + constructor(transformer?: Transformer, writableStrategy?: QueuingStrategy, readableStrategy?: QueuingStrategy); + /** + * The **`readable`** read-only property of the TransformStream interface returns the ReadableStream instance controlled by this `TransformStream`. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/TransformStream/readable) + */ + get readable(): ReadableStream; + /** + * The **`writable`** read-only property of the TransformStream interface returns the WritableStream instance controlled by this `TransformStream`. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/TransformStream/writable) + */ + get writable(): WritableStream; +} +declare class FixedLengthStream extends IdentityTransformStream { + constructor(expectedLength: number | bigint, queuingStrategy?: IdentityTransformStreamQueuingStrategy); +} +declare class IdentityTransformStream extends TransformStream { + constructor(queuingStrategy?: IdentityTransformStreamQueuingStrategy); +} +interface IdentityTransformStreamQueuingStrategy { + highWaterMark?: (number | bigint); +} +interface ReadableStreamValuesOptions { + preventCancel?: boolean; +} +/** + * The **`CompressionStream`** interface of the Compression Streams API is an API for compressing a stream of data. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/CompressionStream) + */ +declare class CompressionStream extends TransformStream { + constructor(format: "gzip" | "deflate" | "deflate-raw"); +} +/** + * The **`DecompressionStream`** interface of the Compression Streams API is an API for decompressing a stream of data. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/DecompressionStream) + */ +declare class DecompressionStream extends TransformStream { + constructor(format: "gzip" | "deflate" | "deflate-raw"); +} +/** + * The **`TextEncoderStream`** interface of the Encoding API converts a stream of strings into bytes in the UTF-8 encoding. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/TextEncoderStream) + */ +declare class TextEncoderStream extends TransformStream { + constructor(); + get encoding(): string; +} +/** + * The **`TextDecoderStream`** interface of the Encoding API converts a stream of text in a binary encoding, such as UTF-8 etc., to a stream of strings. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/TextDecoderStream) + */ +declare class TextDecoderStream extends TransformStream { + constructor(label?: string, options?: TextDecoderStreamTextDecoderStreamInit); + get encoding(): string; + get fatal(): boolean; + get ignoreBOM(): boolean; +} +interface TextDecoderStreamTextDecoderStreamInit { + fatal?: boolean; + ignoreBOM?: boolean; +} +/** + * The **`ByteLengthQueuingStrategy`** interface of the Streams API provides a built-in byte length queuing strategy that can be used when constructing streams. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/ByteLengthQueuingStrategy) + */ +declare class ByteLengthQueuingStrategy implements QueuingStrategy { + constructor(init: QueuingStrategyInit); + /** + * The read-only **`ByteLengthQueuingStrategy.highWaterMark`** property returns the total number of bytes that can be contained in the internal queue before backpressure is applied. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/ByteLengthQueuingStrategy/highWaterMark) + */ + get highWaterMark(): number; + /* [MDN Reference](https://developer.mozilla.org/docs/Web/API/ByteLengthQueuingStrategy/size) */ + get size(): (chunk?: any) => number; +} +/** + * The **`CountQueuingStrategy`** interface of the Streams API provides a built-in chunk counting queuing strategy that can be used when constructing streams. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/CountQueuingStrategy) + */ +declare class CountQueuingStrategy implements QueuingStrategy { + constructor(init: QueuingStrategyInit); + /** + * The read-only **`CountQueuingStrategy.highWaterMark`** property returns the total number of chunks that can be contained in the internal queue before backpressure is applied. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/CountQueuingStrategy/highWaterMark) + */ + get highWaterMark(): number; + /* [MDN Reference](https://developer.mozilla.org/docs/Web/API/CountQueuingStrategy/size) */ + get size(): (chunk?: any) => number; +} +interface QueuingStrategyInit { + /** + * Creates a new ByteLengthQueuingStrategy with the provided high water mark. + * + * Note that the provided high water mark will not be validated ahead of time. Instead, if it is negative, NaN, or not a number, the resulting ByteLengthQueuingStrategy will cause the corresponding stream constructor to throw. + */ + highWaterMark: number; +} +interface ScriptVersion { + id?: string; + tag?: string; + message?: string; +} +declare abstract class TailEvent extends ExtendableEvent { + readonly events: TraceItem[]; + readonly traces: TraceItem[]; +} +interface TraceItem { + readonly event: (TraceItemFetchEventInfo | TraceItemJsRpcEventInfo | TraceItemScheduledEventInfo | TraceItemAlarmEventInfo | TraceItemQueueEventInfo | TraceItemEmailEventInfo | TraceItemTailEventInfo | TraceItemCustomEventInfo | TraceItemHibernatableWebSocketEventInfo) | null; + readonly eventTimestamp: number | null; + readonly logs: TraceLog[]; + readonly exceptions: TraceException[]; + readonly diagnosticsChannelEvents: TraceDiagnosticChannelEvent[]; + readonly scriptName: string | null; + readonly entrypoint?: string; + readonly scriptVersion?: ScriptVersion; + readonly dispatchNamespace?: string; + readonly scriptTags?: string[]; + readonly durableObjectId?: string; + readonly outcome: string; + readonly executionModel: string; + readonly truncated: boolean; + readonly cpuTime: number; + readonly wallTime: number; +} +interface TraceItemAlarmEventInfo { + readonly scheduledTime: Date; +} +interface TraceItemCustomEventInfo { +} +interface TraceItemScheduledEventInfo { + readonly scheduledTime: number; + readonly cron: string; +} +interface TraceItemQueueEventInfo { + readonly queue: string; + readonly batchSize: number; +} +interface TraceItemEmailEventInfo { + readonly mailFrom: string; + readonly rcptTo: string; + readonly rawSize: number; +} +interface TraceItemTailEventInfo { + readonly consumedEvents: TraceItemTailEventInfoTailItem[]; +} +interface TraceItemTailEventInfoTailItem { + readonly scriptName: string | null; +} +interface TraceItemFetchEventInfo { + readonly response?: TraceItemFetchEventInfoResponse; + readonly request: TraceItemFetchEventInfoRequest; +} +interface TraceItemFetchEventInfoRequest { + readonly cf?: any; + readonly headers: Record; + readonly method: string; + readonly url: string; + getUnredacted(): TraceItemFetchEventInfoRequest; +} +interface TraceItemFetchEventInfoResponse { + readonly status: number; +} +interface TraceItemJsRpcEventInfo { + readonly rpcMethod: string; +} +interface TraceItemHibernatableWebSocketEventInfo { + readonly getWebSocketEvent: TraceItemHibernatableWebSocketEventInfoMessage | TraceItemHibernatableWebSocketEventInfoClose | TraceItemHibernatableWebSocketEventInfoError; +} +interface TraceItemHibernatableWebSocketEventInfoMessage { + readonly webSocketEventType: string; +} +interface TraceItemHibernatableWebSocketEventInfoClose { + readonly webSocketEventType: string; + readonly code: number; + readonly wasClean: boolean; +} +interface TraceItemHibernatableWebSocketEventInfoError { + readonly webSocketEventType: string; +} +interface TraceLog { + readonly timestamp: number; + readonly level: string; + readonly message: any; +} +interface TraceException { + readonly timestamp: number; + readonly message: string; + readonly name: string; + readonly stack?: string; +} +interface TraceDiagnosticChannelEvent { + readonly timestamp: number; + readonly channel: string; + readonly message: any; +} +interface TraceMetrics { + readonly cpuTime: number; + readonly wallTime: number; +} +interface UnsafeTraceMetrics { + fromTrace(item: TraceItem): TraceMetrics; +} +/** + * The **`URL`** interface is used to parse, construct, normalize, and encode URL. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/URL) + */ +declare class URL { + constructor(url: string | URL, base?: string | URL); + /** + * The **`origin`** read-only property of the URL interface returns a string containing the Unicode serialization of the origin of the represented URL. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/URL/origin) + */ + get origin(): string; + /** + * The **`href`** property of the URL interface is a string containing the whole URL. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/URL/href) + */ + get href(): string; + /** + * The **`href`** property of the URL interface is a string containing the whole URL. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/URL/href) + */ + set href(value: string); + /** + * The **`protocol`** property of the URL interface is a string containing the protocol or scheme of the URL, including the final `':'`. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/URL/protocol) + */ + get protocol(): string; + /** + * The **`protocol`** property of the URL interface is a string containing the protocol or scheme of the URL, including the final `':'`. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/URL/protocol) + */ + set protocol(value: string); + /** + * The **`username`** property of the URL interface is a string containing the username component of the URL. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/URL/username) + */ + get username(): string; + /** + * The **`username`** property of the URL interface is a string containing the username component of the URL. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/URL/username) + */ + set username(value: string); + /** + * The **`password`** property of the URL interface is a string containing the password component of the URL. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/URL/password) + */ + get password(): string; + /** + * The **`password`** property of the URL interface is a string containing the password component of the URL. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/URL/password) + */ + set password(value: string); + /** + * The **`host`** property of the URL interface is a string containing the host, which is the URL.hostname, and then, if the port of the URL is nonempty, a `':'`, followed by the URL.port of the URL. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/URL/host) + */ + get host(): string; + /** + * The **`host`** property of the URL interface is a string containing the host, which is the URL.hostname, and then, if the port of the URL is nonempty, a `':'`, followed by the URL.port of the URL. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/URL/host) + */ + set host(value: string); + /** + * The **`hostname`** property of the URL interface is a string containing either the domain name or IP address of the URL. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/URL/hostname) + */ + get hostname(): string; + /** + * The **`hostname`** property of the URL interface is a string containing either the domain name or IP address of the URL. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/URL/hostname) + */ + set hostname(value: string); + /** + * The **`port`** property of the URL interface is a string containing the port number of the URL. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/URL/port) + */ + get port(): string; + /** + * The **`port`** property of the URL interface is a string containing the port number of the URL. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/URL/port) + */ + set port(value: string); + /** + * The **`pathname`** property of the URL interface represents a location in a hierarchical structure. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/URL/pathname) + */ + get pathname(): string; + /** + * The **`pathname`** property of the URL interface represents a location in a hierarchical structure. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/URL/pathname) + */ + set pathname(value: string); + /** + * The **`search`** property of the URL interface is a search string, also called a _query string_, that is a string containing a `'?'` followed by the parameters of the URL. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/URL/search) + */ + get search(): string; + /** + * The **`search`** property of the URL interface is a search string, also called a _query string_, that is a string containing a `'?'` followed by the parameters of the URL. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/URL/search) + */ + set search(value: string); + /** + * The **`hash`** property of the URL interface is a string containing a `'#'` followed by the fragment identifier of the URL. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/URL/hash) + */ + get hash(): string; + /** + * The **`hash`** property of the URL interface is a string containing a `'#'` followed by the fragment identifier of the URL. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/URL/hash) + */ + set hash(value: string); + /** + * The **`searchParams`** read-only property of the access to the [MISSING: httpmethod('GET')] decoded query arguments contained in the URL. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/URL/searchParams) + */ + get searchParams(): URLSearchParams; + /** + * The **`toJSON()`** method of the URL interface returns a string containing a serialized version of the URL, although in practice it seems to have the same effect as ```js-nolint toJSON() ``` None. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/URL/toJSON) + */ + toJSON(): string; + /*function toString() { [native code] }*/ + toString(): string; + /** + * The **`URL.canParse()`** static method of the URL interface returns a boolean indicating whether or not an absolute URL, or a relative URL combined with a base URL, are parsable and valid. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/URL/canParse_static) + */ + static canParse(url: string, base?: string): boolean; + /** + * The **`URL.parse()`** static method of the URL interface returns a newly created URL object representing the URL defined by the parameters. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/URL/parse_static) + */ + static parse(url: string, base?: string): URL | null; + /** + * The **`createObjectURL()`** static method of the URL interface creates a string containing a URL representing the object given in the parameter. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/URL/createObjectURL_static) + */ + static createObjectURL(object: File | Blob): string; + /** + * The **`revokeObjectURL()`** static method of the URL interface releases an existing object URL which was previously created by calling Call this method when you've finished using an object URL to let the browser know not to keep the reference to the file any longer. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/URL/revokeObjectURL_static) + */ + static revokeObjectURL(object_url: string): void; +} +/** + * The **`URLSearchParams`** interface defines utility methods to work with the query string of a URL. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/URLSearchParams) + */ +declare class URLSearchParams { + constructor(init?: (Iterable> | Record | string)); + /** + * The **`size`** read-only property of the URLSearchParams interface indicates the total number of search parameter entries. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/URLSearchParams/size) + */ + get size(): number; + /** + * The **`append()`** method of the URLSearchParams interface appends a specified key/value pair as a new search parameter. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/URLSearchParams/append) + */ + append(name: string, value: string): void; + /** + * The **`delete()`** method of the URLSearchParams interface deletes specified parameters and their associated value(s) from the list of all search parameters. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/URLSearchParams/delete) + */ + delete(name: string, value?: string): void; + /** + * The **`get()`** method of the URLSearchParams interface returns the first value associated to the given search parameter. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/URLSearchParams/get) + */ + get(name: string): string | null; + /** + * The **`getAll()`** method of the URLSearchParams interface returns all the values associated with a given search parameter as an array. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/URLSearchParams/getAll) + */ + getAll(name: string): string[]; + /** + * The **`has()`** method of the URLSearchParams interface returns a boolean value that indicates whether the specified parameter is in the search parameters. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/URLSearchParams/has) + */ + has(name: string, value?: string): boolean; + /** + * The **`set()`** method of the URLSearchParams interface sets the value associated with a given search parameter to the given value. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/URLSearchParams/set) + */ + set(name: string, value: string): void; + /** + * The **`URLSearchParams.sort()`** method sorts all key/value pairs contained in this object in place and returns `undefined`. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/URLSearchParams/sort) + */ + sort(): void; + /* Returns an array of key, value pairs for every entry in the search params. */ + entries(): IterableIterator<[ + key: string, + value: string + ]>; + /* Returns a list of keys in the search params. */ + keys(): IterableIterator; + /* Returns a list of values in the search params. */ + values(): IterableIterator; + forEach(callback: (this: This, value: string, key: string, parent: URLSearchParams) => void, thisArg?: This): void; + /*function toString() { [native code] }*/ + toString(): string; + [Symbol.iterator](): IterableIterator<[ + key: string, + value: string + ]>; +} +declare class URLPattern { + constructor(input?: (string | URLPatternInit), baseURL?: (string | URLPatternOptions), patternOptions?: URLPatternOptions); + get protocol(): string; + get username(): string; + get password(): string; + get hostname(): string; + get port(): string; + get pathname(): string; + get search(): string; + get hash(): string; + test(input?: (string | URLPatternInit), baseURL?: string): boolean; + exec(input?: (string | URLPatternInit), baseURL?: string): URLPatternResult | null; +} +interface URLPatternInit { + protocol?: string; + username?: string; + password?: string; + hostname?: string; + port?: string; + pathname?: string; + search?: string; + hash?: string; + baseURL?: string; +} +interface URLPatternComponentResult { + input: string; + groups: Record; +} +interface URLPatternResult { + inputs: (string | URLPatternInit)[]; + protocol: URLPatternComponentResult; + username: URLPatternComponentResult; + password: URLPatternComponentResult; + hostname: URLPatternComponentResult; + port: URLPatternComponentResult; + pathname: URLPatternComponentResult; + search: URLPatternComponentResult; + hash: URLPatternComponentResult; +} +interface URLPatternOptions { + ignoreCase?: boolean; +} +/** + * A `CloseEvent` is sent to clients using WebSockets when the connection is closed. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/CloseEvent) + */ +declare class CloseEvent extends Event { + constructor(type: string, initializer?: CloseEventInit); + /** + * The **`code`** read-only property of the CloseEvent interface returns a WebSocket connection close code indicating the reason the connection was closed. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/CloseEvent/code) + */ + readonly code: number; + /** + * The **`reason`** read-only property of the CloseEvent interface returns the WebSocket connection close reason the server gave for closing the connection; that is, a concise human-readable prose explanation for the closure. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/CloseEvent/reason) + */ + readonly reason: string; + /** + * The **`wasClean`** read-only property of the CloseEvent interface returns `true` if the connection closed cleanly. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/CloseEvent/wasClean) + */ + readonly wasClean: boolean; +} +interface CloseEventInit { + code?: number; + reason?: string; + wasClean?: boolean; +} +type WebSocketEventMap = { + close: CloseEvent; + message: MessageEvent; + open: Event; + error: ErrorEvent; +}; +/** + * The `WebSocket` object provides the API for creating and managing a WebSocket connection to a server, as well as for sending and receiving data on the connection. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/WebSocket) + */ +declare var WebSocket: { + prototype: WebSocket; + new (url: string, protocols?: (string[] | string)): WebSocket; + readonly READY_STATE_CONNECTING: number; + readonly CONNECTING: number; + readonly READY_STATE_OPEN: number; + readonly OPEN: number; + readonly READY_STATE_CLOSING: number; + readonly CLOSING: number; + readonly READY_STATE_CLOSED: number; + readonly CLOSED: number; +}; +/** + * The `WebSocket` object provides the API for creating and managing a WebSocket connection to a server, as well as for sending and receiving data on the connection. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/WebSocket) + */ +interface WebSocket extends EventTarget { + accept(): void; + /** + * The **`WebSocket.send()`** method enqueues the specified data to be transmitted to the server over the WebSocket connection, increasing the value of `bufferedAmount` by the number of bytes needed to contain the data. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/WebSocket/send) + */ + send(message: (ArrayBuffer | ArrayBufferView) | string): void; + /** + * The **`WebSocket.close()`** method closes the already `CLOSED`, this method does nothing. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/WebSocket/close) + */ + close(code?: number, reason?: string): void; + serializeAttachment(attachment: any): void; + deserializeAttachment(): any | null; + /** + * The **`WebSocket.readyState`** read-only property returns the current state of the WebSocket connection. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/WebSocket/readyState) + */ + readyState: number; + /** + * The **`WebSocket.url`** read-only property returns the absolute URL of the WebSocket as resolved by the constructor. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/WebSocket/url) + */ + url: string | null; + /** + * The **`WebSocket.protocol`** read-only property returns the name of the sub-protocol the server selected; this will be one of the strings specified in the `protocols` parameter when creating the WebSocket object, or the empty string if no connection is established. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/WebSocket/protocol) + */ + protocol: string | null; + /** + * The **`WebSocket.extensions`** read-only property returns the extensions selected by the server. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/WebSocket/extensions) + */ + extensions: string | null; +} +declare const WebSocketPair: { + new (): { + 0: WebSocket; + 1: WebSocket; + }; +}; +interface SqlStorage { + exec>(query: string, ...bindings: any[]): SqlStorageCursor; + get databaseSize(): number; + Cursor: typeof SqlStorageCursor; + Statement: typeof SqlStorageStatement; +} +declare abstract class SqlStorageStatement { +} +type SqlStorageValue = ArrayBuffer | string | number | null; +declare abstract class SqlStorageCursor> { + next(): { + done?: false; + value: T; + } | { + done: true; + value?: never; + }; + toArray(): T[]; + one(): T; + raw(): IterableIterator; + columnNames: string[]; + get rowsRead(): number; + get rowsWritten(): number; + [Symbol.iterator](): IterableIterator; +} +interface Socket { + get readable(): ReadableStream; + get writable(): WritableStream; + get closed(): Promise; + get opened(): Promise; + get upgraded(): boolean; + get secureTransport(): "on" | "off" | "starttls"; + close(): Promise; + startTls(options?: TlsOptions): Socket; +} +interface SocketOptions { + secureTransport?: string; + allowHalfOpen: boolean; + highWaterMark?: (number | bigint); +} +interface SocketAddress { + hostname: string; + port: number; +} +interface TlsOptions { + expectedServerHostname?: string; +} +interface SocketInfo { + remoteAddress?: string; + localAddress?: string; +} +/** + * The **`EventSource`** interface is web content's interface to server-sent events. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/EventSource) + */ +declare class EventSource extends EventTarget { + constructor(url: string, init?: EventSourceEventSourceInit); + /** + * The **`close()`** method of the EventSource interface closes the connection, if one is made, and sets the ```js-nolint close() ``` None. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/EventSource/close) + */ + close(): void; + /** + * The **`url`** read-only property of the URL of the source. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/EventSource/url) + */ + get url(): string; + /** + * The **`withCredentials`** read-only property of the the `EventSource` object was instantiated with CORS credentials set. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/EventSource/withCredentials) + */ + get withCredentials(): boolean; + /** + * The **`readyState`** read-only property of the connection. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/EventSource/readyState) + */ + get readyState(): number; + /* [MDN Reference](https://developer.mozilla.org/docs/Web/API/EventSource/open_event) */ + get onopen(): any | null; + /* [MDN Reference](https://developer.mozilla.org/docs/Web/API/EventSource/open_event) */ + set onopen(value: any | null); + /* [MDN Reference](https://developer.mozilla.org/docs/Web/API/EventSource/message_event) */ + get onmessage(): any | null; + /* [MDN Reference](https://developer.mozilla.org/docs/Web/API/EventSource/message_event) */ + set onmessage(value: any | null); + /* [MDN Reference](https://developer.mozilla.org/docs/Web/API/EventSource/error_event) */ + get onerror(): any | null; + /* [MDN Reference](https://developer.mozilla.org/docs/Web/API/EventSource/error_event) */ + set onerror(value: any | null); + static readonly CONNECTING: number; + static readonly OPEN: number; + static readonly CLOSED: number; + static from(stream: ReadableStream): EventSource; +} +interface EventSourceEventSourceInit { + withCredentials?: boolean; + fetcher?: Fetcher; +} +interface Container { + get running(): boolean; + start(options?: ContainerStartupOptions): void; + monitor(): Promise; + destroy(error?: any): Promise; + signal(signo: number): void; + getTcpPort(port: number): Fetcher; + setInactivityTimeout(durationMs: number | bigint): Promise; +} +interface ContainerStartupOptions { + entrypoint?: string[]; + enableInternet: boolean; + env?: Record; + hardTimeout?: (number | bigint); +} +/** + * The **`MessagePort`** interface of the Channel Messaging API represents one of the two ports of a MessageChannel, allowing messages to be sent from one port and listening out for them arriving at the other. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/MessagePort) + */ +interface MessagePort extends EventTarget { + /** + * The **`postMessage()`** method of the transfers ownership of objects to other browsing contexts. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/MessagePort/postMessage) + */ + postMessage(data?: any, options?: (any[] | MessagePortPostMessageOptions)): void; + /** + * The **`close()`** method of the MessagePort interface disconnects the port, so it is no longer active. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/MessagePort/close) + */ + close(): void; + /** + * The **`start()`** method of the MessagePort interface starts the sending of messages queued on the port. + * + * [MDN Reference](https://developer.mozilla.org/docs/Web/API/MessagePort/start) + */ + start(): void; + get onmessage(): any | null; + set onmessage(value: any | null); +} +interface MessagePortPostMessageOptions { + transfer?: any[]; +} +type LoopbackForExport Rpc.EntrypointBranded) | ExportedHandler | undefined = undefined> = T extends new (...args: any[]) => Rpc.WorkerEntrypointBranded ? LoopbackServiceStub> : T extends new (...args: any[]) => Rpc.DurableObjectBranded ? LoopbackDurableObjectClass> : T extends ExportedHandler ? LoopbackServiceStub : undefined; +type LoopbackServiceStub = Fetcher & (T extends CloudflareWorkersModule.WorkerEntrypoint ? (opts: { + props?: Props; +}) => Fetcher : (opts: { + props?: any; +}) => Fetcher); +type LoopbackDurableObjectClass = DurableObjectClass & (T extends CloudflareWorkersModule.DurableObject ? (opts: { + props?: Props; +}) => DurableObjectClass : (opts: { + props?: any; +}) => DurableObjectClass); +interface SyncKvStorage { + get(key: string): T | undefined; + list(options?: SyncKvListOptions): Iterable<[ + string, + T + ]>; + put(key: string, value: T): void; + delete(key: string): boolean; +} +interface SyncKvListOptions { + start?: string; + startAfter?: string; + end?: string; + prefix?: string; + reverse?: boolean; + limit?: number; +} +interface WorkerStub { + getEntrypoint(name?: string, options?: WorkerStubEntrypointOptions): Fetcher; +} +interface WorkerStubEntrypointOptions { + props?: any; +} +interface WorkerLoader { + get(name: string | null, getCode: () => WorkerLoaderWorkerCode | Promise): WorkerStub; +} +interface WorkerLoaderModule { + js?: string; + cjs?: string; + text?: string; + data?: ArrayBuffer; + json?: any; + py?: string; + wasm?: ArrayBuffer; +} +interface WorkerLoaderWorkerCode { + compatibilityDate: string; + compatibilityFlags?: string[]; + allowExperimental?: boolean; + mainModule: string; + modules: Record; + env?: any; + globalOutbound?: (Fetcher | null); + tails?: Fetcher[]; + streamingTails?: Fetcher[]; +} +/** +* The Workers runtime supports a subset of the Performance API, used to measure timing and performance, +* as well as timing of subrequests and other operations. +* +* [Cloudflare Docs Reference](https://developers.cloudflare.com/workers/runtime-apis/performance/) +*/ +declare abstract class Performance { + /* [Cloudflare Docs Reference](https://developers.cloudflare.com/workers/runtime-apis/performance/#performancetimeorigin) */ + get timeOrigin(): number; + /* [Cloudflare Docs Reference](https://developers.cloudflare.com/workers/runtime-apis/performance/#performancenow) */ + now(): number; +} +type AiImageClassificationInput = { + image: number[]; +}; +type AiImageClassificationOutput = { + score?: number; + label?: string; +}[]; +declare abstract class BaseAiImageClassification { + inputs: AiImageClassificationInput; + postProcessedOutputs: AiImageClassificationOutput; +} +type AiImageToTextInput = { + image: number[]; + prompt?: string; + max_tokens?: number; + temperature?: number; + top_p?: number; + top_k?: number; + seed?: number; + repetition_penalty?: number; + frequency_penalty?: number; + presence_penalty?: number; + raw?: boolean; + messages?: RoleScopedChatInput[]; +}; +type AiImageToTextOutput = { + description: string; +}; +declare abstract class BaseAiImageToText { + inputs: AiImageToTextInput; + postProcessedOutputs: AiImageToTextOutput; +} +type AiImageTextToTextInput = { + image: string; + prompt?: string; + max_tokens?: number; + temperature?: number; + ignore_eos?: boolean; + top_p?: number; + top_k?: number; + seed?: number; + repetition_penalty?: number; + frequency_penalty?: number; + presence_penalty?: number; + raw?: boolean; + messages?: RoleScopedChatInput[]; +}; +type AiImageTextToTextOutput = { + description: string; +}; +declare abstract class BaseAiImageTextToText { + inputs: AiImageTextToTextInput; + postProcessedOutputs: AiImageTextToTextOutput; +} +type AiMultimodalEmbeddingsInput = { + image: string; + text: string[]; +}; +type AiIMultimodalEmbeddingsOutput = { + data: number[][]; + shape: number[]; +}; +declare abstract class BaseAiMultimodalEmbeddings { + inputs: AiImageTextToTextInput; + postProcessedOutputs: AiImageTextToTextOutput; +} +type AiObjectDetectionInput = { + image: number[]; +}; +type AiObjectDetectionOutput = { + score?: number; + label?: string; +}[]; +declare abstract class BaseAiObjectDetection { + inputs: AiObjectDetectionInput; + postProcessedOutputs: AiObjectDetectionOutput; +} +type AiSentenceSimilarityInput = { + source: string; + sentences: string[]; +}; +type AiSentenceSimilarityOutput = number[]; +declare abstract class BaseAiSentenceSimilarity { + inputs: AiSentenceSimilarityInput; + postProcessedOutputs: AiSentenceSimilarityOutput; +} +type AiAutomaticSpeechRecognitionInput = { + audio: number[]; +}; +type AiAutomaticSpeechRecognitionOutput = { + text?: string; + words?: { + word: string; + start: number; + end: number; + }[]; + vtt?: string; +}; +declare abstract class BaseAiAutomaticSpeechRecognition { + inputs: AiAutomaticSpeechRecognitionInput; + postProcessedOutputs: AiAutomaticSpeechRecognitionOutput; +} +type AiSummarizationInput = { + input_text: string; + max_length?: number; +}; +type AiSummarizationOutput = { + summary: string; +}; +declare abstract class BaseAiSummarization { + inputs: AiSummarizationInput; + postProcessedOutputs: AiSummarizationOutput; +} +type AiTextClassificationInput = { + text: string; +}; +type AiTextClassificationOutput = { + score?: number; + label?: string; +}[]; +declare abstract class BaseAiTextClassification { + inputs: AiTextClassificationInput; + postProcessedOutputs: AiTextClassificationOutput; +} +type AiTextEmbeddingsInput = { + text: string | string[]; +}; +type AiTextEmbeddingsOutput = { + shape: number[]; + data: number[][]; +}; +declare abstract class BaseAiTextEmbeddings { + inputs: AiTextEmbeddingsInput; + postProcessedOutputs: AiTextEmbeddingsOutput; +} +type RoleScopedChatInput = { + role: "user" | "assistant" | "system" | "tool" | (string & NonNullable); + content: string; + name?: string; +}; +type AiTextGenerationToolLegacyInput = { + name: string; + description: string; + parameters?: { + type: "object" | (string & NonNullable); + properties: { + [key: string]: { + type: string; + description?: string; + }; + }; + required: string[]; + }; +}; +type AiTextGenerationToolInput = { + type: "function" | (string & NonNullable); + function: { + name: string; + description: string; + parameters?: { + type: "object" | (string & NonNullable); + properties: { + [key: string]: { + type: string; + description?: string; + }; + }; + required: string[]; + }; + }; +}; +type AiTextGenerationFunctionsInput = { + name: string; + code: string; +}; +type AiTextGenerationResponseFormat = { + type: string; + json_schema?: any; +}; +type AiTextGenerationInput = { + prompt?: string; + raw?: boolean; + stream?: boolean; + max_tokens?: number; + temperature?: number; + top_p?: number; + top_k?: number; + seed?: number; + repetition_penalty?: number; + frequency_penalty?: number; + presence_penalty?: number; + messages?: RoleScopedChatInput[]; + response_format?: AiTextGenerationResponseFormat; + tools?: AiTextGenerationToolInput[] | AiTextGenerationToolLegacyInput[] | (object & NonNullable); + functions?: AiTextGenerationFunctionsInput[]; +}; +type AiTextGenerationToolLegacyOutput = { + name: string; + arguments: unknown; +}; +type AiTextGenerationToolOutput = { + id: string; + type: "function"; + function: { + name: string; + arguments: string; + }; +}; +type UsageTags = { + prompt_tokens: number; + completion_tokens: number; + total_tokens: number; +}; +type AiTextGenerationOutput = { + response?: string; + tool_calls?: AiTextGenerationToolLegacyOutput[] & AiTextGenerationToolOutput[]; + usage?: UsageTags; +}; +declare abstract class BaseAiTextGeneration { + inputs: AiTextGenerationInput; + postProcessedOutputs: AiTextGenerationOutput; +} +type AiTextToSpeechInput = { + prompt: string; + lang?: string; +}; +type AiTextToSpeechOutput = Uint8Array | { + audio: string; +}; +declare abstract class BaseAiTextToSpeech { + inputs: AiTextToSpeechInput; + postProcessedOutputs: AiTextToSpeechOutput; +} +type AiTextToImageInput = { + prompt: string; + negative_prompt?: string; + height?: number; + width?: number; + image?: number[]; + image_b64?: string; + mask?: number[]; + num_steps?: number; + strength?: number; + guidance?: number; + seed?: number; +}; +type AiTextToImageOutput = ReadableStream; +declare abstract class BaseAiTextToImage { + inputs: AiTextToImageInput; + postProcessedOutputs: AiTextToImageOutput; +} +type AiTranslationInput = { + text: string; + target_lang: string; + source_lang?: string; +}; +type AiTranslationOutput = { + translated_text?: string; +}; +declare abstract class BaseAiTranslation { + inputs: AiTranslationInput; + postProcessedOutputs: AiTranslationOutput; +} +/** + * Workers AI support for OpenAI's Responses API + * Reference: https://github.com/openai/openai-node/blob/master/src/resources/responses/responses.ts + * + * It's a stripped down version from its source. + * It currently supports basic function calling, json mode and accepts images as input. + * + * It does not include types for WebSearch, CodeInterpreter, FileInputs, MCP, CustomTools. + * We plan to add those incrementally as model + platform capabilities evolve. + */ +type ResponsesInput = { + background?: boolean | null; + conversation?: string | ResponseConversationParam | null; + include?: Array | null; + input?: string | ResponseInput; + instructions?: string | null; + max_output_tokens?: number | null; + parallel_tool_calls?: boolean | null; + previous_response_id?: string | null; + prompt_cache_key?: string; + reasoning?: Reasoning | null; + safety_identifier?: string; + service_tier?: "auto" | "default" | "flex" | "scale" | "priority" | null; + stream?: boolean | null; + stream_options?: StreamOptions | null; + temperature?: number | null; + text?: ResponseTextConfig; + tool_choice?: ToolChoiceOptions | ToolChoiceFunction; + tools?: Array; + top_p?: number | null; + truncation?: "auto" | "disabled" | null; +}; +type ResponsesOutput = { + id?: string; + created_at?: number; + output_text?: string; + error?: ResponseError | null; + incomplete_details?: ResponseIncompleteDetails | null; + instructions?: string | Array | null; + object?: "response"; + output?: Array; + parallel_tool_calls?: boolean; + temperature?: number | null; + tool_choice?: ToolChoiceOptions | ToolChoiceFunction; + tools?: Array; + top_p?: number | null; + max_output_tokens?: number | null; + previous_response_id?: string | null; + prompt?: ResponsePrompt | null; + reasoning?: Reasoning | null; + safety_identifier?: string; + service_tier?: "auto" | "default" | "flex" | "scale" | "priority" | null; + status?: ResponseStatus; + text?: ResponseTextConfig; + truncation?: "auto" | "disabled" | null; + usage?: ResponseUsage; +}; +type EasyInputMessage = { + content: string | ResponseInputMessageContentList; + role: "user" | "assistant" | "system" | "developer"; + type?: "message"; +}; +type ResponsesFunctionTool = { + name: string; + parameters: { + [key: string]: unknown; + } | null; + strict: boolean | null; + type: "function"; + description?: string | null; +}; +type ResponseIncompleteDetails = { + reason?: "max_output_tokens" | "content_filter"; +}; +type ResponsePrompt = { + id: string; + variables?: { + [key: string]: string | ResponseInputText | ResponseInputImage; + } | null; + version?: string | null; +}; +type Reasoning = { + effort?: ReasoningEffort | null; + generate_summary?: "auto" | "concise" | "detailed" | null; + summary?: "auto" | "concise" | "detailed" | null; +}; +type ResponseContent = ResponseInputText | ResponseInputImage | ResponseOutputText | ResponseOutputRefusal | ResponseContentReasoningText; +type ResponseContentReasoningText = { + text: string; + type: "reasoning_text"; +}; +type ResponseConversationParam = { + id: string; +}; +type ResponseCreatedEvent = { + response: Response; + sequence_number: number; + type: "response.created"; +}; +type ResponseCustomToolCallOutput = { + call_id: string; + output: string | Array; + type: "custom_tool_call_output"; + id?: string; +}; +type ResponseError = { + code: "server_error" | "rate_limit_exceeded" | "invalid_prompt" | "vector_store_timeout" | "invalid_image" | "invalid_image_format" | "invalid_base64_image" | "invalid_image_url" | "image_too_large" | "image_too_small" | "image_parse_error" | "image_content_policy_violation" | "invalid_image_mode" | "image_file_too_large" | "unsupported_image_media_type" | "empty_image_file" | "failed_to_download_image" | "image_file_not_found"; + message: string; +}; +type ResponseErrorEvent = { + code: string | null; + message: string; + param: string | null; + sequence_number: number; + type: "error"; +}; +type ResponseFailedEvent = { + response: Response; + sequence_number: number; + type: "response.failed"; +}; +type ResponseFormatText = { + type: "text"; +}; +type ResponseFormatJSONObject = { + type: "json_object"; +}; +type ResponseFormatTextConfig = ResponseFormatText | ResponseFormatTextJSONSchemaConfig | ResponseFormatJSONObject; +type ResponseFormatTextJSONSchemaConfig = { + name: string; + schema: { + [key: string]: unknown; + }; + type: "json_schema"; + description?: string; + strict?: boolean | null; +}; +type ResponseFunctionCallArgumentsDeltaEvent = { + delta: string; + item_id: string; + output_index: number; + sequence_number: number; + type: "response.function_call_arguments.delta"; +}; +type ResponseFunctionCallArgumentsDoneEvent = { + arguments: string; + item_id: string; + name: string; + output_index: number; + sequence_number: number; + type: "response.function_call_arguments.done"; +}; +type ResponseFunctionCallOutputItem = ResponseInputTextContent | ResponseInputImageContent; +type ResponseFunctionCallOutputItemList = Array; +type ResponseFunctionToolCall = { + arguments: string; + call_id: string; + name: string; + type: "function_call"; + id?: string; + status?: "in_progress" | "completed" | "incomplete"; +}; +interface ResponseFunctionToolCallItem extends ResponseFunctionToolCall { + id: string; +} +type ResponseFunctionToolCallOutputItem = { + id: string; + call_id: string; + output: string | Array; + type: "function_call_output"; + status?: "in_progress" | "completed" | "incomplete"; +}; +type ResponseIncludable = "message.input_image.image_url" | "message.output_text.logprobs"; +type ResponseIncompleteEvent = { + response: Response; + sequence_number: number; + type: "response.incomplete"; +}; +type ResponseInput = Array; +type ResponseInputContent = ResponseInputText | ResponseInputImage; +type ResponseInputImage = { + detail: "low" | "high" | "auto"; + type: "input_image"; + /** + * Base64 encoded image + */ + image_url?: string | null; +}; +type ResponseInputImageContent = { + type: "input_image"; + detail?: "low" | "high" | "auto" | null; + /** + * Base64 encoded image + */ + image_url?: string | null; +}; +type ResponseInputItem = EasyInputMessage | ResponseInputItemMessage | ResponseOutputMessage | ResponseFunctionToolCall | ResponseInputItemFunctionCallOutput | ResponseReasoningItem; +type ResponseInputItemFunctionCallOutput = { + call_id: string; + output: string | ResponseFunctionCallOutputItemList; + type: "function_call_output"; + id?: string | null; + status?: "in_progress" | "completed" | "incomplete" | null; +}; +type ResponseInputItemMessage = { + content: ResponseInputMessageContentList; + role: "user" | "system" | "developer"; + status?: "in_progress" | "completed" | "incomplete"; + type?: "message"; +}; +type ResponseInputMessageContentList = Array; +type ResponseInputMessageItem = { + id: string; + content: ResponseInputMessageContentList; + role: "user" | "system" | "developer"; + status?: "in_progress" | "completed" | "incomplete"; + type?: "message"; +}; +type ResponseInputText = { + text: string; + type: "input_text"; +}; +type ResponseInputTextContent = { + text: string; + type: "input_text"; +}; +type ResponseItem = ResponseInputMessageItem | ResponseOutputMessage | ResponseFunctionToolCallItem | ResponseFunctionToolCallOutputItem; +type ResponseOutputItem = ResponseOutputMessage | ResponseFunctionToolCall | ResponseReasoningItem; +type ResponseOutputItemAddedEvent = { + item: ResponseOutputItem; + output_index: number; + sequence_number: number; + type: "response.output_item.added"; +}; +type ResponseOutputItemDoneEvent = { + item: ResponseOutputItem; + output_index: number; + sequence_number: number; + type: "response.output_item.done"; +}; +type ResponseOutputMessage = { + id: string; + content: Array; + role: "assistant"; + status: "in_progress" | "completed" | "incomplete"; + type: "message"; +}; +type ResponseOutputRefusal = { + refusal: string; + type: "refusal"; +}; +type ResponseOutputText = { + text: string; + type: "output_text"; + logprobs?: Array; +}; +type ResponseReasoningItem = { + id: string; + summary: Array; + type: "reasoning"; + content?: Array; + encrypted_content?: string | null; + status?: "in_progress" | "completed" | "incomplete"; +}; +type ResponseReasoningSummaryItem = { + text: string; + type: "summary_text"; +}; +type ResponseReasoningContentItem = { + text: string; + type: "reasoning_text"; +}; +type ResponseReasoningTextDeltaEvent = { + content_index: number; + delta: string; + item_id: string; + output_index: number; + sequence_number: number; + type: "response.reasoning_text.delta"; +}; +type ResponseReasoningTextDoneEvent = { + content_index: number; + item_id: string; + output_index: number; + sequence_number: number; + text: string; + type: "response.reasoning_text.done"; +}; +type ResponseRefusalDeltaEvent = { + content_index: number; + delta: string; + item_id: string; + output_index: number; + sequence_number: number; + type: "response.refusal.delta"; +}; +type ResponseRefusalDoneEvent = { + content_index: number; + item_id: string; + output_index: number; + refusal: string; + sequence_number: number; + type: "response.refusal.done"; +}; +type ResponseStatus = "completed" | "failed" | "in_progress" | "cancelled" | "queued" | "incomplete"; +type ResponseStreamEvent = ResponseCompletedEvent | ResponseCreatedEvent | ResponseErrorEvent | ResponseFunctionCallArgumentsDeltaEvent | ResponseFunctionCallArgumentsDoneEvent | ResponseFailedEvent | ResponseIncompleteEvent | ResponseOutputItemAddedEvent | ResponseOutputItemDoneEvent | ResponseReasoningTextDeltaEvent | ResponseReasoningTextDoneEvent | ResponseRefusalDeltaEvent | ResponseRefusalDoneEvent | ResponseTextDeltaEvent | ResponseTextDoneEvent; +type ResponseCompletedEvent = { + response: Response; + sequence_number: number; + type: "response.completed"; +}; +type ResponseTextConfig = { + format?: ResponseFormatTextConfig; + verbosity?: "low" | "medium" | "high" | null; +}; +type ResponseTextDeltaEvent = { + content_index: number; + delta: string; + item_id: string; + logprobs: Array; + output_index: number; + sequence_number: number; + type: "response.output_text.delta"; +}; +type ResponseTextDoneEvent = { + content_index: number; + item_id: string; + logprobs: Array; + output_index: number; + sequence_number: number; + text: string; + type: "response.output_text.done"; +}; +type Logprob = { + token: string; + logprob: number; + top_logprobs?: Array; +}; +type TopLogprob = { + token?: string; + logprob?: number; +}; +type ResponseUsage = { + input_tokens: number; + output_tokens: number; + total_tokens: number; +}; +type Tool = ResponsesFunctionTool; +type ToolChoiceFunction = { + name: string; + type: "function"; +}; +type ToolChoiceOptions = "none"; +type ReasoningEffort = "minimal" | "low" | "medium" | "high" | null; +type StreamOptions = { + include_obfuscation?: boolean; +}; +type Ai_Cf_Baai_Bge_Base_En_V1_5_Input = { + text: string | string[]; + /** + * The pooling method used in the embedding process. `cls` pooling will generate more accurate embeddings on larger inputs - however, embeddings created with cls pooling are not compatible with embeddings generated with mean pooling. The default pooling method is `mean` in order for this to not be a breaking change, but we highly suggest using the new `cls` pooling for better accuracy. + */ + pooling?: "mean" | "cls"; +} | { + /** + * Batch of the embeddings requests to run using async-queue + */ + requests: { + text: string | string[]; + /** + * The pooling method used in the embedding process. `cls` pooling will generate more accurate embeddings on larger inputs - however, embeddings created with cls pooling are not compatible with embeddings generated with mean pooling. The default pooling method is `mean` in order for this to not be a breaking change, but we highly suggest using the new `cls` pooling for better accuracy. + */ + pooling?: "mean" | "cls"; + }[]; +}; +type Ai_Cf_Baai_Bge_Base_En_V1_5_Output = { + shape?: number[]; + /** + * Embeddings of the requested text values + */ + data?: number[][]; + /** + * The pooling method used in the embedding process. + */ + pooling?: "mean" | "cls"; +} | Ai_Cf_Baai_Bge_Base_En_V1_5_AsyncResponse; +interface Ai_Cf_Baai_Bge_Base_En_V1_5_AsyncResponse { + /** + * The async request id that can be used to obtain the results. + */ + request_id?: string; +} +declare abstract class Base_Ai_Cf_Baai_Bge_Base_En_V1_5 { + inputs: Ai_Cf_Baai_Bge_Base_En_V1_5_Input; + postProcessedOutputs: Ai_Cf_Baai_Bge_Base_En_V1_5_Output; +} +type Ai_Cf_Openai_Whisper_Input = string | { + /** + * An array of integers that represent the audio data constrained to 8-bit unsigned integer values + */ + audio: number[]; +}; +interface Ai_Cf_Openai_Whisper_Output { + /** + * The transcription + */ + text: string; + word_count?: number; + words?: { + word?: string; + /** + * The second this word begins in the recording + */ + start?: number; + /** + * The ending second when the word completes + */ + end?: number; + }[]; + vtt?: string; +} +declare abstract class Base_Ai_Cf_Openai_Whisper { + inputs: Ai_Cf_Openai_Whisper_Input; + postProcessedOutputs: Ai_Cf_Openai_Whisper_Output; +} +type Ai_Cf_Meta_M2M100_1_2B_Input = { + /** + * The text to be translated + */ + text: string; + /** + * The language code of the source text (e.g., 'en' for English). Defaults to 'en' if not specified + */ + source_lang?: string; + /** + * The language code to translate the text into (e.g., 'es' for Spanish) + */ + target_lang: string; +} | { + /** + * Batch of the embeddings requests to run using async-queue + */ + requests: { + /** + * The text to be translated + */ + text: string; + /** + * The language code of the source text (e.g., 'en' for English). Defaults to 'en' if not specified + */ + source_lang?: string; + /** + * The language code to translate the text into (e.g., 'es' for Spanish) + */ + target_lang: string; + }[]; +}; +type Ai_Cf_Meta_M2M100_1_2B_Output = { + /** + * The translated text in the target language + */ + translated_text?: string; +} | Ai_Cf_Meta_M2M100_1_2B_AsyncResponse; +interface Ai_Cf_Meta_M2M100_1_2B_AsyncResponse { + /** + * The async request id that can be used to obtain the results. + */ + request_id?: string; +} +declare abstract class Base_Ai_Cf_Meta_M2M100_1_2B { + inputs: Ai_Cf_Meta_M2M100_1_2B_Input; + postProcessedOutputs: Ai_Cf_Meta_M2M100_1_2B_Output; +} +type Ai_Cf_Baai_Bge_Small_En_V1_5_Input = { + text: string | string[]; + /** + * The pooling method used in the embedding process. `cls` pooling will generate more accurate embeddings on larger inputs - however, embeddings created with cls pooling are not compatible with embeddings generated with mean pooling. The default pooling method is `mean` in order for this to not be a breaking change, but we highly suggest using the new `cls` pooling for better accuracy. + */ + pooling?: "mean" | "cls"; +} | { + /** + * Batch of the embeddings requests to run using async-queue + */ + requests: { + text: string | string[]; + /** + * The pooling method used in the embedding process. `cls` pooling will generate more accurate embeddings on larger inputs - however, embeddings created with cls pooling are not compatible with embeddings generated with mean pooling. The default pooling method is `mean` in order for this to not be a breaking change, but we highly suggest using the new `cls` pooling for better accuracy. + */ + pooling?: "mean" | "cls"; + }[]; +}; +type Ai_Cf_Baai_Bge_Small_En_V1_5_Output = { + shape?: number[]; + /** + * Embeddings of the requested text values + */ + data?: number[][]; + /** + * The pooling method used in the embedding process. + */ + pooling?: "mean" | "cls"; +} | Ai_Cf_Baai_Bge_Small_En_V1_5_AsyncResponse; +interface Ai_Cf_Baai_Bge_Small_En_V1_5_AsyncResponse { + /** + * The async request id that can be used to obtain the results. + */ + request_id?: string; +} +declare abstract class Base_Ai_Cf_Baai_Bge_Small_En_V1_5 { + inputs: Ai_Cf_Baai_Bge_Small_En_V1_5_Input; + postProcessedOutputs: Ai_Cf_Baai_Bge_Small_En_V1_5_Output; +} +type Ai_Cf_Baai_Bge_Large_En_V1_5_Input = { + text: string | string[]; + /** + * The pooling method used in the embedding process. `cls` pooling will generate more accurate embeddings on larger inputs - however, embeddings created with cls pooling are not compatible with embeddings generated with mean pooling. The default pooling method is `mean` in order for this to not be a breaking change, but we highly suggest using the new `cls` pooling for better accuracy. + */ + pooling?: "mean" | "cls"; +} | { + /** + * Batch of the embeddings requests to run using async-queue + */ + requests: { + text: string | string[]; + /** + * The pooling method used in the embedding process. `cls` pooling will generate more accurate embeddings on larger inputs - however, embeddings created with cls pooling are not compatible with embeddings generated with mean pooling. The default pooling method is `mean` in order for this to not be a breaking change, but we highly suggest using the new `cls` pooling for better accuracy. + */ + pooling?: "mean" | "cls"; + }[]; +}; +type Ai_Cf_Baai_Bge_Large_En_V1_5_Output = { + shape?: number[]; + /** + * Embeddings of the requested text values + */ + data?: number[][]; + /** + * The pooling method used in the embedding process. + */ + pooling?: "mean" | "cls"; +} | Ai_Cf_Baai_Bge_Large_En_V1_5_AsyncResponse; +interface Ai_Cf_Baai_Bge_Large_En_V1_5_AsyncResponse { + /** + * The async request id that can be used to obtain the results. + */ + request_id?: string; +} +declare abstract class Base_Ai_Cf_Baai_Bge_Large_En_V1_5 { + inputs: Ai_Cf_Baai_Bge_Large_En_V1_5_Input; + postProcessedOutputs: Ai_Cf_Baai_Bge_Large_En_V1_5_Output; +} +type Ai_Cf_Unum_Uform_Gen2_Qwen_500M_Input = string | { + /** + * The input text prompt for the model to generate a response. + */ + prompt?: string; + /** + * If true, a chat template is not applied and you must adhere to the specific model's expected formatting. + */ + raw?: boolean; + /** + * Controls the creativity of the AI's responses by adjusting how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses. + */ + top_p?: number; + /** + * Limits the AI to choose from the top 'k' most probable words. Lower values make responses more focused; higher values introduce more variety and potential surprises. + */ + top_k?: number; + /** + * Random seed for reproducibility of the generation. + */ + seed?: number; + /** + * Penalty for repeated tokens; higher values discourage repetition. + */ + repetition_penalty?: number; + /** + * Decreases the likelihood of the model repeating the same lines verbatim. + */ + frequency_penalty?: number; + /** + * Increases the likelihood of the model introducing new topics. + */ + presence_penalty?: number; + image: number[] | (string & NonNullable); + /** + * The maximum number of tokens to generate in the response. + */ + max_tokens?: number; +}; +interface Ai_Cf_Unum_Uform_Gen2_Qwen_500M_Output { + description?: string; +} +declare abstract class Base_Ai_Cf_Unum_Uform_Gen2_Qwen_500M { + inputs: Ai_Cf_Unum_Uform_Gen2_Qwen_500M_Input; + postProcessedOutputs: Ai_Cf_Unum_Uform_Gen2_Qwen_500M_Output; +} +type Ai_Cf_Openai_Whisper_Tiny_En_Input = string | { + /** + * An array of integers that represent the audio data constrained to 8-bit unsigned integer values + */ + audio: number[]; +}; +interface Ai_Cf_Openai_Whisper_Tiny_En_Output { + /** + * The transcription + */ + text: string; + word_count?: number; + words?: { + word?: string; + /** + * The second this word begins in the recording + */ + start?: number; + /** + * The ending second when the word completes + */ + end?: number; + }[]; + vtt?: string; +} +declare abstract class Base_Ai_Cf_Openai_Whisper_Tiny_En { + inputs: Ai_Cf_Openai_Whisper_Tiny_En_Input; + postProcessedOutputs: Ai_Cf_Openai_Whisper_Tiny_En_Output; +} +interface Ai_Cf_Openai_Whisper_Large_V3_Turbo_Input { + /** + * Base64 encoded value of the audio data. + */ + audio: string; + /** + * Supported tasks are 'translate' or 'transcribe'. + */ + task?: string; + /** + * The language of the audio being transcribed or translated. + */ + language?: string; + /** + * Preprocess the audio with a voice activity detection model. + */ + vad_filter?: boolean; + /** + * A text prompt to help provide context to the model on the contents of the audio. + */ + initial_prompt?: string; + /** + * The prefix it appended the the beginning of the output of the transcription and can guide the transcription result. + */ + prefix?: string; +} +interface Ai_Cf_Openai_Whisper_Large_V3_Turbo_Output { + transcription_info?: { + /** + * The language of the audio being transcribed or translated. + */ + language?: string; + /** + * The confidence level or probability of the detected language being accurate, represented as a decimal between 0 and 1. + */ + language_probability?: number; + /** + * The total duration of the original audio file, in seconds. + */ + duration?: number; + /** + * The duration of the audio after applying Voice Activity Detection (VAD) to remove silent or irrelevant sections, in seconds. + */ + duration_after_vad?: number; + }; + /** + * The complete transcription of the audio. + */ + text: string; + /** + * The total number of words in the transcription. + */ + word_count?: number; + segments?: { + /** + * The starting time of the segment within the audio, in seconds. + */ + start?: number; + /** + * The ending time of the segment within the audio, in seconds. + */ + end?: number; + /** + * The transcription of the segment. + */ + text?: string; + /** + * The temperature used in the decoding process, controlling randomness in predictions. Lower values result in more deterministic outputs. + */ + temperature?: number; + /** + * The average log probability of the predictions for the words in this segment, indicating overall confidence. + */ + avg_logprob?: number; + /** + * The compression ratio of the input to the output, measuring how much the text was compressed during the transcription process. + */ + compression_ratio?: number; + /** + * The probability that the segment contains no speech, represented as a decimal between 0 and 1. + */ + no_speech_prob?: number; + words?: { + /** + * The individual word transcribed from the audio. + */ + word?: string; + /** + * The starting time of the word within the audio, in seconds. + */ + start?: number; + /** + * The ending time of the word within the audio, in seconds. + */ + end?: number; + }[]; + }[]; + /** + * The transcription in WebVTT format, which includes timing and text information for use in subtitles. + */ + vtt?: string; +} +declare abstract class Base_Ai_Cf_Openai_Whisper_Large_V3_Turbo { + inputs: Ai_Cf_Openai_Whisper_Large_V3_Turbo_Input; + postProcessedOutputs: Ai_Cf_Openai_Whisper_Large_V3_Turbo_Output; +} +type Ai_Cf_Baai_Bge_M3_Input = Ai_Cf_Baai_Bge_M3_Input_QueryAnd_Contexts | Ai_Cf_Baai_Bge_M3_Input_Embedding | { + /** + * Batch of the embeddings requests to run using async-queue + */ + requests: (Ai_Cf_Baai_Bge_M3_Input_QueryAnd_Contexts_1 | Ai_Cf_Baai_Bge_M3_Input_Embedding_1)[]; +}; +interface Ai_Cf_Baai_Bge_M3_Input_QueryAnd_Contexts { + /** + * A query you wish to perform against the provided contexts. If no query is provided the model with respond with embeddings for contexts + */ + query?: string; + /** + * List of provided contexts. Note that the index in this array is important, as the response will refer to it. + */ + contexts: { + /** + * One of the provided context content + */ + text?: string; + }[]; + /** + * When provided with too long context should the model error out or truncate the context to fit? + */ + truncate_inputs?: boolean; +} +interface Ai_Cf_Baai_Bge_M3_Input_Embedding { + text: string | string[]; + /** + * When provided with too long context should the model error out or truncate the context to fit? + */ + truncate_inputs?: boolean; +} +interface Ai_Cf_Baai_Bge_M3_Input_QueryAnd_Contexts_1 { + /** + * A query you wish to perform against the provided contexts. If no query is provided the model with respond with embeddings for contexts + */ + query?: string; + /** + * List of provided contexts. Note that the index in this array is important, as the response will refer to it. + */ + contexts: { + /** + * One of the provided context content + */ + text?: string; + }[]; + /** + * When provided with too long context should the model error out or truncate the context to fit? + */ + truncate_inputs?: boolean; +} +interface Ai_Cf_Baai_Bge_M3_Input_Embedding_1 { + text: string | string[]; + /** + * When provided with too long context should the model error out or truncate the context to fit? + */ + truncate_inputs?: boolean; +} +type Ai_Cf_Baai_Bge_M3_Output = Ai_Cf_Baai_Bge_M3_Ouput_Query | Ai_Cf_Baai_Bge_M3_Output_EmbeddingFor_Contexts | Ai_Cf_Baai_Bge_M3_Ouput_Embedding | Ai_Cf_Baai_Bge_M3_AsyncResponse; +interface Ai_Cf_Baai_Bge_M3_Ouput_Query { + response?: { + /** + * Index of the context in the request + */ + id?: number; + /** + * Score of the context under the index. + */ + score?: number; + }[]; +} +interface Ai_Cf_Baai_Bge_M3_Output_EmbeddingFor_Contexts { + response?: number[][]; + shape?: number[]; + /** + * The pooling method used in the embedding process. + */ + pooling?: "mean" | "cls"; +} +interface Ai_Cf_Baai_Bge_M3_Ouput_Embedding { + shape?: number[]; + /** + * Embeddings of the requested text values + */ + data?: number[][]; + /** + * The pooling method used in the embedding process. + */ + pooling?: "mean" | "cls"; +} +interface Ai_Cf_Baai_Bge_M3_AsyncResponse { + /** + * The async request id that can be used to obtain the results. + */ + request_id?: string; +} +declare abstract class Base_Ai_Cf_Baai_Bge_M3 { + inputs: Ai_Cf_Baai_Bge_M3_Input; + postProcessedOutputs: Ai_Cf_Baai_Bge_M3_Output; +} +interface Ai_Cf_Black_Forest_Labs_Flux_1_Schnell_Input { + /** + * A text description of the image you want to generate. + */ + prompt: string; + /** + * The number of diffusion steps; higher values can improve quality but take longer. + */ + steps?: number; +} +interface Ai_Cf_Black_Forest_Labs_Flux_1_Schnell_Output { + /** + * The generated image in Base64 format. + */ + image?: string; +} +declare abstract class Base_Ai_Cf_Black_Forest_Labs_Flux_1_Schnell { + inputs: Ai_Cf_Black_Forest_Labs_Flux_1_Schnell_Input; + postProcessedOutputs: Ai_Cf_Black_Forest_Labs_Flux_1_Schnell_Output; +} +type Ai_Cf_Meta_Llama_3_2_11B_Vision_Instruct_Input = Ai_Cf_Meta_Llama_3_2_11B_Vision_Instruct_Prompt | Ai_Cf_Meta_Llama_3_2_11B_Vision_Instruct_Messages; +interface Ai_Cf_Meta_Llama_3_2_11B_Vision_Instruct_Prompt { + /** + * The input text prompt for the model to generate a response. + */ + prompt: string; + image?: number[] | (string & NonNullable); + /** + * If true, a chat template is not applied and you must adhere to the specific model's expected formatting. + */ + raw?: boolean; + /** + * If true, the response will be streamed back incrementally using SSE, Server Sent Events. + */ + stream?: boolean; + /** + * The maximum number of tokens to generate in the response. + */ + max_tokens?: number; + /** + * Controls the randomness of the output; higher values produce more random results. + */ + temperature?: number; + /** + * Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses. + */ + top_p?: number; + /** + * Limits the AI to choose from the top 'k' most probable words. Lower values make responses more focused; higher values introduce more variety and potential surprises. + */ + top_k?: number; + /** + * Random seed for reproducibility of the generation. + */ + seed?: number; + /** + * Penalty for repeated tokens; higher values discourage repetition. + */ + repetition_penalty?: number; + /** + * Decreases the likelihood of the model repeating the same lines verbatim. + */ + frequency_penalty?: number; + /** + * Increases the likelihood of the model introducing new topics. + */ + presence_penalty?: number; + /** + * Name of the LoRA (Low-Rank Adaptation) model to fine-tune the base model. + */ + lora?: string; +} +interface Ai_Cf_Meta_Llama_3_2_11B_Vision_Instruct_Messages { + /** + * An array of message objects representing the conversation history. + */ + messages: { + /** + * The role of the message sender (e.g., 'user', 'assistant', 'system', 'tool'). + */ + role?: string; + /** + * The tool call id. Must be supplied for tool calls for Mistral-3. If you don't know what to put here you can fall back to 000000001 + */ + tool_call_id?: string; + content?: string | { + /** + * Type of the content provided + */ + type?: string; + text?: string; + image_url?: { + /** + * image uri with data (e.g. data:image/jpeg;base64,/9j/...). HTTP URL will not be accepted + */ + url?: string; + }; + }[] | { + /** + * Type of the content provided + */ + type?: string; + text?: string; + image_url?: { + /** + * image uri with data (e.g. data:image/jpeg;base64,/9j/...). HTTP URL will not be accepted + */ + url?: string; + }; + }; + }[]; + image?: number[] | (string & NonNullable); + functions?: { + name: string; + code: string; + }[]; + /** + * A list of tools available for the assistant to use. + */ + tools?: ({ + /** + * The name of the tool. More descriptive the better. + */ + name: string; + /** + * A brief description of what the tool does. + */ + description: string; + /** + * Schema defining the parameters accepted by the tool. + */ + parameters: { + /** + * The type of the parameters object (usually 'object'). + */ + type: string; + /** + * List of required parameter names. + */ + required?: string[]; + /** + * Definitions of each parameter. + */ + properties: { + [k: string]: { + /** + * The data type of the parameter. + */ + type: string; + /** + * A description of the expected parameter. + */ + description: string; + }; + }; + }; + } | { + /** + * Specifies the type of tool (e.g., 'function'). + */ + type: string; + /** + * Details of the function tool. + */ + function: { + /** + * The name of the function. + */ + name: string; + /** + * A brief description of what the function does. + */ + description: string; + /** + * Schema defining the parameters accepted by the function. + */ + parameters: { + /** + * The type of the parameters object (usually 'object'). + */ + type: string; + /** + * List of required parameter names. + */ + required?: string[]; + /** + * Definitions of each parameter. + */ + properties: { + [k: string]: { + /** + * The data type of the parameter. + */ + type: string; + /** + * A description of the expected parameter. + */ + description: string; + }; + }; + }; + }; + })[]; + /** + * If true, the response will be streamed back incrementally. + */ + stream?: boolean; + /** + * The maximum number of tokens to generate in the response. + */ + max_tokens?: number; + /** + * Controls the randomness of the output; higher values produce more random results. + */ + temperature?: number; + /** + * Controls the creativity of the AI's responses by adjusting how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses. + */ + top_p?: number; + /** + * Limits the AI to choose from the top 'k' most probable words. Lower values make responses more focused; higher values introduce more variety and potential surprises. + */ + top_k?: number; + /** + * Random seed for reproducibility of the generation. + */ + seed?: number; + /** + * Penalty for repeated tokens; higher values discourage repetition. + */ + repetition_penalty?: number; + /** + * Decreases the likelihood of the model repeating the same lines verbatim. + */ + frequency_penalty?: number; + /** + * Increases the likelihood of the model introducing new topics. + */ + presence_penalty?: number; +} +type Ai_Cf_Meta_Llama_3_2_11B_Vision_Instruct_Output = { + /** + * The generated text response from the model + */ + response?: string; + /** + * An array of tool calls requests made during the response generation + */ + tool_calls?: { + /** + * The arguments passed to be passed to the tool call request + */ + arguments?: object; + /** + * The name of the tool to be called + */ + name?: string; + }[]; +}; +declare abstract class Base_Ai_Cf_Meta_Llama_3_2_11B_Vision_Instruct { + inputs: Ai_Cf_Meta_Llama_3_2_11B_Vision_Instruct_Input; + postProcessedOutputs: Ai_Cf_Meta_Llama_3_2_11B_Vision_Instruct_Output; +} +type Ai_Cf_Meta_Llama_3_3_70B_Instruct_Fp8_Fast_Input = Ai_Cf_Meta_Llama_3_3_70B_Instruct_Fp8_Fast_Prompt | Ai_Cf_Meta_Llama_3_3_70B_Instruct_Fp8_Fast_Messages | Ai_Cf_Meta_Llama_3_3_70B_Instruct_Fp8_Fast_Async_Batch; +interface Ai_Cf_Meta_Llama_3_3_70B_Instruct_Fp8_Fast_Prompt { + /** + * The input text prompt for the model to generate a response. + */ + prompt: string; + /** + * Name of the LoRA (Low-Rank Adaptation) model to fine-tune the base model. + */ + lora?: string; + response_format?: Ai_Cf_Meta_Llama_3_3_70B_Instruct_Fp8_Fast_JSON_Mode; + /** + * If true, a chat template is not applied and you must adhere to the specific model's expected formatting. + */ + raw?: boolean; + /** + * If true, the response will be streamed back incrementally using SSE, Server Sent Events. + */ + stream?: boolean; + /** + * The maximum number of tokens to generate in the response. + */ + max_tokens?: number; + /** + * Controls the randomness of the output; higher values produce more random results. + */ + temperature?: number; + /** + * Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses. + */ + top_p?: number; + /** + * Limits the AI to choose from the top 'k' most probable words. Lower values make responses more focused; higher values introduce more variety and potential surprises. + */ + top_k?: number; + /** + * Random seed for reproducibility of the generation. + */ + seed?: number; + /** + * Penalty for repeated tokens; higher values discourage repetition. + */ + repetition_penalty?: number; + /** + * Decreases the likelihood of the model repeating the same lines verbatim. + */ + frequency_penalty?: number; + /** + * Increases the likelihood of the model introducing new topics. + */ + presence_penalty?: number; +} +interface Ai_Cf_Meta_Llama_3_3_70B_Instruct_Fp8_Fast_JSON_Mode { + type?: "json_object" | "json_schema"; + json_schema?: unknown; +} +interface Ai_Cf_Meta_Llama_3_3_70B_Instruct_Fp8_Fast_Messages { + /** + * An array of message objects representing the conversation history. + */ + messages: { + /** + * The role of the message sender (e.g., 'user', 'assistant', 'system', 'tool'). + */ + role: string; + /** + * The content of the message as a string. + */ + content: string; + }[]; + functions?: { + name: string; + code: string; + }[]; + /** + * A list of tools available for the assistant to use. + */ + tools?: ({ + /** + * The name of the tool. More descriptive the better. + */ + name: string; + /** + * A brief description of what the tool does. + */ + description: string; + /** + * Schema defining the parameters accepted by the tool. + */ + parameters: { + /** + * The type of the parameters object (usually 'object'). + */ + type: string; + /** + * List of required parameter names. + */ + required?: string[]; + /** + * Definitions of each parameter. + */ + properties: { + [k: string]: { + /** + * The data type of the parameter. + */ + type: string; + /** + * A description of the expected parameter. + */ + description: string; + }; + }; + }; + } | { + /** + * Specifies the type of tool (e.g., 'function'). + */ + type: string; + /** + * Details of the function tool. + */ + function: { + /** + * The name of the function. + */ + name: string; + /** + * A brief description of what the function does. + */ + description: string; + /** + * Schema defining the parameters accepted by the function. + */ + parameters: { + /** + * The type of the parameters object (usually 'object'). + */ + type: string; + /** + * List of required parameter names. + */ + required?: string[]; + /** + * Definitions of each parameter. + */ + properties: { + [k: string]: { + /** + * The data type of the parameter. + */ + type: string; + /** + * A description of the expected parameter. + */ + description: string; + }; + }; + }; + }; + })[]; + response_format?: Ai_Cf_Meta_Llama_3_3_70B_Instruct_Fp8_Fast_JSON_Mode_1; + /** + * If true, a chat template is not applied and you must adhere to the specific model's expected formatting. + */ + raw?: boolean; + /** + * If true, the response will be streamed back incrementally using SSE, Server Sent Events. + */ + stream?: boolean; + /** + * The maximum number of tokens to generate in the response. + */ + max_tokens?: number; + /** + * Controls the randomness of the output; higher values produce more random results. + */ + temperature?: number; + /** + * Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses. + */ + top_p?: number; + /** + * Limits the AI to choose from the top 'k' most probable words. Lower values make responses more focused; higher values introduce more variety and potential surprises. + */ + top_k?: number; + /** + * Random seed for reproducibility of the generation. + */ + seed?: number; + /** + * Penalty for repeated tokens; higher values discourage repetition. + */ + repetition_penalty?: number; + /** + * Decreases the likelihood of the model repeating the same lines verbatim. + */ + frequency_penalty?: number; + /** + * Increases the likelihood of the model introducing new topics. + */ + presence_penalty?: number; +} +interface Ai_Cf_Meta_Llama_3_3_70B_Instruct_Fp8_Fast_JSON_Mode_1 { + type?: "json_object" | "json_schema"; + json_schema?: unknown; +} +interface Ai_Cf_Meta_Llama_3_3_70B_Instruct_Fp8_Fast_Async_Batch { + requests?: { + /** + * User-supplied reference. This field will be present in the response as well it can be used to reference the request and response. It's NOT validated to be unique. + */ + external_reference?: string; + /** + * Prompt for the text generation model + */ + prompt?: string; + /** + * If true, the response will be streamed back incrementally using SSE, Server Sent Events. + */ + stream?: boolean; + /** + * The maximum number of tokens to generate in the response. + */ + max_tokens?: number; + /** + * Controls the randomness of the output; higher values produce more random results. + */ + temperature?: number; + /** + * Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses. + */ + top_p?: number; + /** + * Random seed for reproducibility of the generation. + */ + seed?: number; + /** + * Penalty for repeated tokens; higher values discourage repetition. + */ + repetition_penalty?: number; + /** + * Decreases the likelihood of the model repeating the same lines verbatim. + */ + frequency_penalty?: number; + /** + * Increases the likelihood of the model introducing new topics. + */ + presence_penalty?: number; + response_format?: Ai_Cf_Meta_Llama_3_3_70B_Instruct_Fp8_Fast_JSON_Mode_2; + }[]; +} +interface Ai_Cf_Meta_Llama_3_3_70B_Instruct_Fp8_Fast_JSON_Mode_2 { + type?: "json_object" | "json_schema"; + json_schema?: unknown; +} +type Ai_Cf_Meta_Llama_3_3_70B_Instruct_Fp8_Fast_Output = { + /** + * The generated text response from the model + */ + response: string; + /** + * Usage statistics for the inference request + */ + usage?: { + /** + * Total number of tokens in input + */ + prompt_tokens?: number; + /** + * Total number of tokens in output + */ + completion_tokens?: number; + /** + * Total number of input and output tokens + */ + total_tokens?: number; + }; + /** + * An array of tool calls requests made during the response generation + */ + tool_calls?: { + /** + * The arguments passed to be passed to the tool call request + */ + arguments?: object; + /** + * The name of the tool to be called + */ + name?: string; + }[]; +} | string | Ai_Cf_Meta_Llama_3_3_70B_Instruct_Fp8_Fast_AsyncResponse; +interface Ai_Cf_Meta_Llama_3_3_70B_Instruct_Fp8_Fast_AsyncResponse { + /** + * The async request id that can be used to obtain the results. + */ + request_id?: string; +} +declare abstract class Base_Ai_Cf_Meta_Llama_3_3_70B_Instruct_Fp8_Fast { + inputs: Ai_Cf_Meta_Llama_3_3_70B_Instruct_Fp8_Fast_Input; + postProcessedOutputs: Ai_Cf_Meta_Llama_3_3_70B_Instruct_Fp8_Fast_Output; +} +interface Ai_Cf_Meta_Llama_Guard_3_8B_Input { + /** + * An array of message objects representing the conversation history. + */ + messages: { + /** + * The role of the message sender must alternate between 'user' and 'assistant'. + */ + role: "user" | "assistant"; + /** + * The content of the message as a string. + */ + content: string; + }[]; + /** + * The maximum number of tokens to generate in the response. + */ + max_tokens?: number; + /** + * Controls the randomness of the output; higher values produce more random results. + */ + temperature?: number; + /** + * Dictate the output format of the generated response. + */ + response_format?: { + /** + * Set to json_object to process and output generated text as JSON. + */ + type?: string; + }; +} +interface Ai_Cf_Meta_Llama_Guard_3_8B_Output { + response?: string | { + /** + * Whether the conversation is safe or not. + */ + safe?: boolean; + /** + * A list of what hazard categories predicted for the conversation, if the conversation is deemed unsafe. + */ + categories?: string[]; + }; + /** + * Usage statistics for the inference request + */ + usage?: { + /** + * Total number of tokens in input + */ + prompt_tokens?: number; + /** + * Total number of tokens in output + */ + completion_tokens?: number; + /** + * Total number of input and output tokens + */ + total_tokens?: number; + }; +} +declare abstract class Base_Ai_Cf_Meta_Llama_Guard_3_8B { + inputs: Ai_Cf_Meta_Llama_Guard_3_8B_Input; + postProcessedOutputs: Ai_Cf_Meta_Llama_Guard_3_8B_Output; +} +interface Ai_Cf_Baai_Bge_Reranker_Base_Input { + /** + * A query you wish to perform against the provided contexts. + */ + /** + * Number of returned results starting with the best score. + */ + top_k?: number; + /** + * List of provided contexts. Note that the index in this array is important, as the response will refer to it. + */ + contexts: { + /** + * One of the provided context content + */ + text?: string; + }[]; +} +interface Ai_Cf_Baai_Bge_Reranker_Base_Output { + response?: { + /** + * Index of the context in the request + */ + id?: number; + /** + * Score of the context under the index. + */ + score?: number; + }[]; +} +declare abstract class Base_Ai_Cf_Baai_Bge_Reranker_Base { + inputs: Ai_Cf_Baai_Bge_Reranker_Base_Input; + postProcessedOutputs: Ai_Cf_Baai_Bge_Reranker_Base_Output; +} +type Ai_Cf_Qwen_Qwen2_5_Coder_32B_Instruct_Input = Ai_Cf_Qwen_Qwen2_5_Coder_32B_Instruct_Prompt | Ai_Cf_Qwen_Qwen2_5_Coder_32B_Instruct_Messages; +interface Ai_Cf_Qwen_Qwen2_5_Coder_32B_Instruct_Prompt { + /** + * The input text prompt for the model to generate a response. + */ + prompt: string; + /** + * Name of the LoRA (Low-Rank Adaptation) model to fine-tune the base model. + */ + lora?: string; + response_format?: Ai_Cf_Qwen_Qwen2_5_Coder_32B_Instruct_JSON_Mode; + /** + * If true, a chat template is not applied and you must adhere to the specific model's expected formatting. + */ + raw?: boolean; + /** + * If true, the response will be streamed back incrementally using SSE, Server Sent Events. + */ + stream?: boolean; + /** + * The maximum number of tokens to generate in the response. + */ + max_tokens?: number; + /** + * Controls the randomness of the output; higher values produce more random results. + */ + temperature?: number; + /** + * Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses. + */ + top_p?: number; + /** + * Limits the AI to choose from the top 'k' most probable words. Lower values make responses more focused; higher values introduce more variety and potential surprises. + */ + top_k?: number; + /** + * Random seed for reproducibility of the generation. + */ + seed?: number; + /** + * Penalty for repeated tokens; higher values discourage repetition. + */ + repetition_penalty?: number; + /** + * Decreases the likelihood of the model repeating the same lines verbatim. + */ + frequency_penalty?: number; + /** + * Increases the likelihood of the model introducing new topics. + */ + presence_penalty?: number; +} +interface Ai_Cf_Qwen_Qwen2_5_Coder_32B_Instruct_JSON_Mode { + type?: "json_object" | "json_schema"; + json_schema?: unknown; +} +interface Ai_Cf_Qwen_Qwen2_5_Coder_32B_Instruct_Messages { + /** + * An array of message objects representing the conversation history. + */ + messages: { + /** + * The role of the message sender (e.g., 'user', 'assistant', 'system', 'tool'). + */ + role: string; + /** + * The content of the message as a string. + */ + content: string; + }[]; + functions?: { + name: string; + code: string; + }[]; + /** + * A list of tools available for the assistant to use. + */ + tools?: ({ + /** + * The name of the tool. More descriptive the better. + */ + name: string; + /** + * A brief description of what the tool does. + */ + description: string; + /** + * Schema defining the parameters accepted by the tool. + */ + parameters: { + /** + * The type of the parameters object (usually 'object'). + */ + type: string; + /** + * List of required parameter names. + */ + required?: string[]; + /** + * Definitions of each parameter. + */ + properties: { + [k: string]: { + /** + * The data type of the parameter. + */ + type: string; + /** + * A description of the expected parameter. + */ + description: string; + }; + }; + }; + } | { + /** + * Specifies the type of tool (e.g., 'function'). + */ + type: string; + /** + * Details of the function tool. + */ + function: { + /** + * The name of the function. + */ + name: string; + /** + * A brief description of what the function does. + */ + description: string; + /** + * Schema defining the parameters accepted by the function. + */ + parameters: { + /** + * The type of the parameters object (usually 'object'). + */ + type: string; + /** + * List of required parameter names. + */ + required?: string[]; + /** + * Definitions of each parameter. + */ + properties: { + [k: string]: { + /** + * The data type of the parameter. + */ + type: string; + /** + * A description of the expected parameter. + */ + description: string; + }; + }; + }; + }; + })[]; + response_format?: Ai_Cf_Qwen_Qwen2_5_Coder_32B_Instruct_JSON_Mode_1; + /** + * If true, a chat template is not applied and you must adhere to the specific model's expected formatting. + */ + raw?: boolean; + /** + * If true, the response will be streamed back incrementally using SSE, Server Sent Events. + */ + stream?: boolean; + /** + * The maximum number of tokens to generate in the response. + */ + max_tokens?: number; + /** + * Controls the randomness of the output; higher values produce more random results. + */ + temperature?: number; + /** + * Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses. + */ + top_p?: number; + /** + * Limits the AI to choose from the top 'k' most probable words. Lower values make responses more focused; higher values introduce more variety and potential surprises. + */ + top_k?: number; + /** + * Random seed for reproducibility of the generation. + */ + seed?: number; + /** + * Penalty for repeated tokens; higher values discourage repetition. + */ + repetition_penalty?: number; + /** + * Decreases the likelihood of the model repeating the same lines verbatim. + */ + frequency_penalty?: number; + /** + * Increases the likelihood of the model introducing new topics. + */ + presence_penalty?: number; +} +interface Ai_Cf_Qwen_Qwen2_5_Coder_32B_Instruct_JSON_Mode_1 { + type?: "json_object" | "json_schema"; + json_schema?: unknown; +} +type Ai_Cf_Qwen_Qwen2_5_Coder_32B_Instruct_Output = { + /** + * The generated text response from the model + */ + response: string; + /** + * Usage statistics for the inference request + */ + usage?: { + /** + * Total number of tokens in input + */ + prompt_tokens?: number; + /** + * Total number of tokens in output + */ + completion_tokens?: number; + /** + * Total number of input and output tokens + */ + total_tokens?: number; + }; + /** + * An array of tool calls requests made during the response generation + */ + tool_calls?: { + /** + * The arguments passed to be passed to the tool call request + */ + arguments?: object; + /** + * The name of the tool to be called + */ + name?: string; + }[]; +}; +declare abstract class Base_Ai_Cf_Qwen_Qwen2_5_Coder_32B_Instruct { + inputs: Ai_Cf_Qwen_Qwen2_5_Coder_32B_Instruct_Input; + postProcessedOutputs: Ai_Cf_Qwen_Qwen2_5_Coder_32B_Instruct_Output; +} +type Ai_Cf_Qwen_Qwq_32B_Input = Ai_Cf_Qwen_Qwq_32B_Prompt | Ai_Cf_Qwen_Qwq_32B_Messages; +interface Ai_Cf_Qwen_Qwq_32B_Prompt { + /** + * The input text prompt for the model to generate a response. + */ + prompt: string; + /** + * JSON schema that should be fulfilled for the response. + */ + guided_json?: object; + /** + * If true, a chat template is not applied and you must adhere to the specific model's expected formatting. + */ + raw?: boolean; + /** + * If true, the response will be streamed back incrementally using SSE, Server Sent Events. + */ + stream?: boolean; + /** + * The maximum number of tokens to generate in the response. + */ + max_tokens?: number; + /** + * Controls the randomness of the output; higher values produce more random results. + */ + temperature?: number; + /** + * Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses. + */ + top_p?: number; + /** + * Limits the AI to choose from the top 'k' most probable words. Lower values make responses more focused; higher values introduce more variety and potential surprises. + */ + top_k?: number; + /** + * Random seed for reproducibility of the generation. + */ + seed?: number; + /** + * Penalty for repeated tokens; higher values discourage repetition. + */ + repetition_penalty?: number; + /** + * Decreases the likelihood of the model repeating the same lines verbatim. + */ + frequency_penalty?: number; + /** + * Increases the likelihood of the model introducing new topics. + */ + presence_penalty?: number; +} +interface Ai_Cf_Qwen_Qwq_32B_Messages { + /** + * An array of message objects representing the conversation history. + */ + messages: { + /** + * The role of the message sender (e.g., 'user', 'assistant', 'system', 'tool'). + */ + role?: string; + /** + * The tool call id. Must be supplied for tool calls for Mistral-3. If you don't know what to put here you can fall back to 000000001 + */ + tool_call_id?: string; + content?: string | { + /** + * Type of the content provided + */ + type?: string; + text?: string; + image_url?: { + /** + * image uri with data (e.g. data:image/jpeg;base64,/9j/...). HTTP URL will not be accepted + */ + url?: string; + }; + }[] | { + /** + * Type of the content provided + */ + type?: string; + text?: string; + image_url?: { + /** + * image uri with data (e.g. data:image/jpeg;base64,/9j/...). HTTP URL will not be accepted + */ + url?: string; + }; + }; + }[]; + functions?: { + name: string; + code: string; + }[]; + /** + * A list of tools available for the assistant to use. + */ + tools?: ({ + /** + * The name of the tool. More descriptive the better. + */ + name: string; + /** + * A brief description of what the tool does. + */ + description: string; + /** + * Schema defining the parameters accepted by the tool. + */ + parameters: { + /** + * The type of the parameters object (usually 'object'). + */ + type: string; + /** + * List of required parameter names. + */ + required?: string[]; + /** + * Definitions of each parameter. + */ + properties: { + [k: string]: { + /** + * The data type of the parameter. + */ + type: string; + /** + * A description of the expected parameter. + */ + description: string; + }; + }; + }; + } | { + /** + * Specifies the type of tool (e.g., 'function'). + */ + type: string; + /** + * Details of the function tool. + */ + function: { + /** + * The name of the function. + */ + name: string; + /** + * A brief description of what the function does. + */ + description: string; + /** + * Schema defining the parameters accepted by the function. + */ + parameters: { + /** + * The type of the parameters object (usually 'object'). + */ + type: string; + /** + * List of required parameter names. + */ + required?: string[]; + /** + * Definitions of each parameter. + */ + properties: { + [k: string]: { + /** + * The data type of the parameter. + */ + type: string; + /** + * A description of the expected parameter. + */ + description: string; + }; + }; + }; + }; + })[]; + /** + * JSON schema that should be fulfilled for the response. + */ + guided_json?: object; + /** + * If true, a chat template is not applied and you must adhere to the specific model's expected formatting. + */ + raw?: boolean; + /** + * If true, the response will be streamed back incrementally using SSE, Server Sent Events. + */ + stream?: boolean; + /** + * The maximum number of tokens to generate in the response. + */ + max_tokens?: number; + /** + * Controls the randomness of the output; higher values produce more random results. + */ + temperature?: number; + /** + * Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses. + */ + top_p?: number; + /** + * Limits the AI to choose from the top 'k' most probable words. Lower values make responses more focused; higher values introduce more variety and potential surprises. + */ + top_k?: number; + /** + * Random seed for reproducibility of the generation. + */ + seed?: number; + /** + * Penalty for repeated tokens; higher values discourage repetition. + */ + repetition_penalty?: number; + /** + * Decreases the likelihood of the model repeating the same lines verbatim. + */ + frequency_penalty?: number; + /** + * Increases the likelihood of the model introducing new topics. + */ + presence_penalty?: number; +} +type Ai_Cf_Qwen_Qwq_32B_Output = { + /** + * The generated text response from the model + */ + response: string; + /** + * Usage statistics for the inference request + */ + usage?: { + /** + * Total number of tokens in input + */ + prompt_tokens?: number; + /** + * Total number of tokens in output + */ + completion_tokens?: number; + /** + * Total number of input and output tokens + */ + total_tokens?: number; + }; + /** + * An array of tool calls requests made during the response generation + */ + tool_calls?: { + /** + * The arguments passed to be passed to the tool call request + */ + arguments?: object; + /** + * The name of the tool to be called + */ + name?: string; + }[]; +}; +declare abstract class Base_Ai_Cf_Qwen_Qwq_32B { + inputs: Ai_Cf_Qwen_Qwq_32B_Input; + postProcessedOutputs: Ai_Cf_Qwen_Qwq_32B_Output; +} +type Ai_Cf_Mistralai_Mistral_Small_3_1_24B_Instruct_Input = Ai_Cf_Mistralai_Mistral_Small_3_1_24B_Instruct_Prompt | Ai_Cf_Mistralai_Mistral_Small_3_1_24B_Instruct_Messages; +interface Ai_Cf_Mistralai_Mistral_Small_3_1_24B_Instruct_Prompt { + /** + * The input text prompt for the model to generate a response. + */ + prompt: string; + /** + * JSON schema that should be fulfilled for the response. + */ + guided_json?: object; + /** + * If true, a chat template is not applied and you must adhere to the specific model's expected formatting. + */ + raw?: boolean; + /** + * If true, the response will be streamed back incrementally using SSE, Server Sent Events. + */ + stream?: boolean; + /** + * The maximum number of tokens to generate in the response. + */ + max_tokens?: number; + /** + * Controls the randomness of the output; higher values produce more random results. + */ + temperature?: number; + /** + * Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses. + */ + top_p?: number; + /** + * Limits the AI to choose from the top 'k' most probable words. Lower values make responses more focused; higher values introduce more variety and potential surprises. + */ + top_k?: number; + /** + * Random seed for reproducibility of the generation. + */ + seed?: number; + /** + * Penalty for repeated tokens; higher values discourage repetition. + */ + repetition_penalty?: number; + /** + * Decreases the likelihood of the model repeating the same lines verbatim. + */ + frequency_penalty?: number; + /** + * Increases the likelihood of the model introducing new topics. + */ + presence_penalty?: number; +} +interface Ai_Cf_Mistralai_Mistral_Small_3_1_24B_Instruct_Messages { + /** + * An array of message objects representing the conversation history. + */ + messages: { + /** + * The role of the message sender (e.g., 'user', 'assistant', 'system', 'tool'). + */ + role?: string; + /** + * The tool call id. Must be supplied for tool calls for Mistral-3. If you don't know what to put here you can fall back to 000000001 + */ + tool_call_id?: string; + content?: string | { + /** + * Type of the content provided + */ + type?: string; + text?: string; + image_url?: { + /** + * image uri with data (e.g. data:image/jpeg;base64,/9j/...). HTTP URL will not be accepted + */ + url?: string; + }; + }[] | { + /** + * Type of the content provided + */ + type?: string; + text?: string; + image_url?: { + /** + * image uri with data (e.g. data:image/jpeg;base64,/9j/...). HTTP URL will not be accepted + */ + url?: string; + }; + }; + }[]; + functions?: { + name: string; + code: string; + }[]; + /** + * A list of tools available for the assistant to use. + */ + tools?: ({ + /** + * The name of the tool. More descriptive the better. + */ + name: string; + /** + * A brief description of what the tool does. + */ + description: string; + /** + * Schema defining the parameters accepted by the tool. + */ + parameters: { + /** + * The type of the parameters object (usually 'object'). + */ + type: string; + /** + * List of required parameter names. + */ + required?: string[]; + /** + * Definitions of each parameter. + */ + properties: { + [k: string]: { + /** + * The data type of the parameter. + */ + type: string; + /** + * A description of the expected parameter. + */ + description: string; + }; + }; + }; + } | { + /** + * Specifies the type of tool (e.g., 'function'). + */ + type: string; + /** + * Details of the function tool. + */ + function: { + /** + * The name of the function. + */ + name: string; + /** + * A brief description of what the function does. + */ + description: string; + /** + * Schema defining the parameters accepted by the function. + */ + parameters: { + /** + * The type of the parameters object (usually 'object'). + */ + type: string; + /** + * List of required parameter names. + */ + required?: string[]; + /** + * Definitions of each parameter. + */ + properties: { + [k: string]: { + /** + * The data type of the parameter. + */ + type: string; + /** + * A description of the expected parameter. + */ + description: string; + }; + }; + }; + }; + })[]; + /** + * JSON schema that should be fulfilled for the response. + */ + guided_json?: object; + /** + * If true, a chat template is not applied and you must adhere to the specific model's expected formatting. + */ + raw?: boolean; + /** + * If true, the response will be streamed back incrementally using SSE, Server Sent Events. + */ + stream?: boolean; + /** + * The maximum number of tokens to generate in the response. + */ + max_tokens?: number; + /** + * Controls the randomness of the output; higher values produce more random results. + */ + temperature?: number; + /** + * Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses. + */ + top_p?: number; + /** + * Limits the AI to choose from the top 'k' most probable words. Lower values make responses more focused; higher values introduce more variety and potential surprises. + */ + top_k?: number; + /** + * Random seed for reproducibility of the generation. + */ + seed?: number; + /** + * Penalty for repeated tokens; higher values discourage repetition. + */ + repetition_penalty?: number; + /** + * Decreases the likelihood of the model repeating the same lines verbatim. + */ + frequency_penalty?: number; + /** + * Increases the likelihood of the model introducing new topics. + */ + presence_penalty?: number; +} +type Ai_Cf_Mistralai_Mistral_Small_3_1_24B_Instruct_Output = { + /** + * The generated text response from the model + */ + response: string; + /** + * Usage statistics for the inference request + */ + usage?: { + /** + * Total number of tokens in input + */ + prompt_tokens?: number; + /** + * Total number of tokens in output + */ + completion_tokens?: number; + /** + * Total number of input and output tokens + */ + total_tokens?: number; + }; + /** + * An array of tool calls requests made during the response generation + */ + tool_calls?: { + /** + * The arguments passed to be passed to the tool call request + */ + arguments?: object; + /** + * The name of the tool to be called + */ + name?: string; + }[]; +}; +declare abstract class Base_Ai_Cf_Mistralai_Mistral_Small_3_1_24B_Instruct { + inputs: Ai_Cf_Mistralai_Mistral_Small_3_1_24B_Instruct_Input; + postProcessedOutputs: Ai_Cf_Mistralai_Mistral_Small_3_1_24B_Instruct_Output; +} +type Ai_Cf_Google_Gemma_3_12B_It_Input = Ai_Cf_Google_Gemma_3_12B_It_Prompt | Ai_Cf_Google_Gemma_3_12B_It_Messages; +interface Ai_Cf_Google_Gemma_3_12B_It_Prompt { + /** + * The input text prompt for the model to generate a response. + */ + prompt: string; + /** + * JSON schema that should be fulfilled for the response. + */ + guided_json?: object; + /** + * If true, a chat template is not applied and you must adhere to the specific model's expected formatting. + */ + raw?: boolean; + /** + * If true, the response will be streamed back incrementally using SSE, Server Sent Events. + */ + stream?: boolean; + /** + * The maximum number of tokens to generate in the response. + */ + max_tokens?: number; + /** + * Controls the randomness of the output; higher values produce more random results. + */ + temperature?: number; + /** + * Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses. + */ + top_p?: number; + /** + * Limits the AI to choose from the top 'k' most probable words. Lower values make responses more focused; higher values introduce more variety and potential surprises. + */ + top_k?: number; + /** + * Random seed for reproducibility of the generation. + */ + seed?: number; + /** + * Penalty for repeated tokens; higher values discourage repetition. + */ + repetition_penalty?: number; + /** + * Decreases the likelihood of the model repeating the same lines verbatim. + */ + frequency_penalty?: number; + /** + * Increases the likelihood of the model introducing new topics. + */ + presence_penalty?: number; +} +interface Ai_Cf_Google_Gemma_3_12B_It_Messages { + /** + * An array of message objects representing the conversation history. + */ + messages: { + /** + * The role of the message sender (e.g., 'user', 'assistant', 'system', 'tool'). + */ + role?: string; + content?: string | { + /** + * Type of the content provided + */ + type?: string; + text?: string; + image_url?: { + /** + * image uri with data (e.g. data:image/jpeg;base64,/9j/...). HTTP URL will not be accepted + */ + url?: string; + }; + }[]; + }[]; + functions?: { + name: string; + code: string; + }[]; + /** + * A list of tools available for the assistant to use. + */ + tools?: ({ + /** + * The name of the tool. More descriptive the better. + */ + name: string; + /** + * A brief description of what the tool does. + */ + description: string; + /** + * Schema defining the parameters accepted by the tool. + */ + parameters: { + /** + * The type of the parameters object (usually 'object'). + */ + type: string; + /** + * List of required parameter names. + */ + required?: string[]; + /** + * Definitions of each parameter. + */ + properties: { + [k: string]: { + /** + * The data type of the parameter. + */ + type: string; + /** + * A description of the expected parameter. + */ + description: string; + }; + }; + }; + } | { + /** + * Specifies the type of tool (e.g., 'function'). + */ + type: string; + /** + * Details of the function tool. + */ + function: { + /** + * The name of the function. + */ + name: string; + /** + * A brief description of what the function does. + */ + description: string; + /** + * Schema defining the parameters accepted by the function. + */ + parameters: { + /** + * The type of the parameters object (usually 'object'). + */ + type: string; + /** + * List of required parameter names. + */ + required?: string[]; + /** + * Definitions of each parameter. + */ + properties: { + [k: string]: { + /** + * The data type of the parameter. + */ + type: string; + /** + * A description of the expected parameter. + */ + description: string; + }; + }; + }; + }; + })[]; + /** + * JSON schema that should be fulfilled for the response. + */ + guided_json?: object; + /** + * If true, a chat template is not applied and you must adhere to the specific model's expected formatting. + */ + raw?: boolean; + /** + * If true, the response will be streamed back incrementally using SSE, Server Sent Events. + */ + stream?: boolean; + /** + * The maximum number of tokens to generate in the response. + */ + max_tokens?: number; + /** + * Controls the randomness of the output; higher values produce more random results. + */ + temperature?: number; + /** + * Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses. + */ + top_p?: number; + /** + * Limits the AI to choose from the top 'k' most probable words. Lower values make responses more focused; higher values introduce more variety and potential surprises. + */ + top_k?: number; + /** + * Random seed for reproducibility of the generation. + */ + seed?: number; + /** + * Penalty for repeated tokens; higher values discourage repetition. + */ + repetition_penalty?: number; + /** + * Decreases the likelihood of the model repeating the same lines verbatim. + */ + frequency_penalty?: number; + /** + * Increases the likelihood of the model introducing new topics. + */ + presence_penalty?: number; +} +type Ai_Cf_Google_Gemma_3_12B_It_Output = { + /** + * The generated text response from the model + */ + response: string; + /** + * Usage statistics for the inference request + */ + usage?: { + /** + * Total number of tokens in input + */ + prompt_tokens?: number; + /** + * Total number of tokens in output + */ + completion_tokens?: number; + /** + * Total number of input and output tokens + */ + total_tokens?: number; + }; + /** + * An array of tool calls requests made during the response generation + */ + tool_calls?: { + /** + * The arguments passed to be passed to the tool call request + */ + arguments?: object; + /** + * The name of the tool to be called + */ + name?: string; + }[]; +}; +declare abstract class Base_Ai_Cf_Google_Gemma_3_12B_It { + inputs: Ai_Cf_Google_Gemma_3_12B_It_Input; + postProcessedOutputs: Ai_Cf_Google_Gemma_3_12B_It_Output; +} +type Ai_Cf_Meta_Llama_4_Scout_17B_16E_Instruct_Input = Ai_Cf_Meta_Llama_4_Scout_17B_16E_Instruct_Prompt | Ai_Cf_Meta_Llama_4_Scout_17B_16E_Instruct_Messages | Ai_Cf_Meta_Llama_4_Scout_17B_16E_Instruct_Async_Batch; +interface Ai_Cf_Meta_Llama_4_Scout_17B_16E_Instruct_Prompt { + /** + * The input text prompt for the model to generate a response. + */ + prompt: string; + /** + * JSON schema that should be fulfilled for the response. + */ + guided_json?: object; + response_format?: Ai_Cf_Meta_Llama_4_Scout_17B_16E_Instruct_JSON_Mode; + /** + * If true, a chat template is not applied and you must adhere to the specific model's expected formatting. + */ + raw?: boolean; + /** + * If true, the response will be streamed back incrementally using SSE, Server Sent Events. + */ + stream?: boolean; + /** + * The maximum number of tokens to generate in the response. + */ + max_tokens?: number; + /** + * Controls the randomness of the output; higher values produce more random results. + */ + temperature?: number; + /** + * Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses. + */ + top_p?: number; + /** + * Limits the AI to choose from the top 'k' most probable words. Lower values make responses more focused; higher values introduce more variety and potential surprises. + */ + top_k?: number; + /** + * Random seed for reproducibility of the generation. + */ + seed?: number; + /** + * Penalty for repeated tokens; higher values discourage repetition. + */ + repetition_penalty?: number; + /** + * Decreases the likelihood of the model repeating the same lines verbatim. + */ + frequency_penalty?: number; + /** + * Increases the likelihood of the model introducing new topics. + */ + presence_penalty?: number; +} +interface Ai_Cf_Meta_Llama_4_Scout_17B_16E_Instruct_JSON_Mode { + type?: "json_object" | "json_schema"; + json_schema?: unknown; +} +interface Ai_Cf_Meta_Llama_4_Scout_17B_16E_Instruct_Messages { + /** + * An array of message objects representing the conversation history. + */ + messages: { + /** + * The role of the message sender (e.g., 'user', 'assistant', 'system', 'tool'). + */ + role?: string; + /** + * The tool call id. If you don't know what to put here you can fall back to 000000001 + */ + tool_call_id?: string; + content?: string | { + /** + * Type of the content provided + */ + type?: string; + text?: string; + image_url?: { + /** + * image uri with data (e.g. data:image/jpeg;base64,/9j/...). HTTP URL will not be accepted + */ + url?: string; + }; + }[] | { + /** + * Type of the content provided + */ + type?: string; + text?: string; + image_url?: { + /** + * image uri with data (e.g. data:image/jpeg;base64,/9j/...). HTTP URL will not be accepted + */ + url?: string; + }; + }; + }[]; + functions?: { + name: string; + code: string; + }[]; + /** + * A list of tools available for the assistant to use. + */ + tools?: ({ + /** + * The name of the tool. More descriptive the better. + */ + name: string; + /** + * A brief description of what the tool does. + */ + description: string; + /** + * Schema defining the parameters accepted by the tool. + */ + parameters: { + /** + * The type of the parameters object (usually 'object'). + */ + type: string; + /** + * List of required parameter names. + */ + required?: string[]; + /** + * Definitions of each parameter. + */ + properties: { + [k: string]: { + /** + * The data type of the parameter. + */ + type: string; + /** + * A description of the expected parameter. + */ + description: string; + }; + }; + }; + } | { + /** + * Specifies the type of tool (e.g., 'function'). + */ + type: string; + /** + * Details of the function tool. + */ + function: { + /** + * The name of the function. + */ + name: string; + /** + * A brief description of what the function does. + */ + description: string; + /** + * Schema defining the parameters accepted by the function. + */ + parameters: { + /** + * The type of the parameters object (usually 'object'). + */ + type: string; + /** + * List of required parameter names. + */ + required?: string[]; + /** + * Definitions of each parameter. + */ + properties: { + [k: string]: { + /** + * The data type of the parameter. + */ + type: string; + /** + * A description of the expected parameter. + */ + description: string; + }; + }; + }; + }; + })[]; + response_format?: Ai_Cf_Meta_Llama_4_Scout_17B_16E_Instruct_JSON_Mode; + /** + * JSON schema that should be fulfilled for the response. + */ + guided_json?: object; + /** + * If true, a chat template is not applied and you must adhere to the specific model's expected formatting. + */ + raw?: boolean; + /** + * If true, the response will be streamed back incrementally using SSE, Server Sent Events. + */ + stream?: boolean; + /** + * The maximum number of tokens to generate in the response. + */ + max_tokens?: number; + /** + * Controls the randomness of the output; higher values produce more random results. + */ + temperature?: number; + /** + * Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses. + */ + top_p?: number; + /** + * Limits the AI to choose from the top 'k' most probable words. Lower values make responses more focused; higher values introduce more variety and potential surprises. + */ + top_k?: number; + /** + * Random seed for reproducibility of the generation. + */ + seed?: number; + /** + * Penalty for repeated tokens; higher values discourage repetition. + */ + repetition_penalty?: number; + /** + * Decreases the likelihood of the model repeating the same lines verbatim. + */ + frequency_penalty?: number; + /** + * Increases the likelihood of the model introducing new topics. + */ + presence_penalty?: number; +} +interface Ai_Cf_Meta_Llama_4_Scout_17B_16E_Instruct_Async_Batch { + requests: (Ai_Cf_Meta_Llama_4_Scout_17B_16E_Instruct_Prompt_Inner | Ai_Cf_Meta_Llama_4_Scout_17B_16E_Instruct_Messages_Inner)[]; +} +interface Ai_Cf_Meta_Llama_4_Scout_17B_16E_Instruct_Prompt_Inner { + /** + * The input text prompt for the model to generate a response. + */ + prompt: string; + /** + * JSON schema that should be fulfilled for the response. + */ + guided_json?: object; + response_format?: Ai_Cf_Meta_Llama_4_Scout_17B_16E_Instruct_JSON_Mode; + /** + * If true, a chat template is not applied and you must adhere to the specific model's expected formatting. + */ + raw?: boolean; + /** + * If true, the response will be streamed back incrementally using SSE, Server Sent Events. + */ + stream?: boolean; + /** + * The maximum number of tokens to generate in the response. + */ + max_tokens?: number; + /** + * Controls the randomness of the output; higher values produce more random results. + */ + temperature?: number; + /** + * Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses. + */ + top_p?: number; + /** + * Limits the AI to choose from the top 'k' most probable words. Lower values make responses more focused; higher values introduce more variety and potential surprises. + */ + top_k?: number; + /** + * Random seed for reproducibility of the generation. + */ + seed?: number; + /** + * Penalty for repeated tokens; higher values discourage repetition. + */ + repetition_penalty?: number; + /** + * Decreases the likelihood of the model repeating the same lines verbatim. + */ + frequency_penalty?: number; + /** + * Increases the likelihood of the model introducing new topics. + */ + presence_penalty?: number; +} +interface Ai_Cf_Meta_Llama_4_Scout_17B_16E_Instruct_Messages_Inner { + /** + * An array of message objects representing the conversation history. + */ + messages: { + /** + * The role of the message sender (e.g., 'user', 'assistant', 'system', 'tool'). + */ + role?: string; + /** + * The tool call id. If you don't know what to put here you can fall back to 000000001 + */ + tool_call_id?: string; + content?: string | { + /** + * Type of the content provided + */ + type?: string; + text?: string; + image_url?: { + /** + * image uri with data (e.g. data:image/jpeg;base64,/9j/...). HTTP URL will not be accepted + */ + url?: string; + }; + }[] | { + /** + * Type of the content provided + */ + type?: string; + text?: string; + image_url?: { + /** + * image uri with data (e.g. data:image/jpeg;base64,/9j/...). HTTP URL will not be accepted + */ + url?: string; + }; + }; + }[]; + functions?: { + name: string; + code: string; + }[]; + /** + * A list of tools available for the assistant to use. + */ + tools?: ({ + /** + * The name of the tool. More descriptive the better. + */ + name: string; + /** + * A brief description of what the tool does. + */ + description: string; + /** + * Schema defining the parameters accepted by the tool. + */ + parameters: { + /** + * The type of the parameters object (usually 'object'). + */ + type: string; + /** + * List of required parameter names. + */ + required?: string[]; + /** + * Definitions of each parameter. + */ + properties: { + [k: string]: { + /** + * The data type of the parameter. + */ + type: string; + /** + * A description of the expected parameter. + */ + description: string; + }; + }; + }; + } | { + /** + * Specifies the type of tool (e.g., 'function'). + */ + type: string; + /** + * Details of the function tool. + */ + function: { + /** + * The name of the function. + */ + name: string; + /** + * A brief description of what the function does. + */ + description: string; + /** + * Schema defining the parameters accepted by the function. + */ + parameters: { + /** + * The type of the parameters object (usually 'object'). + */ + type: string; + /** + * List of required parameter names. + */ + required?: string[]; + /** + * Definitions of each parameter. + */ + properties: { + [k: string]: { + /** + * The data type of the parameter. + */ + type: string; + /** + * A description of the expected parameter. + */ + description: string; + }; + }; + }; + }; + })[]; + response_format?: Ai_Cf_Meta_Llama_4_Scout_17B_16E_Instruct_JSON_Mode; + /** + * JSON schema that should be fulfilled for the response. + */ + guided_json?: object; + /** + * If true, a chat template is not applied and you must adhere to the specific model's expected formatting. + */ + raw?: boolean; + /** + * If true, the response will be streamed back incrementally using SSE, Server Sent Events. + */ + stream?: boolean; + /** + * The maximum number of tokens to generate in the response. + */ + max_tokens?: number; + /** + * Controls the randomness of the output; higher values produce more random results. + */ + temperature?: number; + /** + * Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses. + */ + top_p?: number; + /** + * Limits the AI to choose from the top 'k' most probable words. Lower values make responses more focused; higher values introduce more variety and potential surprises. + */ + top_k?: number; + /** + * Random seed for reproducibility of the generation. + */ + seed?: number; + /** + * Penalty for repeated tokens; higher values discourage repetition. + */ + repetition_penalty?: number; + /** + * Decreases the likelihood of the model repeating the same lines verbatim. + */ + frequency_penalty?: number; + /** + * Increases the likelihood of the model introducing new topics. + */ + presence_penalty?: number; +} +type Ai_Cf_Meta_Llama_4_Scout_17B_16E_Instruct_Output = { + /** + * The generated text response from the model + */ + response: string; + /** + * Usage statistics for the inference request + */ + usage?: { + /** + * Total number of tokens in input + */ + prompt_tokens?: number; + /** + * Total number of tokens in output + */ + completion_tokens?: number; + /** + * Total number of input and output tokens + */ + total_tokens?: number; + }; + /** + * An array of tool calls requests made during the response generation + */ + tool_calls?: { + /** + * The tool call id. + */ + id?: string; + /** + * Specifies the type of tool (e.g., 'function'). + */ + type?: string; + /** + * Details of the function tool. + */ + function?: { + /** + * The name of the tool to be called + */ + name?: string; + /** + * The arguments passed to be passed to the tool call request + */ + arguments?: object; + }; + }[]; +}; +declare abstract class Base_Ai_Cf_Meta_Llama_4_Scout_17B_16E_Instruct { + inputs: Ai_Cf_Meta_Llama_4_Scout_17B_16E_Instruct_Input; + postProcessedOutputs: Ai_Cf_Meta_Llama_4_Scout_17B_16E_Instruct_Output; +} +type Ai_Cf_Qwen_Qwen3_30B_A3B_Fp8_Input = Ai_Cf_Qwen_Qwen3_30B_A3B_Fp8_Prompt | Ai_Cf_Qwen_Qwen3_30B_A3B_Fp8_Messages | Ai_Cf_Qwen_Qwen3_30B_A3B_Fp8_Async_Batch; +interface Ai_Cf_Qwen_Qwen3_30B_A3B_Fp8_Prompt { + /** + * The input text prompt for the model to generate a response. + */ + prompt: string; + /** + * Name of the LoRA (Low-Rank Adaptation) model to fine-tune the base model. + */ + lora?: string; + response_format?: Ai_Cf_Qwen_Qwen3_30B_A3B_Fp8_JSON_Mode; + /** + * If true, a chat template is not applied and you must adhere to the specific model's expected formatting. + */ + raw?: boolean; + /** + * If true, the response will be streamed back incrementally using SSE, Server Sent Events. + */ + stream?: boolean; + /** + * The maximum number of tokens to generate in the response. + */ + max_tokens?: number; + /** + * Controls the randomness of the output; higher values produce more random results. + */ + temperature?: number; + /** + * Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses. + */ + top_p?: number; + /** + * Limits the AI to choose from the top 'k' most probable words. Lower values make responses more focused; higher values introduce more variety and potential surprises. + */ + top_k?: number; + /** + * Random seed for reproducibility of the generation. + */ + seed?: number; + /** + * Penalty for repeated tokens; higher values discourage repetition. + */ + repetition_penalty?: number; + /** + * Decreases the likelihood of the model repeating the same lines verbatim. + */ + frequency_penalty?: number; + /** + * Increases the likelihood of the model introducing new topics. + */ + presence_penalty?: number; +} +interface Ai_Cf_Qwen_Qwen3_30B_A3B_Fp8_JSON_Mode { + type?: "json_object" | "json_schema"; + json_schema?: unknown; +} +interface Ai_Cf_Qwen_Qwen3_30B_A3B_Fp8_Messages { + /** + * An array of message objects representing the conversation history. + */ + messages: { + /** + * The role of the message sender (e.g., 'user', 'assistant', 'system', 'tool'). + */ + role: string; + /** + * The content of the message as a string. + */ + content: string; + }[]; + functions?: { + name: string; + code: string; + }[]; + /** + * A list of tools available for the assistant to use. + */ + tools?: ({ + /** + * The name of the tool. More descriptive the better. + */ + name: string; + /** + * A brief description of what the tool does. + */ + description: string; + /** + * Schema defining the parameters accepted by the tool. + */ + parameters: { + /** + * The type of the parameters object (usually 'object'). + */ + type: string; + /** + * List of required parameter names. + */ + required?: string[]; + /** + * Definitions of each parameter. + */ + properties: { + [k: string]: { + /** + * The data type of the parameter. + */ + type: string; + /** + * A description of the expected parameter. + */ + description: string; + }; + }; + }; + } | { + /** + * Specifies the type of tool (e.g., 'function'). + */ + type: string; + /** + * Details of the function tool. + */ + function: { + /** + * The name of the function. + */ + name: string; + /** + * A brief description of what the function does. + */ + description: string; + /** + * Schema defining the parameters accepted by the function. + */ + parameters: { + /** + * The type of the parameters object (usually 'object'). + */ + type: string; + /** + * List of required parameter names. + */ + required?: string[]; + /** + * Definitions of each parameter. + */ + properties: { + [k: string]: { + /** + * The data type of the parameter. + */ + type: string; + /** + * A description of the expected parameter. + */ + description: string; + }; + }; + }; + }; + })[]; + response_format?: Ai_Cf_Qwen_Qwen3_30B_A3B_Fp8_JSON_Mode_1; + /** + * If true, a chat template is not applied and you must adhere to the specific model's expected formatting. + */ + raw?: boolean; + /** + * If true, the response will be streamed back incrementally using SSE, Server Sent Events. + */ + stream?: boolean; + /** + * The maximum number of tokens to generate in the response. + */ + max_tokens?: number; + /** + * Controls the randomness of the output; higher values produce more random results. + */ + temperature?: number; + /** + * Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses. + */ + top_p?: number; + /** + * Limits the AI to choose from the top 'k' most probable words. Lower values make responses more focused; higher values introduce more variety and potential surprises. + */ + top_k?: number; + /** + * Random seed for reproducibility of the generation. + */ + seed?: number; + /** + * Penalty for repeated tokens; higher values discourage repetition. + */ + repetition_penalty?: number; + /** + * Decreases the likelihood of the model repeating the same lines verbatim. + */ + frequency_penalty?: number; + /** + * Increases the likelihood of the model introducing new topics. + */ + presence_penalty?: number; +} +interface Ai_Cf_Qwen_Qwen3_30B_A3B_Fp8_JSON_Mode_1 { + type?: "json_object" | "json_schema"; + json_schema?: unknown; +} +interface Ai_Cf_Qwen_Qwen3_30B_A3B_Fp8_Async_Batch { + requests: (Ai_Cf_Qwen_Qwen3_30B_A3B_Fp8_Prompt_1 | Ai_Cf_Qwen_Qwen3_30B_A3B_Fp8_Messages_1)[]; +} +interface Ai_Cf_Qwen_Qwen3_30B_A3B_Fp8_Prompt_1 { + /** + * The input text prompt for the model to generate a response. + */ + prompt: string; + /** + * Name of the LoRA (Low-Rank Adaptation) model to fine-tune the base model. + */ + lora?: string; + response_format?: Ai_Cf_Qwen_Qwen3_30B_A3B_Fp8_JSON_Mode_2; + /** + * If true, a chat template is not applied and you must adhere to the specific model's expected formatting. + */ + raw?: boolean; + /** + * If true, the response will be streamed back incrementally using SSE, Server Sent Events. + */ + stream?: boolean; + /** + * The maximum number of tokens to generate in the response. + */ + max_tokens?: number; + /** + * Controls the randomness of the output; higher values produce more random results. + */ + temperature?: number; + /** + * Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses. + */ + top_p?: number; + /** + * Limits the AI to choose from the top 'k' most probable words. Lower values make responses more focused; higher values introduce more variety and potential surprises. + */ + top_k?: number; + /** + * Random seed for reproducibility of the generation. + */ + seed?: number; + /** + * Penalty for repeated tokens; higher values discourage repetition. + */ + repetition_penalty?: number; + /** + * Decreases the likelihood of the model repeating the same lines verbatim. + */ + frequency_penalty?: number; + /** + * Increases the likelihood of the model introducing new topics. + */ + presence_penalty?: number; +} +interface Ai_Cf_Qwen_Qwen3_30B_A3B_Fp8_JSON_Mode_2 { + type?: "json_object" | "json_schema"; + json_schema?: unknown; +} +interface Ai_Cf_Qwen_Qwen3_30B_A3B_Fp8_Messages_1 { + /** + * An array of message objects representing the conversation history. + */ + messages: { + /** + * The role of the message sender (e.g., 'user', 'assistant', 'system', 'tool'). + */ + role: string; + /** + * The content of the message as a string. + */ + content: string; + }[]; + functions?: { + name: string; + code: string; + }[]; + /** + * A list of tools available for the assistant to use. + */ + tools?: ({ + /** + * The name of the tool. More descriptive the better. + */ + name: string; + /** + * A brief description of what the tool does. + */ + description: string; + /** + * Schema defining the parameters accepted by the tool. + */ + parameters: { + /** + * The type of the parameters object (usually 'object'). + */ + type: string; + /** + * List of required parameter names. + */ + required?: string[]; + /** + * Definitions of each parameter. + */ + properties: { + [k: string]: { + /** + * The data type of the parameter. + */ + type: string; + /** + * A description of the expected parameter. + */ + description: string; + }; + }; + }; + } | { + /** + * Specifies the type of tool (e.g., 'function'). + */ + type: string; + /** + * Details of the function tool. + */ + function: { + /** + * The name of the function. + */ + name: string; + /** + * A brief description of what the function does. + */ + description: string; + /** + * Schema defining the parameters accepted by the function. + */ + parameters: { + /** + * The type of the parameters object (usually 'object'). + */ + type: string; + /** + * List of required parameter names. + */ + required?: string[]; + /** + * Definitions of each parameter. + */ + properties: { + [k: string]: { + /** + * The data type of the parameter. + */ + type: string; + /** + * A description of the expected parameter. + */ + description: string; + }; + }; + }; + }; + })[]; + response_format?: Ai_Cf_Qwen_Qwen3_30B_A3B_Fp8_JSON_Mode_3; + /** + * If true, a chat template is not applied and you must adhere to the specific model's expected formatting. + */ + raw?: boolean; + /** + * If true, the response will be streamed back incrementally using SSE, Server Sent Events. + */ + stream?: boolean; + /** + * The maximum number of tokens to generate in the response. + */ + max_tokens?: number; + /** + * Controls the randomness of the output; higher values produce more random results. + */ + temperature?: number; + /** + * Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses. + */ + top_p?: number; + /** + * Limits the AI to choose from the top 'k' most probable words. Lower values make responses more focused; higher values introduce more variety and potential surprises. + */ + top_k?: number; + /** + * Random seed for reproducibility of the generation. + */ + seed?: number; + /** + * Penalty for repeated tokens; higher values discourage repetition. + */ + repetition_penalty?: number; + /** + * Decreases the likelihood of the model repeating the same lines verbatim. + */ + frequency_penalty?: number; + /** + * Increases the likelihood of the model introducing new topics. + */ + presence_penalty?: number; +} +interface Ai_Cf_Qwen_Qwen3_30B_A3B_Fp8_JSON_Mode_3 { + type?: "json_object" | "json_schema"; + json_schema?: unknown; +} +type Ai_Cf_Qwen_Qwen3_30B_A3B_Fp8_Output = Ai_Cf_Qwen_Qwen3_30B_A3B_Fp8_Chat_Completion_Response | Ai_Cf_Qwen_Qwen3_30B_A3B_Fp8_Text_Completion_Response | string | Ai_Cf_Qwen_Qwen3_30B_A3B_Fp8_AsyncResponse; +interface Ai_Cf_Qwen_Qwen3_30B_A3B_Fp8_Chat_Completion_Response { + /** + * Unique identifier for the completion + */ + id?: string; + /** + * Object type identifier + */ + object?: "chat.completion"; + /** + * Unix timestamp of when the completion was created + */ + created?: number; + /** + * Model used for the completion + */ + model?: string; + /** + * List of completion choices + */ + choices?: { + /** + * Index of the choice in the list + */ + index?: number; + /** + * The message generated by the model + */ + message?: { + /** + * Role of the message author + */ + role: string; + /** + * The content of the message + */ + content: string; + /** + * Internal reasoning content (if available) + */ + reasoning_content?: string; + /** + * Tool calls made by the assistant + */ + tool_calls?: { + /** + * Unique identifier for the tool call + */ + id: string; + /** + * Type of tool call + */ + type: "function"; + function: { + /** + * Name of the function to call + */ + name: string; + /** + * JSON string of arguments for the function + */ + arguments: string; + }; + }[]; + }; + /** + * Reason why the model stopped generating + */ + finish_reason?: string; + /** + * Stop reason (may be null) + */ + stop_reason?: string | null; + /** + * Log probabilities (if requested) + */ + logprobs?: {} | null; + }[]; + /** + * Usage statistics for the inference request + */ + usage?: { + /** + * Total number of tokens in input + */ + prompt_tokens?: number; + /** + * Total number of tokens in output + */ + completion_tokens?: number; + /** + * Total number of input and output tokens + */ + total_tokens?: number; + }; + /** + * Log probabilities for the prompt (if requested) + */ + prompt_logprobs?: {} | null; +} +interface Ai_Cf_Qwen_Qwen3_30B_A3B_Fp8_Text_Completion_Response { + /** + * Unique identifier for the completion + */ + id?: string; + /** + * Object type identifier + */ + object?: "text_completion"; + /** + * Unix timestamp of when the completion was created + */ + created?: number; + /** + * Model used for the completion + */ + model?: string; + /** + * List of completion choices + */ + choices?: { + /** + * Index of the choice in the list + */ + index: number; + /** + * The generated text completion + */ + text: string; + /** + * Reason why the model stopped generating + */ + finish_reason: string; + /** + * Stop reason (may be null) + */ + stop_reason?: string | null; + /** + * Log probabilities (if requested) + */ + logprobs?: {} | null; + /** + * Log probabilities for the prompt (if requested) + */ + prompt_logprobs?: {} | null; + }[]; + /** + * Usage statistics for the inference request + */ + usage?: { + /** + * Total number of tokens in input + */ + prompt_tokens?: number; + /** + * Total number of tokens in output + */ + completion_tokens?: number; + /** + * Total number of input and output tokens + */ + total_tokens?: number; + }; +} +interface Ai_Cf_Qwen_Qwen3_30B_A3B_Fp8_AsyncResponse { + /** + * The async request id that can be used to obtain the results. + */ + request_id?: string; +} +declare abstract class Base_Ai_Cf_Qwen_Qwen3_30B_A3B_Fp8 { + inputs: Ai_Cf_Qwen_Qwen3_30B_A3B_Fp8_Input; + postProcessedOutputs: Ai_Cf_Qwen_Qwen3_30B_A3B_Fp8_Output; +} +interface Ai_Cf_Deepgram_Nova_3_Input { + audio: { + body: object; + contentType: string; + }; + /** + * Sets how the model will interpret strings submitted to the custom_topic param. When strict, the model will only return topics submitted using the custom_topic param. When extended, the model will return its own detected topics in addition to those submitted using the custom_topic param. + */ + custom_topic_mode?: "extended" | "strict"; + /** + * Custom topics you want the model to detect within your input audio or text if present Submit up to 100 + */ + custom_topic?: string; + /** + * Sets how the model will interpret intents submitted to the custom_intent param. When strict, the model will only return intents submitted using the custom_intent param. When extended, the model will return its own detected intents in addition those submitted using the custom_intents param + */ + custom_intent_mode?: "extended" | "strict"; + /** + * Custom intents you want the model to detect within your input audio if present + */ + custom_intent?: string; + /** + * Identifies and extracts key entities from content in submitted audio + */ + detect_entities?: boolean; + /** + * Identifies the dominant language spoken in submitted audio + */ + detect_language?: boolean; + /** + * Recognize speaker changes. Each word in the transcript will be assigned a speaker number starting at 0 + */ + diarize?: boolean; + /** + * Identify and extract key entities from content in submitted audio + */ + dictation?: boolean; + /** + * Specify the expected encoding of your submitted audio + */ + encoding?: "linear16" | "flac" | "mulaw" | "amr-nb" | "amr-wb" | "opus" | "speex" | "g729"; + /** + * Arbitrary key-value pairs that are attached to the API response for usage in downstream processing + */ + extra?: string; + /** + * Filler Words can help transcribe interruptions in your audio, like 'uh' and 'um' + */ + filler_words?: boolean; + /** + * Key term prompting can boost or suppress specialized terminology and brands. + */ + keyterm?: string; + /** + * Keywords can boost or suppress specialized terminology and brands. + */ + keywords?: string; + /** + * The BCP-47 language tag that hints at the primary spoken language. Depending on the Model and API endpoint you choose only certain languages are available. + */ + language?: string; + /** + * Spoken measurements will be converted to their corresponding abbreviations. + */ + measurements?: boolean; + /** + * Opts out requests from the Deepgram Model Improvement Program. Refer to our Docs for pricing impacts before setting this to true. https://dpgr.am/deepgram-mip. + */ + mip_opt_out?: boolean; + /** + * Mode of operation for the model representing broad area of topic that will be talked about in the supplied audio + */ + mode?: "general" | "medical" | "finance"; + /** + * Transcribe each audio channel independently. + */ + multichannel?: boolean; + /** + * Numerals converts numbers from written format to numerical format. + */ + numerals?: boolean; + /** + * Splits audio into paragraphs to improve transcript readability. + */ + paragraphs?: boolean; + /** + * Profanity Filter looks for recognized profanity and converts it to the nearest recognized non-profane word or removes it from the transcript completely. + */ + profanity_filter?: boolean; + /** + * Add punctuation and capitalization to the transcript. + */ + punctuate?: boolean; + /** + * Redaction removes sensitive information from your transcripts. + */ + redact?: string; + /** + * Search for terms or phrases in submitted audio and replaces them. + */ + replace?: string; + /** + * Search for terms or phrases in submitted audio. + */ + search?: string; + /** + * Recognizes the sentiment throughout a transcript or text. + */ + sentiment?: boolean; + /** + * Apply formatting to transcript output. When set to true, additional formatting will be applied to transcripts to improve readability. + */ + smart_format?: boolean; + /** + * Detect topics throughout a transcript or text. + */ + topics?: boolean; + /** + * Segments speech into meaningful semantic units. + */ + utterances?: boolean; + /** + * Seconds to wait before detecting a pause between words in submitted audio. + */ + utt_split?: number; + /** + * The number of channels in the submitted audio + */ + channels?: number; + /** + * Specifies whether the streaming endpoint should provide ongoing transcription updates as more audio is received. When set to true, the endpoint sends continuous updates, meaning transcription results may evolve over time. Note: Supported only for webosockets. + */ + interim_results?: boolean; + /** + * Indicates how long model will wait to detect whether a speaker has finished speaking or pauses for a significant period of time. When set to a value, the streaming endpoint immediately finalizes the transcription for the processed time range and returns the transcript with a speech_final parameter set to true. Can also be set to false to disable endpointing + */ + endpointing?: string; + /** + * Indicates that speech has started. You'll begin receiving Speech Started messages upon speech starting. Note: Supported only for webosockets. + */ + vad_events?: boolean; + /** + * Indicates how long model will wait to send an UtteranceEnd message after a word has been transcribed. Use with interim_results. Note: Supported only for webosockets. + */ + utterance_end_ms?: boolean; +} +interface Ai_Cf_Deepgram_Nova_3_Output { + results?: { + channels?: { + alternatives?: { + confidence?: number; + transcript?: string; + words?: { + confidence?: number; + end?: number; + start?: number; + word?: string; + }[]; + }[]; + }[]; + summary?: { + result?: string; + short?: string; + }; + sentiments?: { + segments?: { + text?: string; + start_word?: number; + end_word?: number; + sentiment?: string; + sentiment_score?: number; + }[]; + average?: { + sentiment?: string; + sentiment_score?: number; + }; + }; + }; +} +declare abstract class Base_Ai_Cf_Deepgram_Nova_3 { + inputs: Ai_Cf_Deepgram_Nova_3_Input; + postProcessedOutputs: Ai_Cf_Deepgram_Nova_3_Output; +} +interface Ai_Cf_Qwen_Qwen3_Embedding_0_6B_Input { + queries?: string | string[]; + /** + * Optional instruction for the task + */ + instruction?: string; + documents?: string | string[]; + text?: string | string[]; +} +interface Ai_Cf_Qwen_Qwen3_Embedding_0_6B_Output { + data?: number[][]; + shape?: number[]; +} +declare abstract class Base_Ai_Cf_Qwen_Qwen3_Embedding_0_6B { + inputs: Ai_Cf_Qwen_Qwen3_Embedding_0_6B_Input; + postProcessedOutputs: Ai_Cf_Qwen_Qwen3_Embedding_0_6B_Output; +} +type Ai_Cf_Pipecat_Ai_Smart_Turn_V2_Input = { + /** + * readable stream with audio data and content-type specified for that data + */ + audio: { + body: object; + contentType: string; + }; + /** + * type of data PCM data that's sent to the inference server as raw array + */ + dtype?: "uint8" | "float32" | "float64"; +} | { + /** + * base64 encoded audio data + */ + audio: string; + /** + * type of data PCM data that's sent to the inference server as raw array + */ + dtype?: "uint8" | "float32" | "float64"; +}; +interface Ai_Cf_Pipecat_Ai_Smart_Turn_V2_Output { + /** + * if true, end-of-turn was detected + */ + is_complete?: boolean; + /** + * probability of the end-of-turn detection + */ + probability?: number; +} +declare abstract class Base_Ai_Cf_Pipecat_Ai_Smart_Turn_V2 { + inputs: Ai_Cf_Pipecat_Ai_Smart_Turn_V2_Input; + postProcessedOutputs: Ai_Cf_Pipecat_Ai_Smart_Turn_V2_Output; +} +declare abstract class Base_Ai_Cf_Openai_Gpt_Oss_120B { + inputs: ResponsesInput; + postProcessedOutputs: ResponsesOutput; +} +declare abstract class Base_Ai_Cf_Openai_Gpt_Oss_20B { + inputs: ResponsesInput; + postProcessedOutputs: ResponsesOutput; +} +interface Ai_Cf_Leonardo_Phoenix_1_0_Input { + /** + * A text description of the image you want to generate. + */ + prompt: string; + /** + * Controls how closely the generated image should adhere to the prompt; higher values make the image more aligned with the prompt + */ + guidance?: number; + /** + * Random seed for reproducibility of the image generation + */ + seed?: number; + /** + * The height of the generated image in pixels + */ + height?: number; + /** + * The width of the generated image in pixels + */ + width?: number; + /** + * The number of diffusion steps; higher values can improve quality but take longer + */ + num_steps?: number; + /** + * Specify what to exclude from the generated images + */ + negative_prompt?: string; +} +/** + * The generated image in JPEG format + */ +type Ai_Cf_Leonardo_Phoenix_1_0_Output = string; +declare abstract class Base_Ai_Cf_Leonardo_Phoenix_1_0 { + inputs: Ai_Cf_Leonardo_Phoenix_1_0_Input; + postProcessedOutputs: Ai_Cf_Leonardo_Phoenix_1_0_Output; +} +interface Ai_Cf_Leonardo_Lucid_Origin_Input { + /** + * A text description of the image you want to generate. + */ + prompt: string; + /** + * Controls how closely the generated image should adhere to the prompt; higher values make the image more aligned with the prompt + */ + guidance?: number; + /** + * Random seed for reproducibility of the image generation + */ + seed?: number; + /** + * The height of the generated image in pixels + */ + height?: number; + /** + * The width of the generated image in pixels + */ + width?: number; + /** + * The number of diffusion steps; higher values can improve quality but take longer + */ + num_steps?: number; + /** + * The number of diffusion steps; higher values can improve quality but take longer + */ + steps?: number; +} +interface Ai_Cf_Leonardo_Lucid_Origin_Output { + /** + * The generated image in Base64 format. + */ + image?: string; +} +declare abstract class Base_Ai_Cf_Leonardo_Lucid_Origin { + inputs: Ai_Cf_Leonardo_Lucid_Origin_Input; + postProcessedOutputs: Ai_Cf_Leonardo_Lucid_Origin_Output; +} +interface Ai_Cf_Deepgram_Aura_1_Input { + /** + * Speaker used to produce the audio. + */ + speaker?: "angus" | "asteria" | "arcas" | "orion" | "orpheus" | "athena" | "luna" | "zeus" | "perseus" | "helios" | "hera" | "stella"; + /** + * Encoding of the output audio. + */ + encoding?: "linear16" | "flac" | "mulaw" | "alaw" | "mp3" | "opus" | "aac"; + /** + * Container specifies the file format wrapper for the output audio. The available options depend on the encoding type.. + */ + container?: "none" | "wav" | "ogg"; + /** + * The text content to be converted to speech + */ + text: string; + /** + * Sample Rate specifies the sample rate for the output audio. Based on the encoding, different sample rates are supported. For some encodings, the sample rate is not configurable + */ + sample_rate?: number; + /** + * The bitrate of the audio in bits per second. Choose from predefined ranges or specific values based on the encoding type. + */ + bit_rate?: number; +} +/** + * The generated audio in MP3 format + */ +type Ai_Cf_Deepgram_Aura_1_Output = string; +declare abstract class Base_Ai_Cf_Deepgram_Aura_1 { + inputs: Ai_Cf_Deepgram_Aura_1_Input; + postProcessedOutputs: Ai_Cf_Deepgram_Aura_1_Output; +} +interface Ai_Cf_Ai4Bharat_Indictrans2_En_Indic_1B_Input { + /** + * Input text to translate. Can be a single string or a list of strings. + */ + text: string | string[]; + /** + * Target language to translate to + */ + target_language: "asm_Beng" | "awa_Deva" | "ben_Beng" | "bho_Deva" | "brx_Deva" | "doi_Deva" | "eng_Latn" | "gom_Deva" | "gon_Deva" | "guj_Gujr" | "hin_Deva" | "hne_Deva" | "kan_Knda" | "kas_Arab" | "kas_Deva" | "kha_Latn" | "lus_Latn" | "mag_Deva" | "mai_Deva" | "mal_Mlym" | "mar_Deva" | "mni_Beng" | "mni_Mtei" | "npi_Deva" | "ory_Orya" | "pan_Guru" | "san_Deva" | "sat_Olck" | "snd_Arab" | "snd_Deva" | "tam_Taml" | "tel_Telu" | "urd_Arab" | "unr_Deva"; +} +interface Ai_Cf_Ai4Bharat_Indictrans2_En_Indic_1B_Output { + /** + * Translated texts + */ + translations: string[]; +} +declare abstract class Base_Ai_Cf_Ai4Bharat_Indictrans2_En_Indic_1B { + inputs: Ai_Cf_Ai4Bharat_Indictrans2_En_Indic_1B_Input; + postProcessedOutputs: Ai_Cf_Ai4Bharat_Indictrans2_En_Indic_1B_Output; +} +type Ai_Cf_Aisingapore_Gemma_Sea_Lion_V4_27B_It_Input = Ai_Cf_Aisingapore_Gemma_Sea_Lion_V4_27B_It_Prompt | Ai_Cf_Aisingapore_Gemma_Sea_Lion_V4_27B_It_Messages | Ai_Cf_Aisingapore_Gemma_Sea_Lion_V4_27B_It_Async_Batch; +interface Ai_Cf_Aisingapore_Gemma_Sea_Lion_V4_27B_It_Prompt { + /** + * The input text prompt for the model to generate a response. + */ + prompt: string; + /** + * Name of the LoRA (Low-Rank Adaptation) model to fine-tune the base model. + */ + lora?: string; + response_format?: Ai_Cf_Aisingapore_Gemma_Sea_Lion_V4_27B_It_JSON_Mode; + /** + * If true, a chat template is not applied and you must adhere to the specific model's expected formatting. + */ + raw?: boolean; + /** + * If true, the response will be streamed back incrementally using SSE, Server Sent Events. + */ + stream?: boolean; + /** + * The maximum number of tokens to generate in the response. + */ + max_tokens?: number; + /** + * Controls the randomness of the output; higher values produce more random results. + */ + temperature?: number; + /** + * Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses. + */ + top_p?: number; + /** + * Limits the AI to choose from the top 'k' most probable words. Lower values make responses more focused; higher values introduce more variety and potential surprises. + */ + top_k?: number; + /** + * Random seed for reproducibility of the generation. + */ + seed?: number; + /** + * Penalty for repeated tokens; higher values discourage repetition. + */ + repetition_penalty?: number; + /** + * Decreases the likelihood of the model repeating the same lines verbatim. + */ + frequency_penalty?: number; + /** + * Increases the likelihood of the model introducing new topics. + */ + presence_penalty?: number; +} +interface Ai_Cf_Aisingapore_Gemma_Sea_Lion_V4_27B_It_JSON_Mode { + type?: "json_object" | "json_schema"; + json_schema?: unknown; +} +interface Ai_Cf_Aisingapore_Gemma_Sea_Lion_V4_27B_It_Messages { + /** + * An array of message objects representing the conversation history. + */ + messages: { + /** + * The role of the message sender (e.g., 'user', 'assistant', 'system', 'tool'). + */ + role: string; + /** + * The content of the message as a string. + */ + content: string; + }[]; + functions?: { + name: string; + code: string; + }[]; + /** + * A list of tools available for the assistant to use. + */ + tools?: ({ + /** + * The name of the tool. More descriptive the better. + */ + name: string; + /** + * A brief description of what the tool does. + */ + description: string; + /** + * Schema defining the parameters accepted by the tool. + */ + parameters: { + /** + * The type of the parameters object (usually 'object'). + */ + type: string; + /** + * List of required parameter names. + */ + required?: string[]; + /** + * Definitions of each parameter. + */ + properties: { + [k: string]: { + /** + * The data type of the parameter. + */ + type: string; + /** + * A description of the expected parameter. + */ + description: string; + }; + }; + }; + } | { + /** + * Specifies the type of tool (e.g., 'function'). + */ + type: string; + /** + * Details of the function tool. + */ + function: { + /** + * The name of the function. + */ + name: string; + /** + * A brief description of what the function does. + */ + description: string; + /** + * Schema defining the parameters accepted by the function. + */ + parameters: { + /** + * The type of the parameters object (usually 'object'). + */ + type: string; + /** + * List of required parameter names. + */ + required?: string[]; + /** + * Definitions of each parameter. + */ + properties: { + [k: string]: { + /** + * The data type of the parameter. + */ + type: string; + /** + * A description of the expected parameter. + */ + description: string; + }; + }; + }; + }; + })[]; + response_format?: Ai_Cf_Aisingapore_Gemma_Sea_Lion_V4_27B_It_JSON_Mode_1; + /** + * If true, a chat template is not applied and you must adhere to the specific model's expected formatting. + */ + raw?: boolean; + /** + * If true, the response will be streamed back incrementally using SSE, Server Sent Events. + */ + stream?: boolean; + /** + * The maximum number of tokens to generate in the response. + */ + max_tokens?: number; + /** + * Controls the randomness of the output; higher values produce more random results. + */ + temperature?: number; + /** + * Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses. + */ + top_p?: number; + /** + * Limits the AI to choose from the top 'k' most probable words. Lower values make responses more focused; higher values introduce more variety and potential surprises. + */ + top_k?: number; + /** + * Random seed for reproducibility of the generation. + */ + seed?: number; + /** + * Penalty for repeated tokens; higher values discourage repetition. + */ + repetition_penalty?: number; + /** + * Decreases the likelihood of the model repeating the same lines verbatim. + */ + frequency_penalty?: number; + /** + * Increases the likelihood of the model introducing new topics. + */ + presence_penalty?: number; +} +interface Ai_Cf_Aisingapore_Gemma_Sea_Lion_V4_27B_It_JSON_Mode_1 { + type?: "json_object" | "json_schema"; + json_schema?: unknown; +} +interface Ai_Cf_Aisingapore_Gemma_Sea_Lion_V4_27B_It_Async_Batch { + requests: (Ai_Cf_Aisingapore_Gemma_Sea_Lion_V4_27B_It_Prompt_1 | Ai_Cf_Aisingapore_Gemma_Sea_Lion_V4_27B_It_Messages_1)[]; +} +interface Ai_Cf_Aisingapore_Gemma_Sea_Lion_V4_27B_It_Prompt_1 { + /** + * The input text prompt for the model to generate a response. + */ + prompt: string; + /** + * Name of the LoRA (Low-Rank Adaptation) model to fine-tune the base model. + */ + lora?: string; + response_format?: Ai_Cf_Aisingapore_Gemma_Sea_Lion_V4_27B_It_JSON_Mode_2; + /** + * If true, a chat template is not applied and you must adhere to the specific model's expected formatting. + */ + raw?: boolean; + /** + * If true, the response will be streamed back incrementally using SSE, Server Sent Events. + */ + stream?: boolean; + /** + * The maximum number of tokens to generate in the response. + */ + max_tokens?: number; + /** + * Controls the randomness of the output; higher values produce more random results. + */ + temperature?: number; + /** + * Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses. + */ + top_p?: number; + /** + * Limits the AI to choose from the top 'k' most probable words. Lower values make responses more focused; higher values introduce more variety and potential surprises. + */ + top_k?: number; + /** + * Random seed for reproducibility of the generation. + */ + seed?: number; + /** + * Penalty for repeated tokens; higher values discourage repetition. + */ + repetition_penalty?: number; + /** + * Decreases the likelihood of the model repeating the same lines verbatim. + */ + frequency_penalty?: number; + /** + * Increases the likelihood of the model introducing new topics. + */ + presence_penalty?: number; +} +interface Ai_Cf_Aisingapore_Gemma_Sea_Lion_V4_27B_It_JSON_Mode_2 { + type?: "json_object" | "json_schema"; + json_schema?: unknown; +} +interface Ai_Cf_Aisingapore_Gemma_Sea_Lion_V4_27B_It_Messages_1 { + /** + * An array of message objects representing the conversation history. + */ + messages: { + /** + * The role of the message sender (e.g., 'user', 'assistant', 'system', 'tool'). + */ + role: string; + /** + * The content of the message as a string. + */ + content: string; + }[]; + functions?: { + name: string; + code: string; + }[]; + /** + * A list of tools available for the assistant to use. + */ + tools?: ({ + /** + * The name of the tool. More descriptive the better. + */ + name: string; + /** + * A brief description of what the tool does. + */ + description: string; + /** + * Schema defining the parameters accepted by the tool. + */ + parameters: { + /** + * The type of the parameters object (usually 'object'). + */ + type: string; + /** + * List of required parameter names. + */ + required?: string[]; + /** + * Definitions of each parameter. + */ + properties: { + [k: string]: { + /** + * The data type of the parameter. + */ + type: string; + /** + * A description of the expected parameter. + */ + description: string; + }; + }; + }; + } | { + /** + * Specifies the type of tool (e.g., 'function'). + */ + type: string; + /** + * Details of the function tool. + */ + function: { + /** + * The name of the function. + */ + name: string; + /** + * A brief description of what the function does. + */ + description: string; + /** + * Schema defining the parameters accepted by the function. + */ + parameters: { + /** + * The type of the parameters object (usually 'object'). + */ + type: string; + /** + * List of required parameter names. + */ + required?: string[]; + /** + * Definitions of each parameter. + */ + properties: { + [k: string]: { + /** + * The data type of the parameter. + */ + type: string; + /** + * A description of the expected parameter. + */ + description: string; + }; + }; + }; + }; + })[]; + response_format?: Ai_Cf_Aisingapore_Gemma_Sea_Lion_V4_27B_It_JSON_Mode_3; + /** + * If true, a chat template is not applied and you must adhere to the specific model's expected formatting. + */ + raw?: boolean; + /** + * If true, the response will be streamed back incrementally using SSE, Server Sent Events. + */ + stream?: boolean; + /** + * The maximum number of tokens to generate in the response. + */ + max_tokens?: number; + /** + * Controls the randomness of the output; higher values produce more random results. + */ + temperature?: number; + /** + * Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses. + */ + top_p?: number; + /** + * Limits the AI to choose from the top 'k' most probable words. Lower values make responses more focused; higher values introduce more variety and potential surprises. + */ + top_k?: number; + /** + * Random seed for reproducibility of the generation. + */ + seed?: number; + /** + * Penalty for repeated tokens; higher values discourage repetition. + */ + repetition_penalty?: number; + /** + * Decreases the likelihood of the model repeating the same lines verbatim. + */ + frequency_penalty?: number; + /** + * Increases the likelihood of the model introducing new topics. + */ + presence_penalty?: number; +} +interface Ai_Cf_Aisingapore_Gemma_Sea_Lion_V4_27B_It_JSON_Mode_3 { + type?: "json_object" | "json_schema"; + json_schema?: unknown; +} +type Ai_Cf_Aisingapore_Gemma_Sea_Lion_V4_27B_It_Output = Ai_Cf_Aisingapore_Gemma_Sea_Lion_V4_27B_It_Chat_Completion_Response | Ai_Cf_Aisingapore_Gemma_Sea_Lion_V4_27B_It_Text_Completion_Response | string | Ai_Cf_Aisingapore_Gemma_Sea_Lion_V4_27B_It_AsyncResponse; +interface Ai_Cf_Aisingapore_Gemma_Sea_Lion_V4_27B_It_Chat_Completion_Response { + /** + * Unique identifier for the completion + */ + id?: string; + /** + * Object type identifier + */ + object?: "chat.completion"; + /** + * Unix timestamp of when the completion was created + */ + created?: number; + /** + * Model used for the completion + */ + model?: string; + /** + * List of completion choices + */ + choices?: { + /** + * Index of the choice in the list + */ + index?: number; + /** + * The message generated by the model + */ + message?: { + /** + * Role of the message author + */ + role: string; + /** + * The content of the message + */ + content: string; + /** + * Internal reasoning content (if available) + */ + reasoning_content?: string; + /** + * Tool calls made by the assistant + */ + tool_calls?: { + /** + * Unique identifier for the tool call + */ + id: string; + /** + * Type of tool call + */ + type: "function"; + function: { + /** + * Name of the function to call + */ + name: string; + /** + * JSON string of arguments for the function + */ + arguments: string; + }; + }[]; + }; + /** + * Reason why the model stopped generating + */ + finish_reason?: string; + /** + * Stop reason (may be null) + */ + stop_reason?: string | null; + /** + * Log probabilities (if requested) + */ + logprobs?: {} | null; + }[]; + /** + * Usage statistics for the inference request + */ + usage?: { + /** + * Total number of tokens in input + */ + prompt_tokens?: number; + /** + * Total number of tokens in output + */ + completion_tokens?: number; + /** + * Total number of input and output tokens + */ + total_tokens?: number; + }; + /** + * Log probabilities for the prompt (if requested) + */ + prompt_logprobs?: {} | null; +} +interface Ai_Cf_Aisingapore_Gemma_Sea_Lion_V4_27B_It_Text_Completion_Response { + /** + * Unique identifier for the completion + */ + id?: string; + /** + * Object type identifier + */ + object?: "text_completion"; + /** + * Unix timestamp of when the completion was created + */ + created?: number; + /** + * Model used for the completion + */ + model?: string; + /** + * List of completion choices + */ + choices?: { + /** + * Index of the choice in the list + */ + index: number; + /** + * The generated text completion + */ + text: string; + /** + * Reason why the model stopped generating + */ + finish_reason: string; + /** + * Stop reason (may be null) + */ + stop_reason?: string | null; + /** + * Log probabilities (if requested) + */ + logprobs?: {} | null; + /** + * Log probabilities for the prompt (if requested) + */ + prompt_logprobs?: {} | null; + }[]; + /** + * Usage statistics for the inference request + */ + usage?: { + /** + * Total number of tokens in input + */ + prompt_tokens?: number; + /** + * Total number of tokens in output + */ + completion_tokens?: number; + /** + * Total number of input and output tokens + */ + total_tokens?: number; + }; +} +interface Ai_Cf_Aisingapore_Gemma_Sea_Lion_V4_27B_It_AsyncResponse { + /** + * The async request id that can be used to obtain the results. + */ + request_id?: string; +} +declare abstract class Base_Ai_Cf_Aisingapore_Gemma_Sea_Lion_V4_27B_It { + inputs: Ai_Cf_Aisingapore_Gemma_Sea_Lion_V4_27B_It_Input; + postProcessedOutputs: Ai_Cf_Aisingapore_Gemma_Sea_Lion_V4_27B_It_Output; +} +interface Ai_Cf_Pfnet_Plamo_Embedding_1B_Input { + /** + * Input text to embed. Can be a single string or a list of strings. + */ + text: string | string[]; +} +interface Ai_Cf_Pfnet_Plamo_Embedding_1B_Output { + /** + * Embedding vectors, where each vector is a list of floats. + */ + data: number[][]; + /** + * Shape of the embedding data as [number_of_embeddings, embedding_dimension]. + * + * @minItems 2 + * @maxItems 2 + */ + shape: [ + number, + number + ]; +} +declare abstract class Base_Ai_Cf_Pfnet_Plamo_Embedding_1B { + inputs: Ai_Cf_Pfnet_Plamo_Embedding_1B_Input; + postProcessedOutputs: Ai_Cf_Pfnet_Plamo_Embedding_1B_Output; +} +interface Ai_Cf_Deepgram_Flux_Input { + /** + * Encoding of the audio stream. Currently only supports raw signed little-endian 16-bit PCM. + */ + encoding: "linear16"; + /** + * Sample rate of the audio stream in Hz. + */ + sample_rate: string; + /** + * End-of-turn confidence required to fire an eager end-of-turn event. When set, enables EagerEndOfTurn and TurnResumed events. Valid Values 0.3 - 0.9. + */ + eager_eot_threshold?: string; + /** + * End-of-turn confidence required to finish a turn. Valid Values 0.5 - 0.9. + */ + eot_threshold?: string; + /** + * A turn will be finished when this much time has passed after speech, regardless of EOT confidence. + */ + eot_timeout_ms?: string; + /** + * Keyterm prompting can improve recognition of specialized terminology. Pass multiple keyterm query parameters to boost multiple keyterms. + */ + keyterm?: string; + /** + * Opts out requests from the Deepgram Model Improvement Program. Refer to Deepgram Docs for pricing impacts before setting this to true. https://dpgr.am/deepgram-mip + */ + mip_opt_out?: "true" | "false"; + /** + * Label your requests for the purpose of identification during usage reporting + */ + tag?: string; +} +/** + * Output will be returned as websocket messages. + */ +interface Ai_Cf_Deepgram_Flux_Output { + /** + * The unique identifier of the request (uuid) + */ + request_id?: string; + /** + * Starts at 0 and increments for each message the server sends to the client. + */ + sequence_id?: number; + /** + * The type of event being reported. + */ + event?: "Update" | "StartOfTurn" | "EagerEndOfTurn" | "TurnResumed" | "EndOfTurn"; + /** + * The index of the current turn + */ + turn_index?: number; + /** + * Start time in seconds of the audio range that was transcribed + */ + audio_window_start?: number; + /** + * End time in seconds of the audio range that was transcribed + */ + audio_window_end?: number; + /** + * Text that was said over the course of the current turn + */ + transcript?: string; + /** + * The words in the transcript + */ + words?: { + /** + * The individual punctuated, properly-cased word from the transcript + */ + word: string; + /** + * Confidence that this word was transcribed correctly + */ + confidence: number; + }[]; + /** + * Confidence that no more speech is coming in this turn + */ + end_of_turn_confidence?: number; +} +declare abstract class Base_Ai_Cf_Deepgram_Flux { + inputs: Ai_Cf_Deepgram_Flux_Input; + postProcessedOutputs: Ai_Cf_Deepgram_Flux_Output; +} +interface Ai_Cf_Deepgram_Aura_2_En_Input { + /** + * Speaker used to produce the audio. + */ + speaker?: "amalthea" | "andromeda" | "apollo" | "arcas" | "aries" | "asteria" | "athena" | "atlas" | "aurora" | "callista" | "cora" | "cordelia" | "delia" | "draco" | "electra" | "harmonia" | "helena" | "hera" | "hermes" | "hyperion" | "iris" | "janus" | "juno" | "jupiter" | "luna" | "mars" | "minerva" | "neptune" | "odysseus" | "ophelia" | "orion" | "orpheus" | "pandora" | "phoebe" | "pluto" | "saturn" | "thalia" | "theia" | "vesta" | "zeus"; + /** + * Encoding of the output audio. + */ + encoding?: "linear16" | "flac" | "mulaw" | "alaw" | "mp3" | "opus" | "aac"; + /** + * Container specifies the file format wrapper for the output audio. The available options depend on the encoding type.. + */ + container?: "none" | "wav" | "ogg"; + /** + * The text content to be converted to speech + */ + text: string; + /** + * Sample Rate specifies the sample rate for the output audio. Based on the encoding, different sample rates are supported. For some encodings, the sample rate is not configurable + */ + sample_rate?: number; + /** + * The bitrate of the audio in bits per second. Choose from predefined ranges or specific values based on the encoding type. + */ + bit_rate?: number; +} +/** + * The generated audio in MP3 format + */ +type Ai_Cf_Deepgram_Aura_2_En_Output = string; +declare abstract class Base_Ai_Cf_Deepgram_Aura_2_En { + inputs: Ai_Cf_Deepgram_Aura_2_En_Input; + postProcessedOutputs: Ai_Cf_Deepgram_Aura_2_En_Output; +} +interface Ai_Cf_Deepgram_Aura_2_Es_Input { + /** + * Speaker used to produce the audio. + */ + speaker?: "sirio" | "nestor" | "carina" | "celeste" | "alvaro" | "diana" | "aquila" | "selena" | "estrella" | "javier"; + /** + * Encoding of the output audio. + */ + encoding?: "linear16" | "flac" | "mulaw" | "alaw" | "mp3" | "opus" | "aac"; + /** + * Container specifies the file format wrapper for the output audio. The available options depend on the encoding type.. + */ + container?: "none" | "wav" | "ogg"; + /** + * The text content to be converted to speech + */ + text: string; + /** + * Sample Rate specifies the sample rate for the output audio. Based on the encoding, different sample rates are supported. For some encodings, the sample rate is not configurable + */ + sample_rate?: number; + /** + * The bitrate of the audio in bits per second. Choose from predefined ranges or specific values based on the encoding type. + */ + bit_rate?: number; +} +/** + * The generated audio in MP3 format + */ +type Ai_Cf_Deepgram_Aura_2_Es_Output = string; +declare abstract class Base_Ai_Cf_Deepgram_Aura_2_Es { + inputs: Ai_Cf_Deepgram_Aura_2_Es_Input; + postProcessedOutputs: Ai_Cf_Deepgram_Aura_2_Es_Output; +} +interface AiModels { + "@cf/huggingface/distilbert-sst-2-int8": BaseAiTextClassification; + "@cf/stabilityai/stable-diffusion-xl-base-1.0": BaseAiTextToImage; + "@cf/runwayml/stable-diffusion-v1-5-inpainting": BaseAiTextToImage; + "@cf/runwayml/stable-diffusion-v1-5-img2img": BaseAiTextToImage; + "@cf/lykon/dreamshaper-8-lcm": BaseAiTextToImage; + "@cf/bytedance/stable-diffusion-xl-lightning": BaseAiTextToImage; + "@cf/myshell-ai/melotts": BaseAiTextToSpeech; + "@cf/google/embeddinggemma-300m": BaseAiTextEmbeddings; + "@cf/microsoft/resnet-50": BaseAiImageClassification; + "@cf/meta/llama-2-7b-chat-int8": BaseAiTextGeneration; + "@cf/mistral/mistral-7b-instruct-v0.1": BaseAiTextGeneration; + "@cf/meta/llama-2-7b-chat-fp16": BaseAiTextGeneration; + "@hf/thebloke/llama-2-13b-chat-awq": BaseAiTextGeneration; + "@hf/thebloke/mistral-7b-instruct-v0.1-awq": BaseAiTextGeneration; + "@hf/thebloke/zephyr-7b-beta-awq": BaseAiTextGeneration; + "@hf/thebloke/openhermes-2.5-mistral-7b-awq": BaseAiTextGeneration; + "@hf/thebloke/neural-chat-7b-v3-1-awq": BaseAiTextGeneration; + "@hf/thebloke/llamaguard-7b-awq": BaseAiTextGeneration; + "@hf/thebloke/deepseek-coder-6.7b-base-awq": BaseAiTextGeneration; + "@hf/thebloke/deepseek-coder-6.7b-instruct-awq": BaseAiTextGeneration; + "@cf/deepseek-ai/deepseek-math-7b-instruct": BaseAiTextGeneration; + "@cf/defog/sqlcoder-7b-2": BaseAiTextGeneration; + "@cf/openchat/openchat-3.5-0106": BaseAiTextGeneration; + "@cf/tiiuae/falcon-7b-instruct": BaseAiTextGeneration; + "@cf/thebloke/discolm-german-7b-v1-awq": BaseAiTextGeneration; + "@cf/qwen/qwen1.5-0.5b-chat": BaseAiTextGeneration; + "@cf/qwen/qwen1.5-7b-chat-awq": BaseAiTextGeneration; + "@cf/qwen/qwen1.5-14b-chat-awq": BaseAiTextGeneration; + "@cf/tinyllama/tinyllama-1.1b-chat-v1.0": BaseAiTextGeneration; + "@cf/microsoft/phi-2": BaseAiTextGeneration; + "@cf/qwen/qwen1.5-1.8b-chat": BaseAiTextGeneration; + "@cf/mistral/mistral-7b-instruct-v0.2-lora": BaseAiTextGeneration; + "@hf/nousresearch/hermes-2-pro-mistral-7b": BaseAiTextGeneration; + "@hf/nexusflow/starling-lm-7b-beta": BaseAiTextGeneration; + "@hf/google/gemma-7b-it": BaseAiTextGeneration; + "@cf/meta-llama/llama-2-7b-chat-hf-lora": BaseAiTextGeneration; + "@cf/google/gemma-2b-it-lora": BaseAiTextGeneration; + "@cf/google/gemma-7b-it-lora": BaseAiTextGeneration; + "@hf/mistral/mistral-7b-instruct-v0.2": BaseAiTextGeneration; + "@cf/meta/llama-3-8b-instruct": BaseAiTextGeneration; + "@cf/fblgit/una-cybertron-7b-v2-bf16": BaseAiTextGeneration; + "@cf/meta/llama-3-8b-instruct-awq": BaseAiTextGeneration; + "@cf/meta/llama-3.1-8b-instruct-fp8": BaseAiTextGeneration; + "@cf/meta/llama-3.1-8b-instruct-awq": BaseAiTextGeneration; + "@cf/meta/llama-3.2-3b-instruct": BaseAiTextGeneration; + "@cf/meta/llama-3.2-1b-instruct": BaseAiTextGeneration; + "@cf/deepseek-ai/deepseek-r1-distill-qwen-32b": BaseAiTextGeneration; + "@cf/ibm-granite/granite-4.0-h-micro": BaseAiTextGeneration; + "@cf/facebook/bart-large-cnn": BaseAiSummarization; + "@cf/llava-hf/llava-1.5-7b-hf": BaseAiImageToText; + "@cf/baai/bge-base-en-v1.5": Base_Ai_Cf_Baai_Bge_Base_En_V1_5; + "@cf/openai/whisper": Base_Ai_Cf_Openai_Whisper; + "@cf/meta/m2m100-1.2b": Base_Ai_Cf_Meta_M2M100_1_2B; + "@cf/baai/bge-small-en-v1.5": Base_Ai_Cf_Baai_Bge_Small_En_V1_5; + "@cf/baai/bge-large-en-v1.5": Base_Ai_Cf_Baai_Bge_Large_En_V1_5; + "@cf/unum/uform-gen2-qwen-500m": Base_Ai_Cf_Unum_Uform_Gen2_Qwen_500M; + "@cf/openai/whisper-tiny-en": Base_Ai_Cf_Openai_Whisper_Tiny_En; + "@cf/openai/whisper-large-v3-turbo": Base_Ai_Cf_Openai_Whisper_Large_V3_Turbo; + "@cf/baai/bge-m3": Base_Ai_Cf_Baai_Bge_M3; + "@cf/black-forest-labs/flux-1-schnell": Base_Ai_Cf_Black_Forest_Labs_Flux_1_Schnell; + "@cf/meta/llama-3.2-11b-vision-instruct": Base_Ai_Cf_Meta_Llama_3_2_11B_Vision_Instruct; + "@cf/meta/llama-3.3-70b-instruct-fp8-fast": Base_Ai_Cf_Meta_Llama_3_3_70B_Instruct_Fp8_Fast; + "@cf/meta/llama-guard-3-8b": Base_Ai_Cf_Meta_Llama_Guard_3_8B; + "@cf/baai/bge-reranker-base": Base_Ai_Cf_Baai_Bge_Reranker_Base; + "@cf/qwen/qwen2.5-coder-32b-instruct": Base_Ai_Cf_Qwen_Qwen2_5_Coder_32B_Instruct; + "@cf/qwen/qwq-32b": Base_Ai_Cf_Qwen_Qwq_32B; + "@cf/mistralai/mistral-small-3.1-24b-instruct": Base_Ai_Cf_Mistralai_Mistral_Small_3_1_24B_Instruct; + "@cf/google/gemma-3-12b-it": Base_Ai_Cf_Google_Gemma_3_12B_It; + "@cf/meta/llama-4-scout-17b-16e-instruct": Base_Ai_Cf_Meta_Llama_4_Scout_17B_16E_Instruct; + "@cf/qwen/qwen3-30b-a3b-fp8": Base_Ai_Cf_Qwen_Qwen3_30B_A3B_Fp8; + "@cf/deepgram/nova-3": Base_Ai_Cf_Deepgram_Nova_3; + "@cf/qwen/qwen3-embedding-0.6b": Base_Ai_Cf_Qwen_Qwen3_Embedding_0_6B; + "@cf/pipecat-ai/smart-turn-v2": Base_Ai_Cf_Pipecat_Ai_Smart_Turn_V2; + "@cf/openai/gpt-oss-120b": Base_Ai_Cf_Openai_Gpt_Oss_120B; + "@cf/openai/gpt-oss-20b": Base_Ai_Cf_Openai_Gpt_Oss_20B; + "@cf/leonardo/phoenix-1.0": Base_Ai_Cf_Leonardo_Phoenix_1_0; + "@cf/leonardo/lucid-origin": Base_Ai_Cf_Leonardo_Lucid_Origin; + "@cf/deepgram/aura-1": Base_Ai_Cf_Deepgram_Aura_1; + "@cf/ai4bharat/indictrans2-en-indic-1B": Base_Ai_Cf_Ai4Bharat_Indictrans2_En_Indic_1B; + "@cf/aisingapore/gemma-sea-lion-v4-27b-it": Base_Ai_Cf_Aisingapore_Gemma_Sea_Lion_V4_27B_It; + "@cf/pfnet/plamo-embedding-1b": Base_Ai_Cf_Pfnet_Plamo_Embedding_1B; + "@cf/deepgram/flux": Base_Ai_Cf_Deepgram_Flux; + "@cf/deepgram/aura-2-en": Base_Ai_Cf_Deepgram_Aura_2_En; + "@cf/deepgram/aura-2-es": Base_Ai_Cf_Deepgram_Aura_2_Es; +} +type AiOptions = { + /** + * Send requests as an asynchronous batch job, only works for supported models + * https://developers.cloudflare.com/workers-ai/features/batch-api + */ + queueRequest?: boolean; + /** + * Establish websocket connections, only works for supported models + */ + websocket?: boolean; + /** + * Tag your requests to group and view them in Cloudflare dashboard. + * + * Rules: + * Tags must only contain letters, numbers, and the symbols: : - . / @ + * Each tag can have maximum 50 characters. + * Maximum 5 tags are allowed each request. + * Duplicate tags will removed. + */ + tags?: string[]; + gateway?: GatewayOptions; + returnRawResponse?: boolean; + prefix?: string; + extraHeaders?: object; +}; +type AiModelsSearchParams = { + author?: string; + hide_experimental?: boolean; + page?: number; + per_page?: number; + search?: string; + source?: number; + task?: string; +}; +type AiModelsSearchObject = { + id: string; + source: number; + name: string; + description: string; + task: { + id: string; + name: string; + description: string; + }; + tags: string[]; + properties: { + property_id: string; + value: string; + }[]; +}; +interface InferenceUpstreamError extends Error { +} +interface AiInternalError extends Error { +} +type AiModelListType = Record; +declare abstract class Ai { + aiGatewayLogId: string | null; + gateway(gatewayId: string): AiGateway; + autorag(autoragId: string): AutoRAG; + run(model: Name, inputs: InputOptions, options?: Options): Promise; + models(params?: AiModelsSearchParams): Promise; + toMarkdown(): ToMarkdownService; + toMarkdown(files: MarkdownDocument[], options?: ConversionRequestOptions): Promise; + toMarkdown(files: MarkdownDocument, options?: ConversionRequestOptions): Promise; +} +type GatewayRetries = { + maxAttempts?: 1 | 2 | 3 | 4 | 5; + retryDelayMs?: number; + backoff?: 'constant' | 'linear' | 'exponential'; +}; +type GatewayOptions = { + id: string; + cacheKey?: string; + cacheTtl?: number; + skipCache?: boolean; + metadata?: Record; + collectLog?: boolean; + eventId?: string; + requestTimeoutMs?: number; + retries?: GatewayRetries; +}; +type UniversalGatewayOptions = Exclude & { + /** + ** @deprecated + */ + id?: string; +}; +type AiGatewayPatchLog = { + score?: number | null; + feedback?: -1 | 1 | null; + metadata?: Record | null; +}; +type AiGatewayLog = { + id: string; + provider: string; + model: string; + model_type?: string; + path: string; + duration: number; + request_type?: string; + request_content_type?: string; + status_code: number; + response_content_type?: string; + success: boolean; + cached: boolean; + tokens_in?: number; + tokens_out?: number; + metadata?: Record; + step?: number; + cost?: number; + custom_cost?: boolean; + request_size: number; + request_head?: string; + request_head_complete: boolean; + response_size: number; + response_head?: string; + response_head_complete: boolean; + created_at: Date; +}; +type AIGatewayProviders = 'workers-ai' | 'anthropic' | 'aws-bedrock' | 'azure-openai' | 'google-vertex-ai' | 'huggingface' | 'openai' | 'perplexity-ai' | 'replicate' | 'groq' | 'cohere' | 'google-ai-studio' | 'mistral' | 'grok' | 'openrouter' | 'deepseek' | 'cerebras' | 'cartesia' | 'elevenlabs' | 'adobe-firefly'; +type AIGatewayHeaders = { + 'cf-aig-metadata': Record | string; + 'cf-aig-custom-cost': { + per_token_in?: number; + per_token_out?: number; + } | { + total_cost?: number; + } | string; + 'cf-aig-cache-ttl': number | string; + 'cf-aig-skip-cache': boolean | string; + 'cf-aig-cache-key': string; + 'cf-aig-event-id': string; + 'cf-aig-request-timeout': number | string; + 'cf-aig-max-attempts': number | string; + 'cf-aig-retry-delay': number | string; + 'cf-aig-backoff': string; + 'cf-aig-collect-log': boolean | string; + Authorization: string; + 'Content-Type': string; + [key: string]: string | number | boolean | object; +}; +type AIGatewayUniversalRequest = { + provider: AIGatewayProviders | string; // eslint-disable-line + endpoint: string; + headers: Partial; + query: unknown; +}; +interface AiGatewayInternalError extends Error { +} +interface AiGatewayLogNotFound extends Error { +} +declare abstract class AiGateway { + patchLog(logId: string, data: AiGatewayPatchLog): Promise; + getLog(logId: string): Promise; + run(data: AIGatewayUniversalRequest | AIGatewayUniversalRequest[], options?: { + gateway?: UniversalGatewayOptions; + extraHeaders?: object; + }): Promise; + getUrl(provider?: AIGatewayProviders | string): Promise; // eslint-disable-line +} +interface AutoRAGInternalError extends Error { +} +interface AutoRAGNotFoundError extends Error { +} +interface AutoRAGUnauthorizedError extends Error { +} +interface AutoRAGNameNotSetError extends Error { +} +type ComparisonFilter = { + key: string; + type: 'eq' | 'ne' | 'gt' | 'gte' | 'lt' | 'lte'; + value: string | number | boolean; +}; +type CompoundFilter = { + type: 'and' | 'or'; + filters: ComparisonFilter[]; +}; +type AutoRagSearchRequest = { + query: string; + filters?: CompoundFilter | ComparisonFilter; + max_num_results?: number; + ranking_options?: { + ranker?: string; + score_threshold?: number; + }; + reranking?: { + enabled?: boolean; + model?: string; + }; + rewrite_query?: boolean; +}; +type AutoRagAiSearchRequest = AutoRagSearchRequest & { + stream?: boolean; + system_prompt?: string; +}; +type AutoRagAiSearchRequestStreaming = Omit & { + stream: true; +}; +type AutoRagSearchResponse = { + object: 'vector_store.search_results.page'; + search_query: string; + data: { + file_id: string; + filename: string; + score: number; + attributes: Record; + content: { + type: 'text'; + text: string; + }[]; + }[]; + has_more: boolean; + next_page: string | null; +}; +type AutoRagListResponse = { + id: string; + enable: boolean; + type: string; + source: string; + vectorize_name: string; + paused: boolean; + status: string; +}[]; +type AutoRagAiSearchResponse = AutoRagSearchResponse & { + response: string; +}; +declare abstract class AutoRAG { + list(): Promise; + search(params: AutoRagSearchRequest): Promise; + aiSearch(params: AutoRagAiSearchRequestStreaming): Promise; + aiSearch(params: AutoRagAiSearchRequest): Promise; + aiSearch(params: AutoRagAiSearchRequest): Promise; +} +interface BasicImageTransformations { + /** + * Maximum width in image pixels. The value must be an integer. + */ + width?: number; + /** + * Maximum height in image pixels. The value must be an integer. + */ + height?: number; + /** + * Resizing mode as a string. It affects interpretation of width and height + * options: + * - scale-down: Similar to contain, but the image is never enlarged. If + * the image is larger than given width or height, it will be resized. + * Otherwise its original size will be kept. + * - contain: Resizes to maximum size that fits within the given width and + * height. If only a single dimension is given (e.g. only width), the + * image will be shrunk or enlarged to exactly match that dimension. + * Aspect ratio is always preserved. + * - cover: Resizes (shrinks or enlarges) to fill the entire area of width + * and height. If the image has an aspect ratio different from the ratio + * of width and height, it will be cropped to fit. + * - crop: The image will be shrunk and cropped to fit within the area + * specified by width and height. The image will not be enlarged. For images + * smaller than the given dimensions it's the same as scale-down. For + * images larger than the given dimensions, it's the same as cover. + * See also trim. + * - pad: Resizes to the maximum size that fits within the given width and + * height, and then fills the remaining area with a background color + * (white by default). Use of this mode is not recommended, as the same + * effect can be more efficiently achieved with the contain mode and the + * CSS object-fit: contain property. + * - squeeze: Stretches and deforms to the width and height given, even if it + * breaks aspect ratio + */ + fit?: "scale-down" | "contain" | "cover" | "crop" | "pad" | "squeeze"; + /** + * Image segmentation using artificial intelligence models. Sets pixels not + * within selected segment area to transparent e.g "foreground" sets every + * background pixel as transparent. + */ + segment?: "foreground"; + /** + * When cropping with fit: "cover", this defines the side or point that should + * be left uncropped. The value is either a string + * "left", "right", "top", "bottom", "auto", or "center" (the default), + * or an object {x, y} containing focal point coordinates in the original + * image expressed as fractions ranging from 0.0 (top or left) to 1.0 + * (bottom or right), 0.5 being the center. {fit: "cover", gravity: "top"} will + * crop bottom or left and right sides as necessary, but won’t crop anything + * from the top. {fit: "cover", gravity: {x:0.5, y:0.2}} will crop each side to + * preserve as much as possible around a point at 20% of the height of the + * source image. + */ + gravity?: 'face' | 'left' | 'right' | 'top' | 'bottom' | 'center' | 'auto' | 'entropy' | BasicImageTransformationsGravityCoordinates; + /** + * Background color to add underneath the image. Applies only to images with + * transparency (such as PNG). Accepts any CSS color (#RRGGBB, rgba(…), + * hsl(…), etc.) + */ + background?: string; + /** + * Number of degrees (90, 180, 270) to rotate the image by. width and height + * options refer to axes after rotation. + */ + rotate?: 0 | 90 | 180 | 270 | 360; +} +interface BasicImageTransformationsGravityCoordinates { + x?: number; + y?: number; + mode?: 'remainder' | 'box-center'; +} +/** + * In addition to the properties you can set in the RequestInit dict + * that you pass as an argument to the Request constructor, you can + * set certain properties of a `cf` object to control how Cloudflare + * features are applied to that new Request. + * + * Note: Currently, these properties cannot be tested in the + * playground. + */ +interface RequestInitCfProperties extends Record { + cacheEverything?: boolean; + /** + * A request's cache key is what determines if two requests are + * "the same" for caching purposes. If a request has the same cache key + * as some previous request, then we can serve the same cached response for + * both. (e.g. 'some-key') + * + * Only available for Enterprise customers. + */ + cacheKey?: string; + /** + * This allows you to append additional Cache-Tag response headers + * to the origin response without modifications to the origin server. + * This will allow for greater control over the Purge by Cache Tag feature + * utilizing changes only in the Workers process. + * + * Only available for Enterprise customers. + */ + cacheTags?: string[]; + /** + * Force response to be cached for a given number of seconds. (e.g. 300) + */ + cacheTtl?: number; + /** + * Force response to be cached for a given number of seconds based on the Origin status code. + * (e.g. { '200-299': 86400, '404': 1, '500-599': 0 }) + */ + cacheTtlByStatus?: Record; + scrapeShield?: boolean; + apps?: boolean; + image?: RequestInitCfPropertiesImage; + minify?: RequestInitCfPropertiesImageMinify; + mirage?: boolean; + polish?: "lossy" | "lossless" | "off"; + r2?: RequestInitCfPropertiesR2; + /** + * Redirects the request to an alternate origin server. You can use this, + * for example, to implement load balancing across several origins. + * (e.g.us-east.example.com) + * + * Note - For security reasons, the hostname set in resolveOverride must + * be proxied on the same Cloudflare zone of the incoming request. + * Otherwise, the setting is ignored. CNAME hosts are allowed, so to + * resolve to a host under a different domain or a DNS only domain first + * declare a CNAME record within your own zone’s DNS mapping to the + * external hostname, set proxy on Cloudflare, then set resolveOverride + * to point to that CNAME record. + */ + resolveOverride?: string; +} +interface RequestInitCfPropertiesImageDraw extends BasicImageTransformations { + /** + * Absolute URL of the image file to use for the drawing. It can be any of + * the supported file formats. For drawing of watermarks or non-rectangular + * overlays we recommend using PNG or WebP images. + */ + url: string; + /** + * Floating-point number between 0 (transparent) and 1 (opaque). + * For example, opacity: 0.5 makes overlay semitransparent. + */ + opacity?: number; + /** + * - If set to true, the overlay image will be tiled to cover the entire + * area. This is useful for stock-photo-like watermarks. + * - If set to "x", the overlay image will be tiled horizontally only + * (form a line). + * - If set to "y", the overlay image will be tiled vertically only + * (form a line). + */ + repeat?: true | "x" | "y"; + /** + * Position of the overlay image relative to a given edge. Each property is + * an offset in pixels. 0 aligns exactly to the edge. For example, left: 10 + * positions left side of the overlay 10 pixels from the left edge of the + * image it's drawn over. bottom: 0 aligns bottom of the overlay with bottom + * of the background image. + * + * Setting both left & right, or both top & bottom is an error. + * + * If no position is specified, the image will be centered. + */ + top?: number; + left?: number; + bottom?: number; + right?: number; +} +interface RequestInitCfPropertiesImage extends BasicImageTransformations { + /** + * Device Pixel Ratio. Default 1. Multiplier for width/height that makes it + * easier to specify higher-DPI sizes in . + */ + dpr?: number; + /** + * Allows you to trim your image. Takes dpr into account and is performed before + * resizing or rotation. + * + * It can be used as: + * - left, top, right, bottom - it will specify the number of pixels to cut + * off each side + * - width, height - the width/height you'd like to end up with - can be used + * in combination with the properties above + * - border - this will automatically trim the surroundings of an image based on + * it's color. It consists of three properties: + * - color: rgb or hex representation of the color you wish to trim (todo: verify the rgba bit) + * - tolerance: difference from color to treat as color + * - keep: the number of pixels of border to keep + */ + trim?: "border" | { + top?: number; + bottom?: number; + left?: number; + right?: number; + width?: number; + height?: number; + border?: boolean | { + color?: string; + tolerance?: number; + keep?: number; + }; + }; + /** + * Quality setting from 1-100 (useful values are in 60-90 range). Lower values + * make images look worse, but load faster. The default is 85. It applies only + * to JPEG and WebP images. It doesn’t have any effect on PNG. + */ + quality?: number | "low" | "medium-low" | "medium-high" | "high"; + /** + * Output format to generate. It can be: + * - avif: generate images in AVIF format. + * - webp: generate images in Google WebP format. Set quality to 100 to get + * the WebP-lossless format. + * - json: instead of generating an image, outputs information about the + * image, in JSON format. The JSON object will contain image size + * (before and after resizing), source image’s MIME type, file size, etc. + * - jpeg: generate images in JPEG format. + * - png: generate images in PNG format. + */ + format?: "avif" | "webp" | "json" | "jpeg" | "png" | "baseline-jpeg" | "png-force" | "svg"; + /** + * Whether to preserve animation frames from input files. Default is true. + * Setting it to false reduces animations to still images. This setting is + * recommended when enlarging images or processing arbitrary user content, + * because large GIF animations can weigh tens or even hundreds of megabytes. + * It is also useful to set anim:false when using format:"json" to get the + * response quicker without the number of frames. + */ + anim?: boolean; + /** + * What EXIF data should be preserved in the output image. Note that EXIF + * rotation and embedded color profiles are always applied ("baked in" into + * the image), and aren't affected by this option. Note that if the Polish + * feature is enabled, all metadata may have been removed already and this + * option may have no effect. + * - keep: Preserve most of EXIF metadata, including GPS location if there's + * any. + * - copyright: Only keep the copyright tag, and discard everything else. + * This is the default behavior for JPEG files. + * - none: Discard all invisible EXIF metadata. Currently WebP and PNG + * output formats always discard metadata. + */ + metadata?: "keep" | "copyright" | "none"; + /** + * Strength of sharpening filter to apply to the image. Floating-point + * number between 0 (no sharpening, default) and 10 (maximum). 1.0 is a + * recommended value for downscaled images. + */ + sharpen?: number; + /** + * Radius of a blur filter (approximate gaussian). Maximum supported radius + * is 250. + */ + blur?: number; + /** + * Overlays are drawn in the order they appear in the array (last array + * entry is the topmost layer). + */ + draw?: RequestInitCfPropertiesImageDraw[]; + /** + * Fetching image from authenticated origin. Setting this property will + * pass authentication headers (Authorization, Cookie, etc.) through to + * the origin. + */ + "origin-auth"?: "share-publicly"; + /** + * Adds a border around the image. The border is added after resizing. Border + * width takes dpr into account, and can be specified either using a single + * width property, or individually for each side. + */ + border?: { + color: string; + width: number; + } | { + color: string; + top: number; + right: number; + bottom: number; + left: number; + }; + /** + * Increase brightness by a factor. A value of 1.0 equals no change, a value + * of 0.5 equals half brightness, and a value of 2.0 equals twice as bright. + * 0 is ignored. + */ + brightness?: number; + /** + * Increase contrast by a factor. A value of 1.0 equals no change, a value of + * 0.5 equals low contrast, and a value of 2.0 equals high contrast. 0 is + * ignored. + */ + contrast?: number; + /** + * Increase exposure by a factor. A value of 1.0 equals no change, a value of + * 0.5 darkens the image, and a value of 2.0 lightens the image. 0 is ignored. + */ + gamma?: number; + /** + * Increase contrast by a factor. A value of 1.0 equals no change, a value of + * 0.5 equals low contrast, and a value of 2.0 equals high contrast. 0 is + * ignored. + */ + saturation?: number; + /** + * Flips the images horizontally, vertically, or both. Flipping is applied before + * rotation, so if you apply flip=h,rotate=90 then the image will be flipped + * horizontally, then rotated by 90 degrees. + */ + flip?: 'h' | 'v' | 'hv'; + /** + * Slightly reduces latency on a cache miss by selecting a + * quickest-to-compress file format, at a cost of increased file size and + * lower image quality. It will usually override the format option and choose + * JPEG over WebP or AVIF. We do not recommend using this option, except in + * unusual circumstances like resizing uncacheable dynamically-generated + * images. + */ + compression?: "fast"; +} +interface RequestInitCfPropertiesImageMinify { + javascript?: boolean; + css?: boolean; + html?: boolean; +} +interface RequestInitCfPropertiesR2 { + /** + * Colo id of bucket that an object is stored in + */ + bucketColoId?: number; +} +/** + * Request metadata provided by Cloudflare's edge. + */ +type IncomingRequestCfProperties = IncomingRequestCfPropertiesBase & IncomingRequestCfPropertiesBotManagementEnterprise & IncomingRequestCfPropertiesCloudflareForSaaSEnterprise & IncomingRequestCfPropertiesGeographicInformation & IncomingRequestCfPropertiesCloudflareAccessOrApiShield; +interface IncomingRequestCfPropertiesBase extends Record { + /** + * [ASN](https://www.iana.org/assignments/as-numbers/as-numbers.xhtml) of the incoming request. + * + * @example 395747 + */ + asn?: number; + /** + * The organization which owns the ASN of the incoming request. + * + * @example "Google Cloud" + */ + asOrganization?: string; + /** + * The original value of the `Accept-Encoding` header if Cloudflare modified it. + * + * @example "gzip, deflate, br" + */ + clientAcceptEncoding?: string; + /** + * The number of milliseconds it took for the request to reach your worker. + * + * @example 22 + */ + clientTcpRtt?: number; + /** + * The three-letter [IATA](https://en.wikipedia.org/wiki/IATA_airport_code) + * airport code of the data center that the request hit. + * + * @example "DFW" + */ + colo: string; + /** + * Represents the upstream's response to a + * [TCP `keepalive` message](https://tldp.org/HOWTO/TCP-Keepalive-HOWTO/overview.html) + * from cloudflare. + * + * For workers with no upstream, this will always be `1`. + * + * @example 3 + */ + edgeRequestKeepAliveStatus: IncomingRequestCfPropertiesEdgeRequestKeepAliveStatus; + /** + * The HTTP Protocol the request used. + * + * @example "HTTP/2" + */ + httpProtocol: string; + /** + * The browser-requested prioritization information in the request object. + * + * If no information was set, defaults to the empty string `""` + * + * @example "weight=192;exclusive=0;group=3;group-weight=127" + * @default "" + */ + requestPriority: string; + /** + * The TLS version of the connection to Cloudflare. + * In requests served over plaintext (without TLS), this property is the empty string `""`. + * + * @example "TLSv1.3" + */ + tlsVersion: string; + /** + * The cipher for the connection to Cloudflare. + * In requests served over plaintext (without TLS), this property is the empty string `""`. + * + * @example "AEAD-AES128-GCM-SHA256" + */ + tlsCipher: string; + /** + * Metadata containing the [`HELLO`](https://www.rfc-editor.org/rfc/rfc5246#section-7.4.1.2) and [`FINISHED`](https://www.rfc-editor.org/rfc/rfc5246#section-7.4.9) messages from this request's TLS handshake. + * + * If the incoming request was served over plaintext (without TLS) this field is undefined. + */ + tlsExportedAuthenticator?: IncomingRequestCfPropertiesExportedAuthenticatorMetadata; +} +interface IncomingRequestCfPropertiesBotManagementBase { + /** + * Cloudflare’s [level of certainty](https://developers.cloudflare.com/bots/concepts/bot-score/) that a request comes from a bot, + * represented as an integer percentage between `1` (almost certainly a bot) and `99` (almost certainly human). + * + * @example 54 + */ + score: number; + /** + * A boolean value that is true if the request comes from a good bot, like Google or Bing. + * Most customers choose to allow this traffic. For more details, see [Traffic from known bots](https://developers.cloudflare.com/firewall/known-issues-and-faq/#how-does-firewall-rules-handle-traffic-from-known-bots). + */ + verifiedBot: boolean; + /** + * A boolean value that is true if the request originates from a + * Cloudflare-verified proxy service. + */ + corporateProxy: boolean; + /** + * A boolean value that's true if the request matches [file extensions](https://developers.cloudflare.com/bots/reference/static-resources/) for many types of static resources. + */ + staticResource: boolean; + /** + * List of IDs that correlate to the Bot Management heuristic detections made on a request (you can have multiple heuristic detections on the same request). + */ + detectionIds: number[]; +} +interface IncomingRequestCfPropertiesBotManagement { + /** + * Results of Cloudflare's Bot Management analysis + */ + botManagement: IncomingRequestCfPropertiesBotManagementBase; + /** + * Duplicate of `botManagement.score`. + * + * @deprecated + */ + clientTrustScore: number; +} +interface IncomingRequestCfPropertiesBotManagementEnterprise extends IncomingRequestCfPropertiesBotManagement { + /** + * Results of Cloudflare's Bot Management analysis + */ + botManagement: IncomingRequestCfPropertiesBotManagementBase & { + /** + * A [JA3 Fingerprint](https://developers.cloudflare.com/bots/concepts/ja3-fingerprint/) to help profile specific SSL/TLS clients + * across different destination IPs, Ports, and X509 certificates. + */ + ja3Hash: string; + }; +} +interface IncomingRequestCfPropertiesCloudflareForSaaSEnterprise { + /** + * Custom metadata set per-host in [Cloudflare for SaaS](https://developers.cloudflare.com/cloudflare-for-platforms/cloudflare-for-saas/). + * + * This field is only present if you have Cloudflare for SaaS enabled on your account + * and you have followed the [required steps to enable it]((https://developers.cloudflare.com/cloudflare-for-platforms/cloudflare-for-saas/domain-support/custom-metadata/)). + */ + hostMetadata?: HostMetadata; +} +interface IncomingRequestCfPropertiesCloudflareAccessOrApiShield { + /** + * Information about the client certificate presented to Cloudflare. + * + * This is populated when the incoming request is served over TLS using + * either Cloudflare Access or API Shield (mTLS) + * and the presented SSL certificate has a valid + * [Certificate Serial Number](https://ldapwiki.com/wiki/Certificate%20Serial%20Number) + * (i.e., not `null` or `""`). + * + * Otherwise, a set of placeholder values are used. + * + * The property `certPresented` will be set to `"1"` when + * the object is populated (i.e. the above conditions were met). + */ + tlsClientAuth: IncomingRequestCfPropertiesTLSClientAuth | IncomingRequestCfPropertiesTLSClientAuthPlaceholder; +} +/** + * Metadata about the request's TLS handshake + */ +interface IncomingRequestCfPropertiesExportedAuthenticatorMetadata { + /** + * The client's [`HELLO` message](https://www.rfc-editor.org/rfc/rfc5246#section-7.4.1.2), encoded in hexadecimal + * + * @example "44372ba35fa1270921d318f34c12f155dc87b682cf36a790cfaa3ba8737a1b5d" + */ + clientHandshake: string; + /** + * The server's [`HELLO` message](https://www.rfc-editor.org/rfc/rfc5246#section-7.4.1.2), encoded in hexadecimal + * + * @example "44372ba35fa1270921d318f34c12f155dc87b682cf36a790cfaa3ba8737a1b5d" + */ + serverHandshake: string; + /** + * The client's [`FINISHED` message](https://www.rfc-editor.org/rfc/rfc5246#section-7.4.9), encoded in hexadecimal + * + * @example "084ee802fe1348f688220e2a6040a05b2199a761f33cf753abb1b006792d3f8b" + */ + clientFinished: string; + /** + * The server's [`FINISHED` message](https://www.rfc-editor.org/rfc/rfc5246#section-7.4.9), encoded in hexadecimal + * + * @example "084ee802fe1348f688220e2a6040a05b2199a761f33cf753abb1b006792d3f8b" + */ + serverFinished: string; +} +/** + * Geographic data about the request's origin. + */ +interface IncomingRequestCfPropertiesGeographicInformation { + /** + * The [ISO 3166-1 Alpha 2](https://www.iso.org/iso-3166-country-codes.html) country code the request originated from. + * + * If your worker is [configured to accept TOR connections](https://support.cloudflare.com/hc/en-us/articles/203306930-Understanding-Cloudflare-Tor-support-and-Onion-Routing), this may also be `"T1"`, indicating a request that originated over TOR. + * + * If Cloudflare is unable to determine where the request originated this property is omitted. + * + * The country code `"T1"` is used for requests originating on TOR. + * + * @example "GB" + */ + country?: Iso3166Alpha2Code | "T1"; + /** + * If present, this property indicates that the request originated in the EU + * + * @example "1" + */ + isEUCountry?: "1"; + /** + * A two-letter code indicating the continent the request originated from. + * + * @example "AN" + */ + continent?: ContinentCode; + /** + * The city the request originated from + * + * @example "Austin" + */ + city?: string; + /** + * Postal code of the incoming request + * + * @example "78701" + */ + postalCode?: string; + /** + * Latitude of the incoming request + * + * @example "30.27130" + */ + latitude?: string; + /** + * Longitude of the incoming request + * + * @example "-97.74260" + */ + longitude?: string; + /** + * Timezone of the incoming request + * + * @example "America/Chicago" + */ + timezone?: string; + /** + * If known, the ISO 3166-2 name for the first level region associated with + * the IP address of the incoming request + * + * @example "Texas" + */ + region?: string; + /** + * If known, the ISO 3166-2 code for the first-level region associated with + * the IP address of the incoming request + * + * @example "TX" + */ + regionCode?: string; + /** + * Metro code (DMA) of the incoming request + * + * @example "635" + */ + metroCode?: string; +} +/** Data about the incoming request's TLS certificate */ +interface IncomingRequestCfPropertiesTLSClientAuth { + /** Always `"1"`, indicating that the certificate was presented */ + certPresented: "1"; + /** + * Result of certificate verification. + * + * @example "FAILED:self signed certificate" + */ + certVerified: Exclude; + /** The presented certificate's revokation status. + * + * - A value of `"1"` indicates the certificate has been revoked + * - A value of `"0"` indicates the certificate has not been revoked + */ + certRevoked: "1" | "0"; + /** + * The certificate issuer's [distinguished name](https://knowledge.digicert.com/generalinformation/INFO1745.html) + * + * @example "CN=cloudflareaccess.com, C=US, ST=Texas, L=Austin, O=Cloudflare" + */ + certIssuerDN: string; + /** + * The certificate subject's [distinguished name](https://knowledge.digicert.com/generalinformation/INFO1745.html) + * + * @example "CN=*.cloudflareaccess.com, C=US, ST=Texas, L=Austin, O=Cloudflare" + */ + certSubjectDN: string; + /** + * The certificate issuer's [distinguished name](https://knowledge.digicert.com/generalinformation/INFO1745.html) ([RFC 2253](https://www.rfc-editor.org/rfc/rfc2253.html) formatted) + * + * @example "CN=cloudflareaccess.com, C=US, ST=Texas, L=Austin, O=Cloudflare" + */ + certIssuerDNRFC2253: string; + /** + * The certificate subject's [distinguished name](https://knowledge.digicert.com/generalinformation/INFO1745.html) ([RFC 2253](https://www.rfc-editor.org/rfc/rfc2253.html) formatted) + * + * @example "CN=*.cloudflareaccess.com, C=US, ST=Texas, L=Austin, O=Cloudflare" + */ + certSubjectDNRFC2253: string; + /** The certificate issuer's distinguished name (legacy policies) */ + certIssuerDNLegacy: string; + /** The certificate subject's distinguished name (legacy policies) */ + certSubjectDNLegacy: string; + /** + * The certificate's serial number + * + * @example "00936EACBE07F201DF" + */ + certSerial: string; + /** + * The certificate issuer's serial number + * + * @example "2489002934BDFEA34" + */ + certIssuerSerial: string; + /** + * The certificate's Subject Key Identifier + * + * @example "BB:AF:7E:02:3D:FA:A6:F1:3C:84:8E:AD:EE:38:98:EC:D9:32:32:D4" + */ + certSKI: string; + /** + * The certificate issuer's Subject Key Identifier + * + * @example "BB:AF:7E:02:3D:FA:A6:F1:3C:84:8E:AD:EE:38:98:EC:D9:32:32:D4" + */ + certIssuerSKI: string; + /** + * The certificate's SHA-1 fingerprint + * + * @example "6b9109f323999e52259cda7373ff0b4d26bd232e" + */ + certFingerprintSHA1: string; + /** + * The certificate's SHA-256 fingerprint + * + * @example "acf77cf37b4156a2708e34c4eb755f9b5dbbe5ebb55adfec8f11493438d19e6ad3f157f81fa3b98278453d5652b0c1fd1d71e5695ae4d709803a4d3f39de9dea" + */ + certFingerprintSHA256: string; + /** + * The effective starting date of the certificate + * + * @example "Dec 22 19:39:00 2018 GMT" + */ + certNotBefore: string; + /** + * The effective expiration date of the certificate + * + * @example "Dec 22 19:39:00 2018 GMT" + */ + certNotAfter: string; +} +/** Placeholder values for TLS Client Authorization */ +interface IncomingRequestCfPropertiesTLSClientAuthPlaceholder { + certPresented: "0"; + certVerified: "NONE"; + certRevoked: "0"; + certIssuerDN: ""; + certSubjectDN: ""; + certIssuerDNRFC2253: ""; + certSubjectDNRFC2253: ""; + certIssuerDNLegacy: ""; + certSubjectDNLegacy: ""; + certSerial: ""; + certIssuerSerial: ""; + certSKI: ""; + certIssuerSKI: ""; + certFingerprintSHA1: ""; + certFingerprintSHA256: ""; + certNotBefore: ""; + certNotAfter: ""; +} +/** Possible outcomes of TLS verification */ +declare type CertVerificationStatus = +/** Authentication succeeded */ +"SUCCESS" +/** No certificate was presented */ + | "NONE" +/** Failed because the certificate was self-signed */ + | "FAILED:self signed certificate" +/** Failed because the certificate failed a trust chain check */ + | "FAILED:unable to verify the first certificate" +/** Failed because the certificate not yet valid */ + | "FAILED:certificate is not yet valid" +/** Failed because the certificate is expired */ + | "FAILED:certificate has expired" +/** Failed for another unspecified reason */ + | "FAILED"; +/** + * An upstream endpoint's response to a TCP `keepalive` message from Cloudflare. + */ +declare type IncomingRequestCfPropertiesEdgeRequestKeepAliveStatus = 0 /** Unknown */ | 1 /** no keepalives (not found) */ | 2 /** no connection re-use, opening keepalive connection failed */ | 3 /** no connection re-use, keepalive accepted and saved */ | 4 /** connection re-use, refused by the origin server (`TCP FIN`) */ | 5; /** connection re-use, accepted by the origin server */ +/** ISO 3166-1 Alpha-2 codes */ +declare type Iso3166Alpha2Code = "AD" | "AE" | "AF" | "AG" | "AI" | "AL" | "AM" | "AO" | "AQ" | "AR" | "AS" | "AT" | "AU" | "AW" | "AX" | "AZ" | "BA" | "BB" | "BD" | "BE" | "BF" | "BG" | "BH" | "BI" | "BJ" | "BL" | "BM" | "BN" | "BO" | "BQ" | "BR" | "BS" | "BT" | "BV" | "BW" | "BY" | "BZ" | "CA" | "CC" | "CD" | "CF" | "CG" | "CH" | "CI" | "CK" | "CL" | "CM" | "CN" | "CO" | "CR" | "CU" | "CV" | "CW" | "CX" | "CY" | "CZ" | "DE" | "DJ" | "DK" | "DM" | "DO" | "DZ" | "EC" | "EE" | "EG" | "EH" | "ER" | "ES" | "ET" | "FI" | "FJ" | "FK" | "FM" | "FO" | "FR" | "GA" | "GB" | "GD" | "GE" | "GF" | "GG" | "GH" | "GI" | "GL" | "GM" | "GN" | "GP" | "GQ" | "GR" | "GS" | "GT" | "GU" | "GW" | "GY" | "HK" | "HM" | "HN" | "HR" | "HT" | "HU" | "ID" | "IE" | "IL" | "IM" | "IN" | "IO" | "IQ" | "IR" | "IS" | "IT" | "JE" | "JM" | "JO" | "JP" | "KE" | "KG" | "KH" | "KI" | "KM" | "KN" | "KP" | "KR" | "KW" | "KY" | "KZ" | "LA" | "LB" | "LC" | "LI" | "LK" | "LR" | "LS" | "LT" | "LU" | "LV" | "LY" | "MA" | "MC" | "MD" | "ME" | "MF" | "MG" | "MH" | "MK" | "ML" | "MM" | "MN" | "MO" | "MP" | "MQ" | "MR" | "MS" | "MT" | "MU" | "MV" | "MW" | "MX" | "MY" | "MZ" | "NA" | "NC" | "NE" | "NF" | "NG" | "NI" | "NL" | "NO" | "NP" | "NR" | "NU" | "NZ" | "OM" | "PA" | "PE" | "PF" | "PG" | "PH" | "PK" | "PL" | "PM" | "PN" | "PR" | "PS" | "PT" | "PW" | "PY" | "QA" | "RE" | "RO" | "RS" | "RU" | "RW" | "SA" | "SB" | "SC" | "SD" | "SE" | "SG" | "SH" | "SI" | "SJ" | "SK" | "SL" | "SM" | "SN" | "SO" | "SR" | "SS" | "ST" | "SV" | "SX" | "SY" | "SZ" | "TC" | "TD" | "TF" | "TG" | "TH" | "TJ" | "TK" | "TL" | "TM" | "TN" | "TO" | "TR" | "TT" | "TV" | "TW" | "TZ" | "UA" | "UG" | "UM" | "US" | "UY" | "UZ" | "VA" | "VC" | "VE" | "VG" | "VI" | "VN" | "VU" | "WF" | "WS" | "YE" | "YT" | "ZA" | "ZM" | "ZW"; +/** The 2-letter continent codes Cloudflare uses */ +declare type ContinentCode = "AF" | "AN" | "AS" | "EU" | "NA" | "OC" | "SA"; +type CfProperties = IncomingRequestCfProperties | RequestInitCfProperties; +interface D1Meta { + duration: number; + size_after: number; + rows_read: number; + rows_written: number; + last_row_id: number; + changed_db: boolean; + changes: number; + /** + * The region of the database instance that executed the query. + */ + served_by_region?: string; + /** + * The three letters airport code of the colo that executed the query. + */ + served_by_colo?: string; + /** + * True if-and-only-if the database instance that executed the query was the primary. + */ + served_by_primary?: boolean; + timings?: { + /** + * The duration of the SQL query execution by the database instance. It doesn't include any network time. + */ + sql_duration_ms: number; + }; + /** + * Number of total attempts to execute the query, due to automatic retries. + * Note: All other fields in the response like `timings` only apply to the last attempt. + */ + total_attempts?: number; +} +interface D1Response { + success: true; + meta: D1Meta & Record; + error?: never; +} +type D1Result = D1Response & { + results: T[]; +}; +interface D1ExecResult { + count: number; + duration: number; +} +type D1SessionConstraint = +// Indicates that the first query should go to the primary, and the rest queries +// using the same D1DatabaseSession will go to any replica that is consistent with +// the bookmark maintained by the session (returned by the first query). +'first-primary' +// Indicates that the first query can go anywhere (primary or replica), and the rest queries +// using the same D1DatabaseSession will go to any replica that is consistent with +// the bookmark maintained by the session (returned by the first query). + | 'first-unconstrained'; +type D1SessionBookmark = string; +declare abstract class D1Database { + prepare(query: string): D1PreparedStatement; + batch(statements: D1PreparedStatement[]): Promise[]>; + exec(query: string): Promise; + /** + * Creates a new D1 Session anchored at the given constraint or the bookmark. + * All queries executed using the created session will have sequential consistency, + * meaning that all writes done through the session will be visible in subsequent reads. + * + * @param constraintOrBookmark Either the session constraint or the explicit bookmark to anchor the created session. + */ + withSession(constraintOrBookmark?: D1SessionBookmark | D1SessionConstraint): D1DatabaseSession; + /** + * @deprecated dump() will be removed soon, only applies to deprecated alpha v1 databases. + */ + dump(): Promise; +} +declare abstract class D1DatabaseSession { + prepare(query: string): D1PreparedStatement; + batch(statements: D1PreparedStatement[]): Promise[]>; + /** + * @returns The latest session bookmark across all executed queries on the session. + * If no query has been executed yet, `null` is returned. + */ + getBookmark(): D1SessionBookmark | null; +} +declare abstract class D1PreparedStatement { + bind(...values: unknown[]): D1PreparedStatement; + first(colName: string): Promise; + first>(): Promise; + run>(): Promise>; + all>(): Promise>; + raw(options: { + columnNames: true; + }): Promise<[ + string[], + ...T[] + ]>; + raw(options?: { + columnNames?: false; + }): Promise; +} +// `Disposable` was added to TypeScript's standard lib types in version 5.2. +// To support older TypeScript versions, define an empty `Disposable` interface. +// Users won't be able to use `using`/`Symbol.dispose` without upgrading to 5.2, +// but this will ensure type checking on older versions still passes. +// TypeScript's interface merging will ensure our empty interface is effectively +// ignored when `Disposable` is included in the standard lib. +interface Disposable { +} +/** + * The returned data after sending an email + */ +interface EmailSendResult { + /** + * The Email Message ID + */ + messageId: string; +} +/** + * An email message that can be sent from a Worker. + */ +interface EmailMessage { + /** + * Envelope From attribute of the email message. + */ + readonly from: string; + /** + * Envelope To attribute of the email message. + */ + readonly to: string; +} +/** + * An email message that is sent to a consumer Worker and can be rejected/forwarded. + */ +interface ForwardableEmailMessage extends EmailMessage { + /** + * Stream of the email message content. + */ + readonly raw: ReadableStream; + /** + * An [Headers object](https://developer.mozilla.org/en-US/docs/Web/API/Headers). + */ + readonly headers: Headers; + /** + * Size of the email message content. + */ + readonly rawSize: number; + /** + * Reject this email message by returning a permanent SMTP error back to the connecting client including the given reason. + * @param reason The reject reason. + * @returns void + */ + setReject(reason: string): void; + /** + * Forward this email message to a verified destination address of the account. + * @param rcptTo Verified destination address. + * @param headers A [Headers object](https://developer.mozilla.org/en-US/docs/Web/API/Headers). + * @returns A promise that resolves when the email message is forwarded. + */ + forward(rcptTo: string, headers?: Headers): Promise; + /** + * Reply to the sender of this email message with a new EmailMessage object. + * @param message The reply message. + * @returns A promise that resolves when the email message is replied. + */ + reply(message: EmailMessage): Promise; +} +/** A file attachment for an email message */ +type EmailAttachment = { + disposition: 'inline'; + contentId: string; + filename: string; + type: string; + content: string | ArrayBuffer | ArrayBufferView; +} | { + disposition: 'attachment'; + contentId?: undefined; + filename: string; + type: string; + content: string | ArrayBuffer | ArrayBufferView; +}; +/** An Email Address */ +interface EmailAddress { + name: string; + email: string; +} +/** + * A binding that allows a Worker to send email messages. + */ +interface SendEmail { + send(message: EmailMessage): Promise; + send(builder: { + from: string | EmailAddress; + to: string | string[]; + subject: string; + replyTo?: string | EmailAddress; + cc?: string | string[]; + bcc?: string | string[]; + headers?: Record; + text?: string; + html?: string; + attachments?: EmailAttachment[]; + }): Promise; +} +declare abstract class EmailEvent extends ExtendableEvent { + readonly message: ForwardableEmailMessage; +} +declare type EmailExportedHandler = (message: ForwardableEmailMessage, env: Env, ctx: ExecutionContext) => void | Promise; +declare module "cloudflare:email" { + let _EmailMessage: { + prototype: EmailMessage; + new (from: string, to: string, raw: ReadableStream | string): EmailMessage; + }; + export { _EmailMessage as EmailMessage }; +} +/** + * Hello World binding to serve as an explanatory example. DO NOT USE + */ +interface HelloWorldBinding { + /** + * Retrieve the current stored value + */ + get(): Promise<{ + value: string; + ms?: number; + }>; + /** + * Set a new stored value + */ + set(value: string): Promise; +} +interface Hyperdrive { + /** + * Connect directly to Hyperdrive as if it's your database, returning a TCP socket. + * + * Calling this method returns an identical socket to if you call + * `connect("host:port")` using the `host` and `port` fields from this object. + * Pick whichever approach works better with your preferred DB client library. + * + * Note that this socket is not yet authenticated -- it's expected that your + * code (or preferably, the client library of your choice) will authenticate + * using the information in this class's readonly fields. + */ + connect(): Socket; + /** + * A valid DB connection string that can be passed straight into the typical + * client library/driver/ORM. This will typically be the easiest way to use + * Hyperdrive. + */ + readonly connectionString: string; + /* + * A randomly generated hostname that is only valid within the context of the + * currently running Worker which, when passed into `connect()` function from + * the "cloudflare:sockets" module, will connect to the Hyperdrive instance + * for your database. + */ + readonly host: string; + /* + * The port that must be paired the the host field when connecting. + */ + readonly port: number; + /* + * The username to use when authenticating to your database via Hyperdrive. + * Unlike the host and password, this will be the same every time + */ + readonly user: string; + /* + * The randomly generated password to use when authenticating to your + * database via Hyperdrive. Like the host field, this password is only valid + * within the context of the currently running Worker instance from which + * it's read. + */ + readonly password: string; + /* + * The name of the database to connect to. + */ + readonly database: string; +} +// Copyright (c) 2024 Cloudflare, Inc. +// Licensed under the Apache 2.0 license found in the LICENSE file or at: +// https://opensource.org/licenses/Apache-2.0 +type ImageInfoResponse = { + format: 'image/svg+xml'; +} | { + format: string; + fileSize: number; + width: number; + height: number; +}; +type ImageTransform = { + width?: number; + height?: number; + background?: string; + blur?: number; + border?: { + color?: string; + width?: number; + } | { + top?: number; + bottom?: number; + left?: number; + right?: number; + }; + brightness?: number; + contrast?: number; + fit?: 'scale-down' | 'contain' | 'pad' | 'squeeze' | 'cover' | 'crop'; + flip?: 'h' | 'v' | 'hv'; + gamma?: number; + segment?: 'foreground'; + gravity?: 'face' | 'left' | 'right' | 'top' | 'bottom' | 'center' | 'auto' | 'entropy' | { + x?: number; + y?: number; + mode: 'remainder' | 'box-center'; + }; + rotate?: 0 | 90 | 180 | 270; + saturation?: number; + sharpen?: number; + trim?: 'border' | { + top?: number; + bottom?: number; + left?: number; + right?: number; + width?: number; + height?: number; + border?: boolean | { + color?: string; + tolerance?: number; + keep?: number; + }; + }; +}; +type ImageDrawOptions = { + opacity?: number; + repeat?: boolean | string; + top?: number; + left?: number; + bottom?: number; + right?: number; +}; +type ImageInputOptions = { + encoding?: 'base64'; +}; +type ImageOutputOptions = { + format: 'image/jpeg' | 'image/png' | 'image/gif' | 'image/webp' | 'image/avif' | 'rgb' | 'rgba'; + quality?: number; + background?: string; + anim?: boolean; +}; +interface ImagesBinding { + /** + * Get image metadata (type, width and height) + * @throws {@link ImagesError} with code 9412 if input is not an image + * @param stream The image bytes + */ + info(stream: ReadableStream, options?: ImageInputOptions): Promise; + /** + * Begin applying a series of transformations to an image + * @param stream The image bytes + * @returns A transform handle + */ + input(stream: ReadableStream, options?: ImageInputOptions): ImageTransformer; +} +interface ImageTransformer { + /** + * Apply transform next, returning a transform handle. + * You can then apply more transformations, draw, or retrieve the output. + * @param transform + */ + transform(transform: ImageTransform): ImageTransformer; + /** + * Draw an image on this transformer, returning a transform handle. + * You can then apply more transformations, draw, or retrieve the output. + * @param image The image (or transformer that will give the image) to draw + * @param options The options configuring how to draw the image + */ + draw(image: ReadableStream | ImageTransformer, options?: ImageDrawOptions): ImageTransformer; + /** + * Retrieve the image that results from applying the transforms to the + * provided input + * @param options Options that apply to the output e.g. output format + */ + output(options: ImageOutputOptions): Promise; +} +type ImageTransformationOutputOptions = { + encoding?: 'base64'; +}; +interface ImageTransformationResult { + /** + * The image as a response, ready to store in cache or return to users + */ + response(): Response; + /** + * The content type of the returned image + */ + contentType(): string; + /** + * The bytes of the response + */ + image(options?: ImageTransformationOutputOptions): ReadableStream; +} +interface ImagesError extends Error { + readonly code: number; + readonly message: string; + readonly stack?: string; +} +/** + * Media binding for transforming media streams. + * Provides the entry point for media transformation operations. + */ +interface MediaBinding { + /** + * Creates a media transformer from an input stream. + * @param media - The input media bytes + * @returns A MediaTransformer instance for applying transformations + */ + input(media: ReadableStream): MediaTransformer; +} +/** + * Media transformer for applying transformation operations to media content. + * Handles sizing, fitting, and other input transformation parameters. + */ +interface MediaTransformer { + /** + * Applies transformation options to the media content. + * @param transform - Configuration for how the media should be transformed + * @returns A generator for producing the transformed media output + */ + transform(transform: MediaTransformationInputOptions): MediaTransformationGenerator; +} +/** + * Generator for producing media transformation results. + * Configures the output format and parameters for the transformed media. + */ +interface MediaTransformationGenerator { + /** + * Generates the final media output with specified options. + * @param output - Configuration for the output format and parameters + * @returns The final transformation result containing the transformed media + */ + output(output: MediaTransformationOutputOptions): MediaTransformationResult; +} +/** + * Result of a media transformation operation. + * Provides multiple ways to access the transformed media content. + */ +interface MediaTransformationResult { + /** + * Returns the transformed media as a readable stream of bytes. + * @returns A stream containing the transformed media data + */ + media(): ReadableStream; + /** + * Returns the transformed media as an HTTP response object. + * @returns The transformed media as a Response, ready to store in cache or return to users + */ + response(): Response; + /** + * Returns the MIME type of the transformed media. + * @returns The content type string (e.g., 'image/jpeg', 'video/mp4') + */ + contentType(): string; +} +/** + * Configuration options for transforming media input. + * Controls how the media should be resized and fitted. + */ +type MediaTransformationInputOptions = { + /** How the media should be resized to fit the specified dimensions */ + fit?: 'contain' | 'cover' | 'scale-down'; + /** Target width in pixels */ + width?: number; + /** Target height in pixels */ + height?: number; +}; +/** + * Configuration options for Media Transformations output. + * Controls the format, timing, and type of the generated output. + */ +type MediaTransformationOutputOptions = { + /** + * Output mode determining the type of media to generate + */ + mode?: 'video' | 'spritesheet' | 'frame' | 'audio'; + /** Whether to include audio in the output */ + audio?: boolean; + /** + * Starting timestamp for frame extraction or start time for clips. (e.g. '2s'). + */ + time?: string; + /** + * Duration for video clips, audio extraction, and spritesheet generation (e.g. '5s'). + */ + duration?: string; + /** + * Number of frames in the spritesheet. + */ + imageCount?: number; + /** + * Output format for the generated media. + */ + format?: 'jpg' | 'png' | 'm4a'; +}; +/** + * Error object for media transformation operations. + * Extends the standard Error interface with additional media-specific information. + */ +interface MediaError extends Error { + readonly code: number; + readonly message: string; + readonly stack?: string; +} +declare module 'cloudflare:node' { + interface NodeStyleServer { + listen(...args: unknown[]): this; + address(): { + port?: number | null | undefined; + }; + } + export function httpServerHandler(port: number): ExportedHandler; + export function httpServerHandler(options: { + port: number; + }): ExportedHandler; + export function httpServerHandler(server: NodeStyleServer): ExportedHandler; +} +type Params

= Record; +type EventContext = { + request: Request>; + functionPath: string; + waitUntil: (promise: Promise) => void; + passThroughOnException: () => void; + next: (input?: Request | string, init?: RequestInit) => Promise; + env: Env & { + ASSETS: { + fetch: typeof fetch; + }; + }; + params: Params

; + data: Data; +}; +type PagesFunction = Record> = (context: EventContext) => Response | Promise; +type EventPluginContext = { + request: Request>; + functionPath: string; + waitUntil: (promise: Promise) => void; + passThroughOnException: () => void; + next: (input?: Request | string, init?: RequestInit) => Promise; + env: Env & { + ASSETS: { + fetch: typeof fetch; + }; + }; + params: Params

; + data: Data; + pluginArgs: PluginArgs; +}; +type PagesPluginFunction = Record, PluginArgs = unknown> = (context: EventPluginContext) => Response | Promise; +declare module "assets:*" { + export const onRequest: PagesFunction; +} +// Copyright (c) 2022-2023 Cloudflare, Inc. +// Licensed under the Apache 2.0 license found in the LICENSE file or at: +// https://opensource.org/licenses/Apache-2.0 +declare module "cloudflare:pipelines" { + export abstract class PipelineTransformationEntrypoint { + protected env: Env; + protected ctx: ExecutionContext; + constructor(ctx: ExecutionContext, env: Env); + /** + * run receives an array of PipelineRecord which can be + * transformed and returned to the pipeline + * @param records Incoming records from the pipeline to be transformed + * @param metadata Information about the specific pipeline calling the transformation entrypoint + * @returns A promise containing the transformed PipelineRecord array + */ + public run(records: I[], metadata: PipelineBatchMetadata): Promise; + } + export type PipelineRecord = Record; + export type PipelineBatchMetadata = { + pipelineId: string; + pipelineName: string; + }; + export interface Pipeline { + /** + * The Pipeline interface represents the type of a binding to a Pipeline + * + * @param records The records to send to the pipeline + */ + send(records: T[]): Promise; + } +} +// PubSubMessage represents an incoming PubSub message. +// The message includes metadata about the broker, the client, and the payload +// itself. +// https://developers.cloudflare.com/pub-sub/ +interface PubSubMessage { + // Message ID + readonly mid: number; + // MQTT broker FQDN in the form mqtts://BROKER.NAMESPACE.cloudflarepubsub.com:PORT + readonly broker: string; + // The MQTT topic the message was sent on. + readonly topic: string; + // The client ID of the client that published this message. + readonly clientId: string; + // The unique identifier (JWT ID) used by the client to authenticate, if token + // auth was used. + readonly jti?: string; + // A Unix timestamp (seconds from Jan 1, 1970), set when the Pub/Sub Broker + // received the message from the client. + readonly receivedAt: number; + // An (optional) string with the MIME type of the payload, if set by the + // client. + readonly contentType: string; + // Set to 1 when the payload is a UTF-8 string + // https://docs.oasis-open.org/mqtt/mqtt/v5.0/os/mqtt-v5.0-os.html#_Toc3901063 + readonly payloadFormatIndicator: number; + // Pub/Sub (MQTT) payloads can be UTF-8 strings, or byte arrays. + // You can use payloadFormatIndicator to inspect this before decoding. + payload: string | Uint8Array; +} +// JsonWebKey extended by kid parameter +interface JsonWebKeyWithKid extends JsonWebKey { + // Key Identifier of the JWK + readonly kid: string; +} +interface RateLimitOptions { + key: string; +} +interface RateLimitOutcome { + success: boolean; +} +interface RateLimit { + /** + * Rate limit a request based on the provided options. + * @see https://developers.cloudflare.com/workers/runtime-apis/bindings/rate-limit/ + * @returns A promise that resolves with the outcome of the rate limit. + */ + limit(options: RateLimitOptions): Promise; +} +// Namespace for RPC utility types. Unfortunately, we can't use a `module` here as these types need +// to referenced by `Fetcher`. This is included in the "importable" version of the types which +// strips all `module` blocks. +declare namespace Rpc { + // Branded types for identifying `WorkerEntrypoint`/`DurableObject`/`Target`s. + // TypeScript uses *structural* typing meaning anything with the same shape as type `T` is a `T`. + // For the classes exported by `cloudflare:workers` we want *nominal* typing (i.e. we only want to + // accept `WorkerEntrypoint` from `cloudflare:workers`, not any other class with the same shape) + export const __RPC_STUB_BRAND: '__RPC_STUB_BRAND'; + export const __RPC_TARGET_BRAND: '__RPC_TARGET_BRAND'; + export const __WORKER_ENTRYPOINT_BRAND: '__WORKER_ENTRYPOINT_BRAND'; + export const __DURABLE_OBJECT_BRAND: '__DURABLE_OBJECT_BRAND'; + export const __WORKFLOW_ENTRYPOINT_BRAND: '__WORKFLOW_ENTRYPOINT_BRAND'; + export interface RpcTargetBranded { + [__RPC_TARGET_BRAND]: never; + } + export interface WorkerEntrypointBranded { + [__WORKER_ENTRYPOINT_BRAND]: never; + } + export interface DurableObjectBranded { + [__DURABLE_OBJECT_BRAND]: never; + } + export interface WorkflowEntrypointBranded { + [__WORKFLOW_ENTRYPOINT_BRAND]: never; + } + export type EntrypointBranded = WorkerEntrypointBranded | DurableObjectBranded | WorkflowEntrypointBranded; + // Types that can be used through `Stub`s + export type Stubable = RpcTargetBranded | ((...args: any[]) => any); + // Types that can be passed over RPC + // The reason for using a generic type here is to build a serializable subset of structured + // cloneable composite types. This allows types defined with the "interface" keyword to pass the + // serializable check as well. Otherwise, only types defined with the "type" keyword would pass. + type Serializable = + // Structured cloneables + BaseType + // Structured cloneable composites + | Map ? Serializable : never, T extends Map ? Serializable : never> | Set ? Serializable : never> | ReadonlyArray ? Serializable : never> | { + [K in keyof T]: K extends number | string ? Serializable : never; + } + // Special types + | Stub + // Serialized as stubs, see `Stubify` + | Stubable; + // Base type for all RPC stubs, including common memory management methods. + // `T` is used as a marker type for unwrapping `Stub`s later. + interface StubBase extends Disposable { + [__RPC_STUB_BRAND]: T; + dup(): this; + } + export type Stub = Provider & StubBase; + // This represents all the types that can be sent as-is over an RPC boundary + type BaseType = void | undefined | null | boolean | number | bigint | string | TypedArray | ArrayBuffer | DataView | Date | Error | RegExp | ReadableStream | WritableStream | Request | Response | Headers; + // Recursively rewrite all `Stubable` types with `Stub`s + // prettier-ignore + type Stubify = T extends Stubable ? Stub : T extends Map ? Map, Stubify> : T extends Set ? Set> : T extends Array ? Array> : T extends ReadonlyArray ? ReadonlyArray> : T extends BaseType ? T : T extends { + [key: string | number]: any; + } ? { + [K in keyof T]: Stubify; + } : T; + // Recursively rewrite all `Stub`s with the corresponding `T`s. + // Note we use `StubBase` instead of `Stub` here to avoid circular dependencies: + // `Stub` depends on `Provider`, which depends on `Unstubify`, which would depend on `Stub`. + // prettier-ignore + type Unstubify = T extends StubBase ? V : T extends Map ? Map, Unstubify> : T extends Set ? Set> : T extends Array ? Array> : T extends ReadonlyArray ? ReadonlyArray> : T extends BaseType ? T : T extends { + [key: string | number]: unknown; + } ? { + [K in keyof T]: Unstubify; + } : T; + type UnstubifyAll = { + [I in keyof A]: Unstubify; + }; + // Utility type for adding `Provider`/`Disposable`s to `object` types only. + // Note `unknown & T` is equivalent to `T`. + type MaybeProvider = T extends object ? Provider : unknown; + type MaybeDisposable = T extends object ? Disposable : unknown; + // Type for method return or property on an RPC interface. + // - Stubable types are replaced by stubs. + // - Serializable types are passed by value, with stubable types replaced by stubs + // and a top-level `Disposer`. + // Everything else can't be passed over PRC. + // Technically, we use custom thenables here, but they quack like `Promise`s. + // Intersecting with `(Maybe)Provider` allows pipelining. + // prettier-ignore + type Result = R extends Stubable ? Promise> & Provider : R extends Serializable ? Promise & MaybeDisposable> & MaybeProvider : never; + // Type for method or property on an RPC interface. + // For methods, unwrap `Stub`s in parameters, and rewrite returns to be `Result`s. + // Unwrapping `Stub`s allows calling with `Stubable` arguments. + // For properties, rewrite types to be `Result`s. + // In each case, unwrap `Promise`s. + type MethodOrProperty = V extends (...args: infer P) => infer R ? (...args: UnstubifyAll

) => Result> : Result>; + // Type for the callable part of an `Provider` if `T` is callable. + // This is intersected with methods/properties. + type MaybeCallableProvider = T extends (...args: any[]) => any ? MethodOrProperty : unknown; + // Base type for all other types providing RPC-like interfaces. + // Rewrites all methods/properties to be `MethodOrProperty`s, while preserving callable types. + // `Reserved` names (e.g. stub method names like `dup()`) and symbols can't be accessed over RPC. + export type Provider = MaybeCallableProvider & Pick<{ + [K in keyof T]: MethodOrProperty; + }, Exclude>>; +} +declare namespace Cloudflare { + // Type of `env`. + // + // The specific project can extend `Env` by redeclaring it in project-specific files. Typescript + // will merge all declarations. + // + // You can use `wrangler types` to generate the `Env` type automatically. + interface Env { + } + // Project-specific parameters used to inform types. + // + // This interface is, again, intended to be declared in project-specific files, and then that + // declaration will be merged with this one. + // + // A project should have a declaration like this: + // + // interface GlobalProps { + // // Declares the main module's exports. Used to populate Cloudflare.Exports aka the type + // // of `ctx.exports`. + // mainModule: typeof import("my-main-module"); + // + // // Declares which of the main module's exports are configured with durable storage, and + // // thus should behave as Durable Object namsepace bindings. + // durableNamespaces: "MyDurableObject" | "AnotherDurableObject"; + // } + // + // You can use `wrangler types` to generate `GlobalProps` automatically. + interface GlobalProps { + } + // Evaluates to the type of a property in GlobalProps, defaulting to `Default` if it is not + // present. + type GlobalProp = K extends keyof GlobalProps ? GlobalProps[K] : Default; + // The type of the program's main module exports, if known. Requires `GlobalProps` to declare the + // `mainModule` property. + type MainModule = GlobalProp<"mainModule", {}>; + // The type of ctx.exports, which contains loopback bindings for all top-level exports. + type Exports = { + [K in keyof MainModule]: LoopbackForExport + // If the export is listed in `durableNamespaces`, then it is also a + // DurableObjectNamespace. + & (K extends GlobalProp<"durableNamespaces", never> ? MainModule[K] extends new (...args: any[]) => infer DoInstance ? DoInstance extends Rpc.DurableObjectBranded ? DurableObjectNamespace : DurableObjectNamespace : DurableObjectNamespace : {}); + }; +} +declare namespace CloudflareWorkersModule { + export type RpcStub = Rpc.Stub; + export const RpcStub: { + new (value: T): Rpc.Stub; + }; + export abstract class RpcTarget implements Rpc.RpcTargetBranded { + [Rpc.__RPC_TARGET_BRAND]: never; + } + // `protected` fields don't appear in `keyof`s, so can't be accessed over RPC + export abstract class WorkerEntrypoint implements Rpc.WorkerEntrypointBranded { + [Rpc.__WORKER_ENTRYPOINT_BRAND]: never; + protected ctx: ExecutionContext; + protected env: Env; + constructor(ctx: ExecutionContext, env: Env); + email?(message: ForwardableEmailMessage): void | Promise; + fetch?(request: Request): Response | Promise; + queue?(batch: MessageBatch): void | Promise; + scheduled?(controller: ScheduledController): void | Promise; + tail?(events: TraceItem[]): void | Promise; + tailStream?(event: TailStream.TailEvent): TailStream.TailEventHandlerType | Promise; + test?(controller: TestController): void | Promise; + trace?(traces: TraceItem[]): void | Promise; + } + export abstract class DurableObject implements Rpc.DurableObjectBranded { + [Rpc.__DURABLE_OBJECT_BRAND]: never; + protected ctx: DurableObjectState; + protected env: Env; + constructor(ctx: DurableObjectState, env: Env); + alarm?(alarmInfo?: AlarmInvocationInfo): void | Promise; + fetch?(request: Request): Response | Promise; + webSocketMessage?(ws: WebSocket, message: string | ArrayBuffer): void | Promise; + webSocketClose?(ws: WebSocket, code: number, reason: string, wasClean: boolean): void | Promise; + webSocketError?(ws: WebSocket, error: unknown): void | Promise; + } + export type WorkflowDurationLabel = 'second' | 'minute' | 'hour' | 'day' | 'week' | 'month' | 'year'; + export type WorkflowSleepDuration = `${number} ${WorkflowDurationLabel}${'s' | ''}` | number; + export type WorkflowDelayDuration = WorkflowSleepDuration; + export type WorkflowTimeoutDuration = WorkflowSleepDuration; + export type WorkflowRetentionDuration = WorkflowSleepDuration; + export type WorkflowBackoff = 'constant' | 'linear' | 'exponential'; + export type WorkflowStepConfig = { + retries?: { + limit: number; + delay: WorkflowDelayDuration | number; + backoff?: WorkflowBackoff; + }; + timeout?: WorkflowTimeoutDuration | number; + }; + export type WorkflowEvent = { + payload: Readonly; + timestamp: Date; + instanceId: string; + }; + export type WorkflowStepEvent = { + payload: Readonly; + timestamp: Date; + type: string; + }; + export abstract class WorkflowStep { + do>(name: string, callback: () => Promise): Promise; + do>(name: string, config: WorkflowStepConfig, callback: () => Promise): Promise; + sleep: (name: string, duration: WorkflowSleepDuration) => Promise; + sleepUntil: (name: string, timestamp: Date | number) => Promise; + waitForEvent>(name: string, options: { + type: string; + timeout?: WorkflowTimeoutDuration | number; + }): Promise>; + } + export abstract class WorkflowEntrypoint | unknown = unknown> implements Rpc.WorkflowEntrypointBranded { + [Rpc.__WORKFLOW_ENTRYPOINT_BRAND]: never; + protected ctx: ExecutionContext; + protected env: Env; + constructor(ctx: ExecutionContext, env: Env); + run(event: Readonly>, step: WorkflowStep): Promise; + } + export function waitUntil(promise: Promise): void; + export function withEnv(newEnv: unknown, fn: () => unknown): unknown; + export function withExports(newExports: unknown, fn: () => unknown): unknown; + export function withEnvAndExports(newEnv: unknown, newExports: unknown, fn: () => unknown): unknown; + export const env: Cloudflare.Env; + export const exports: Cloudflare.Exports; +} +declare module 'cloudflare:workers' { + export = CloudflareWorkersModule; +} +interface SecretsStoreSecret { + /** + * Get a secret from the Secrets Store, returning a string of the secret value + * if it exists, or throws an error if it does not exist + */ + get(): Promise; +} +declare module "cloudflare:sockets" { + function _connect(address: string | SocketAddress, options?: SocketOptions): Socket; + export { _connect as connect }; +} +type MarkdownDocument = { + name: string; + blob: Blob; +}; +type ConversionResponse = { + name: string; + mimeType: string; + format: 'markdown'; + tokens: number; + data: string; +} | { + name: string; + mimeType: string; + format: 'error'; + error: string; +}; +type ImageConversionOptions = { + descriptionLanguage?: 'en' | 'es' | 'fr' | 'it' | 'pt' | 'de'; +}; +type EmbeddedImageConversionOptions = ImageConversionOptions & { + convert?: boolean; + maxConvertedImages?: number; +}; +type ConversionOptions = { + html?: { + images?: EmbeddedImageConversionOptions & { + convertOGImage?: boolean; + }; + }; + docx?: { + images?: EmbeddedImageConversionOptions; + }; + image?: ImageConversionOptions; + pdf?: { + images?: EmbeddedImageConversionOptions; + metadata?: boolean; + }; +}; +type ConversionRequestOptions = { + gateway?: GatewayOptions; + extraHeaders?: object; + conversionOptions?: ConversionOptions; +}; +type SupportedFileFormat = { + mimeType: string; + extension: string; +}; +declare abstract class ToMarkdownService { + transform(files: MarkdownDocument[], options?: ConversionRequestOptions): Promise; + transform(files: MarkdownDocument, options?: ConversionRequestOptions): Promise; + supported(): Promise; +} +declare namespace TailStream { + interface Header { + readonly name: string; + readonly value: string; + } + interface FetchEventInfo { + readonly type: "fetch"; + readonly method: string; + readonly url: string; + readonly cfJson?: object; + readonly headers: Header[]; + } + interface JsRpcEventInfo { + readonly type: "jsrpc"; + } + interface ScheduledEventInfo { + readonly type: "scheduled"; + readonly scheduledTime: Date; + readonly cron: string; + } + interface AlarmEventInfo { + readonly type: "alarm"; + readonly scheduledTime: Date; + } + interface QueueEventInfo { + readonly type: "queue"; + readonly queueName: string; + readonly batchSize: number; + } + interface EmailEventInfo { + readonly type: "email"; + readonly mailFrom: string; + readonly rcptTo: string; + readonly rawSize: number; + } + interface TraceEventInfo { + readonly type: "trace"; + readonly traces: (string | null)[]; + } + interface HibernatableWebSocketEventInfoMessage { + readonly type: "message"; + } + interface HibernatableWebSocketEventInfoError { + readonly type: "error"; + } + interface HibernatableWebSocketEventInfoClose { + readonly type: "close"; + readonly code: number; + readonly wasClean: boolean; + } + interface HibernatableWebSocketEventInfo { + readonly type: "hibernatableWebSocket"; + readonly info: HibernatableWebSocketEventInfoClose | HibernatableWebSocketEventInfoError | HibernatableWebSocketEventInfoMessage; + } + interface CustomEventInfo { + readonly type: "custom"; + } + interface FetchResponseInfo { + readonly type: "fetch"; + readonly statusCode: number; + } + type EventOutcome = "ok" | "canceled" | "exception" | "unknown" | "killSwitch" | "daemonDown" | "exceededCpu" | "exceededMemory" | "loadShed" | "responseStreamDisconnected" | "scriptNotFound"; + interface ScriptVersion { + readonly id: string; + readonly tag?: string; + readonly message?: string; + } + interface Onset { + readonly type: "onset"; + readonly attributes: Attribute[]; + // id for the span being opened by this Onset event. + readonly spanId: string; + readonly dispatchNamespace?: string; + readonly entrypoint?: string; + readonly executionModel: string; + readonly scriptName?: string; + readonly scriptTags?: string[]; + readonly scriptVersion?: ScriptVersion; + readonly info: FetchEventInfo | JsRpcEventInfo | ScheduledEventInfo | AlarmEventInfo | QueueEventInfo | EmailEventInfo | TraceEventInfo | HibernatableWebSocketEventInfo | CustomEventInfo; + } + interface Outcome { + readonly type: "outcome"; + readonly outcome: EventOutcome; + readonly cpuTime: number; + readonly wallTime: number; + } + interface SpanOpen { + readonly type: "spanOpen"; + readonly name: string; + // id for the span being opened by this SpanOpen event. + readonly spanId: string; + readonly info?: FetchEventInfo | JsRpcEventInfo | Attributes; + } + interface SpanClose { + readonly type: "spanClose"; + readonly outcome: EventOutcome; + } + interface DiagnosticChannelEvent { + readonly type: "diagnosticChannel"; + readonly channel: string; + readonly message: any; + } + interface Exception { + readonly type: "exception"; + readonly name: string; + readonly message: string; + readonly stack?: string; + } + interface Log { + readonly type: "log"; + readonly level: "debug" | "error" | "info" | "log" | "warn"; + readonly message: object; + } + // This marks the worker handler return information. + // This is separate from Outcome because the worker invocation can live for a long time after + // returning. For example - Websockets that return an http upgrade response but then continue + // streaming information or SSE http connections. + interface Return { + readonly type: "return"; + readonly info?: FetchResponseInfo; + } + interface Attribute { + readonly name: string; + readonly value: string | string[] | boolean | boolean[] | number | number[] | bigint | bigint[]; + } + interface Attributes { + readonly type: "attributes"; + readonly info: Attribute[]; + } + type EventType = Onset | Outcome | SpanOpen | SpanClose | DiagnosticChannelEvent | Exception | Log | Return | Attributes; + // Context in which this trace event lives. + interface SpanContext { + // Single id for the entire top-level invocation + // This should be a new traceId for the first worker stage invoked in the eyeball request and then + // same-account service-bindings should reuse the same traceId but cross-account service-bindings + // should use a new traceId. + readonly traceId: string; + // spanId in which this event is handled + // for Onset and SpanOpen events this would be the parent span id + // for Outcome and SpanClose these this would be the span id of the opening Onset and SpanOpen events + // For Hibernate and Mark this would be the span under which they were emitted. + // spanId is not set ONLY if: + // 1. This is an Onset event + // 2. We are not inheriting any SpanContext. (e.g. this is a cross-account service binding or a new top-level invocation) + readonly spanId?: string; + } + interface TailEvent { + // invocation id of the currently invoked worker stage. + // invocation id will always be unique to every Onset event and will be the same until the Outcome event. + readonly invocationId: string; + // Inherited spanContext for this event. + readonly spanContext: SpanContext; + readonly timestamp: Date; + readonly sequence: number; + readonly event: Event; + } + type TailEventHandler = (event: TailEvent) => void | Promise; + type TailEventHandlerObject = { + outcome?: TailEventHandler; + spanOpen?: TailEventHandler; + spanClose?: TailEventHandler; + diagnosticChannel?: TailEventHandler; + exception?: TailEventHandler; + log?: TailEventHandler; + return?: TailEventHandler; + attributes?: TailEventHandler; + }; + type TailEventHandlerType = TailEventHandler | TailEventHandlerObject; +} +// Copyright (c) 2022-2023 Cloudflare, Inc. +// Licensed under the Apache 2.0 license found in the LICENSE file or at: +// https://opensource.org/licenses/Apache-2.0 +/** + * Data types supported for holding vector metadata. + */ +type VectorizeVectorMetadataValue = string | number | boolean | string[]; +/** + * Additional information to associate with a vector. + */ +type VectorizeVectorMetadata = VectorizeVectorMetadataValue | Record; +type VectorFloatArray = Float32Array | Float64Array; +interface VectorizeError { + code?: number; + error: string; +} +/** + * Comparison logic/operation to use for metadata filtering. + * + * This list is expected to grow as support for more operations are released. + */ +type VectorizeVectorMetadataFilterOp = '$eq' | '$ne' | '$lt' | '$lte' | '$gt' | '$gte'; +type VectorizeVectorMetadataFilterCollectionOp = '$in' | '$nin'; +/** + * Filter criteria for vector metadata used to limit the retrieved query result set. + */ +type VectorizeVectorMetadataFilter = { + [field: string]: Exclude | null | { + [Op in VectorizeVectorMetadataFilterOp]?: Exclude | null; + } | { + [Op in VectorizeVectorMetadataFilterCollectionOp]?: Exclude[]; + }; +}; +/** + * Supported distance metrics for an index. + * Distance metrics determine how other "similar" vectors are determined. + */ +type VectorizeDistanceMetric = "euclidean" | "cosine" | "dot-product"; +/** + * Metadata return levels for a Vectorize query. + * + * Default to "none". + * + * @property all Full metadata for the vector return set, including all fields (including those un-indexed) without truncation. This is a more expensive retrieval, as it requires additional fetching & reading of un-indexed data. + * @property indexed Return all metadata fields configured for indexing in the vector return set. This level of retrieval is "free" in that no additional overhead is incurred returning this data. However, note that indexed metadata is subject to truncation (especially for larger strings). + * @property none No indexed metadata will be returned. + */ +type VectorizeMetadataRetrievalLevel = "all" | "indexed" | "none"; +interface VectorizeQueryOptions { + topK?: number; + namespace?: string; + returnValues?: boolean; + returnMetadata?: boolean | VectorizeMetadataRetrievalLevel; + filter?: VectorizeVectorMetadataFilter; +} +/** + * Information about the configuration of an index. + */ +type VectorizeIndexConfig = { + dimensions: number; + metric: VectorizeDistanceMetric; +} | { + preset: string; // keep this generic, as we'll be adding more presets in the future and this is only in a read capacity +}; +/** + * Metadata about an existing index. + * + * This type is exclusively for the Vectorize **beta** and will be deprecated once Vectorize RC is released. + * See {@link VectorizeIndexInfo} for its post-beta equivalent. + */ +interface VectorizeIndexDetails { + /** The unique ID of the index */ + readonly id: string; + /** The name of the index. */ + name: string; + /** (optional) A human readable description for the index. */ + description?: string; + /** The index configuration, including the dimension size and distance metric. */ + config: VectorizeIndexConfig; + /** The number of records containing vectors within the index. */ + vectorsCount: number; +} +/** + * Metadata about an existing index. + */ +interface VectorizeIndexInfo { + /** The number of records containing vectors within the index. */ + vectorCount: number; + /** Number of dimensions the index has been configured for. */ + dimensions: number; + /** ISO 8601 datetime of the last processed mutation on in the index. All changes before this mutation will be reflected in the index state. */ + processedUpToDatetime: number; + /** UUIDv4 of the last mutation processed by the index. All changes before this mutation will be reflected in the index state. */ + processedUpToMutation: number; +} +/** + * Represents a single vector value set along with its associated metadata. + */ +interface VectorizeVector { + /** The ID for the vector. This can be user-defined, and must be unique. It should uniquely identify the object, and is best set based on the ID of what the vector represents. */ + id: string; + /** The vector values */ + values: VectorFloatArray | number[]; + /** The namespace this vector belongs to. */ + namespace?: string; + /** Metadata associated with the vector. Includes the values of other fields and potentially additional details. */ + metadata?: Record; +} +/** + * Represents a matched vector for a query along with its score and (if specified) the matching vector information. + */ +type VectorizeMatch = Pick, "values"> & Omit & { + /** The score or rank for similarity, when returned as a result */ + score: number; +}; +/** + * A set of matching {@link VectorizeMatch} for a particular query. + */ +interface VectorizeMatches { + matches: VectorizeMatch[]; + count: number; +} +/** + * Results of an operation that performed a mutation on a set of vectors. + * Here, `ids` is a list of vectors that were successfully processed. + * + * This type is exclusively for the Vectorize **beta** and will be deprecated once Vectorize RC is released. + * See {@link VectorizeAsyncMutation} for its post-beta equivalent. + */ +interface VectorizeVectorMutation { + /* List of ids of vectors that were successfully processed. */ + ids: string[]; + /* Total count of the number of processed vectors. */ + count: number; +} +/** + * Result type indicating a mutation on the Vectorize Index. + * Actual mutations are processed async where the `mutationId` is the unique identifier for the operation. + */ +interface VectorizeAsyncMutation { + /** The unique identifier for the async mutation operation containing the changeset. */ + mutationId: string; +} +/** + * A Vectorize Vector Search Index for querying vectors/embeddings. + * + * This type is exclusively for the Vectorize **beta** and will be deprecated once Vectorize RC is released. + * See {@link Vectorize} for its new implementation. + */ +declare abstract class VectorizeIndex { + /** + * Get information about the currently bound index. + * @returns A promise that resolves with information about the current index. + */ + public describe(): Promise; + /** + * Use the provided vector to perform a similarity search across the index. + * @param vector Input vector that will be used to drive the similarity search. + * @param options Configuration options to massage the returned data. + * @returns A promise that resolves with matched and scored vectors. + */ + public query(vector: VectorFloatArray | number[], options?: VectorizeQueryOptions): Promise; + /** + * Insert a list of vectors into the index dataset. If a provided id exists, an error will be thrown. + * @param vectors List of vectors that will be inserted. + * @returns A promise that resolves with the ids & count of records that were successfully processed. + */ + public insert(vectors: VectorizeVector[]): Promise; + /** + * Upsert a list of vectors into the index dataset. If a provided id exists, it will be replaced with the new values. + * @param vectors List of vectors that will be upserted. + * @returns A promise that resolves with the ids & count of records that were successfully processed. + */ + public upsert(vectors: VectorizeVector[]): Promise; + /** + * Delete a list of vectors with a matching id. + * @param ids List of vector ids that should be deleted. + * @returns A promise that resolves with the ids & count of records that were successfully processed (and thus deleted). + */ + public deleteByIds(ids: string[]): Promise; + /** + * Get a list of vectors with a matching id. + * @param ids List of vector ids that should be returned. + * @returns A promise that resolves with the raw unscored vectors matching the id set. + */ + public getByIds(ids: string[]): Promise; +} +/** + * A Vectorize Vector Search Index for querying vectors/embeddings. + * + * Mutations in this version are async, returning a mutation id. + */ +declare abstract class Vectorize { + /** + * Get information about the currently bound index. + * @returns A promise that resolves with information about the current index. + */ + public describe(): Promise; + /** + * Use the provided vector to perform a similarity search across the index. + * @param vector Input vector that will be used to drive the similarity search. + * @param options Configuration options to massage the returned data. + * @returns A promise that resolves with matched and scored vectors. + */ + public query(vector: VectorFloatArray | number[], options?: VectorizeQueryOptions): Promise; + /** + * Use the provided vector-id to perform a similarity search across the index. + * @param vectorId Id for a vector in the index against which the index should be queried. + * @param options Configuration options to massage the returned data. + * @returns A promise that resolves with matched and scored vectors. + */ + public queryById(vectorId: string, options?: VectorizeQueryOptions): Promise; + /** + * Insert a list of vectors into the index dataset. If a provided id exists, an error will be thrown. + * @param vectors List of vectors that will be inserted. + * @returns A promise that resolves with a unique identifier of a mutation containing the insert changeset. + */ + public insert(vectors: VectorizeVector[]): Promise; + /** + * Upsert a list of vectors into the index dataset. If a provided id exists, it will be replaced with the new values. + * @param vectors List of vectors that will be upserted. + * @returns A promise that resolves with a unique identifier of a mutation containing the upsert changeset. + */ + public upsert(vectors: VectorizeVector[]): Promise; + /** + * Delete a list of vectors with a matching id. + * @param ids List of vector ids that should be deleted. + * @returns A promise that resolves with a unique identifier of a mutation containing the delete changeset. + */ + public deleteByIds(ids: string[]): Promise; + /** + * Get a list of vectors with a matching id. + * @param ids List of vector ids that should be returned. + * @returns A promise that resolves with the raw unscored vectors matching the id set. + */ + public getByIds(ids: string[]): Promise; +} +/** + * The interface for "version_metadata" binding + * providing metadata about the Worker Version using this binding. + */ +type WorkerVersionMetadata = { + /** The ID of the Worker Version using this binding */ + id: string; + /** The tag of the Worker Version using this binding */ + tag: string; + /** The timestamp of when the Worker Version was uploaded */ + timestamp: string; +}; +interface DynamicDispatchLimits { + /** + * Limit CPU time in milliseconds. + */ + cpuMs?: number; + /** + * Limit number of subrequests. + */ + subRequests?: number; +} +interface DynamicDispatchOptions { + /** + * Limit resources of invoked Worker script. + */ + limits?: DynamicDispatchLimits; + /** + * Arguments for outbound Worker script, if configured. + */ + outbound?: { + [key: string]: any; + }; +} +interface DispatchNamespace { + /** + * @param name Name of the Worker script. + * @param args Arguments to Worker script. + * @param options Options for Dynamic Dispatch invocation. + * @returns A Fetcher object that allows you to send requests to the Worker script. + * @throws If the Worker script does not exist in this dispatch namespace, an error will be thrown. + */ + get(name: string, args?: { + [key: string]: any; + }, options?: DynamicDispatchOptions): Fetcher; +} +declare module 'cloudflare:workflows' { + /** + * NonRetryableError allows for a user to throw a fatal error + * that makes a Workflow instance fail immediately without triggering a retry + */ + export class NonRetryableError extends Error { + public constructor(message: string, name?: string); + } +} +declare abstract class Workflow { + /** + * Get a handle to an existing instance of the Workflow. + * @param id Id for the instance of this Workflow + * @returns A promise that resolves with a handle for the Instance + */ + public get(id: string): Promise; + /** + * Create a new instance and return a handle to it. If a provided id exists, an error will be thrown. + * @param options Options when creating an instance including id and params + * @returns A promise that resolves with a handle for the Instance + */ + public create(options?: WorkflowInstanceCreateOptions): Promise; + /** + * Create a batch of instances and return handle for all of them. If a provided id exists, an error will be thrown. + * `createBatch` is limited at 100 instances at a time or when the RPC limit for the batch (1MiB) is reached. + * @param batch List of Options when creating an instance including name and params + * @returns A promise that resolves with a list of handles for the created instances. + */ + public createBatch(batch: WorkflowInstanceCreateOptions[]): Promise; +} +type WorkflowDurationLabel = 'second' | 'minute' | 'hour' | 'day' | 'week' | 'month' | 'year'; +type WorkflowSleepDuration = `${number} ${WorkflowDurationLabel}${'s' | ''}` | number; +type WorkflowRetentionDuration = WorkflowSleepDuration; +interface WorkflowInstanceCreateOptions { + /** + * An id for your Workflow instance. Must be unique within the Workflow. + */ + id?: string; + /** + * The event payload the Workflow instance is triggered with + */ + params?: PARAMS; + /** + * The retention policy for Workflow instance. + * Defaults to the maximum retention period available for the owner's account. + */ + retention?: { + successRetention?: WorkflowRetentionDuration; + errorRetention?: WorkflowRetentionDuration; + }; +} +type InstanceStatus = { + status: 'queued' // means that instance is waiting to be started (see concurrency limits) + | 'running' | 'paused' | 'errored' | 'terminated' // user terminated the instance while it was running + | 'complete' | 'waiting' // instance is hibernating and waiting for sleep or event to finish + | 'waitingForPause' // instance is finishing the current work to pause + | 'unknown'; + error?: { + name: string; + message: string; + }; + output?: unknown; +}; +interface WorkflowError { + code?: number; + message: string; +} +declare abstract class WorkflowInstance { + public id: string; + /** + * Pause the instance. + */ + public pause(): Promise; + /** + * Resume the instance. If it is already running, an error will be thrown. + */ + public resume(): Promise; + /** + * Terminate the instance. If it is errored, terminated or complete, an error will be thrown. + */ + public terminate(): Promise; + /** + * Restart the instance. + */ + public restart(): Promise; + /** + * Returns the current status of the instance. + */ + public status(): Promise; + /** + * Send an event to this instance. + */ + public sendEvent({ type, payload, }: { + type: string; + payload: unknown; + }): Promise; +} diff --git a/ipfs-retriever/wrangler.toml b/ipfs-retriever/wrangler.toml new file mode 100644 index 00000000..d67f2278 --- /dev/null +++ b/ipfs-retriever/wrangler.toml @@ -0,0 +1,83 @@ +name = "filbeam-ipfs-retriever" +main = "bin/ipfs-retriever.js" +compatibility_date = "2024-12-05" +compatibility_flags = ["nodejs_compat", "enable_request_signal"] +logpush = true +upload_source_maps = true + +[[d1_databases]] +binding = "DB" +database_name = "test-db" +database_id = "8cc92155-16f6-426a-b782-2965e0daf100" + +[[kv_namespaces]] +binding = "BAD_BITS_KV" +id = "2f2e5486ea0c48e993f6dff87a4aa102" + +[vars] +ORIGIN_CACHE_TTL = 86400 +CLIENT_CACHE_TTL = 31536000 +DNS_ROOT = ".localhost" +BOT_TOKENS = "" +ENFORCE_EGRESS_QUOTA = false + +[env.dev] +tail_consumers = [{ service = "filbeam-tail-handler-dev" }] + +[env.dev.vars] +ENVIRONMENT = "dev" +ORIGIN_CACHE_TTL = 86400 +CLIENT_CACHE_TTL = 31536000 +DNS_ROOT = ".localhost" +BOT_TOKENS = "" +ENFORCE_EGRESS_QUOTA = false + +[[env.dev.d1_databases]] +binding = "DB" +database_name = "dev-db" +database_id = "8cc92155-16f6-426a-b782-2965e0daf101" + +[[env.dev.kv_namespaces]] +binding = "BAD_BITS_KV" +id = "2f2e5486ea0c48e993f6dff87a4aa102" + +[env.calibration] +tail_consumers = [{ service = "filbeam-tail-handler-calibration" }] + +[env.calibration.vars] +ENVIRONMENT = "calibration" +ORIGIN_CACHE_TTL = 86400 +CLIENT_CACHE_TTL = 31536000 +DNS_ROOT = ".ipfs.calibration.filbeam.io" +ENFORCE_EGRESS_QUOTA = true + +[[env.calibration.d1_databases]] +binding = "DB" +database_name = "filcdn-calibration-db" +database_id = "78f15bbb-391f-4797-9016-a6cb86c0b9b8" + +[[env.calibration.kv_namespaces]] +binding = "BAD_BITS_KV" +id = "178592ee0a3b4b00894a23186b3a0179" + +[env.mainnet] +tail_consumers = [{ service = "filbeam-tail-handler-mainnet" }] + +[env.mainnet.vars] +ENVIRONMENT = "mainnet" +ORIGIN_CACHE_TTL = 86400 +CLIENT_CACHE_TTL = 31536000 +DNS_ROOT = ".ipfs.filbeam.io" +ENFORCE_EGRESS_QUOTA = true + +[[env.mainnet.d1_databases]] +binding = "DB" +database_name = "filcdn-mainnet-db" +database_id = "e8de6418-2cb7-4413-9ba0-a9c8aacf9a66" + +[[env.mainnet.kv_namespaces]] +binding = "BAD_BITS_KV" +id = "7b03c39d53a041fdbe973c20285e16e9" + +[limits] +cpu_ms = 300000 # maximum diff --git a/package-lock.json b/package-lock.json index b96a7be0..52f501c4 100644 --- a/package-lock.json +++ b/package-lock.json @@ -10,6 +10,7 @@ "workspaces": [ "bad-bits", "indexer", + "ipfs-retriever", "piece-retriever", "tail-handler", "terminator", @@ -53,6 +54,17 @@ "@types/validator": "^13.15.10" } }, + "ipfs-retriever": { + "name": "@filbeam/ipfs-retriever", + "version": "1.0.0", + "dependencies": { + "@filbeam/retrieval": "^1.0.0", + "@ipld/car": "^5.4.2", + "@web3-storage/car-block-validator": "^1.2.2", + "ipfs-unixfs-exporter": "^13.7.3", + "multiformats": "^13.4.1" + } + }, "node_modules/@adraffy/ens-normalize": { "version": "1.11.1", "license": "MIT" @@ -1562,6 +1574,10 @@ "resolved": "indexer", "link": true }, + "node_modules/@filbeam/ipfs-retriever": { + "resolved": "ipfs-retriever", + "link": true + }, "node_modules/@filbeam/payment-settler": { "resolved": "payment-settler", "link": true @@ -2152,6 +2168,73 @@ "url": "https://opencollective.com/libvips" } }, + "node_modules/@ipld/car": { + "version": "5.4.6", + "resolved": "https://registry.npmjs.org/@ipld/car/-/car-5.4.6.tgz", + "integrity": "sha512-f1Iyb9ci8B/TGCyjAuSQ6BOLiy2u8en7rB+SumbyweqXDu3gxYJwYbGzoposKjOzc4CeGlkEbwEw8REoekYAKQ==", + "license": "Apache-2.0 OR MIT", + "dependencies": { + "@ipld/dag-cbor": "^10.0.1", + "cborg": "^5.0.0", + "multiformats": "^14.0.0", + "varint": "^6.0.0" + }, + "engines": { + "node": ">=16.0.0", + "npm": ">=7.0.0" + } + }, + "node_modules/@ipld/car/node_modules/multiformats": { + "version": "14.0.0", + "resolved": "https://registry.npmjs.org/multiformats/-/multiformats-14.0.0.tgz", + "integrity": "sha512-iWK1RrAS58p2NDfeZFuSUSv3ZPewTIhsGbh/5NgeGGJwJmRljLxGtjRR3nkn+loG3zl+IrfR/W1590QnrSK+Gg==", + "license": "Apache-2.0 OR MIT" + }, + "node_modules/@ipld/dag-cbor": { + "version": "10.0.1", + "resolved": "https://registry.npmjs.org/@ipld/dag-cbor/-/dag-cbor-10.0.1.tgz", + "integrity": "sha512-nF07iiZPqduSXyMxc0jGANArRHFa9hjMQpQlgLOV2O/3xI1CNb5sXhvbmigbMiz5owSGi0Oq10VtauupMojuiw==", + "license": "Apache-2.0 OR MIT", + "dependencies": { + "cborg": "^5.0.1", + "multiformats": "^14.0.0" + } + }, + "node_modules/@ipld/dag-cbor/node_modules/multiformats": { + "version": "14.0.0", + "resolved": "https://registry.npmjs.org/multiformats/-/multiformats-14.0.0.tgz", + "integrity": "sha512-iWK1RrAS58p2NDfeZFuSUSv3ZPewTIhsGbh/5NgeGGJwJmRljLxGtjRR3nkn+loG3zl+IrfR/W1590QnrSK+Gg==", + "license": "Apache-2.0 OR MIT" + }, + "node_modules/@ipld/dag-json": { + "version": "10.2.9", + "resolved": "https://registry.npmjs.org/@ipld/dag-json/-/dag-json-10.2.9.tgz", + "integrity": "sha512-opNPQQsTuCFZkaJCAqXrB/n9OqUD6W2Boz/Au5HjhLQyczmT8lxoOZObqQ5S5hhnV8p6sgKAimNhUB2W6y0Mzg==", + "license": "Apache-2.0 OR MIT", + "dependencies": { + "cborg": "^5.0.0", + "multiformats": "^13.1.0" + } + }, + "node_modules/@ipld/dag-pb": { + "version": "4.1.7", + "resolved": "https://registry.npmjs.org/@ipld/dag-pb/-/dag-pb-4.1.7.tgz", + "integrity": "sha512-/i/13trFihjWfDyXlylRwhuYjtzYjvOFw0vlRjYGnZuv7d7MOgA2lV/vRuL5RfeUajM03aZfFLdq4S7cTbbTRg==", + "license": "Apache-2.0 OR MIT", + "dependencies": { + "multiformats": "^14.0.0" + }, + "engines": { + "node": ">=16.0.0", + "npm": ">=7.0.0" + } + }, + "node_modules/@ipld/dag-pb/node_modules/multiformats": { + "version": "14.0.0", + "resolved": "https://registry.npmjs.org/multiformats/-/multiformats-14.0.0.tgz", + "integrity": "sha512-iWK1RrAS58p2NDfeZFuSUSv3ZPewTIhsGbh/5NgeGGJwJmRljLxGtjRR3nkn+loG3zl+IrfR/W1590QnrSK+Gg==", + "license": "Apache-2.0 OR MIT" + }, "node_modules/@isaacs/cliui": { "version": "8.0.2", "dev": true, @@ -2738,6 +2821,45 @@ "uuid": "dist/bin/uuid" } }, + "node_modules/@multiformats/blake2": { + "version": "2.0.2", + "resolved": "https://registry.npmjs.org/@multiformats/blake2/-/blake2-2.0.2.tgz", + "integrity": "sha512-AOWu6Tyuk5UoT5m4faB6ntVnPB8EmuD6rn18s4cCgHNEGgsamT8GdvjP9DYjzFHQVaP/0L3CaKqWQqJlXx9ecw==", + "license": "Apache-2.0 OR MIT", + "dependencies": { + "blakejs": "^1.2.1", + "multiformats": "^13.0.0" + } + }, + "node_modules/@multiformats/murmur3": { + "version": "2.2.5", + "resolved": "https://registry.npmjs.org/@multiformats/murmur3/-/murmur3-2.2.5.tgz", + "integrity": "sha512-M+VwV8hEx5qB8i7b8fljMwZETJsqyLo8RAXA+JAn+QF9NnH46ZWvGErNukJVlxXSR2KQpuOtHIYIzZlbhhdFvw==", + "license": "Apache-2.0 OR MIT", + "dependencies": { + "multiformats": "^14.0.0" + }, + "engines": { + "node": ">=16.0.0", + "npm": ">=7.0.0" + } + }, + "node_modules/@multiformats/murmur3/node_modules/multiformats": { + "version": "14.0.0", + "resolved": "https://registry.npmjs.org/multiformats/-/multiformats-14.0.0.tgz", + "integrity": "sha512-iWK1RrAS58p2NDfeZFuSUSv3ZPewTIhsGbh/5NgeGGJwJmRljLxGtjRR3nkn+loG3zl+IrfR/W1590QnrSK+Gg==", + "license": "Apache-2.0 OR MIT" + }, + "node_modules/@multiformats/sha3": { + "version": "3.0.3", + "resolved": "https://registry.npmjs.org/@multiformats/sha3/-/sha3-3.0.3.tgz", + "integrity": "sha512-FugQMj5ts/aUFx2cWuE9Y3O/YN/puVJWaRoFQSSKTc5jm89o0UUowvwxLkkURx9MQiR5GJsoG9d6ZITWi8FiBA==", + "license": "Apache-2.0 OR MIT", + "dependencies": { + "js-sha3": "^0.9.1", + "multiformats": "^13.0.0" + } + }, "node_modules/@noble/ciphers": { "version": "1.3.0", "license": "MIT", @@ -6628,6 +6750,28 @@ "integrity": "sha512-Xni35NKzjgMrwevysHTCArtLDpPvye8zV/0E4EyYn43P7/7qvQwPh9BGkHewbMulVntbigmcT7rdX3BNo9wRJg==", "license": "0BSD" }, + "node_modules/@web3-storage/car-block-validator": { + "version": "1.2.2", + "resolved": "https://registry.npmjs.org/@web3-storage/car-block-validator/-/car-block-validator-1.2.2.tgz", + "integrity": "sha512-lR9l+ZszhTid5HfZE8ohnGf2RJp2kaBOnoejmsACs3iTNiy+3K09dnPm8MhgBE9RCIgPBKM0CCWXO9l+I6jrKA==", + "license": "Apache-2.0 OR MIT", + "dependencies": { + "@multiformats/blake2": "^2.0.2", + "@multiformats/murmur3": "^2.1.8", + "@multiformats/sha3": "^3.0.2", + "multiformats": "^13.3.1", + "uint8arrays": "^5.1.0" + } + }, + "node_modules/@web3-storage/car-block-validator/node_modules/uint8arrays": { + "version": "5.1.1", + "resolved": "https://registry.npmjs.org/uint8arrays/-/uint8arrays-5.1.1.tgz", + "integrity": "sha512-9muQwa4wZG4dKi9gMAIBtnk2Pw87SRpvWTH6lOGm19V2Uqxr4uomUf2PGqPnWc+qs06sN8owUU4jfcoWOcfwVQ==", + "license": "Apache-2.0 OR MIT", + "dependencies": { + "multiformats": "^13.0.0" + } + }, "node_modules/abitype": { "version": "1.2.3", "license": "MIT", @@ -6647,6 +6791,12 @@ } } }, + "node_modules/abort-error": { + "version": "1.0.2", + "resolved": "https://registry.npmjs.org/abort-error/-/abort-error-1.0.2.tgz", + "integrity": "sha512-lVgvB2NyPLqbXXhVmXcYFTC1x5K7CiVdPgdY7LGgFQWC8506oN01sPN3i9cl9ynuwF4iJ0TS9exnR7cZ9FuX4w==", + "license": "Apache-2.0 OR MIT" + }, "node_modules/acorn": { "version": "8.15.0", "dev": true, @@ -7010,6 +7160,12 @@ "dev": true, "license": "MIT" }, + "node_modules/blakejs": { + "version": "1.2.1", + "resolved": "https://registry.npmjs.org/blakejs/-/blakejs-1.2.1.tgz", + "integrity": "sha512-QXUSXI3QVc/gJME0dBpXrag1kbzOqCjCX8/b54ntNyW6sjtoqxqRk3LTmXzaJoh71zMsDCjM+47jS7XiwN/+fQ==", + "license": "MIT" + }, "node_modules/bn.js": { "version": "5.2.2", "resolved": "https://registry.npmjs.org/bn.js/-/bn.js-5.2.2.tgz", @@ -7155,6 +7311,15 @@ "node": ">=6" } }, + "node_modules/cborg": { + "version": "5.1.1", + "resolved": "https://registry.npmjs.org/cborg/-/cborg-5.1.1.tgz", + "integrity": "sha512-BDbSRIp6XrQXkTc7g+DN0RB9RrDPTUfals2ecWUlt3juPLjbAvy/V72mJcXY0Ehu0Dq/3WpNCOCT68HUTbW+lw==", + "license": "Apache-2.0", + "bin": { + "cborg": "lib/bin.js" + } + }, "node_modules/cbw-sdk": { "name": "@coinbase/wallet-sdk", "version": "3.9.3", @@ -9059,6 +9224,31 @@ "uncrypto": "^0.1.3" } }, + "node_modules/hamt-sharding": { + "version": "3.0.8", + "resolved": "https://registry.npmjs.org/hamt-sharding/-/hamt-sharding-3.0.8.tgz", + "integrity": "sha512-pEh4hYa4ZyDJ2jWBTz7MVqTTGC3va3iEIMDJfbCKskR9rzHRd8HVwUyH5nLiLxK6VCFWIJ4vnlAEn4Hwx49Rbg==", + "license": "Apache-2.0 OR MIT", + "dependencies": { + "sparse-array": "^1.3.1", + "uint8arrays": "^6.1.1" + } + }, + "node_modules/hamt-sharding/node_modules/multiformats": { + "version": "14.0.0", + "resolved": "https://registry.npmjs.org/multiformats/-/multiformats-14.0.0.tgz", + "integrity": "sha512-iWK1RrAS58p2NDfeZFuSUSv3ZPewTIhsGbh/5NgeGGJwJmRljLxGtjRR3nkn+loG3zl+IrfR/W1590QnrSK+Gg==", + "license": "Apache-2.0 OR MIT" + }, + "node_modules/hamt-sharding/node_modules/uint8arrays": { + "version": "6.1.1", + "resolved": "https://registry.npmjs.org/uint8arrays/-/uint8arrays-6.1.1.tgz", + "integrity": "sha512-iz7JN0XCSZYA111lhFG2Ui9EhFvTNekqSRHw3lvMHq+dzwWy1OQftxFQREEh4rffU0oSoXdQHsk2TiHKVm4fsA==", + "license": "Apache-2.0 OR MIT", + "dependencies": { + "multiformats": "^14.0.0" + } + }, "node_modules/has-bigints": { "version": "1.1.0", "dev": true, @@ -9238,6 +9428,22 @@ "node": "^18.17.0 || >=20.5.0" } }, + "node_modules/interface-blockstore": { + "version": "5.3.2", + "resolved": "https://registry.npmjs.org/interface-blockstore/-/interface-blockstore-5.3.2.tgz", + "integrity": "sha512-oA9Pjkxun/JHAsZrYEyKX+EoPjLciTzidE7wipLc/3YoHDjzsnXRJzAzFJXNUvogtY4g7hIwxArx8+WKJs2RIg==", + "license": "Apache-2.0 OR MIT", + "dependencies": { + "interface-store": "^6.0.0", + "multiformats": "^13.3.6" + } + }, + "node_modules/interface-store": { + "version": "6.0.3", + "resolved": "https://registry.npmjs.org/interface-store/-/interface-store-6.0.3.tgz", + "integrity": "sha512-+WvfEZnFUhRwFxgz+QCQi7UC6o9AM0EHM9bpIe2Nhqb100NHCsTvNAn4eJgvgV2/tmLo1MP9nGxQKEcZTAueLA==", + "license": "Apache-2.0 OR MIT" + }, "node_modules/internal-slot": { "version": "1.1.0", "dev": true, @@ -9251,6 +9457,50 @@ "node": ">= 0.4" } }, + "node_modules/ipfs-unixfs": { + "version": "11.2.5", + "resolved": "https://registry.npmjs.org/ipfs-unixfs/-/ipfs-unixfs-11.2.5.tgz", + "integrity": "sha512-uasYJ0GLPbViaTFsOLnL9YPjX5VmhnqtWRriogAHOe4ApmIi9VAOFBzgDHsUW2ub4pEa/EysbtWk126g2vkU/g==", + "license": "Apache-2.0 OR MIT", + "dependencies": { + "protons-runtime": "^5.5.0", + "uint8arraylist": "^2.4.8" + } + }, + "node_modules/ipfs-unixfs-exporter": { + "version": "13.7.3", + "resolved": "https://registry.npmjs.org/ipfs-unixfs-exporter/-/ipfs-unixfs-exporter-13.7.3.tgz", + "integrity": "sha512-sTFjAEnsPu5irh9rvT1j5mNf7nXnW78x5SJrCIrNZb1UqkXQtNX81RjAnTBShUtZ5ujSOc/yrC9Az8il8NVkKQ==", + "license": "Apache-2.0 OR MIT", + "dependencies": { + "@ipld/dag-cbor": "^9.2.4", + "@ipld/dag-json": "^10.2.5", + "@ipld/dag-pb": "^4.1.5", + "@multiformats/murmur3": "^2.1.8", + "hamt-sharding": "^3.0.6", + "interface-blockstore": "^5.3.2", + "ipfs-unixfs": "^11.0.0", + "it-filter": "^3.1.4", + "it-last": "^3.0.9", + "it-map": "^3.1.4", + "it-parallel": "^3.0.13", + "it-pipe": "^3.0.1", + "it-pushable": "^3.2.3", + "multiformats": "^13.3.7", + "p-queue": "^8.1.0", + "progress-events": "^1.0.1" + } + }, + "node_modules/ipfs-unixfs-exporter/node_modules/@ipld/dag-cbor": { + "version": "9.2.7", + "resolved": "https://registry.npmjs.org/@ipld/dag-cbor/-/dag-cbor-9.2.7.tgz", + "integrity": "sha512-ZmfXmElRWATr+hoUTSAOr6HUcjVhOcNHDqgczc76qte2DHHFEK0ZhNzUcdTDQhF/VSIvf2ioaRTRLWwLc83sNw==", + "license": "Apache-2.0 OR MIT", + "dependencies": { + "cborg": "^5.0.1", + "multiformats": "^13.1.0" + } + }, "node_modules/iron-webcrypto": { "version": "1.2.1", "resolved": "https://registry.npmjs.org/iron-webcrypto/-/iron-webcrypto-1.2.1.tgz", @@ -9700,6 +9950,95 @@ "ws": "*" } }, + "node_modules/it-filter": { + "version": "3.1.6", + "resolved": "https://registry.npmjs.org/it-filter/-/it-filter-3.1.6.tgz", + "integrity": "sha512-yXiGPAvJn/exXjVFSCMQc3+J/7RLpOMwKoY2DH1yMhF4lYkdRoAdOwU0vnDACAlRAexf7AZvESZIc9mzhEoi/A==", + "license": "Apache-2.0 OR MIT", + "dependencies": { + "it-peekable": "^3.0.0" + } + }, + "node_modules/it-last": { + "version": "3.0.11", + "resolved": "https://registry.npmjs.org/it-last/-/it-last-3.0.11.tgz", + "integrity": "sha512-Fg571l81nPzhZsiYjkw4dkhRqAK4oqIamTPEfAOnXI/5pYXz+dIfMVYmh9ncZs58oFNMkdF3bYFuCBTw/xJK0w==", + "license": "Apache-2.0 OR MIT" + }, + "node_modules/it-map": { + "version": "3.1.6", + "resolved": "https://registry.npmjs.org/it-map/-/it-map-3.1.6.tgz", + "integrity": "sha512-wCix0FXImtIPIxhCnbz35RqWs00e/CReSZX9nZq1j46JcAzBBp57ob9/2l1WnDYEaUURIR8xCyg2NsWbOwBJFQ==", + "license": "Apache-2.0 OR MIT", + "dependencies": { + "it-peekable": "^3.0.0" + } + }, + "node_modules/it-merge": { + "version": "3.0.14", + "resolved": "https://registry.npmjs.org/it-merge/-/it-merge-3.0.14.tgz", + "integrity": "sha512-D3t1Go2G2SQMkTujaA6EVojJPJKA9pFksxlSPDRBfrHKhWl6O40vEP7Itr5eCAjyCQH5p9+BFFVIy9bhLM4ZuQ==", + "license": "Apache-2.0 OR MIT", + "dependencies": { + "it-queueless-pushable": "^2.0.0" + } + }, + "node_modules/it-parallel": { + "version": "3.0.15", + "resolved": "https://registry.npmjs.org/it-parallel/-/it-parallel-3.0.15.tgz", + "integrity": "sha512-1iUV4wg7cDy40N32/XosK7mcwKM+oeSGq0r7czxNaUGGSQvbdSmkIoK4Vu/XPsXZIqBLt9tO+LDPi8RJBJ/Qwg==", + "license": "Apache-2.0 OR MIT", + "dependencies": { + "p-defer": "^4.0.1" + } + }, + "node_modules/it-peekable": { + "version": "3.0.10", + "resolved": "https://registry.npmjs.org/it-peekable/-/it-peekable-3.0.10.tgz", + "integrity": "sha512-2E6+p1pelZOhzp69aaiiBuEybWzAl10uYbIdCR3Pxy8bFNnS/kgpbLtGbNbIZ6RVdU7yHHkmATYwjy52GfFEKA==", + "license": "Apache-2.0 OR MIT" + }, + "node_modules/it-pipe": { + "version": "3.0.1", + "resolved": "https://registry.npmjs.org/it-pipe/-/it-pipe-3.0.1.tgz", + "integrity": "sha512-sIoNrQl1qSRg2seYSBH/3QxWhJFn9PKYvOf/bHdtCBF0bnghey44VyASsWzn5dAx0DCDDABq1hZIuzKmtBZmKA==", + "license": "Apache-2.0 OR MIT", + "dependencies": { + "it-merge": "^3.0.0", + "it-pushable": "^3.1.2", + "it-stream-types": "^2.0.1" + }, + "engines": { + "node": ">=16.0.0", + "npm": ">=7.0.0" + } + }, + "node_modules/it-pushable": { + "version": "3.2.4", + "resolved": "https://registry.npmjs.org/it-pushable/-/it-pushable-3.2.4.tgz", + "integrity": "sha512-WSD7Ss4oCRfDZJT4ldLWr0Bom/muY90xxoJ5PQnU3uSKf0kxCOeehqZtiJX1ARqn+ymXGh1bxpDW9bDNHp2ivQ==", + "license": "Apache-2.0 OR MIT", + "dependencies": { + "p-defer": "^4.0.0" + } + }, + "node_modules/it-queueless-pushable": { + "version": "2.0.5", + "resolved": "https://registry.npmjs.org/it-queueless-pushable/-/it-queueless-pushable-2.0.5.tgz", + "integrity": "sha512-BaKqGLL1AQMR1AEaxiM09vzJQVXHHhfhh9UV0qPqORw/8Rm8igDQqT1qHregfHIb1NIW9jxQ/aXBibHJyuivuQ==", + "license": "Apache-2.0 OR MIT", + "dependencies": { + "abort-error": "^1.0.2", + "p-defer": "^4.0.1", + "race-signal": "^2.0.0" + } + }, + "node_modules/it-stream-types": { + "version": "2.0.4", + "resolved": "https://registry.npmjs.org/it-stream-types/-/it-stream-types-2.0.4.tgz", + "integrity": "sha512-tsX+klvMQ53J4Jm2B52vCIs7WD609ck+VS9X2TKMEv7VPY9VwaYKmSWyHek5QS0wHBtP0bWj9KMqCtAHgVKiXw==", + "license": "Apache-2.0 OR MIT" + }, "node_modules/iterator.prototype": { "version": "1.1.5", "dev": true, @@ -9798,6 +10137,12 @@ "url": "https://github.com/sponsors/panva" } }, + "node_modules/js-sha3": { + "version": "0.9.3", + "resolved": "https://registry.npmjs.org/js-sha3/-/js-sha3-0.9.3.tgz", + "integrity": "sha512-BcJPCQeLg6WjEx3FE591wVAevlli8lxsxm9/FzV4HXkV49TmBH38Yvrpce6fjbADGMKFrBMGTqrVz3qPIZ88Gg==", + "license": "MIT" + }, "node_modules/js-tokens": { "version": "4.0.0", "license": "MIT" @@ -11111,6 +11456,18 @@ } } }, + "node_modules/p-defer": { + "version": "4.0.1", + "resolved": "https://registry.npmjs.org/p-defer/-/p-defer-4.0.1.tgz", + "integrity": "sha512-Mr5KC5efvAK5VUptYEIopP1bakB85k2IWXaRC0rsh1uwn1L6M0LVml8OIQ4Gudg4oyZakf7FmeRLkMMtZW1i5A==", + "license": "MIT", + "engines": { + "node": ">=12" + }, + "funding": { + "url": "https://github.com/sponsors/sindresorhus" + } + }, "node_modules/p-limit": { "version": "3.1.0", "dev": true, @@ -11141,6 +11498,22 @@ "url": "https://github.com/sponsors/sindresorhus" } }, + "node_modules/p-queue": { + "version": "8.1.1", + "resolved": "https://registry.npmjs.org/p-queue/-/p-queue-8.1.1.tgz", + "integrity": "sha512-aNZ+VfjobsWryoiPnEApGGmf5WmNsCo9xu8dfaYamG5qaLP7ClhLN6NgsFe6SwJ2UbLEBK5dv9x8Mn5+RVhMWQ==", + "license": "MIT", + "dependencies": { + "eventemitter3": "^5.0.1", + "p-timeout": "^6.1.2" + }, + "engines": { + "node": ">=18" + }, + "funding": { + "url": "https://github.com/sponsors/sindresorhus" + } + }, "node_modules/p-retry": { "version": "7.1.1", "license": "MIT", @@ -11154,6 +11527,18 @@ "url": "https://github.com/sponsors/sindresorhus" } }, + "node_modules/p-timeout": { + "version": "6.1.4", + "resolved": "https://registry.npmjs.org/p-timeout/-/p-timeout-6.1.4.tgz", + "integrity": "sha512-MyIV3ZA/PmyBN/ud8vV9XzwTrNtR4jFrObymZYnZqMmW0zA8Z17vnT0rBgFE/TlohB+YCHqXMgZzb3Csp49vqg==", + "license": "MIT", + "engines": { + "node": ">=14.16" + }, + "funding": { + "url": "https://github.com/sponsors/sindresorhus" + } + }, "node_modules/p-try": { "version": "2.2.0", "resolved": "https://registry.npmjs.org/p-try/-/p-try-2.2.0.tgz", @@ -11541,6 +11926,12 @@ "integrity": "sha512-du4wfLyj4yCZq1VupnVSZmRsPJsNuxoDQFdCFHLaYiEbFBD7QE0a+I4D7hOxrVnh78QE/YipFAj9lXHiXocV+Q==", "license": "MIT" }, + "node_modules/progress-events": { + "version": "1.1.0", + "resolved": "https://registry.npmjs.org/progress-events/-/progress-events-1.1.0.tgz", + "integrity": "sha512-82DVc5tI36neVB3IjdXR11ztwGuoBc98em9ijzubeZKxI47OlV2Znq6mlPqE5xPDzO2Uw98GHiQSjj2favBCRQ==", + "license": "Apache-2.0 OR MIT" + }, "node_modules/promise-retry": { "version": "2.0.1", "dev": true, @@ -11563,6 +11954,26 @@ "react-is": "^16.13.1" } }, + "node_modules/protons-runtime": { + "version": "5.6.0", + "resolved": "https://registry.npmjs.org/protons-runtime/-/protons-runtime-5.6.0.tgz", + "integrity": "sha512-/Kde+sB9DsMFrddJT/UZWe6XqvL7SL5dbag/DBCElFKhkwDj7XKt53S+mzLyaDP5OqS0wXjV5SA572uWDaT0Hg==", + "license": "Apache-2.0 OR MIT", + "dependencies": { + "uint8-varint": "^2.0.2", + "uint8arraylist": "^2.4.3", + "uint8arrays": "^5.0.1" + } + }, + "node_modules/protons-runtime/node_modules/uint8arrays": { + "version": "5.1.1", + "resolved": "https://registry.npmjs.org/uint8arrays/-/uint8arrays-5.1.1.tgz", + "integrity": "sha512-9muQwa4wZG4dKi9gMAIBtnk2Pw87SRpvWTH6lOGm19V2Uqxr4uomUf2PGqPnWc+qs06sN8owUU4jfcoWOcfwVQ==", + "license": "Apache-2.0 OR MIT", + "dependencies": { + "multiformats": "^13.0.0" + } + }, "node_modules/proxy-compare": { "version": "2.6.0", "resolved": "https://registry.npmjs.org/proxy-compare/-/proxy-compare-2.6.0.tgz", @@ -11636,6 +12047,12 @@ "integrity": "sha512-tYC1Q1hgyRuHgloV/YXs2w15unPVh8qfu/qCTfhTYamaw7fyhumKa2yGpdSo87vY32rIclj+4fWYQXUMs9EHvg==", "license": "MIT" }, + "node_modules/race-signal": { + "version": "2.0.0", + "resolved": "https://registry.npmjs.org/race-signal/-/race-signal-2.0.0.tgz", + "integrity": "sha512-P31bLhE4ByBX/70QDXMutxnqgwrF1WUXea1O8DXuviAgkdbQ1iQMQotNgzJIBC9yUSn08u/acZrMUhgw7w6GpA==", + "license": "Apache-2.0 OR MIT" + }, "node_modules/radix3": { "version": "1.1.2", "resolved": "https://registry.npmjs.org/radix3/-/radix3-1.1.2.tgz", @@ -12242,6 +12659,12 @@ "node": ">=0.10.0" } }, + "node_modules/sparse-array": { + "version": "1.3.2", + "resolved": "https://registry.npmjs.org/sparse-array/-/sparse-array-1.3.2.tgz", + "integrity": "sha512-ZT711fePGn3+kQyLuv1fpd3rNSkNF8vd5Kv2D+qnOANeyKs3fx6bUMGWRPvgTTcYV64QMqZKZwcuaQSP3AZ0tg==", + "license": "ISC" + }, "node_modules/spdx-correct": { "version": "3.2.0", "dev": true, @@ -12860,6 +13283,43 @@ "integrity": "sha512-yDJTmhydvl5lJzBmy/hyOAA0d+aqCBuwl818haVdYCRrWV84o7YyeVm4QlVHStqNrrJSTb6jKuFAVqAFsr+K3Q==", "license": "MIT" }, + "node_modules/uint8-varint": { + "version": "2.0.5", + "resolved": "https://registry.npmjs.org/uint8-varint/-/uint8-varint-2.0.5.tgz", + "integrity": "sha512-jeFLbL/x30wBRnWjKE1qVBXeumG46r7XmYkpis955lTQ+blccGKFrOsSMHlxePwYB1pI7L8YPHz1t4jLxEs3nA==", + "license": "Apache-2.0 OR MIT", + "dependencies": { + "uint8arraylist": "^2.0.0", + "uint8arrays": "^5.0.0" + } + }, + "node_modules/uint8-varint/node_modules/uint8arrays": { + "version": "5.1.1", + "resolved": "https://registry.npmjs.org/uint8arrays/-/uint8arrays-5.1.1.tgz", + "integrity": "sha512-9muQwa4wZG4dKi9gMAIBtnk2Pw87SRpvWTH6lOGm19V2Uqxr4uomUf2PGqPnWc+qs06sN8owUU4jfcoWOcfwVQ==", + "license": "Apache-2.0 OR MIT", + "dependencies": { + "multiformats": "^13.0.0" + } + }, + "node_modules/uint8arraylist": { + "version": "2.4.9", + "resolved": "https://registry.npmjs.org/uint8arraylist/-/uint8arraylist-2.4.9.tgz", + "integrity": "sha512-KxWjyEFzchzik3aoQlK66oaoxIReoMo5bQRm1fcjBUZvE8xv/tyR3CTKhjh6K/faV8VaF6hd5pjr45CzbwuwkA==", + "license": "Apache-2.0 OR MIT", + "dependencies": { + "uint8arrays": "^5.0.1" + } + }, + "node_modules/uint8arraylist/node_modules/uint8arrays": { + "version": "5.1.1", + "resolved": "https://registry.npmjs.org/uint8arrays/-/uint8arrays-5.1.1.tgz", + "integrity": "sha512-9muQwa4wZG4dKi9gMAIBtnk2Pw87SRpvWTH6lOGm19V2Uqxr4uomUf2PGqPnWc+qs06sN8owUU4jfcoWOcfwVQ==", + "license": "Apache-2.0 OR MIT", + "dependencies": { + "multiformats": "^13.0.0" + } + }, "node_modules/uint8arrays": { "version": "3.1.0", "resolved": "https://registry.npmjs.org/uint8arrays/-/uint8arrays-3.1.0.tgz", @@ -13163,6 +13623,12 @@ "react": "^16.8.0 || ^17.0.0 || ^18.0.0" } }, + "node_modules/varint": { + "version": "6.0.0", + "resolved": "https://registry.npmjs.org/varint/-/varint-6.0.0.tgz", + "integrity": "sha512-cXEIW6cfr15lFv563k4GuVuW/fiwjknytD37jIOLSdSWuOI6WnO/oKwmP2FQTU2l01LP8/M5TSAJpzUaGe3uWg==", + "license": "MIT" + }, "node_modules/viem": { "version": "2.47.6", "resolved": "https://registry.npmjs.org/viem/-/viem-2.47.6.tgz", diff --git a/package.json b/package.json index 96bdde53..a0cda96f 100644 --- a/package.json +++ b/package.json @@ -8,6 +8,7 @@ "workspaces": [ "bad-bits", "indexer", + "ipfs-retriever", "piece-retriever", "tail-handler", "terminator", diff --git a/piece-retriever/bin/piece-retriever.js b/piece-retriever/bin/piece-retriever.js index 060f6875..17712a82 100644 --- a/piece-retriever/bin/piece-retriever.js +++ b/piece-retriever/bin/piece-retriever.js @@ -1,10 +1,8 @@ import { - isValidEthereumAddress, httpAssert, - setContentSecurityPolicy, - getBadBitsEntry, - updateDataSetStats, - logRetrievalResult, + assertCidNotDenied, + handleFetchRequest, + selectRetrievalCandidate, } from '@filbeam/retrieval' import { parseRequest } from '../lib/request.js' @@ -23,15 +21,10 @@ export default { * @param {typeof defaultRetrieveFile} [options.retrieveFile] * @returns */ - async fetch(request, env, ctx, { retrieveFile = defaultRetrieveFile } = {}) { - request.signal.addEventListener('abort', () => { - console.log('The request was aborted!', { url: request.url }) - }) - try { - return await this._fetch(request, env, ctx, { retrieveFile }) - } catch (error) { - return this._handleError(error) - } + async fetch(request, env, ctx, options) { + return handleFetchRequest(request, env, ctx, (context) => + this._fetch(request, env, ctx, options, context), + ) }, /** @@ -40,82 +33,53 @@ export default { * @param {ExecutionContext} ctx * @param {object} options * @param {typeof defaultRetrieveFile} [options.retrieveFile] - * @returns + * @param {import('@filbeam/retrieval').RequestContext} context + * @returns {Promise} */ - async _fetch(request, env, ctx, { retrieveFile = defaultRetrieveFile } = {}) { - httpAssert( - ['GET', 'HEAD'].includes(request.method), - 405, - 'Method Not Allowed', - ) + async _fetch( + request, + env, + ctx, + { retrieveFile = defaultRetrieveFile } = {}, + context, + ) { if (URL.parse(request.url)?.pathname === '/') { return Response.redirect('https://filbeam.com/', 302) } - if (URL.parse(request.url)?.hostname.endsWith('filcdn.io')) { - return Response.redirect( - request.url.replace('filcdn.io', 'filbeam.io'), - 301, - ) - } - const requestTimestamp = new Date().toISOString() - const workerStartedAt = performance.now() - const requestCountryCode = request.headers.get('CF-IPCountry') - - const { payerWalletAddress, pieceCid, botName, validateCacheMissResponse } = + const { payerWalletAddress, pieceCid, validateCacheMissResponse } = parseRequest(request, env) - httpAssert(payerWalletAddress && pieceCid, 400, 'Missing required fields') - httpAssert( - isValidEthereumAddress(payerWalletAddress), - 400, - `Invalid address: ${payerWalletAddress}. Address must be a valid ethereum address.`, - ) - - try { + return async () => { // Timestamp to measure file retrieval performance (from cache and from SP) const fetchStartedAt = performance.now() - const [retrievalCandidates, isBadBit] = await Promise.all([ + const [retrievalCandidates] = await Promise.all([ getRetrievalCandidatesAndValidatePayer( env, payerWalletAddress, pieceCid, env.ENFORCE_EGRESS_QUOTA, ), - env.BAD_BITS_KV.get(`bad-bits:${await getBadBitsEntry(pieceCid)}`, { - type: 'json', - }), + assertCidNotDenied(env, pieceCid), ]) - httpAssert( - !isBadBit, - 404, - 'The requested CID was flagged by the Bad Bits Denylist at https://badbits.dwebops.pub', - ) - httpAssert( retrievalCandidates.length > 0, 500, 'Service provider lookup failed', ) - let retrievalCandidate - let retrievalResult - const retrievalAttempts = [] - - while (retrievalCandidates.length > 0) { - const retrievalCandidateIndex = Math.floor( - Math.random() * retrievalCandidates.length, - ) - retrievalCandidate = retrievalCandidates[retrievalCandidateIndex] - retrievalAttempts.push(retrievalCandidate) - retrievalCandidates.splice(retrievalCandidateIndex, 1) - console.log(`Attempting retrieval via ${retrievalCandidate.serviceUrl}`) - try { - retrievalResult = await retrieveFile( + const { + failureResponse, + candidate: retrievalCandidate, + result: retrievalResult, + } = await selectRetrievalCandidate( + retrievalCandidates, + (candidate) => + retrieveFile( ctx, - retrievalCandidate.serviceUrl, + candidate.serviceUrl, pieceCid, request, env.ORIGIN_CACHE_TTL, @@ -123,282 +87,42 @@ export default { signal: request.signal, addCacheMissResponseValidation: validateCacheMissResponse, }, - ) - if (retrievalResult.response.ok) { - console.log( - `Retrieval attempt succeeded (cache ${retrievalResult.cacheMiss ? 'miss' : 'hit'})`, - ) - break - } - console.log( - `Retrieval attempt failed: HTTP ${retrievalResult.response.status}`, - { - retrievalCandidate, - willRetry: retrievalCandidates.length > 0, - }, - ) - } catch (err) { - const msg = - typeof err === 'object' && err !== null && 'message' in err - ? err.message - : String(err) - console.log(`Retrieval attempt failed: ${msg}`, { - retrievalCandidate, - willRetry: retrievalCandidates.length > 0, - }) - } - } - - httpAssert(retrievalCandidate, 500, 'should never happen') - - if (!retrievalResult || retrievalResult.response.status >= 500) { - ctx.waitUntil( - logRetrievalResult(env, { - cacheMiss: retrievalResult?.cacheMiss || null, - cacheMissResponseValid: null, - responseStatus: 502, - egressBytes: 0, - requestCountryCode, - timestamp: requestTimestamp, - dataSetId: retrievalCandidate.dataSetId, - botName, - }), - ) - const response = new Response( - `No available service provider found. Attempted: ${retrievalAttempts.map((a) => `ID=${a.serviceProviderId} (Service URL=${a.serviceUrl})`).join(', ')}`, - { - status: 502, - headers: new Headers({ - 'X-Data-Set-ID': retrievalAttempts - .map((a) => a.dataSetId) - .join(','), - }), - }, - ) - setContentSecurityPolicy(response) - return response - } - - if (!retrievalResult.response.body) { - // The upstream response does not have any readable body - // There is no need to measure response body size, we can - // return the original response object. - ctx.waitUntil( - logRetrievalResult(env, { - cacheMiss: retrievalResult.cacheMiss, - cacheMissResponseValid: false, - responseStatus: retrievalResult.response.status, - egressBytes: 0, - requestCountryCode, - timestamp: requestTimestamp, - dataSetId: retrievalCandidate.dataSetId, - botName, - }), - ) - const response = new Response( - retrievalResult.response.body, - retrievalResult.response, - ) - setContentSecurityPolicy(response) - response.headers.set('X-Data-Set-ID', retrievalCandidate.dataSetId) - response.headers.set( - 'Cache-Control', - `public, max-age=${env.CLIENT_CACHE_TTL}`, - ) - return response - } - - // Stream, count bytes and validate (a cache miss) - let egressBytes = 0 - /** @type {number | null} */ - let firstByteAt = null - - /** @type {number | null} */ - let minChunkSize = null - /** @type {number | null} */ - let maxChunkSize = null - let bytesReceived = 0 - - const logStreamStats = () => { - console.log( - 'Stream stats ' + - `minChunkSize=${minChunkSize} ` + - `maxChunkSize=${maxChunkSize} ` + - `bytesReceived=${bytesReceived} ` + - `url=${request.url} ` + - `cf-ray=${request.headers.get('cf-ray')}`, - ) - minChunkSize = null - maxChunkSize = null - bytesReceived = 0 - } - - const iv = setInterval(logStreamStats, 10_000) - - const measureStream = new TransformStream({ - transform(chunk, controller) { - if (firstByteAt === null) { - console.log('First byte received') - firstByteAt = performance.now() - } - egressBytes += chunk.length - bytesReceived += chunk.length - if (minChunkSize === null || chunk.length < minChunkSize) { - minChunkSize = chunk.length - } - if (maxChunkSize === null || chunk.length > maxChunkSize) { - maxChunkSize = chunk.length - } - controller.enqueue(chunk) - }, - flush() { - logStreamStats() - clearInterval(iv) + ), + { + env, + ctx, + requestCountryCode: context.requestCountryCode, + timestamp: context.requestTimestamp, + botName: context.botName, }, - }) - - const returnedStream = new TransformStream() - - ctx.waitUntil( - (async () => { - try { - httpAssert( - retrievalResult.response.body, - 500, - 'Should never happen', - ) - await Promise.all([ - retrievalResult.response.body.pipeTo(measureStream.writable), - measureStream.readable.pipeTo(returnedStream.writable), - ]) - console.log('Response finished') - - const cacheMissResponseValid = - typeof retrievalResult.validate === 'function' - ? retrievalResult.validate() - : null - httpAssert(firstByteAt, 500, 'Should never happen') - const lastByteFetchedAt = performance.now() - - if (cacheMissResponseValid === false) { - await caches.default.delete( - getRetrievalUrl(retrievalCandidate.serviceUrl, pieceCid), - ) - } - - await logRetrievalResult(env, { - cacheMiss: retrievalResult.cacheMiss, - cacheMissResponseValid, - responseStatus: retrievalResult.response.status, - egressBytes, - requestCountryCode, - timestamp: requestTimestamp, - performanceStats: { - fetchTtfb: firstByteAt - fetchStartedAt, - fetchTtlb: lastByteFetchedAt - fetchStartedAt, - workerTtfb: firstByteAt - workerStartedAt, - }, - dataSetId: retrievalCandidate.dataSetId, - botName, - }) - - await updateDataSetStats(env, { - dataSetId: retrievalCandidate.dataSetId, - egressBytes, - cacheMiss: retrievalResult.cacheMiss, - cacheMissResponseValid, - enforceEgressQuota: env.ENFORCE_EGRESS_QUOTA, - }) - } catch (err) { - console.error('Error in server stream:', err) - logStreamStats() - clearInterval(iv) - - await logRetrievalResult(env, { - cacheMiss: retrievalResult.cacheMiss, - cacheMissResponseValid: null, - responseStatus: 900, - egressBytes, - requestCountryCode, - timestamp: requestTimestamp, - dataSetId: retrievalCandidate.dataSetId, - botName, - }) - } - })(), - ) - - // Return immediately, proxying the transformed response - const response = new Response(returnedStream.readable, { - status: retrievalResult.response.status, - statusText: retrievalResult.response.statusText, - headers: retrievalResult.response.headers, - }) - setContentSecurityPolicy(response) - response.headers.set('X-Data-Set-ID', retrievalCandidate.dataSetId) - response.headers.set( - 'Cache-Control', - `public, max-age=${env.CLIENT_CACHE_TTL}`, ) - return response - } catch (error) { - const { status } = getErrorHttpStatusMessage(error) - - ctx.waitUntil( - logRetrievalResult(env, { - cacheMiss: null, - cacheMissResponseValid: null, - responseStatus: status, - egressBytes: null, - requestCountryCode, - timestamp: requestTimestamp, - dataSetId: null, - botName, - }), + if (failureResponse) return failureResponse + httpAssert( + retrievalCandidate && retrievalResult, + 500, + 'should never happen', ) - throw error - } - }, - - /** - * @param {unknown} error - * @returns - */ - _handleError(error) { - const { status, message } = getErrorHttpStatusMessage(error) - - if (status >= 500) { - console.error(error) + return { + response: retrievalResult.response, + cacheMiss: retrievalResult.cacheMiss, + dataSetId: retrievalCandidate.dataSetId, + fetchStartedAt, + // Validate the cache-miss response (a `?validate` request) once it has + // streamed, and drop the cache entry when it fails validation. + finalizeCacheMiss: async () => { + const cacheMissResponseValid = + typeof retrievalResult.validate === 'function' + ? retrievalResult.validate() + : null + if (cacheMissResponseValid === false) { + await caches.default.delete( + getRetrievalUrl(retrievalCandidate.serviceUrl, pieceCid), + ) + } + return { cacheMissResponseValid } + }, + } } - return new Response(message, { status }) }, } - -/** - * Extracts status and message from an error object. - * - * - If the error has a numeric `status`, it is used; otherwise, defaults to 500. - * - If the status is < 500 and a string `message` exists, it's used; otherwise, a - * generic message is returned. - * - * @param {unknown} error - The error object to extract from. - * @returns {{ status: number; message: string }} - */ -function getErrorHttpStatusMessage(error) { - const isObject = typeof error === 'object' && error !== null - const status = - isObject && 'status' in error && typeof error.status === 'number' - ? error.status - : 500 - - const message = - isObject && - status < 500 && - 'message' in error && - typeof error.message === 'string' - ? error.message - : 'Internal Server Error' - - return { status, message } -} diff --git a/piece-retriever/lib/request.js b/piece-retriever/lib/request.js index b26e42db..847b1287 100644 --- a/piece-retriever/lib/request.js +++ b/piece-retriever/lib/request.js @@ -1,4 +1,4 @@ -import { httpAssert } from '@filbeam/retrieval' +import { httpAssert, isValidEthereumAddress } from '@filbeam/retrieval' /** * Parse params found in path of the request URL @@ -6,15 +6,13 @@ import { httpAssert } from '@filbeam/retrieval' * @param {Request} request * @param {object} options * @param {string} options.DNS_ROOT - * @param {string} options.BOT_TOKENS * @returns {{ - * payerWalletAddress?: string - * pieceCid?: string - * botName?: string + * payerWalletAddress: string + * pieceCid: string * validateCacheMissResponse: boolean * }} */ -export function parseRequest(request, { DNS_ROOT, BOT_TOKENS }) { +export function parseRequest(request, { DNS_ROOT }) { const url = new URL(request.url) console.log('retrieval request', { DNS_ROOT, url }) @@ -34,37 +32,14 @@ export function parseRequest(request, { DNS_ROOT, BOT_TOKENS }) { `Invalid CID: ${pieceCid}. It is not a valid CommP (v1 or v2).`, ) - const botName = checkBotAuthorization(request, { BOT_TOKENS }) - const validateCacheMissResponse = url.searchParams.has('validate') - - return { payerWalletAddress, pieceCid, botName, validateCacheMissResponse } -} - -/** - * @param {Request} request - * @param {object} args - * @param {string} args.BOT_TOKENS - * @returns {string | undefined} Bot name or the access token - */ -export function checkBotAuthorization(request, { BOT_TOKENS }) { - const botTokens = JSON.parse(BOT_TOKENS) - - const auth = request.headers.get('authorization') - if (!auth) return undefined - - const [prefix, token, ...rest] = auth.split(' ') - + httpAssert(payerWalletAddress && pieceCid, 400, 'Missing required fields') httpAssert( - prefix === 'Bearer' && token && rest.length === 0, - 401, - 'Unauthorized: Authorization header must use Bearer scheme', + isValidEthereumAddress(payerWalletAddress), + 400, + `Invalid address: ${payerWalletAddress}. Address must be a valid ethereum address.`, ) - httpAssert( - token in botTokens, - 401, - `Unauthorized: Invalid Access Token ${token.slice(0, 1)}...${token.slice(-1)}`, - ) + const validateCacheMissResponse = url.searchParams.has('validate') - return botTokens[token] + return { payerWalletAddress, pieceCid, validateCacheMissResponse } } diff --git a/piece-retriever/lib/retrieval.js b/piece-retriever/lib/retrieval.js index 5fede541..f60d37ef 100644 --- a/piece-retriever/lib/retrieval.js +++ b/piece-retriever/lib/retrieval.js @@ -1,4 +1,5 @@ import assert from 'node:assert/strict' +import { originCacheOptions } from '@filbeam/retrieval' import { createPieceCIDStream } from './piece.js' /** @@ -41,14 +42,7 @@ export async function retrieveFile( cacheMiss = false } else { response = await fetch(url, { - cf: { - cacheTtlByStatus: { - '200-299': cacheTtl, - 404: 0, - '500-599': 0, - }, - cacheEverything: true, - }, + cf: originCacheOptions(cacheTtl), signal, }) if (response.ok) { diff --git a/piece-retriever/lib/store.js b/piece-retriever/lib/store.js index 053ce329..2a76d970 100644 --- a/piece-retriever/lib/store.js +++ b/piece-retriever/lib/store.js @@ -1,4 +1,7 @@ -import { httpAssert } from '@filbeam/retrieval' +import { + filterAuthorizedRetrievalCandidates, + buildRetrievalCandidateQuery, +} from '@filbeam/retrieval' /** * Retrieves the provider and data set id for a given root CID. @@ -26,29 +29,9 @@ export async function getRetrievalCandidatesAndValidatePayer( pieceCid, enforceEgressQuota = false, ) { - const query = ` - SELECT - pieces.data_set_id, - data_sets.service_provider_id, - data_sets.payer_address, - data_sets.with_cdn, - data_set_egress_quotas.cdn_egress_quota, - data_set_egress_quotas.cache_miss_egress_quota, - service_providers.service_url, - service_providers.is_deleted as service_provider_is_deleted, - wallet_details.is_sanctioned - FROM pieces - LEFT OUTER JOIN data_sets - ON pieces.data_set_id = data_sets.id - LEFT OUTER JOIN data_set_egress_quotas - ON pieces.data_set_id = data_set_egress_quotas.data_set_id - LEFT OUTER JOIN service_providers - ON data_sets.service_provider_id = service_providers.id - LEFT OUTER JOIN wallet_details - ON data_sets.payer_address = wallet_details.address - WHERE - pieces.cid = ? AND pieces.is_deleted IS FALSE - ` + const query = buildRetrievalCandidateQuery({ + where: 'pieces.cid = ?', + }) const results = /** * @type {{ @@ -67,86 +50,12 @@ export async function getRetrievalCandidatesAndValidatePayer( (await env.DB.prepare(query).bind(pieceCid).all()).results ) ) - httpAssert( - results && results.length > 0, - 404, - `Piece_cid '${pieceCid}' does not exist or may not have been indexed yet.`, + const authorizedRetrievalCandidates = filterAuthorizedRetrievalCandidates( + results, + { payerAddress, enforceEgressQuota }, ) - const withServiceProvider = results.filter( - (row) => - row && - row.service_provider_id != null && - !row.service_provider_is_deleted, - ) - httpAssert( - withServiceProvider.length > 0, - 404, - `Piece_cid '${pieceCid}' exists but has no associated service provider.`, - ) - - const withPaymentRail = withServiceProvider.filter( - (row) => - row.payer_address && row.payer_address.toLowerCase() === payerAddress, - ) - httpAssert( - withPaymentRail.length > 0, - 402, - `There is no Filecoin Warm Storage Service deal for payer '${payerAddress}' and piece_cid '${pieceCid}'.`, - ) - - const withCDN = withPaymentRail.filter( - (row) => row.with_cdn && row.with_cdn === 1, - ) - httpAssert( - withCDN.length > 0, - 402, - `The Filecoin Warm Storage Service deal for payer '${payerAddress}' and piece_cid '${pieceCid}' has withCDN=false.`, - ) - - const withPayerNotSanctioned = withCDN.filter((row) => !row.is_sanctioned) - httpAssert( - withPayerNotSanctioned.length > 0, - 403, - `Wallet '${payerAddress}' is sanctioned and cannot retrieve piece_cid '${pieceCid}'.`, - ) - - const withApprovedProvider = withPayerNotSanctioned.filter( - (row) => row.service_url, - ) - httpAssert( - withApprovedProvider.length > 0, - 404, - `No approved service provider found for payer '${payerAddress}' and piece_cid '${pieceCid}'.`, - ) - - // Check CDN quota first - const withSufficientCDNQuota = enforceEgressQuota - ? withApprovedProvider.filter((row) => { - return BigInt(row.cdn_egress_quota ?? '0') > 0n - }) - : withApprovedProvider - - httpAssert( - withSufficientCDNQuota.length > 0, - 402, - `CDN egress quota exhausted for payer '${payerAddress}' and data set '${withApprovedProvider[0]?.data_set_id}'. Please top up your CDN egress quota.`, - ) - - // Check cache-miss quota - const withSufficientCacheMissQuota = enforceEgressQuota - ? withSufficientCDNQuota.filter((row) => { - return BigInt(row.cache_miss_egress_quota ?? '0') > 0n - }) - : withSufficientCDNQuota - - httpAssert( - withSufficientCacheMissQuota.length > 0, - 402, - `Cache miss egress quota exhausted for payer '${payerAddress}' and data set '${withSufficientCDNQuota[0]?.data_set_id}'. Please top up your cache miss egress quota.`, - ) - - const retrievalCandidates = withSufficientCacheMissQuota.map((row) => ({ + const retrievalCandidates = authorizedRetrievalCandidates.map((row) => ({ dataSetId: row.data_set_id, serviceProviderId: row.service_provider_id, // We need this cast to supress a TypeScript error. The compiler is not able to infer that diff --git a/piece-retriever/test/request.test.js b/piece-retriever/test/request.test.js index 66cfec58..0624ae00 100644 --- a/piece-retriever/test/request.test.js +++ b/piece-retriever/test/request.test.js @@ -1,8 +1,8 @@ import { describe, it, expect } from 'vitest' -import { parseRequest, checkBotAuthorization } from '../lib/request.js' +import { parseRequest } from '../lib/request.js' const DNS_ROOT = '.filbeam.io' -const TEST_WALLET = 'abc123' +const TEST_WALLET = '0x1234567890abcdef1234567890abcdef12345678' const TEST_CID = 'baga123' const BOT_TOKENS = JSON.stringify({ secret: 'bot1' }) @@ -43,6 +43,13 @@ describe('parseRequest', () => { ) }) + it('throws for an invalid payer wallet address', () => { + const request = new Request(`https://notanaddress${DNS_ROOT}/${TEST_CID}`) + expect(() => parseRequest(request, { DNS_ROOT, BOT_TOKENS })).toThrowError( + 'Invalid address: notanaddress. Address must be a valid ethereum address.', + ) + }) + it('should ignore query parameters', () => { const request = new Request( `https://${TEST_WALLET}${DNS_ROOT}/${TEST_CID}?foo=bar`, @@ -55,53 +62,3 @@ describe('parseRequest', () => { }) }) }) - -describe('checkBotAuthorization', () => { - it('should return undefined when no authorization header is present', () => { - const request = new Request('https://example.com', { - headers: {}, - }) - const result = checkBotAuthorization(request, { BOT_TOKENS }) - expect(result).toBeUndefined() - }) - - it('should throw 401 error when authorization header is not Bearer format', () => { - const request = new Request('https://example.com', { - headers: { authorization: 'Basic sometoken' }, - }) - expect(() => checkBotAuthorization(request, { BOT_TOKENS })).toThrowError( - 'Unauthorized: Authorization header must use Bearer scheme', - ) - }) - - it('should throw 401 error when authorization header has no token after Bearer', () => { - const request = new Request('https://example.com', { - headers: { authorization: 'Bearer' }, - }) - expect(() => checkBotAuthorization(request, { BOT_TOKENS })).toThrowError( - 'Unauthorized: Authorization header must use Bearer scheme', - ) - }) - - it('should throw 401 error when token is not in BOT_TOKENS list', () => { - const request = new Request('https://example.com', { - headers: { authorization: 'Bearer invalid_token' }, - }) - expect(() => checkBotAuthorization(request, { BOT_TOKENS })).toThrowError( - 'Unauthorized: Invalid Access Token i...n', - ) - }) - - it('should return token prefix when valid token is provided', () => { - const request = new Request('https://example.com', { - headers: { authorization: 'Bearer secret' }, - }) - const result = checkBotAuthorization(request, { - BOT_TOKENS: JSON.stringify({ - secret: 'bot1', - secret_2: 'bot2', - }), - }) - expect(result).toBe('bot1') - }) -}) diff --git a/piece-retriever/test/retriever.test.js b/piece-retriever/test/retriever.test.js index 5702bcef..b337282f 100644 --- a/piece-retriever/test/retriever.test.js +++ b/piece-retriever/test/retriever.test.js @@ -582,7 +582,7 @@ describe('piece-retriever.fetch', () => { // Expect an error because no URL was found expect(res.status).toBe(404) expect(await res.text()).toBe( - `No approved service provider found for payer '0x2a06d234246ed18b6c91de8349ff34c22c7268e8' and piece_cid 'bagaTest'.`, + `No approved service provider found for payer '0x2a06d234246ed18b6c91de8349ff34c22c7268e8' and the requested content.`, ) }) diff --git a/retrieval/index.js b/retrieval/index.js index 26448ecf..f2014af2 100644 --- a/retrieval/index.js +++ b/retrieval/index.js @@ -1,7 +1,13 @@ +export * from './lib/access.js' export * from './lib/address.js' export * from './lib/bad-bits-util.js' +export * from './lib/bot-auth.js' +export * from './lib/candidate-selection.js' export * from './lib/content-security-policy.js' +export * from './lib/fetch-handler.js' export * from './lib/http-assert.js' +export * from './lib/http-error.js' +export * from './lib/origin-cache.js' export * from './lib/stats.js' export default { diff --git a/retrieval/lib/access.js b/retrieval/lib/access.js new file mode 100644 index 00000000..1afe5157 --- /dev/null +++ b/retrieval/lib/access.js @@ -0,0 +1,161 @@ +import { httpAssert } from './http-assert.js' + +/** + * The columns the authorization cascade reads from a candidate row. Callers may + * pass rows with additional columns, which are preserved in the return value. + * Quotas are stored as integers but D1 may surface them as strings, so both are + * accepted. + * + * @typedef {object} RetrievalCandidateRow + * @property {string | null} [service_provider_id] + * @property {boolean | null} [service_provider_is_deleted] + * @property {string | null} [payer_address] + * @property {number | null} [with_cdn] + * @property {number | boolean | null} [is_sanctioned] + * @property {string | null} [service_url] + * @property {string | number | null} [cdn_egress_quota] + * @property {string | number | null} [cache_miss_egress_quota] + */ + +/** + * Runs the shared retrieval authorization cascade over candidate rows joined + * from pieces, data_sets, service_providers and wallet_details. Each check + * filters the rows and throws an httpAssert error when no row survives. Returns + * the rows that pass every check. + * + * The checks run in order: indexed, has a (non-deleted) service provider, has a + * payment rail for the payer, has CDN enabled, payer is not sanctioned, the + * service provider is approved, and (when `enforceEgressQuota` is set) the data + * set has CDN and cache-miss egress quota remaining. + * + * @template {RetrievalCandidateRow} Row + * @param {Row[]} rows + * @param {object} options + * @param {string} options.payerAddress - Lower-cased payer address to match. + * @param {boolean} [options.enforceEgressQuota] - Also require remaining CDN + * and cache-miss egress quota. + * @returns {Row[]} The rows passing every check. + */ +export function filterAuthorizedRetrievalCandidates( + rows, + { payerAddress, enforceEgressQuota = false }, +) { + httpAssert( + rows && rows.length > 0, + 404, + 'The requested content does not exist or may not have been indexed yet.', + ) + + const withServiceProvider = rows.filter( + (row) => + row && + row.service_provider_id != null && + !row.service_provider_is_deleted, + ) + httpAssert( + withServiceProvider.length > 0, + 404, + 'The requested content exists but has no associated service provider.', + ) + + const withPaymentRail = withServiceProvider.filter( + (row) => + row.payer_address && row.payer_address.toLowerCase() === payerAddress, + ) + httpAssert( + withPaymentRail.length > 0, + 402, + `There is no Filecoin Warm Storage Service deal for payer '${payerAddress}' and the requested content.`, + ) + + const withCDN = withPaymentRail.filter( + (row) => row.with_cdn && row.with_cdn === 1, + ) + httpAssert( + withCDN.length > 0, + 402, + `The Filecoin Warm Storage Service deal for payer '${payerAddress}' and the requested content has withCDN=false.`, + ) + + const withPayerNotSanctioned = withCDN.filter((row) => !row.is_sanctioned) + httpAssert( + withPayerNotSanctioned.length > 0, + 403, + `Wallet '${payerAddress}' is sanctioned and cannot retrieve the requested content.`, + ) + + const authorizedRetrievalCandidates = withPayerNotSanctioned.filter( + (row) => row.service_url, + ) + httpAssert( + authorizedRetrievalCandidates.length > 0, + 404, + `No approved service provider found for payer '${payerAddress}' and the requested content.`, + ) + + if (!enforceEgressQuota) return authorizedRetrievalCandidates + + const withSufficientCDNQuota = authorizedRetrievalCandidates.filter( + (row) => BigInt(row.cdn_egress_quota ?? '0') > 0n, + ) + httpAssert( + withSufficientCDNQuota.length > 0, + 402, + `CDN egress quota exhausted for payer '${payerAddress}' and the requested content. Please top up your CDN egress quota.`, + ) + + const withSufficientCacheMissQuota = withSufficientCDNQuota.filter( + (row) => BigInt(row.cache_miss_egress_quota ?? '0') > 0n, + ) + httpAssert( + withSufficientCacheMissQuota.length > 0, + 402, + `Cache miss egress quota exhausted for payer '${payerAddress}' and the requested content. Please top up your cache miss egress quota.`, + ) + + return withSufficientCacheMissQuota +} + +/** + * Builds the SELECT that joins pieces, data_sets, data_set_egress_quotas, + * service_providers and wallet_details and returns the columns the + * authorization cascade in {@link filterAuthorizedRetrievalCandidates} reads. + * Callers supply the lookup-specific `where` clause and any extra columns. + * + * Deleted pieces are always excluded. + * + * @param {object} options + * @param {string[]} [options.extraColumns] - Extra columns to select, in + * addition to the ones the cascade reads. + * @param {string} options.where - The lookup condition, without the `WHERE` + * keyword. It is combined with a filter that excludes deleted pieces. + * @returns {string} + */ +export function buildRetrievalCandidateQuery({ extraColumns = [], where }) { + const columns = [ + 'pieces.data_set_id', + 'data_sets.service_provider_id', + 'data_sets.payer_address', + 'data_sets.with_cdn', + 'data_set_egress_quotas.cdn_egress_quota', + 'data_set_egress_quotas.cache_miss_egress_quota', + 'service_providers.service_url', + 'service_providers.is_deleted AS service_provider_is_deleted', + 'wallet_details.is_sanctioned', + ...extraColumns, + ] + + return ` + SELECT ${columns.join(', ')} + FROM pieces + LEFT OUTER JOIN data_sets + ON pieces.data_set_id = data_sets.id + LEFT OUTER JOIN data_set_egress_quotas + ON pieces.data_set_id = data_set_egress_quotas.data_set_id + LEFT OUTER JOIN service_providers + ON data_sets.service_provider_id = service_providers.id + LEFT OUTER JOIN wallet_details + ON data_sets.payer_address = wallet_details.address + WHERE (${where}) AND pieces.is_deleted IS FALSE + ` +} diff --git a/retrieval/lib/bad-bits-util.js b/retrieval/lib/bad-bits-util.js index 7801ae29..7e76d52f 100644 --- a/retrieval/lib/bad-bits-util.js +++ b/retrieval/lib/bad-bits-util.js @@ -1,3 +1,5 @@ +import { httpAssert } from './http-assert.js' + /** * @param {string} cid * @returns {Promise} Bad Bits entry in the legacy double-hash format @@ -10,3 +12,33 @@ export async function getBadBitsEntry(cid) { .join('') return hashHex } + +/** + * Looks up whether a CID is on the Bad Bits denylist stored in KV. + * + * @param {{ BAD_BITS_KV: KVNamespace }} env + * @param {string} cid + * @returns {Promise} + */ +export async function isCidDenied(env, cid) { + const entry = await env.BAD_BITS_KV.get( + `bad-bits:${await getBadBitsEntry(cid)}`, + { type: 'json' }, + ) + return Boolean(entry) +} + +/** + * Throws a `404` when the CID is on the Bad Bits denylist. + * + * @param {{ BAD_BITS_KV: KVNamespace }} env + * @param {string} cid + * @returns {Promise} + */ +export async function assertCidNotDenied(env, cid) { + httpAssert( + !(await isCidDenied(env, cid)), + 404, + 'The requested CID was flagged by the Bad Bits Denylist at https://badbits.dwebops.pub', + ) +} diff --git a/retrieval/lib/bot-auth.js b/retrieval/lib/bot-auth.js new file mode 100644 index 00000000..d132612e --- /dev/null +++ b/retrieval/lib/bot-auth.js @@ -0,0 +1,36 @@ +import { httpAssert } from './http-assert.js' + +/** + * Resolves the bot name for a request's Bearer token, or `undefined` for + * anonymous requests. Throws a 401 for a malformed Authorization header or an + * unknown token. + * + * @param {Request} request + * @param {object} args + * @param {string} args.BOT_TOKENS - JSON object mapping access token to bot + * name. An empty or unset value is treated as an empty mapping. + * @returns {string | undefined} The resolved bot name, or `undefined` for + * anonymous requests. + */ +export function checkBotAuthorization(request, { BOT_TOKENS }) { + const auth = request.headers.get('authorization') + if (!auth) return undefined + + const botTokens = BOT_TOKENS ? JSON.parse(BOT_TOKENS) : {} + + const [prefix, token, ...rest] = auth.split(' ') + + httpAssert( + prefix === 'Bearer' && token && rest.length === 0, + 401, + 'Unauthorized: Authorization header must use Bearer scheme', + ) + + httpAssert( + token in botTokens, + 401, + `Unauthorized: Invalid Access Token ${token.slice(0, 1)}...${token.slice(-1)}`, + ) + + return botTokens[token] +} diff --git a/retrieval/lib/candidate-selection.js b/retrieval/lib/candidate-selection.js new file mode 100644 index 00000000..5979b2cc --- /dev/null +++ b/retrieval/lib/candidate-selection.js @@ -0,0 +1,116 @@ +import { logRetrievalResult } from './stats.js' +import { setContentSecurityPolicy } from './content-security-policy.js' +import { httpAssert } from './http-assert.js' + +/** @typedef {{ response: Response; cacheMiss: boolean }} RetrievalResult */ + +/** + * Attempt retrieval from the given candidates in random order until one returns + * a usable response (status `< 500`). Candidates whose retrieval throws or + * returns a `5xx` response are skipped. + * + * On success, returns the selected candidate and its result. When every + * candidate fails (throws or returns a `5xx`), logs the failure and returns the + * `502` "No available service provider found" response instead. + * + * The input array is not mutated. + * + * @template {{ + * serviceUrl: string + * serviceProviderId: string + * dataSetId: string + * }} Candidate + * @template {RetrievalResult} Result + * @param {Candidate[]} candidates - The candidates to attempt, in any order. + * @param {(candidate: Candidate) => Promise} attemptRetrieval - + * Performs the retrieval for a single candidate. + * @param {object} options - Context for logging a failed retrieval. + * @param {{ DB: D1Database }} options.env + * @param {ExecutionContext} options.ctx + * @param {string | null} options.requestCountryCode + * @param {string} options.timestamp + * @param {string | undefined} options.botName + * @returns {Promise< + * | { candidate: Candidate; result: Result; failureResponse?: undefined } + * | { candidate?: undefined; result?: undefined; failureResponse: Response } + * >} + * - On success, `candidate` and its `result`. + * - On failure, the `502` `failureResponse`. + */ +export async function selectRetrievalCandidate( + candidates, + attemptRetrieval, + { env, ctx, requestCountryCode, timestamp, botName }, +) { + const remaining = [...candidates] + /** @type {Candidate | undefined} */ + let candidate + /** @type {Result | undefined} */ + let result + /** @type {Candidate[]} */ + const attempts = [] + + while (remaining.length > 0) { + const index = Math.floor(Math.random() * remaining.length) + candidate = remaining[index] + attempts.push(candidate) + remaining.splice(index, 1) + console.log(`Attempting retrieval via ${candidate.serviceUrl}`) + try { + result = await attemptRetrieval(candidate) + if (result.response.ok) { + console.log( + `Retrieval attempt succeeded (cache ${result.cacheMiss ? 'miss' : 'hit'})`, + ) + break + } + console.log(`Retrieval attempt failed: HTTP ${result.response.status}`, { + candidate, + willRetry: remaining.length > 0, + }) + } catch (err) { + const msg = + typeof err === 'object' && err !== null && 'message' in err + ? err.message + : String(err) + console.log(`Retrieval attempt failed: ${msg}`, { + candidate, + willRetry: remaining.length > 0, + }) + } + } + + httpAssert(candidate, 500, 'should never happen') + + if (result && result.response.status < 500) { + return { candidate, result } + } + + ctx.waitUntil( + logRetrievalResult(env, { + cacheMiss: result?.cacheMiss ?? null, + cacheMissResponseValid: null, + responseStatus: 502, + egressBytes: 0, + cacheMissEgressBytes: 0, + requestCountryCode, + timestamp, + dataSetId: candidate.dataSetId, + botName, + }), + ) + + const failureResponse = new Response( + `No available service provider found. Attempted: ${attempts + .map((a) => `ID=${a.serviceProviderId} (Service URL=${a.serviceUrl})`) + .join(', ')}`, + { + status: 502, + headers: new Headers({ + 'X-Data-Set-ID': attempts.map((a) => a.dataSetId).join(','), + }), + }, + ) + setContentSecurityPolicy(failureResponse) + return { failureResponse } +} diff --git a/retrieval/lib/fetch-handler.js b/retrieval/lib/fetch-handler.js new file mode 100644 index 00000000..575f004a --- /dev/null +++ b/retrieval/lib/fetch-handler.js @@ -0,0 +1,344 @@ +import { handleError, getErrorHttpStatusMessage } from './http-error.js' +import { httpAssert } from './http-assert.js' +import { checkBotAuthorization } from './bot-auth.js' +import { setContentSecurityPolicy } from './content-security-policy.js' +import { recordRetrieval, logRetrievalResult } from './stats.js' + +/** + * The successful retrieval outcome a worker hands back to + * {@link handleFetchRequest}: the response to serve plus the metadata needed to + * measure egress and log the retrieval. + * + * @typedef {object} RetrievalOutcome + * @property {Response} response - The response to serve. Its body is streamed + * to the client and measured; a `null` body is logged as a zero-egress result + * and returned unchanged. + * @property {boolean} cacheMiss + * @property {string} dataSetId + * @property {number} fetchStartedAt + * @property {(egressBytes: number) => Promise<{ + * cacheMissEgressBytes?: number + * cacheMissResponseValid: boolean | null + * }>} finalizeCacheMiss + * - Computes the worker-specific cache-miss accounting once the bytes served to + * the client are known. Runs after the response has streamed, before the + * retrieval is logged. + */ + +/** + * A worker's retrieval step: looks up and retrieves the content. Returns the + * {@link RetrievalOutcome} to serve, or a {@link Response} when no service + * provider could serve the content. An error thrown here is logged as a + * retrieval error. + * + * @typedef {() => Promise} Retrieve + */ + +/** + * Per-request telemetry shared by the success and error logging paths. + * + * @typedef {object} RequestContext + * @property {string} requestTimestamp - ISO timestamp of the request. + * @property {string | null} requestCountryCode - The request's `CF-IPCountry`. + * @property {number} workerStartedAt - `performance.now()` when the worker + * started handling the request. + * @property {string} [botName] - The bot name resolved from the request's + * Authorization header, or `undefined` for anonymous requests. + */ + +/** + * Runs a worker's retrieval handler with the shared request lifecycle: log when + * the request is aborted, reject non-GET/HEAD methods with a `405`, redirect + * legacy `*.filcdn.io` requests to `*.filbeam.io`, and turn thrown errors into + * HTTP responses via {@link handleError}. + * + * The handler returns either a plain {@link Response} (redirects), served as-is, + * or a {@link Retrieve} function. The retrieval is run inside a try/catch that + * logs a retrieval error on failure; its result is then served: a + * {@link Response} (the no-service-provider response) as-is, or a + * {@link RetrievalOutcome} whose body is streamed while egress is measured and + * the retrieval is logged. + * + * @param {Request} request + * @param {{ + * DB: D1Database + * CLIENT_CACHE_TTL: number + * ENFORCE_EGRESS_QUOTA: boolean + * BOT_TOKENS: string + * }} env + * @param {ExecutionContext} ctx + * @param {(context: RequestContext) => Promise} run + * + * - Invokes the worker handler with the per-request telemetry context. + * + * @returns {Promise} + */ +export async function handleFetchRequest(request, env, ctx, run) { + request.signal.addEventListener('abort', () => { + console.log('The request was aborted!', { url: request.url }) + }) + + /** @type {RequestContext} */ + const context = { + requestTimestamp: new Date().toISOString(), + requestCountryCode: request.headers.get('CF-IPCountry'), + workerStartedAt: performance.now(), + } + + try { + httpAssert( + ['GET', 'HEAD'].includes(request.method), + 405, + 'Method Not Allowed', + ) + const legacyRedirect = redirectLegacyDomain(request) + if (legacyRedirect) return legacyRedirect + + context.botName = checkBotAuthorization(request, { + BOT_TOKENS: env.BOT_TOKENS, + }) + + const retrieve = await run(context) + if (retrieve instanceof Response) return retrieve + + let outcome + try { + outcome = await retrieve() + } catch (error) { + logRetrievalError(env, ctx, error, { + requestCountryCode: context.requestCountryCode, + timestamp: context.requestTimestamp, + botName: context.botName, + }) + throw error + } + + if (outcome instanceof Response) return outcome + return serveRetrievalOutcome(env, ctx, outcome, context) + } catch (error) { + return handleError(error) + } +} + +/** + * Streams a retrieval response to the client while measuring egress and logging + * the result on the execution context. Returns the response immediately. + * + * @param {{ + * DB: D1Database + * CLIENT_CACHE_TTL: number + * ENFORCE_EGRESS_QUOTA: boolean + * }} env + * @param {ExecutionContext} ctx + * @param {RetrievalOutcome} result + * @param {RequestContext} context + * @returns {Response} + */ +function serveRetrievalOutcome( + env, + ctx, + result, + { requestCountryCode, requestTimestamp: timestamp, workerStartedAt, botName }, +) { + const { response, cacheMiss, dataSetId, fetchStartedAt, finalizeCacheMiss } = + result + + // No readable body (e.g. a HEAD request or an error status): nothing to + // stream or measure. + if (!response.body) { + return handleEmptyBodyResponse(env, ctx, { + response, + cacheMiss, + dataSetId, + requestCountryCode, + timestamp, + botName, + }) + } + + // Measure egress by piping the body through a counting transform on its way + // to the client. This preserves backpressure: the origin is pulled only as + // fast as the client reads, so a slow client cannot make the worker buffer + // the whole response in memory. + const responseBody = response.body + let egressBytes = 0 + /** @type {number | null} */ + let firstByteAt = null + const measureStream = new TransformStream({ + transform(chunk, controller) { + if (firstByteAt === null) firstByteAt = performance.now() + egressBytes += chunk.length + controller.enqueue(chunk) + }, + }) + const returnedStream = new TransformStream() + + ctx.waitUntil( + (async () => { + try { + await Promise.all([ + responseBody.pipeTo(measureStream.writable), + measureStream.readable.pipeTo(returnedStream.writable), + ]) + const lastByteFetchedAt = performance.now() + const startedAt = firstByteAt ?? lastByteFetchedAt + + const { cacheMissEgressBytes, cacheMissResponseValid } = + await finalizeCacheMiss(egressBytes) + + await recordRetrieval(env, { + cacheMiss, + cacheMissResponseValid, + cacheMissEgressBytes, + responseStatus: response.status, + egressBytes, + requestCountryCode, + timestamp, + performanceStats: { + fetchTtfb: startedAt - fetchStartedAt, + fetchTtlb: lastByteFetchedAt - fetchStartedAt, + workerTtfb: startedAt - workerStartedAt, + }, + dataSetId, + botName, + enforceEgressQuota: env.ENFORCE_EGRESS_QUOTA, + }) + } catch (err) { + console.error('Error in server stream:', err) + + await logRetrievalResult(env, { + cacheMiss, + cacheMissResponseValid: null, + responseStatus: 900, + egressBytes, + requestCountryCode, + timestamp, + dataSetId, + botName, + }) + } + })(), + ) + + const proxied = new Response(returnedStream.readable, { + status: response.status, + statusText: response.statusText, + headers: response.headers, + }) + setRetrievalResponseHeaders(proxied, { + dataSetId, + clientCacheTtl: env.CLIENT_CACHE_TTL, + }) + return proxied +} + +/** + * Logs a zero-egress retrieval result and returns the upstream response + * unchanged. Used when the upstream response carries no readable body (e.g. a + * non-OK status or a `HEAD` request), so there is nothing to stream or + * measure. + * + * @param {{ DB: D1Database; CLIENT_CACHE_TTL: number }} env + * @param {ExecutionContext} ctx + * @param {object} params + * @param {Response} params.response - The upstream response to return as-is. + * @param {boolean} params.cacheMiss + * @param {string} params.dataSetId + * @param {string | null} params.requestCountryCode + * @param {string} params.timestamp + * @param {string | undefined} params.botName + * @returns {Response} + */ +function handleEmptyBodyResponse( + env, + ctx, + { response, cacheMiss, dataSetId, requestCountryCode, timestamp, botName }, +) { + ctx.waitUntil( + logRetrievalResult(env, { + cacheMiss, + cacheMissResponseValid: null, + responseStatus: response.status, + egressBytes: 0, + cacheMissEgressBytes: 0, + requestCountryCode, + timestamp, + dataSetId, + botName, + }), + ) + + const emptyResponse = new Response(response.body, response) + setRetrievalResponseHeaders(emptyResponse, { + dataSetId, + clientCacheTtl: env.CLIENT_CACHE_TTL, + }) + return emptyResponse +} + +/** + * Redirects legacy `*.filcdn.io` requests to the equivalent `*.filbeam.io` URL + * with a 301. + * + * @param {Request} request + * @returns {Response | undefined} A redirect response, or `undefined` when the + * request is not for a legacy domain. + */ +function redirectLegacyDomain(request) { + if (URL.parse(request.url)?.hostname.endsWith('filcdn.io')) { + return Response.redirect( + request.url.replace('filcdn.io', 'filbeam.io'), + 301, + ) + } +} + +/** + * Applies the standard headers for a successful retrieval response: the content + * security policy, the data set id, and the client cache policy. + * + * @param {Response} response + * @param {object} options + * @param {string} options.dataSetId + * @param {number} options.clientCacheTtl - `Cache-Control` max-age in seconds. + */ +function setRetrievalResponseHeaders(response, { dataSetId, clientCacheTtl }) { + setContentSecurityPolicy(response) + response.headers.set('X-Data-Set-ID', dataSetId) + response.headers.set('Cache-Control', `public, max-age=${clientCacheTtl}`) +} + +/** + * Records a failed retrieval: logs the resolved HTTP status with no egress and + * no data set, scheduled on the execution context. Intended for a worker's + * request error handler. + * + * @param {{ DB: D1Database }} env - Worker environment (contains D1 binding). + * @param {ExecutionContext} ctx + * @param {unknown} error - The error thrown while handling the request. + * @param {object} context + * @param {string | null} context.requestCountryCode + * @param {string} context.timestamp + * @param {string | undefined} context.botName + */ +function logRetrievalError( + env, + ctx, + error, + { requestCountryCode, timestamp, botName }, +) { + const { status } = getErrorHttpStatusMessage(error) + + ctx.waitUntil( + logRetrievalResult(env, { + cacheMiss: null, + cacheMissResponseValid: null, + responseStatus: status, + egressBytes: null, + requestCountryCode, + timestamp, + dataSetId: null, + botName, + }), + ) +} diff --git a/retrieval/lib/http-error.js b/retrieval/lib/http-error.js new file mode 100644 index 00000000..28c7efa7 --- /dev/null +++ b/retrieval/lib/http-error.js @@ -0,0 +1,43 @@ +/** + * Extracts status and message from an error object. + * + * - If the error has a numeric `status`, it is used; otherwise, defaults to 500. + * - If the status is < 500 and a string `message` exists, it's used; otherwise, a + * generic message is returned. + * + * @param {unknown} error - The error object to extract from. + * @returns {{ status: number; message: string }} + */ +export function getErrorHttpStatusMessage(error) { + const isObject = typeof error === 'object' && error !== null + const status = + isObject && 'status' in error && typeof error.status === 'number' + ? error.status + : 500 + + const message = + isObject && + status < 500 && + 'message' in error && + typeof error.message === 'string' + ? error.message + : 'Internal Server Error' + + return { status, message } +} + +/** + * Builds a Response for an error thrown while handling a request, logging + * server errors (status >= 500) to the console. + * + * @param {unknown} error - The error to turn into a response. + * @returns {Response} + */ +export function handleError(error) { + const { status, message } = getErrorHttpStatusMessage(error) + + if (status >= 500) { + console.error(error) + } + return new Response(message, { status }) +} diff --git a/retrieval/lib/origin-cache.js b/retrieval/lib/origin-cache.js new file mode 100644 index 00000000..72a5c093 --- /dev/null +++ b/retrieval/lib/origin-cache.js @@ -0,0 +1,13 @@ +/** + * Cloudflare `fetch` cache options for retrieving content from a service + * provider: cache successful responses for `cacheTtl` seconds and never cache + * 404 or 5xx responses. + * + * @param {number} cacheTtl - Cache TTL in seconds for 2xx responses. + */ +export function originCacheOptions(cacheTtl) { + return { + cacheTtlByStatus: { '200-299': cacheTtl, 404: 0, '500-599': 0 }, + cacheEverything: true, + } +} diff --git a/retrieval/lib/stats.js b/retrieval/lib/stats.js index e6e46bc8..847ba2b1 100644 --- a/retrieval/lib/stats.js +++ b/retrieval/lib/stats.js @@ -2,10 +2,17 @@ * @param {{ DB: D1Database }} env - Worker environment (contains D1 binding). * @param {object} params - Parameters for the data set update. * @param {string} params.dataSetId - The ID of the data set to update. - * @param {number} params.egressBytes - The egress bytes used for the response. + * @param {number} params.egressBytes - The egress bytes sent to the client. + * This is what the CDN egress quota is charged for. + * @param {number} [params.cacheMissEgressBytes] - The egress bytes fetched from + * the service provider on a cache miss. This is what the cache-miss egress + * quota is charged for. Defaults to `egressBytes`, which is correct whenever + * the bytes served to the client equal the bytes fetched from the origin + * (e.g. raw piece retrievals). For IPFS retrievals the origin response is a + * CAR that is larger than the raw bytes served to the client. * @param {boolean} params.cacheMiss - Whether this was a cache miss (true) or * cache hit (false). - * @param {boolean} [params.cacheMissResponseValid] + * @param {boolean | null} [params.cacheMissResponseValid] * @param {boolean} [params.enforceEgressQuota=false] - Whether to decrement * egress quotas. Default is `false` * @param {boolean} [params.isBotTraffic=false] - Whether the egress traffic @@ -16,6 +23,7 @@ export async function updateDataSetStats( { dataSetId, egressBytes, + cacheMissEgressBytes = egressBytes, cacheMiss, cacheMissResponseValid, enforceEgressQuota = false, @@ -42,7 +50,7 @@ export async function updateDataSetStats( ) .bind( egressBytes, - cacheMiss && cacheMissResponseValid ? egressBytes : 0, + cacheMiss && cacheMissResponseValid ? cacheMissEgressBytes : 0, dataSetId, ) .run() @@ -54,7 +62,13 @@ export async function updateDataSetStats( * * @param {{ DB: D1Database }} env - Worker environment (contains D1 binding). * @param {object} params - Parameters for the retrieval log. - * @param {number | null} params.egressBytes - The egress bytes of the response. + * @param {number | null} params.egressBytes - The egress bytes sent to the + * client. + * @param {number | null} [params.cacheMissEgressBytes] - The egress bytes + * fetched from the service provider on a cache miss. Defaults to + * `egressBytes`, which is correct whenever the bytes served to the client + * equal the bytes fetched from the origin (e.g. raw piece retrievals). For + * IPFS retrievals the origin CAR is larger than the raw bytes served. * @param {number} params.responseStatus - The HTTP response status code. * @param {boolean | null} params.cacheMiss - Whether the retrieval was a cache * miss. @@ -80,6 +94,7 @@ export async function logRetrievalResult(env, params) { cacheMiss, cacheMissResponseValid, egressBytes, + cacheMissEgressBytes = egressBytes, responseStatus, timestamp, performanceStats, @@ -95,6 +110,7 @@ export async function logRetrievalResult(env, params) { timestamp, response_status, egress_bytes, + cache_miss_egress_bytes, cache_miss, cache_miss_response_valid, fetch_ttfb, @@ -104,13 +120,14 @@ export async function logRetrievalResult(env, params) { data_set_id, bot_name ) - VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?) + VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?) `, ) .bind( timestamp, responseStatus, egressBytes, + cacheMissEgressBytes, cacheMiss, cacheMissResponseValid, performanceStats?.fetchTtfb ?? null, @@ -127,3 +144,39 @@ export async function logRetrievalResult(env, params) { throw error } } + +/** + * Records a completed retrieval: writes the retrieval log and updates the data + * set egress stats and quotas. These are always performed together for a + * successful streamed response. + * + * @param {{ DB: D1Database }} env - Worker environment (contains D1 binding). + * @param {object} params - Combined parameters for {@link logRetrievalResult} + * and {@link updateDataSetStats}. + * @param {string} params.dataSetId + * @param {number} params.egressBytes + * @param {number} [params.cacheMissEgressBytes] + * @param {boolean} params.cacheMiss + * @param {boolean | null} params.cacheMissResponseValid + * @param {number} params.responseStatus + * @param {string | null} params.requestCountryCode + * @param {string} params.timestamp + * @param {{ + * fetchTtfb: number + * fetchTtlb: number + * workerTtfb: number + * }} [params.performanceStats] + * @param {string | undefined} params.botName + * @param {boolean} [params.enforceEgressQuota] + */ +export async function recordRetrieval(env, params) { + await logRetrievalResult(env, params) + await updateDataSetStats(env, { + dataSetId: params.dataSetId, + egressBytes: params.egressBytes, + cacheMissEgressBytes: params.cacheMissEgressBytes, + cacheMiss: params.cacheMiss, + cacheMissResponseValid: params.cacheMissResponseValid, + enforceEgressQuota: params.enforceEgressQuota, + }) +} diff --git a/retrieval/test/access.test.js b/retrieval/test/access.test.js new file mode 100644 index 00000000..c424fa27 --- /dev/null +++ b/retrieval/test/access.test.js @@ -0,0 +1,214 @@ +import { describe, it, expect } from 'vitest' +import { + filterAuthorizedRetrievalCandidates, + buildRetrievalCandidateQuery, +} from '../lib/access.js' + +const payerAddress = '0xabcdef' + +/** A row that passes every check. payer_address is upper-cased on purpose. */ +function authorizedRow(overrides = {}) { + return { + service_provider_id: 'sp1', + service_provider_is_deleted: 0, + payer_address: '0xABCDEF', + with_cdn: 1, + is_sanctioned: 0, + service_url: 'https://sp.example/', + cdn_egress_quota: '100', + cache_miss_egress_quota: '100', + ...overrides, + } +} + +/** @param {() => unknown} fn */ +function expectHttpError(fn, status, message) { + let error + try { + fn() + } catch (err) { + error = err + } + expect(error).toBeInstanceOf(Error) + expect(error.status).toBe(status) + expect(error.message).toBe(message) +} + +describe('filterAuthorizedRetrievalCandidates', () => { + it('returns the rows passing every check, matching the payer case-insensitively', () => { + const rows = [authorizedRow()] + expect(filterAuthorizedRetrievalCandidates(rows, { payerAddress })).toEqual( + rows, + ) + }) + + it('throws 404 when there are no rows', () => { + expectHttpError( + () => filterAuthorizedRetrievalCandidates([], { payerAddress }), + 404, + 'The requested content does not exist or may not have been indexed yet.', + ) + }) + + it('throws 404 when no row has a service provider', () => { + expectHttpError( + () => + filterAuthorizedRetrievalCandidates( + [authorizedRow({ service_provider_id: null })], + { payerAddress }, + ), + 404, + 'The requested content exists but has no associated service provider.', + ) + }) + + it('always excludes soft-deleted service providers', () => { + expectHttpError( + () => + filterAuthorizedRetrievalCandidates( + [authorizedRow({ service_provider_is_deleted: 1 })], + { payerAddress }, + ), + 404, + 'The requested content exists but has no associated service provider.', + ) + }) + + it('throws 402 when the payer has no payment rail', () => { + expectHttpError( + () => + filterAuthorizedRetrievalCandidates( + [authorizedRow({ payer_address: '0xother' })], + { payerAddress }, + ), + 402, + `There is no Filecoin Warm Storage Service deal for payer '${payerAddress}' and the requested content.`, + ) + }) + + it('throws 402 when CDN is disabled', () => { + expectHttpError( + () => + filterAuthorizedRetrievalCandidates([authorizedRow({ with_cdn: 0 })], { + payerAddress, + }), + 402, + `The Filecoin Warm Storage Service deal for payer '${payerAddress}' and the requested content has withCDN=false.`, + ) + }) + + it('throws 403 when the payer is sanctioned', () => { + expectHttpError( + () => + filterAuthorizedRetrievalCandidates( + [authorizedRow({ is_sanctioned: 1 })], + { payerAddress }, + ), + 403, + `Wallet '${payerAddress}' is sanctioned and cannot retrieve the requested content.`, + ) + }) + + it('throws 404 when no service provider is approved (no service_url)', () => { + expectHttpError( + () => + filterAuthorizedRetrievalCandidates( + [authorizedRow({ service_url: null })], + { payerAddress }, + ), + 404, + `No approved service provider found for payer '${payerAddress}' and the requested content.`, + ) + }) + + it('does not check egress quota when enforceEgressQuota is false', () => { + const rows = [ + authorizedRow({ cdn_egress_quota: '0', cache_miss_egress_quota: '0' }), + ] + expect(filterAuthorizedRetrievalCandidates(rows, { payerAddress })).toEqual( + rows, + ) + }) + + it('returns the rows when enforceEgressQuota is set and both quotas have budget', () => { + const rows = [authorizedRow()] + expect( + filterAuthorizedRetrievalCandidates(rows, { + payerAddress, + enforceEgressQuota: true, + }), + ).toEqual(rows) + }) + + it('throws 402 when enforcing and the CDN egress quota is exhausted', () => { + expectHttpError( + () => + filterAuthorizedRetrievalCandidates( + [authorizedRow({ cdn_egress_quota: '0' })], + { payerAddress, enforceEgressQuota: true }, + ), + 402, + `CDN egress quota exhausted for payer '${payerAddress}' and the requested content. Please top up your CDN egress quota.`, + ) + }) + + it('throws 402 when enforcing and the cache-miss egress quota is exhausted', () => { + expectHttpError( + () => + filterAuthorizedRetrievalCandidates( + [authorizedRow({ cache_miss_egress_quota: '0' })], + { payerAddress, enforceEgressQuota: true }, + ), + 402, + `Cache miss egress quota exhausted for payer '${payerAddress}' and the requested content. Please top up your cache miss egress quota.`, + ) + }) +}) + +describe('buildRetrievalCandidateQuery', () => { + it('selects the cascade columns, the joins, and the given where clause', () => { + const query = buildRetrievalCandidateQuery({ + where: 'pieces.cid = ?', + }) + + for (const column of [ + 'pieces.data_set_id', + 'data_sets.service_provider_id', + 'data_sets.payer_address', + 'data_sets.with_cdn', + 'data_set_egress_quotas.cdn_egress_quota', + 'data_set_egress_quotas.cache_miss_egress_quota', + 'service_providers.service_url', + 'service_providers.is_deleted AS service_provider_is_deleted', + 'wallet_details.is_sanctioned', + ]) { + expect(query).toContain(column) + } + expect(query).toContain('FROM pieces') + expect(query).toContain( + 'LEFT OUTER JOIN data_sets\n ON pieces.data_set_id = data_sets.id', + ) + expect(query).toContain('WHERE (pieces.cid = ?)') + }) + + it('always excludes deleted pieces', () => { + const query = buildRetrievalCandidateQuery({ + where: 'pieces.cid = ?', + }) + + expect(query).toContain( + 'WHERE (pieces.cid = ?) AND pieces.is_deleted IS FALSE', + ) + }) + + it('appends the extra columns', () => { + const query = buildRetrievalCandidateQuery({ + extraColumns: ['pieces.id AS piece_id', 'pieces.ipfs_root_cid'], + where: 'pieces.ipfs_root_cid = ?', + }) + + expect(query).toContain('pieces.id AS piece_id') + expect(query).toContain('pieces.ipfs_root_cid') + expect(query).toContain('WHERE (pieces.ipfs_root_cid = ?)') + }) +}) diff --git a/retrieval/test/bad-bits-util.test.js b/retrieval/test/bad-bits-util.test.js index a27f7804..c2c88e52 100644 --- a/retrieval/test/bad-bits-util.test.js +++ b/retrieval/test/bad-bits-util.test.js @@ -1,5 +1,9 @@ import { describe, it, expect } from 'vitest' -import { getBadBitsEntry } from '../lib/bad-bits-util.js' +import { + getBadBitsEntry, + isCidDenied, + assertCidNotDenied, +} from '../lib/bad-bits-util.js' describe('getBadBitsEntry', () => { it('creates entry in the legacy double-hash format', async () => { @@ -12,3 +16,44 @@ describe('getBadBitsEntry', () => { ) }) }) + +describe('isCidDenied', () => { + it('returns true when the denylist has an entry for the CID, querying by double-hash key', async () => { + const cid = 'bafybeiefwqslmf6zyyrxodaxx4vwqircuxpza5ri45ws3y5a62ypxti42e' + const expectedKey = `bad-bits:${await getBadBitsEntry(cid)}` + /** @type {string[]} */ + const queriedKeys = [] + const env = { + BAD_BITS_KV: { + get: async (/** @type {string} */ key) => { + queriedKeys.push(key) + return key === expectedKey ? {} : null + }, + }, + } + + expect(await isCidDenied(env, cid)).toBe(true) + expect(queriedKeys).toEqual([expectedKey]) + }) + + it('returns false when the CID is not on the denylist', async () => { + const env = { BAD_BITS_KV: { get: async () => null } } + expect(await isCidDenied(env, 'bafytest')).toBe(false) + }) +}) + +describe('assertCidNotDenied', () => { + it('throws a 404 when the CID is on the denylist', async () => { + const env = { BAD_BITS_KV: { get: async () => ({}) } } + await expect(assertCidNotDenied(env, 'bafytest')).rejects.toMatchObject({ + status: 404, + message: + 'The requested CID was flagged by the Bad Bits Denylist at https://badbits.dwebops.pub', + }) + }) + + it('resolves when the CID is not on the denylist', async () => { + const env = { BAD_BITS_KV: { get: async () => null } } + await expect(assertCidNotDenied(env, 'bafytest')).resolves.toBeUndefined() + }) +}) diff --git a/retrieval/test/bot-auth.test.js b/retrieval/test/bot-auth.test.js new file mode 100644 index 00000000..81166747 --- /dev/null +++ b/retrieval/test/bot-auth.test.js @@ -0,0 +1,50 @@ +import { describe, it, expect } from 'vitest' +import { checkBotAuthorization } from '../lib/bot-auth.js' + +const BOT_TOKENS = JSON.stringify({ secret: 'bot1', secret_2: 'bot2' }) + +describe('checkBotAuthorization', () => { + it('returns undefined when no authorization header is present', () => { + const request = new Request('https://example.com', { headers: {} }) + expect(checkBotAuthorization(request, { BOT_TOKENS })).toBeUndefined() + }) + + it('returns undefined for anonymous requests even when BOT_TOKENS is empty', () => { + const request = new Request('https://example.com', { headers: {} }) + expect(checkBotAuthorization(request, { BOT_TOKENS: '' })).toBeUndefined() + }) + + it('throws 401 when the authorization header is not Bearer format', () => { + const request = new Request('https://example.com', { + headers: { authorization: 'Basic sometoken' }, + }) + expect(() => checkBotAuthorization(request, { BOT_TOKENS })).toThrowError( + 'Unauthorized: Authorization header must use Bearer scheme', + ) + }) + + it('throws 401 when there is no token after Bearer', () => { + const request = new Request('https://example.com', { + headers: { authorization: 'Bearer' }, + }) + expect(() => checkBotAuthorization(request, { BOT_TOKENS })).toThrowError( + 'Unauthorized: Authorization header must use Bearer scheme', + ) + }) + + it('throws 401 when the token is not in BOT_TOKENS', () => { + const request = new Request('https://example.com', { + headers: { authorization: 'Bearer invalid_token' }, + }) + expect(() => checkBotAuthorization(request, { BOT_TOKENS })).toThrowError( + 'Unauthorized: Invalid Access Token i...n', + ) + }) + + it('returns the bot name for a valid token', () => { + const request = new Request('https://example.com', { + headers: { authorization: 'Bearer secret' }, + }) + expect(checkBotAuthorization(request, { BOT_TOKENS })).toBe('bot1') + }) +}) diff --git a/retrieval/test/candidate-selection.test.js b/retrieval/test/candidate-selection.test.js new file mode 100644 index 00000000..6b07ab97 --- /dev/null +++ b/retrieval/test/candidate-selection.test.js @@ -0,0 +1,258 @@ +import { describe, it, expect } from 'vitest' +import { selectRetrievalCandidate } from '../lib/candidate-selection.js' +import { + env, + createExecutionContext, + waitOnExecutionContext, +} from 'cloudflare:test' + +const ok = () => ({ + response: new Response('ok', { status: 200 }), + cacheMiss: true, +}) +const notFound = () => ({ + response: new Response('nope', { status: 404 }), + cacheMiss: false, +}) +const serverError = () => ({ + response: new Response(null, { status: 503 }), + cacheMiss: true, +}) + +function candidate(overrides = {}) { + return { + serviceUrl: 'https://a.example/', + serviceProviderId: '1', + dataSetId: '10', + ...overrides, + } +} + +function context(ctx) { + return { + env, + ctx, + requestCountryCode: 'US', + timestamp: new Date().toISOString(), + botName: undefined, + } +} + +describe('selectRetrievalCandidate', () => { + it('throws when the candidate list is empty', async () => { + const ctx = createExecutionContext() + await expect( + selectRetrievalCandidate([], async () => ok(), context(ctx)), + ).rejects.toThrow('should never happen') + }) + + it('returns the first candidate that responds OK without retrying', async () => { + const ctx = createExecutionContext() + let calls = 0 + const selection = await selectRetrievalCandidate( + [ + candidate({ serviceUrl: 'https://a.example/' }), + candidate({ serviceUrl: 'https://b.example/' }), + ], + async () => { + calls++ + return ok() + }, + context(ctx), + ) + + expect(calls).toBe(1) + expect(selection.failureResponse).toBeUndefined() + expect(selection.candidate).toBeDefined() + expect(selection.result?.response.ok).toBe(true) + }) + + it('retries after a 5xx response until one succeeds', async () => { + const ctx = createExecutionContext() + let calls = 0 + const selection = await selectRetrievalCandidate( + [ + candidate({ serviceUrl: 'https://a.example/' }), + candidate({ serviceUrl: 'https://b.example/' }), + candidate({ serviceUrl: 'https://c.example/' }), + ], + async () => { + calls++ + return calls < 2 ? serverError() : ok() + }, + context(ctx), + ) + + expect(calls).toBe(2) + expect(selection.failureResponse).toBeUndefined() + expect(selection.result?.response.ok).toBe(true) + }) + + it('retries after a thrown error until one succeeds', async () => { + const ctx = createExecutionContext() + let calls = 0 + const selection = await selectRetrievalCandidate( + [ + candidate({ serviceUrl: 'https://a.example/' }), + candidate({ serviceUrl: 'https://b.example/' }), + ], + async () => { + calls++ + if (calls === 1) throw new Error('boom') + return ok() + }, + context(ctx), + ) + + expect(calls).toBe(2) + expect(selection.result?.response.ok).toBe(true) + }) + + it('treats a 4xx response as a success', async () => { + const ctx = createExecutionContext() + const selection = await selectRetrievalCandidate( + [candidate()], + async () => notFound(), + context(ctx), + ) + + expect(selection.failureResponse).toBeUndefined() + expect(selection.result?.response.status).toBe(404) + }) + + it('returns a 502 failure response when every candidate returns a 5xx', async () => { + const ctx = createExecutionContext() + const selection = await selectRetrievalCandidate( + [ + candidate({ + serviceProviderId: '1', + serviceUrl: 'https://a.example/', + dataSetId: '10', + }), + ], + async () => serverError(), + context(ctx), + ) + await waitOnExecutionContext(ctx) + + expect(selection.candidate).toBeUndefined() + expect(selection.result).toBeUndefined() + expect(selection.failureResponse?.status).toBe(502) + expect(selection.failureResponse?.headers.get('X-Data-Set-ID')).toBe('10') + expect(await selection.failureResponse?.text()).toBe( + 'No available service provider found. Attempted: ID=1 (Service URL=https://a.example/)', + ) + }) + + it('lists every attempted provider in the failure response', async () => { + const ctx = createExecutionContext() + const selection = await selectRetrievalCandidate( + [ + candidate({ + serviceProviderId: '1', + serviceUrl: 'https://a.example/', + dataSetId: '10', + }), + candidate({ + serviceProviderId: '2', + serviceUrl: 'https://b.example/', + dataSetId: '11', + }), + ], + async () => serverError(), + context(ctx), + ) + await waitOnExecutionContext(ctx) + + const body = await selection.failureResponse?.text() + expect(body).toContain('ID=1 (Service URL=https://a.example/)') + expect(body).toContain('ID=2 (Service URL=https://b.example/)') + const dataSetIds = selection.failureResponse?.headers.get('X-Data-Set-ID') + expect(dataSetIds?.split(',').sort()).toEqual(['10', '11']) + }) + + it('sets a content security policy on the failure response', async () => { + const ctx = createExecutionContext() + const selection = await selectRetrievalCandidate( + [candidate()], + async () => serverError(), + context(ctx), + ) + await waitOnExecutionContext(ctx) + + expect( + selection.failureResponse?.headers.get('Content-Security-Policy'), + ).toBeTruthy() + }) + + it('logs the failure with the cache miss flag and zero egress', async () => { + const dataSetId = 'no-sp-cache-miss' + const ctx = createExecutionContext() + await selectRetrievalCandidate( + [candidate({ dataSetId })], + async () => serverError(), + { + env, + ctx, + requestCountryCode: 'US', + timestamp: new Date().toISOString(), + botName: 'bot1', + }, + ) + await waitOnExecutionContext(ctx) + + const log = await env.DB.prepare( + `SELECT response_status, egress_bytes, cache_miss_egress_bytes, cache_miss, bot_name + FROM retrieval_logs WHERE data_set_id = ?`, + ) + .bind(dataSetId) + .first() + + expect(log).toEqual({ + response_status: 502, + egress_bytes: 0, + cache_miss_egress_bytes: 0, + cache_miss: 1, + bot_name: 'bot1', + }) + }) + + it('logs a null cache miss when every attempt threw', async () => { + const dataSetId = 'no-sp-all-threw' + const ctx = createExecutionContext() + await selectRetrievalCandidate( + [candidate({ dataSetId })], + async () => { + throw new Error('boom') + }, + context(ctx), + ) + await waitOnExecutionContext(ctx) + + const log = await env.DB.prepare( + `SELECT response_status, cache_miss, bot_name + FROM retrieval_logs WHERE data_set_id = ?`, + ) + .bind(dataSetId) + .first() + + expect(log).toEqual({ + response_status: 502, + cache_miss: null, + bot_name: null, + }) + }) + + it('does not mutate the input candidate list', async () => { + const ctx = createExecutionContext() + const candidates = [ + candidate({ serviceUrl: 'https://a.example/' }), + candidate({ serviceUrl: 'https://b.example/' }), + ] + const snapshot = [...candidates] + + await selectRetrievalCandidate(candidates, async () => ok(), context(ctx)) + + expect(candidates).toEqual(snapshot) + }) +}) diff --git a/retrieval/test/fetch-handler.test.js b/retrieval/test/fetch-handler.test.js new file mode 100644 index 00000000..e641f39a --- /dev/null +++ b/retrieval/test/fetch-handler.test.js @@ -0,0 +1,304 @@ +import { describe, it, expect } from 'vitest' +import { handleFetchRequest } from '../lib/fetch-handler.js' +import { + env, + createExecutionContext, + waitOnExecutionContext, +} from 'cloudflare:test' + +const testEnv = { + ...env, + CLIENT_CACHE_TTL: 31536000, + ENFORCE_EGRESS_QUOTA: false, + BOT_TOKENS: '{}', +} + +function retrievalResult(overrides = {}) { + return { + response: new Response('hello world', { status: 200 }), + cacheMiss: true, + dataSetId: 'fh-test', + fetchStartedAt: performance.now(), + finalizeCacheMiss: async () => ({ cacheMissResponseValid: true }), + ...overrides, + } +} + +/** A `run` that resolves to a retrieve function yielding the given outcome. */ +function runYielding(overrides = {}) { + return async () => async () => retrievalResult(overrides) +} + +describe('handleFetchRequest', () => { + it('returns a plain handler response unchanged', async () => { + const ctx = createExecutionContext() + const res = await handleFetchRequest( + new Request('https://example.com/'), + testEnv, + ctx, + async () => new Response('ok', { status: 200 }), + ) + + expect(res.status).toBe(200) + expect(await res.text()).toBe('ok') + }) + + it('rejects non-GET/HEAD methods with 405 without running the handler', async () => { + const ctx = createExecutionContext() + let ran = false + const res = await handleFetchRequest( + new Request('https://example.com/', { method: 'POST' }), + testEnv, + ctx, + async () => { + ran = true + return new Response('ok') + }, + ) + + expect(res.status).toBe(405) + expect(await res.text()).toBe('Method Not Allowed') + expect(ran).toBe(false) + }) + + it('allows HEAD requests', async () => { + const ctx = createExecutionContext() + const res = await handleFetchRequest( + new Request('https://example.com/', { method: 'HEAD' }), + testEnv, + ctx, + async () => new Response('ok', { status: 200 }), + ) + + expect(res.status).toBe(200) + }) + + it('redirects legacy *.filcdn.io requests before running the handler', async () => { + const ctx = createExecutionContext() + let ran = false + const res = await handleFetchRequest( + new Request('https://0xabc.filcdn.io/baga123?format=car'), + testEnv, + ctx, + async () => { + ran = true + return new Response('ok') + }, + ) + + expect(res.status).toBe(301) + expect(res.headers.get('Location')).toBe( + 'https://0xabc.filbeam.io/baga123?format=car', + ) + expect(ran).toBe(false) + }) + + it('turns a thrown error into a response via handleError', async () => { + const ctx = createExecutionContext() + const res = await handleFetchRequest( + new Request('https://example.com/'), + testEnv, + ctx, + async () => { + throw Object.assign(new Error('Bad Request'), { status: 400 }) + }, + ) + + expect(res.status).toBe(400) + expect(await res.text()).toBe('Bad Request') + }) + + it('hides the message for server errors', async () => { + const ctx = createExecutionContext() + const res = await handleFetchRequest( + new Request('https://example.com/'), + testEnv, + ctx, + async () => { + throw new Error('boom') + }, + ) + + expect(res.status).toBe(500) + expect(await res.text()).toBe('Internal Server Error') + }) + + it('logs the error status with no egress and no data set when the retrieval throws', async () => { + const ctx = createExecutionContext() + const res = await handleFetchRequest( + new Request('https://example.com/', { + headers: { 'CF-IPCountry': 'US' }, + }), + testEnv, + ctx, + async () => async () => { + throw Object.assign(new Error("I'm a teapot"), { status: 418 }) + }, + ) + expect(res.status).toBe(418) + await waitOnExecutionContext(ctx) + + const log = await env.DB.prepare( + `SELECT response_status, egress_bytes, cache_miss, data_set_id + FROM retrieval_logs WHERE response_status = 418`, + ).first() + expect(log).toEqual({ + response_status: 418, + egress_bytes: null, + cache_miss: null, + data_set_id: null, + }) + }) + + it('streams a retrieval result, measuring egress and logging it', async () => { + const ctx = createExecutionContext() + const dataSetId = 'fh-stream' + let finalizedEgress + const res = await handleFetchRequest( + new Request('https://example.com/', { + headers: { 'CF-IPCountry': 'US' }, + }), + testEnv, + ctx, + runYielding({ + dataSetId, + finalizeCacheMiss: async (egressBytes) => { + finalizedEgress = egressBytes + return { cacheMissResponseValid: true } + }, + }), + ) + + expect(res.status).toBe(200) + expect(res.headers.get('X-Data-Set-ID')).toBe(dataSetId) + expect(res.headers.get('Content-Security-Policy')).toMatch( + /^default-src 'self'/, + ) + expect(await res.text()).toBe('hello world') + await waitOnExecutionContext(ctx) + + expect(finalizedEgress).toBe('hello world'.length) + const log = await env.DB.prepare( + `SELECT response_status, egress_bytes, cache_miss, request_country_code + FROM retrieval_logs WHERE data_set_id = ?`, + ) + .bind(dataSetId) + .first() + // request_country_code is sourced by handleFetchRequest from the request. + expect(log).toEqual({ + response_status: 200, + egress_bytes: 'hello world'.length, + cache_miss: 1, + request_country_code: 'US', + }) + }) + + it('logs a zero-egress result for a response without a body', async () => { + const ctx = createExecutionContext() + const dataSetId = 'fh-empty' + const res = await handleFetchRequest( + new Request('https://example.com/', { + headers: { 'CF-IPCountry': 'US', authorization: 'Bearer tok' }, + }), + { ...testEnv, BOT_TOKENS: JSON.stringify({ tok: 'bot-1' }) }, + ctx, + runYielding({ + dataSetId, + response: new Response(null, { status: 404 }), + }), + ) + await waitOnExecutionContext(ctx) + + expect(res.status).toBe(404) + expect(res.body).toBeNull() + expect(res.headers.get('X-Data-Set-ID')).toBe(dataSetId) + expect(res.headers.get('Cache-Control')).toBe( + `public, max-age=${testEnv.CLIENT_CACHE_TTL}`, + ) + const log = await env.DB.prepare( + `SELECT response_status, egress_bytes, cache_miss_egress_bytes, cache_miss, cache_miss_response_valid, bot_name + FROM retrieval_logs WHERE data_set_id = ?`, + ) + .bind(dataSetId) + .first() + expect(log).toEqual({ + response_status: 404, + egress_bytes: 0, + cache_miss_egress_bytes: 0, + cache_miss: 1, + cache_miss_response_valid: null, + bot_name: 'bot-1', + }) + }) + + it('logs a 900 result when streaming the body errors', async () => { + const ctx = createExecutionContext() + const dataSetId = 'fh-stream-error' + const erroringBody = new ReadableStream({ + pull(controller) { + controller.enqueue(new Uint8Array([1, 2, 3])) + controller.error(new Error('stream boom')) + }, + }) + const res = await handleFetchRequest( + new Request('https://example.com/'), + testEnv, + ctx, + runYielding({ + dataSetId, + response: new Response(erroringBody, { status: 200 }), + }), + ) + expect(res.status).toBe(200) + await waitOnExecutionContext(ctx) + + const log = await env.DB.prepare( + 'SELECT response_status FROM retrieval_logs WHERE data_set_id = ? AND response_status = 900', + ) + .bind(dataSetId) + .first() + expect(log).toEqual({ response_status: 900 }) + }) + + it('resolves the bot name from the Authorization header and logs it', async () => { + const ctx = createExecutionContext() + const dataSetId = 'fh-bot' + const res = await handleFetchRequest( + new Request('https://example.com/', { + headers: { authorization: 'Bearer tok' }, + }), + { ...testEnv, BOT_TOKENS: JSON.stringify({ tok: 'bot-1' }) }, + ctx, + runYielding({ dataSetId }), + ) + expect(res.status).toBe(200) + expect(await res.text()).toBe('hello world') + await waitOnExecutionContext(ctx) + + const log = await env.DB.prepare( + 'SELECT bot_name FROM retrieval_logs WHERE data_set_id = ?', + ) + .bind(dataSetId) + .first() + expect(log).toEqual({ bot_name: 'bot-1' }) + }) + + it('rejects an unknown bot token with 401 without running the retrieval', async () => { + const ctx = createExecutionContext() + let ran = false + const res = await handleFetchRequest( + new Request('https://example.com/', { + headers: { authorization: 'Bearer wrong' }, + }), + { ...testEnv, BOT_TOKENS: JSON.stringify({ tok: 'bot-1' }) }, + ctx, + () => { + ran = true + return Promise.resolve(async () => retrievalResult()) + }, + ) + + expect(res.status).toBe(401) + expect(ran).toBe(false) + }) +}) diff --git a/retrieval/test/http-error.test.js b/retrieval/test/http-error.test.js new file mode 100644 index 00000000..828c4ccb --- /dev/null +++ b/retrieval/test/http-error.test.js @@ -0,0 +1,54 @@ +import { describe, it, expect } from 'vitest' +import { getErrorHttpStatusMessage, handleError } from '../lib/http-error.js' + +describe('getErrorHttpStatusMessage', () => { + it('uses the numeric status and message for client errors', () => { + expect( + getErrorHttpStatusMessage( + Object.assign(new Error('Not Found'), { + status: 404, + }), + ), + ).toEqual({ status: 404, message: 'Not Found' }) + }) + + it('hides the message for server errors', () => { + expect( + getErrorHttpStatusMessage( + Object.assign(new Error('boom'), { + status: 500, + }), + ), + ).toEqual({ status: 500, message: 'Internal Server Error' }) + }) + + it('defaults to 500 when the error has no numeric status', () => { + expect(getErrorHttpStatusMessage(new Error('boom'))).toEqual({ + status: 500, + message: 'Internal Server Error', + }) + }) + + it('defaults to 500 for non-object errors', () => { + expect(getErrorHttpStatusMessage('boom')).toEqual({ + status: 500, + message: 'Internal Server Error', + }) + }) +}) + +describe('handleError', () => { + it('returns the status and message for client errors', async () => { + const res = handleError( + Object.assign(new Error('Bad Request'), { status: 400 }), + ) + expect(res.status).toBe(400) + expect(await res.text()).toBe('Bad Request') + }) + + it('returns a generic 500 response for server errors', async () => { + const res = handleError(new Error('boom')) + expect(res.status).toBe(500) + expect(await res.text()).toBe('Internal Server Error') + }) +}) diff --git a/retrieval/test/origin-cache.test.js b/retrieval/test/origin-cache.test.js new file mode 100644 index 00000000..cda5ff19 --- /dev/null +++ b/retrieval/test/origin-cache.test.js @@ -0,0 +1,11 @@ +import { describe, it, expect } from 'vitest' +import { originCacheOptions } from '../lib/origin-cache.js' + +describe('originCacheOptions', () => { + it('caches 2xx for the given TTL and never caches 404 or 5xx', () => { + expect(originCacheOptions(86400)).toEqual({ + cacheTtlByStatus: { '200-299': 86400, 404: 0, '500-599': 0 }, + cacheEverything: true, + }) + }) +}) diff --git a/retrieval/test/stats.test.js b/retrieval/test/stats.test.js index 9faa8ca2..4df60a5e 100644 --- a/retrieval/test/stats.test.js +++ b/retrieval/test/stats.test.js @@ -1,5 +1,9 @@ import { describe, it, expect } from 'vitest' -import { updateDataSetStats, logRetrievalResult } from '../lib/stats' +import { + updateDataSetStats, + logRetrievalResult, + recordRetrieval, +} from '../lib/stats' import { withDataSet } from './test-helpers' import { env } from 'cloudflare:test' @@ -218,6 +222,51 @@ describe('updateDataSetStats', () => { expect(quotaResult.cdn_egress_quota).toBe(initialCdnQuota - EGRESS_BYTES) expect(quotaResult.cache_miss_egress_quota).toBe(initialCacheMissQuota) }) + + it('charges the cache miss quota by cacheMissEgressBytes when it differs from egressBytes', async () => { + const DATA_SET_ID = 'test-data-set-car' + // Raw bytes served to the client. + const EGRESS_BYTES = 100 + // Larger CAR fetched from the service provider on a cache miss. + const CACHE_MISS_EGRESS_BYTES = 250 + const initialCdnQuota = 500 + const initialCacheMissQuota = 300 + + await withDataSet(env, { + dataSetId: DATA_SET_ID, + cdnEgressQuota: initialCdnQuota, + cacheMissEgressQuota: initialCacheMissQuota, + }) + + await updateDataSetStats(env, { + dataSetId: DATA_SET_ID, + egressBytes: EGRESS_BYTES, + cacheMissEgressBytes: CACHE_MISS_EGRESS_BYTES, + cacheMiss: true, + cacheMissResponseValid: true, + enforceEgressQuota: true, + }) + + const dataSetResult = await env.DB.prepare( + `SELECT total_egress_bytes_used FROM data_sets WHERE id = ?`, + ) + .bind(DATA_SET_ID) + .first() + + const quotaResult = await env.DB.prepare( + `SELECT cdn_egress_quota, cache_miss_egress_quota FROM data_set_egress_quotas WHERE data_set_id = ?`, + ) + .bind(DATA_SET_ID) + .first() + + // Total egress and the CDN quota are charged the raw bytes served. + expect(dataSetResult.total_egress_bytes_used).toBe(EGRESS_BYTES) + expect(quotaResult.cdn_egress_quota).toBe(initialCdnQuota - EGRESS_BYTES) + // The cache-miss quota is charged the larger CAR fetched from the SP. + expect(quotaResult.cache_miss_egress_quota).toBe( + initialCacheMissQuota - CACHE_MISS_EGRESS_BYTES, + ) + }) }) describe('logRetrievalResult', () => { @@ -255,4 +304,112 @@ describe('logRetrievalResult', () => { }, ]) }) + + it('persists cache_miss_egress_bytes distinct from egress_bytes', async () => { + const DATA_SET_ID = 'cme-distinct' + + await logRetrievalResult(env, { + dataSetId: DATA_SET_ID, + cacheMiss: true, + cacheMissResponseValid: true, + egressBytes: 1234, + cacheMissEgressBytes: 5678, + responseStatus: 200, + timestamp: new Date().toISOString(), + requestCountryCode: 'US', + }) + + const result = await env.DB.prepare( + `SELECT data_set_id, egress_bytes, cache_miss_egress_bytes + FROM retrieval_logs + WHERE data_set_id = ?`, + ) + .bind(DATA_SET_ID) + .all() + + expect(result.results).toEqual([ + { + data_set_id: DATA_SET_ID, + egress_bytes: 1234, + cache_miss_egress_bytes: 5678, + }, + ]) + }) + + it('defaults cache_miss_egress_bytes to egress_bytes when not provided', async () => { + const DATA_SET_ID = 'cme-default' + + await logRetrievalResult(env, { + dataSetId: DATA_SET_ID, + cacheMiss: false, + cacheMissResponseValid: null, + egressBytes: 999, + responseStatus: 200, + timestamp: new Date().toISOString(), + requestCountryCode: 'US', + }) + + const result = await env.DB.prepare( + `SELECT egress_bytes, cache_miss_egress_bytes + FROM retrieval_logs + WHERE data_set_id = ?`, + ) + .bind(DATA_SET_ID) + .first() + + expect(result).toEqual({ egress_bytes: 999, cache_miss_egress_bytes: 999 }) + }) +}) + +describe('recordRetrieval', () => { + it('writes the retrieval log and updates the data set egress stats', async () => { + const DATA_SET_ID = 'record-retrieval' + await withDataSet(env, { + dataSetId: DATA_SET_ID, + cdnEgressQuota: 1000, + cacheMissEgressQuota: 1000, + }) + + await recordRetrieval(env, { + dataSetId: DATA_SET_ID, + cacheMiss: true, + cacheMissResponseValid: true, + egressBytes: 100, + cacheMissEgressBytes: 250, + responseStatus: 200, + requestCountryCode: 'US', + timestamp: new Date().toISOString(), + enforceEgressQuota: true, + }) + + const log = await env.DB.prepare( + `SELECT egress_bytes, cache_miss_egress_bytes, cache_miss + FROM retrieval_logs WHERE data_set_id = ?`, + ) + .bind(DATA_SET_ID) + .first() + expect(log).toEqual({ + egress_bytes: 100, + cache_miss_egress_bytes: 250, + cache_miss: 1, + }) + + const dataSet = await env.DB.prepare( + `SELECT total_egress_bytes_used FROM data_sets WHERE id = ?`, + ) + .bind(DATA_SET_ID) + .first() + expect(dataSet.total_egress_bytes_used).toBe(100) + + const quota = await env.DB.prepare( + `SELECT cdn_egress_quota, cache_miss_egress_quota + FROM data_set_egress_quotas WHERE data_set_id = ?`, + ) + .bind(DATA_SET_ID) + .first() + expect(quota).toEqual({ + cdn_egress_quota: 1000 - 100, + cache_miss_egress_quota: 1000 - 250, + }) + }) }) diff --git a/stats-api/lib/handlers.js b/stats-api/lib/handlers.js index f32614f8..da98826b 100644 --- a/stats-api/lib/handlers.js +++ b/stats-api/lib/handlers.js @@ -60,7 +60,7 @@ export async function handleGetPayerStats(env, payerAddress) { COUNT(rl.id) AS total_requests, SUM(CASE WHEN rl.cache_miss THEN 1 ELSE 0 END) AS cache_miss_requests, SUM(rl.egress_bytes) AS total_egress_bytes, - SUM(CASE WHEN rl.cache_miss THEN rl.egress_bytes ELSE 0 END) AS cache_miss_egress_bytes + SUM(CASE WHEN rl.cache_miss THEN COALESCE(rl.cache_miss_egress_bytes, rl.egress_bytes) ELSE 0 END) AS cache_miss_egress_bytes FROM retrieval_logs rl JOIN diff --git a/stats-api/test/handlers.test.js b/stats-api/test/handlers.test.js index dc973104..72601a9a 100644 --- a/stats-api/test/handlers.test.js +++ b/stats-api/test/handlers.test.js @@ -99,6 +99,41 @@ describe('stats-handlers', () => { }) }) + it('reports the origin CAR size as cache-miss egress and raw bytes as total egress', async () => { + const payerAddress = '0xcarpayer' + const dataSetId = '1' + await withDataSet(env, { + dataSetId, + serviceProviderId: '1', + payerAddress, + withCDN: true, + cdnEgressQuota: 3000, + cacheMissEgressQuota: 6000, + }) + await withRetrievalLog(env, { + timestamp: new Date().toISOString(), + dataSetId, + egressBytes: 100, + cacheMissEgressBytes: 250, + cacheMiss: true, + }) + + const res = await handleGetPayerStats(env, payerAddress) + + expect(res.status).toBe(200) + const data = await res.json() + expect(data).toStrictEqual({ + // Total egress is the raw bytes served to the client. + totalEgressBytes: '100', + // Cache-miss egress is the larger CAR fetched from the SP. + cacheMissEgressBytes: '250', + cacheMissRequests: '1', + remainingCDNEgressBytes: '3000', + remainingCacheMissEgressBytes: '6000', + totalRequests: '1', + }) + }) + it('returns 404 for non-existent payer', async () => { const res = await handleGetPayerStats(env, 'non-existent') diff --git a/stats-api/test/test-helpers.js b/stats-api/test/test-helpers.js index e56acd4d..b87d7ff7 100644 --- a/stats-api/test/test-helpers.js +++ b/stats-api/test/test-helpers.js @@ -46,6 +46,8 @@ export async function withDataSet( * @param {string} params.dataSetId - Data set ID * @param {number} params.responseStatus - HTTP response status (default: 200) * @param {number | null} params.egressBytes - Egress bytes (default: null) + * @param {number | null} params.cacheMissEgressBytes - CAR bytes fetched from + * the SP on a cache miss (default: null) * @param {number} params.cacheMiss - Cache miss flag (0 or 1, default: 0) */ export async function withRetrievalLog( @@ -55,13 +57,21 @@ export async function withRetrievalLog( dataSetId, responseStatus = 200, egressBytes = null, + cacheMissEgressBytes = null, cacheMiss = 0, }, ) { return await env.DB.prepare( - `INSERT INTO retrieval_logs (timestamp, data_set_id, response_status, egress_bytes, cache_miss) - VALUES (datetime(?), ?, ?, ?, ?)`, + `INSERT INTO retrieval_logs (timestamp, data_set_id, response_status, egress_bytes, cache_miss_egress_bytes, cache_miss) + VALUES (datetime(?), ?, ?, ?, ?, ?)`, ) - .bind(timestamp, dataSetId, responseStatus, egressBytes, cacheMiss) + .bind( + timestamp, + dataSetId, + responseStatus, + egressBytes, + cacheMissEgressBytes, + cacheMiss, + ) .run() } diff --git a/tsconfig.json b/tsconfig.json index b9c35895..ff9b3827 100644 --- a/tsconfig.json +++ b/tsconfig.json @@ -12,6 +12,9 @@ { "path": "indexer/tsconfig.json" }, + { + "path": "ipfs-retriever/tsconfig.json" + }, { "path": "piece-retriever/tsconfig.json" }, diff --git a/usage-reporter/lib/usage-report.js b/usage-reporter/lib/usage-report.js index 9a018f9c..bb178158 100644 --- a/usage-reporter/lib/usage-report.js +++ b/usage-reporter/lib/usage-report.js @@ -20,10 +20,14 @@ export async function aggregateUsageData(db, upToTimestampMs) { const query = ` SELECT rl.data_set_id, - -- Note: cdn_bytes tracks all egress (cache hits + cache misses) - -- cache_miss_bytes tracks only cache misses (subset of cdn_bytes) + -- Note: cdn_bytes tracks the bytes served to clients across all egress + -- (cache hits + cache misses). cache_miss_bytes tracks the bytes fetched + -- from the service provider on cache misses. For IPFS retrievals the + -- origin CAR (cache_miss_egress_bytes) is larger than the raw bytes served + -- to the client (egress_bytes). cache_miss_egress_bytes falls back to + -- egress_bytes for rows logged before that column existed. SUM(rl.egress_bytes) as cdn_bytes, - SUM(CASE WHEN rl.cache_miss = 1 AND rl.cache_miss_response_valid = 1 THEN rl.egress_bytes ELSE 0 END) as cache_miss_bytes + SUM(CASE WHEN rl.cache_miss = 1 AND rl.cache_miss_response_valid = 1 THEN COALESCE(rl.cache_miss_egress_bytes, rl.egress_bytes) ELSE 0 END) as cache_miss_bytes FROM retrieval_logs rl INNER JOIN data_sets ds ON rl.data_set_id = ds.id WHERE rl.timestamp > datetime(ds.usage_reported_until) diff --git a/usage-reporter/test/test-helpers.js b/usage-reporter/test/test-helpers.js index cfc0078e..1327faa1 100644 --- a/usage-reporter/test/test-helpers.js +++ b/usage-reporter/test/test-helpers.js @@ -97,8 +97,11 @@ export const randomId = () => String(Math.ceil(Math.random() * 1e10)) * @param {string} params.dataSetId - Data set ID * @param {number} params.responseStatus - HTTP response status (default: 200) * @param {number | null} params.egressBytes - Egress bytes (default: null) + * @param {number | null} params.cacheMissEgressBytes - CAR bytes fetched from + * the SP on a cache miss (default: null) * @param {number} params.cacheMiss - Cache miss flag (0 or 1, default: 0) - * @param {boolena} params.cacheMissResponseValid + * @param {number} params.cacheMissResponseValid - Cache-miss response valid + * flag (0 or 1, default: 0) */ export async function withRetrievalLog( env, @@ -107,19 +110,21 @@ export async function withRetrievalLog( dataSetId, responseStatus = 200, egressBytes = null, + cacheMissEgressBytes = null, cacheMiss = 0, cacheMissResponseValid = 0, }, ) { return await env.DB.prepare( - `INSERT INTO retrieval_logs (timestamp, data_set_id, response_status, egress_bytes, cache_miss, cache_miss_response_valid) - VALUES (datetime(?), ?, ?, ?, ?, ?)`, + `INSERT INTO retrieval_logs (timestamp, data_set_id, response_status, egress_bytes, cache_miss_egress_bytes, cache_miss, cache_miss_response_valid) + VALUES (datetime(?), ?, ?, ?, ?, ?, ?)`, ) .bind( timestamp, dataSetId, responseStatus, egressBytes, + cacheMissEgressBytes, cacheMiss, cacheMissResponseValid, ) diff --git a/usage-reporter/test/usage-report.test.js b/usage-reporter/test/usage-report.test.js index 47a7973f..9386c69c 100644 --- a/usage-reporter/test/usage-report.test.js +++ b/usage-reporter/test/usage-report.test.js @@ -95,6 +95,48 @@ describe('usage report', () => { ]) }) + it('charges cache-miss bytes from cache_miss_egress_bytes (CAR) and cdn bytes from egress_bytes (raw)', async () => { + await withDataSet(env, { + id: '1', + usageReportedUntil: EPOCH_99_TIMESTAMP_ISO, + }) + + // IPFS cache miss: client served raw bytes (egress_bytes), but a larger + // CAR was fetched from the SP (cache_miss_egress_bytes). + await withRetrievalLog(env, { + timestamp: EPOCH_100_TIMESTAMP_ISO, + dataSetId: '1', + egressBytes: 1000, + cacheMissEgressBytes: 2500, + cacheMiss: 1, + cacheMissResponseValid: 1, + }) + + // Row without the column set (e.g. logged before the column existed) + // falls back to egress_bytes for cache-miss accounting. + await withRetrievalLog(env, { + timestamp: EPOCH_100_TIMESTAMP_ISO, + dataSetId: '1', + egressBytes: 400, + cacheMissEgressBytes: null, + cacheMiss: 1, + cacheMissResponseValid: 1, + }) + + const usageData = await aggregateUsageData( + env.DB, + EPOCH_100_TIMESTAMP_MS, + ) + + expect(usageData).toStrictEqual([ + { + data_set_id: '1', + cdn_bytes: 1400, + cache_miss_bytes: 2900, + }, + ]) + }) + it('should include non-200 responses but filter out null egress_bytes', async () => { await withDataSet(env, { id: '1', diff --git a/vitest.workspace.js b/vitest.workspace.js index 81f644c0..d3972ec3 100644 --- a/vitest.workspace.js +++ b/vitest.workspace.js @@ -2,6 +2,7 @@ import { defineWorkspace } from 'vitest/config' export default defineWorkspace([ 'indexer', + 'ipfs-retriever', 'piece-retriever', 'tail-handler', 'bad-bits', diff --git a/x402-piece-gateway/bin/x402-piece-gateway.js b/x402-piece-gateway/bin/x402-piece-gateway.js index 36e83cd2..4abe93b8 100644 --- a/x402-piece-gateway/bin/x402-piece-gateway.js +++ b/x402-piece-gateway/bin/x402-piece-gateway.js @@ -1,4 +1,4 @@ -import { httpAssert } from '@filbeam/retrieval' +import { httpAssert, handleError } from '@filbeam/retrieval' import { buildForwardUrl, parseRequest } from '../lib/request.js' import { useFacilitator as defaultUseFacilitator } from 'x402/verify' import { settleResponseHeader } from 'x402/types' @@ -32,7 +32,7 @@ export default { try { return await this._fetch(request, env, ctx, { useFacilitator }) } catch (error) { - return this._handleError(error) + return handleError(error) } }, @@ -164,45 +164,4 @@ export default { ) } }, - - /** - * @param {unknown} error - * @returns {Response} - */ - _handleError(error) { - const { status, message } = getErrorHttpStatusMessage(error) - - if (status >= 500) { - console.error(error) - } - return new Response(message, { status }) - }, -} - -/** - * Extracts status and message from an error object. - * - * - If the error has a numeric `status`, it is used; otherwise, defaults to 500. - * - If the status is < 500 and a string `message` exists, it's used; otherwise, a - * generic message is returned. - * - * @param {unknown} error - The error object to extract from. - * @returns {{ status: number; message: string }} - */ -function getErrorHttpStatusMessage(error) { - const isObject = typeof error === 'object' && error !== null - const status = - isObject && 'status' in error && typeof error.status === 'number' - ? error.status - : 500 - - const message = - isObject && - status < 500 && - 'message' in error && - typeof error.message === 'string' - ? error.message - : 'Internal Server Error' - - return { status, message } }