fix(go): set ResponseIsOptional when response type is Optional/Nullable

[devin-ai-integration]

Description

Fixes a bug where endpoints declaring both a 200 (with body) and 204 (no body) success response would return an error on 204 instead of a successful nil response. When the API returns a real 204, the JSON decoder hits io.EOF and the SDK returns:

expected a *teams.CreateMembersTenantsResponseContent response, but the server responded with nothing

Changes Made

The OpenAPI-to-IR converter already correctly handles mixed 200/204 responses by wrapping the response body type in an Optional container. The v1 Go generator detects this via getOptionalOrNullableContainer(typeReference) != nil at sdk.go:3111, but v2 (which generates raw_client.go) had no equivalent check.

• HttpEndpointGenerator.ts: Added hasOptionalOrNullableResponse() that inspects endpoint.response.body.json.responseBodyType for Optional or Nullable container wrapping. Passes responseIsOptional: true to the Caller when detected.
• Caller.ts: Added responseIsOptional field to CallArgs interface and emits ResponseIsOptional: true in generated CallParams when set.
• seed.yml: Added no-content-response fixture to go-sdk seed tests.
• Seed fixtures updated: no-content-response and exhaustive fixtures regenerated to include ResponseIsOptional: true in affected endpoints.

Generated CallParams before:

&internal.CallParams{
    URL:      endpointURL,
    Method:   http.MethodPost,
    Response: &response,
}

After (for endpoints with Optional/Nullable response type):

&internal.CallParams{
    URL:                endpointURL,
    Method:             http.MethodPost,
    Response:           &response,
    ResponseIsOptional: true,
}

Testing

• Seed test no-content-response passes with ResponseIsOptional: true in generated contacts/raw_client.go
• Seed test exhaustive passes — GetAndReturnOptional endpoint correctly gets ResponseIsOptional: true
• Regression tests pass: plain-text, oauth-client-credentials fixtures unchanged
• TypeScript compilation passes (pnpm turbo run compile --filter @fern-api/go-sdk)
• Formatting check passes

Link to Devin session: https://app.devin.ai/sessions/2ec9715faa89458c893d26e94b9043eb

---

[devin-ai-integration]

🤖 Devin AI Engineer

I'll be helping with this pull request! Here's what you should know:

✅ I will automatically:

• Address comments on this PR. Add '(aside)' to your comment to have me ignore it.
• Look at CI failures and help fix them

Note: I can only respond to comments from users who have write access to this repository.

⚙️ Control Options:

• Disable automatic comment, CI, and merge conflict monitoring

[devin-ai-integration]

Devin Review found 0 potential issues.

[github-actions]

SDK Generation Benchmark Results

Comparing PR branch against median of 5 nightly run(s) on main (latest: 2026-06-24T05:18:39Z).

Full benchmark table (click to expand)

Generator	Spec	main (generator)	main (E2E)	PR (generator)	Delta
go-sdk	square	134s (n=5)	283s (n=5)	128s	-6s (-4.5%)

main (generator): generator-only time via --skip-scripts (includes Docker image build, container startup, IR parsing, and code generation — this is the same Docker-based flow customers use via fern generate). main (E2E): full customer-observable time including build/test scripts (nightly baseline, informational). Delta is computed against generator-only baseline.
⚠️ = generation exited with a non-zero exit code (timing may not reflect a successful run).
Baseline from nightly runs on main (latest: 2026-06-24T05:18:39Z). Trigger benchmark-baseline to refresh.
Last updated: 2026-06-24 13:22 UTC