Add IGN site

[Edlison]

Mirrored site

IGN — https://www.ign.com/

Added as the 17th WebHarbor site, served on port 40016.

What's included

sites/ign/ — Flask + SQLAlchemy app modeling IGN's editorial and account surfaces: news, reviews, videos, guides/wiki pages, local utility pages, scored token-overlap News search, review filtering/sorting controls, article/review/video/guide detail pages, accounts, saved stories, playlist, comments, alerts, profile editing, and guide checklist progress. 17 Jinja2 templates with an IGN-style dark left rail, content cards, review-score treatment, and responsive utility pages.

109 seeded content items: 36 articles, 9 reviews, 40 videos, and 24 guides. Real IGN article/video/guide/review images are included under static/images/content/ with 114 local image files.

Idempotent seeding with function-level gates and the 4 canonical benchmark users (alice.j@test.com, bob.c@test.com, carol.d@test.com, david.k@test.com; password TestPass123!). Byte-identical reset verified.

19 WebVoyager tasks in sites/ign/tasks.jsonl.

Registered in the three required places: websyn_start.sh, control_server.py, Dockerfile (EXPOSE ... 40016).

Seeded rows per major model

model	rows	model	rows
sections	7	comments	8
content_items	109	alert_subscriptions	12
users	4	digests	4
saved_items	20	guide_progress	16
playlist_entries	16

Content breakdown: Articles 36 · Reviews 9 · Videos 40 · Guides 24.

Content by section: Games 27 · Reviews 3 · Guides 24 · Videos 32 · Movies 5 · TV 12 · Deals 6.

Tasks

19 tasks in sites/ign/tasks.jsonl (WebVoyager schema), spanning search, news/category browsing, reviews/filtering, guide checklist progress, login, profile editing, saved stories, playlist add/update/remove, comments, alerts, and multi-step comparison/detail-page reading.

The task set includes several auth-gated state-changing workflows and multi-step tasks that require disambiguation between near-miss IGN content, such as same-topic article/video results, physical-media stories, TV streaming stories, and tech review filters.

Byte-identical reset

$ curl -X POST http://localhost:8201/reset/ign
{"pid":245,"ready":true,"site":"ign"}

$ docker exec wh-submit md5sum /opt/WebSyn/ign/instance/ign.db \
                            /opt/WebSyn/ign/instance_seed/ign.db
da9379739f89373db60e2c32e801bc60  /opt/WebSyn/ign/instance/ign.db
da9379739f89373db60e2c32e801bc60  /opt/WebSyn/ign/instance_seed/ign.db

Both seed functions early-return as a whole. Fresh container verification passed with all 17 sites alive, ports 41000-41016 returning 200, and POST /reset-all returning every site ready:true.

Visual fidelity vs. the real site

Mirror rendered at localhost:40016 (compare against https://www.ign.com/). Real IGN content images are harvested and stored locally.

Home (IGN Mirror)	Home (IGN)

Review (IGN Mirror)	Review (IGN)

News (IGN Mirror)	News (IGN)

Paired HuggingFace assets PR

https://huggingface.co/datasets/ChilleD/WebHarbor/discussions/41 — uploads ign.tar.gz (instance_seed/ign.db + static/images: 114 local IGN content images).

.assets-revision is temporarily pinned to revision: refs/pr/41 so this GitHub PR builds end-to-end before the assets PR is merged:

repo: ChilleD/WebHarbor
# Pinned to HF PR #41 (adds ign.tar.gz on top of current main) so this
# PR builds end-to-end before the assets PR is merged. This ref contains all
# 17 site tarballs. Maintainer: once
# https://huggingface.co/datasets/ChilleD/WebHarbor/discussions/41 is merged,
# bump this to the merged commit SHA on main.
revision: refs/pr/41

Merge the HF dataset PR before this one so a clean-clone build can fetch the assets, or keep this GitHub PR pinned to refs/pr/41 until the asset PR is merged. Once the dataset PR merges, bump .assets-revision to the merged commit SHA on main.

Verification

./scripts/fetch_assets.sh ign
python -m py_compile sites/ign/app.py sites/ign/seed_data.py sites/ign/content_seed.py sites/ign/_health.py
./scripts/build.sh webharbor:dev
docker run -d --rm --name wh-submit -p 8201:8101 -p 41000-41016:40000-40016 webharbor:dev
curl -s http://127.0.0.1:8201/health
curl -s -X POST http://127.0.0.1:8201/reset/ign
docker exec wh-submit md5sum /opt/WebSyn/ign/instance/ign.db /opt/WebSyn/ign/instance_seed/ign.db
curl -s -X POST http://127.0.0.1:8201/reset-all

Verified:

• ./scripts/fetch_assets.sh ign succeeds from refs/pr/41.
• py_compile passes for app.py, seed_data.py, content_seed.py, and _health.py.
• Docker build succeeds.
• http://127.0.0.1:8201/health returns ok:true with all 17 sites alive.
• Ports 41000-41016 return HTTP 200.
• POST /reset/ign returns ready:true.
• Runtime DB md5 matches seed DB md5: da9379739f89373db60e2c32e801bc60.
• POST /reset-all returns ok:true and every site ready:true.

Notes for human review

• IGN is a broad editorial/media site rather than a commerce catalog, so the main seeded entity is content_items rather than products. The implementation seeds 109 content rows across News, Reviews, Videos, and Guides to preserve browsing depth.
• Search intentionally targets News categories only, matching the current task design and avoiding a leaked type selector. Reviews have their own database-backed controls for sort, score filter, genre filter, review-score visibility, and Editor's Choice filtering.
• Sidebar utility links (Interactive Maps, Discover, Store, Rewards, Privacy Policy, Terms of Use, More) are implemented as local IGN-style replicas. They are visual/navigation surfaces, not real checkout/reward/legal-service integrations.
• Review task hardening was applied after review: tasks that previously leaked answers via listing cards were rewritten to require detail-page fields or state-changing account workflows.
• A submit-blocking syntax issue in content_seed.py was fixed before this PR; the file now compiles and contains 109 seed content items.