# Recipe page snapshots

This folder contains raw HTML snapshots and metadata for public recipe pages, saved for offline inspection and to support unit tests without hitting third-party sites.

Generated by:
- `python -m scripts.save_recipe_pages` (downloads pages listed in `scripts/recipe_urls.py`)
- `python -m scripts.generate_expected_from_snapshots` (parses HTML into structured recipe JSON)

File layout per URL:
- `<slug>.html` — raw page HTML (only for successful 2xx responses)
- `<slug>.meta.json` — metadata with original URL, final URL, status code, fetch time, and any error
- `<slug>.recipe.json` — expected parsed Recipe model (Pydantic-serialized, by_alias=True). This is the canonical expectation file used by offline tests.

Notes:
- Some domains may block requests (e.g., 403) or fail TLS verification. In those cases, only the `.meta.json` is written.
- Keep `scripts/recipe_urls.py` curated to stable, public URLs to minimize churn.