content-hash-cache-pattern

Cache expensive PDF text extraction results by using SHA-256 content hashes as cache keys. This approach stores file processing outputs in a frozen CacheEntry dataclass, enabling stable reuse across runs without re-parsing unchanged files.

A content-hash cache key is a SHA-256 hash generated from file contents to store and retrieve processed results. It decouples caching from file paths, ensuring cache validity persists across file moves and renames while auto-invalidating on content changes.

Path-based cache keys break on file moves because the path changes. Using SHA-256 content hashes as cache keys provides path independence, allowing your PDF and text extraction caches to survive renames and automatically invalidate only when content changes.

Auto-invalidate cached file extraction data by switching to content-hash keys derived from SHA-256. When a file's content changes, its hash changes, naturally bypassing the old cache entry and triggering fresh processing without manual cleanup.

Yes, you can wrap existing pure processing functions like extract_text with a service-layer wrapper such as extract_with_cache. This separation of concerns applies content-hash caching to PDF, image, and text extraction pipelines without modifying core processing logic.