adding phrase matching to ebook rag engen #36

Merged
Richie merged 36 commits from feature/adding-phrase-matching-to-ebook-rag-engen into main 2026-07-25 10:17:10 -04:00
Owner
No description provided.
Richie force-pushed feature/adding-phrase-matching-to-ebook-rag-engen from fdb20732b2 to 39ab7358bb 2026-07-09 11:05:10 -04:00 Compare
Richie force-pushed feature/adding-phrase-matching-to-ebook-rag-engen from 659b1e2798 to 1841ac69fa 2026-07-12 13:34:47 -04:00 Compare
Richie force-pushed feature/adding-phrase-matching-to-ebook-rag-engen from 1841ac69fa to 40f707798e 2026-07-12 17:53:49 -04:00 Compare
Richie force-pushed feature/adding-phrase-matching-to-ebook-rag-engen from f250c07ae0 to 19201fa13c 2026-07-16 13:57:10 -04:00 Compare
Richie added 36 commits 2026-07-24 11:39:00 -04:00
Introduce four ORM models and their Alembic migration to support
phrase-based query matching in the ebook RAG engine:

- EbookCandidatePhrase: high-recall phrase candidates extracted per book,
  with source flags (ngram/yake/spacy/capitalized/metadata), scoring, and
  LLM judge results.
- EbookProtectedPhrase: phrases accepted by the LLM judge, with canonical
  id, importance, and nesting controls.
- EbookPhraseAlias: normalized aliases mapping to protected phrases.
- EbookChunkPhraseMention: precomputed phrase occurrences within chunks.

Export the new models from python.orm.richie and add a JSON_DOCUMENT
helper (JSON with JSONB postgres variant) for storing sample contexts.
Refactor protected phrase handling from a single module into a
python/ebook_search/protected_phrases package covering extraction,
storage, and runtime matching. Phrase filtering is now data-driven via
bundled TOML files: ignored_phrases, bad_starts, bad_ends, and
most_common_words.

Add phrase-tuning settings to EbookSearchConfig so candidate generation,
scoring, LLM judging, and matching are configurable rather than hardcoded:
token bounds, entity token limit, raw n-gram min count, frequency and
chapter-spread score thresholds, candidate/LLM/target caps, confidence
threshold, nesting defaults, and the phrase hit boost.
Expose the protected phrase extraction pipeline through the web UI:

- Admin routes: POST /admin/build-phrases, /admin/generate-ngrams, and
  /admin/judge-ngrams, each wrapping the protected_phrases.lib backfill
  helpers, committing on success, rolling back and rendering an error
  partial on failure, and reporting per-book/candidate/mention counts.
- Book detail page: show candidate, judged, and protected phrase counts,
  list top candidate n-grams (with kept/rejected status) and protected
  phrases, and add a POST /books/{id}/recalculate-phrases action that
  clears and regenerates candidates, then redirects back with a status
  message.
- Admin template: add Generate/Judge n-gram buttons.

Also reflows admin.html to 2-space HTML formatting.
- Introduced dataclasses for phrase candidates, judgments, and matches in `models.py`.
- Implemented database operations for candidate and protected phrases in `store.py`, including loading, saving, and deleting phrases.
- Enhanced text normalization functions in `text_normalization.py` with detailed docstrings.
- Refactored search functionality to utilize new models and methods for detecting protected phrases.
Convert the ebook-search web app to async end to end and add concurrency
to the protected-phrase extraction and judging pipeline so large books no
longer block the event loop or the UI.

ORM / infra:
- Add get_async_postgres_engine and factor shared URL/connect_args building
  into build_postgres_url (reused by the sync and async engine builders)
- Add async FastAPI session helpers (get_async_db, AsyncDbSession) with
  expire_on_commit=False to avoid implicit IO under asyncio

App:
- Use AsyncEngine/AsyncSession throughout routes, search, ingest, embeddings,
  answer, rerank and LLM calls; convert handlers to async
- Share a single httpx.AsyncClient in app state for LLM requests; size the
  connection pool for concurrent phrase-judging workers
- Add judge_tasks: run per-book judging as tracked background tasks so a
  book already being judged isn't double-queued

Protected phrases:
- Add a process pool (pool.py) and worker-count config
  (extraction/judge book/phrase workers) to parallelize candidate generation
  and judging
- Split admin actions into all/missing variants for generation and judging

Config:
- Add protected_phrase_extraction_workers, phrase_judge_book_workers,
  phrase_judge_phrase_workers
Add a two-column grid layout for the admin protected-phrases actions
and submit the search form on Enter (Shift+Enter for newline).
Add test_protected_phrases.py covering phrase-matching behavior in the
RAG engine, and update the existing ebook_search tests to use the async
SQLAlchemy engine/session (create_async_engine, AsyncSession) and async
HTTP paths.
Run full-book candidate generation inside worker-owned sessions so each book commits independently during backfills. Abort recalculation when a book has no indexed chapters to preserve existing phrase data, and update admin/UI tests for the new generation flow.
Move protected phrase detection into the retrieval gather so it runs
concurrently with vector and BM25 candidates instead of sequentially
before them. Make the search API accept real bool form fields for
rerank/phrase_matching, gate phrase matching on both the request and
config kill switch, and reflow log f-strings for readability.
feat(chat): add optional response format parameter to request_chat_completion
treefmt / nix fmt (pull_request) Successful in 5s
pytest / pytest (pull_request) Successful in 30s
test ebook search / test-ebook-search (pull_request) Successful in 36s
build_systems / build-brain (pull_request) Successful in 51s
build_systems / build-bob (pull_request) Successful in 52s
build_systems / build-jeeves (pull_request) Successful in 2m23s
build_systems / build-rhapsody-in-green (pull_request) Successful in 1m3s
treefmt / nix fmt (push) Successful in 6s
pytest / pytest (push) Successful in 31s
build_systems / build-brain (push) Successful in 37s
test ebook search / test-ebook-search (push) Successful in 38s
build_systems / build-bob (push) Successful in 40s
build_systems / build-rhapsody-in-green (push) Successful in 53s
build_systems / build-jeeves (push) Successful in 2m11s
8eee5faf72
Richie force-pushed feature/adding-phrase-matching-to-ebook-rag-engen from 19201fa13c to 8eee5faf72 2026-07-24 11:39:00 -04:00 Compare
Richie merged commit 8eee5faf72 into main 2026-07-25 10:17:10 -04:00
Richie deleted branch feature/adding-phrase-matching-to-ebook-rag-engen 2026-07-25 10:17:11 -04:00
Sign in to join this conversation.