kert 1b81aeac92 feat(llm): in-page PDF viewer for comment and article attachments — /ui/pdf/<key>, /pdf/<key>[/file], self-hosted PDF.js, viewer links on chat sources and search hits ...

Rules keep their federalregister.gov links; this is for the files the
library holds itself: a regulations.gov comment's downloaded
attachments (.state/comments/<docket>/<id>/attachment_N.*), an item's
bib attachments (data/storage/<att>/<name>) and Zotero-only storage
PDFs (zotero.sqlite read immutable, one indexed lookup per request).

llm.pdfs lists an item's attachments from those three places (comment
first, de-duplicated by name) and resolves a file by *name* only,
refusing anything outside the storage roots. GET /pdf/<key> lists
them (name, size, source, renderable, media type); GET
/pdf/<key>/file?name= serves a PDF inline with range requests and
other formats as downloads. /ui/vendor/<name> serves the vendored
pdf.js 4.10.38 (same-origin worker; no CDN dependency).

/ui/pdf/<key>?file=&page=&q=: continuous scroll, pages rendered lazily
(IntersectionObserver, ~1.5 screens ahead) at device pixel ratio,
re-fit on resize/orientation (ResizeObserver, debounced), fit-to-width
and zoom, page input with hash tracking, a text layer for selection,
whole-document find with highlights (seeded from the cited snippet),
keyboard shortcuts, a file selector when an item has several, and a
download fallback when a file is not a PDF or cannot be rendered.

as_source carries attachment + page from the chunk metadata; the chat
sources and search hits show a 'PDF p.N' link into the viewer for
comment/corpus sources. compose: the llm service mounts
./.state/comments read-only.

2026-09-24 18:56:10 -04:00

..

__init__.py

feat(llm): module scaffold, stack[llm] extra, [llm] config (refs #563)

2026-07-17 08:15:31 -04:00

golden_lineage.yaml

fix(pfs,llm): descriptor stems trimmed to the code's own parenthetical (#700: G0502/G0503 no longer bridge into CCM); excerpts using a family acronym for something else are dropped (CCM = cardiac contractility modulation in the CY2027 NPRM) (refs #700 #691)

2026-09-10 17:48:30 -04:00

golden_themes.yaml

feat(llm): golden-set gate for theme tagging — llm.tageval, tests/llm/golden_themes.yaml, stack llm eval-tags; vocab v2 (refs #577)

2026-09-22 16:58:41 -04:00

test_api.py

feat(llm): in-page PDF viewer for comment and article attachments — /ui/pdf/<key>, /pdf/<key>[/file], self-hosted PDF.js, viewer links on chat sources and search hits

2026-09-24 18:56:10 -04:00

test_chunk.py

fix(llm,pfs): thread one code_index into chunk_doc so index and restamp agree (F1)

2026-09-09 20:57:43 -04:00

test_classify.py

test: cover P49 degrade paths so the 99% coverage gate holds (refs #721)

2026-09-11 15:37:56 -04:00

test_client.py

feat(llm,notebooks): httpx-only LlmClient for notebooks and an llm_search example notebook (refs #573)

2026-09-11 17:26:54 -04:00

test_config.py

feat(llm): lineage evidence — timeline events, element diffs and guidance as a lineage SSE event + cited prompt block (refs #691)

2026-09-10 10:26:34 -04:00

test_evidence.py

test: cover P49 degrade paths so the 99% coverage gate holds (refs #721)

2026-09-11 15:37:56 -04:00

test_golden.py

fix(ci,llm): golden runner exits 0 when filing issues; workflow keeps the report on failure; cmd_run tests (refs #692)

2026-09-10 12:12:59 -04:00

test_index_refs.py

fix(llm,pfs): thread one code_index into chunk_doc so index and restamp agree (F1)

2026-09-09 20:57:43 -04:00

test_index.py

fix(llm): mock _code_index in index_docs unit tests

2026-09-09 21:03:46 -04:00

test_init_lazy.py

feat(llm,notebooks): httpx-only LlmClient for notebooks and an llm_search example notebook (refs #573)

2026-09-11 17:26:54 -04:00

test_lineage.py

feat(pfs,llm): resolve MLN products/articles to bib items; locate IOM section pages in chapter PDFs (refs #705)

2026-09-11 17:51:20 -04:00

test_links.py

feat(llm): in-page PDF viewer for comment and article attachments — /ui/pdf/<key>, /pdf/<key>[/file], self-hosted PDF.js, viewer links on chat sources and search hits

2026-09-24 18:56:10 -04:00

test_metrics.py

feat(llm,infra): OTel metrics for dispatch/embeddings/chat/indexing + LLM Grafana dashboard (refs #579)

2026-09-11 18:54:27 -04:00

test_migrate.py

fix(llm): restamp must not rebuild the HNSW index (F6)

2026-09-09 21:03:56 -04:00

test_pages.py

feat(pfs,llm): resolve MLN products/articles to bib items; locate IOM section pages in chapter PDFs (refs #705)

2026-09-11 17:51:20 -04:00

test_pdfs.py

feat(llm): in-page PDF viewer for comment and article attachments — /ui/pdf/<key>, /pdf/<key>[/file], self-hosted PDF.js, viewer links on chat sources and search hits

2026-09-24 18:56:10 -04:00

test_pool.py

fix(llm): drop a pool host that dies mid-run and retry the batch on the survivors (refs #796)

2026-09-22 15:52:07 -04:00

test_rag.py

test: cover P49 degrade paths so the 99% coverage gate holds (refs #721)

2026-09-11 15:37:56 -04:00

test_rerank.py

feat(llm): recency-blended re-ranking of retrieval hits

2026-09-03 10:32:14 -04:00

test_restamp.py

test(llm): restamp pagination-loop mechanics with a fake engine (F8)

2026-09-09 21:04:52 -04:00

test_search.py

feat(llm): /search and /similar retrieval endpoints with metadata filters and pagination (refs #570)

2026-09-11 17:17:07 -04:00

test_source_refs.py

feat(llm): stack llm index --key also limits the corpus collection

2026-09-09 21:28:04 -04:00

test_source.py

feat(rex,pfs): EPUB attachments readable by the corpus indexer (F2)

2026-09-09 21:11:04 -04:00

test_tageval.py

feat(llm): golden-set gate for theme tagging — llm.tageval, tests/llm/golden_themes.yaml, stack llm eval-tags; vocab v2 (refs #577)

2026-09-22 16:58:41 -04:00

test_tagger.py

fix(llm): the tagging judge drops a host that dies mid-run and retries on the next largest, like embed_texts (refs #575, #796)

2026-09-22 17:11:31 -04:00

test_vocab.py

feat(llm): closed theme vocabulary for comment tagging — themes.yaml v1, llm.vocab loader, stack llm vocab, P35 design spec (refs #574)

2026-09-22 16:48:51 -04:00