Repository navigation
feat(index): E26 file-level chunk augmentation behind flag (default OFF) - #73
Merged
Merged
Conversation
Phase 1 of the approved stack migration. New pure helper chunk_augment.py (FILE/SYMBOLS/DOC = frozen E26 template, parity 27/27); single wiring point in index_parser.parse_one_file covering AST+fallback paths; MSCODEBASE_AUGMENT_CHUNKS default off (byte-identical output). Flip changes chunk_hash=sha256(text) so vectors recompute instead of going stale. Tests: test_chunk_augment.py 10/10 (template pin, flag on/off, wiring both modes); adjacent suites 47/47. No prod reindex triggered.
|
Important
This repository does not receive automatic reviews because it has fewer than 10 stars. ⚙️ Run configuration
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
Phase 2 of the approved stack migration (E25). New pure ondemand_rerank.py (cosine, stable ties, skips text-less); MSCODEBASE_ONDEMAND_RERANK (+TOP_N) config mirroring late_enrichment; hybrid hook reranks top-N fused pool with live embeddings INSTEAD of BGE only when flagged. Default OFF: prod path byte-identical. Tests: test_ondemand_rerank.py 6/6 exact math on fake vectors; adjacent 36/36. Live proof with flag: quality h1 3/10 h5 4/10 = baseline, rerank_timing.mode=ondemand_cosine in 1041ms; fast untouched.
Phase 3 as a measurement, not a deletion. Flag skips the whole dense stage (query embedding + LanceDB vector search); default OFF so the prod path is byte-identical. EXP-28: frozen 33-case panel (sha256 f31e594e, union of three pre-existing frozen sets, frozen before the run), 2 runs per arm, same session and index. A dense ON: hit1 11/33, hit5 15/33, 0 errors, mean 3359/3534ms. B dense OFF: hit1 12/33, hit5 16/33, 0 errors, mean 3163/2492ms. Per case 2 better / 1 worse / 30 same; speedup mean-of-means 18% but per-run spread 5.8-29.5%. Verdict: H1/H2/H6 refuted as expected, H4 confirmed (0 errors), H3 PARTIAL, H5 untested (no symbol/NL labelling in the panel). Harness proven by a _vector_search_async call counter 1 -> 0 with the embedder alive. Key finding: of 52 table-API call sites in 13 src files only 5 are vector search; bm25.py:93 builds BM25 from table.to_pandas(), so LanceDB is the chunk primary store, not a vector index. Dense removal NOT recommended; the flag stays as an operational kill-switch. KNOWN_ISSUES untouched (foreign session edits it) - proposals in experiments/e28_denseoff/PENDING_LEDGER.md. Tests: 3 flag tests + adjacent suites 29/29, ruff clean.
…up claim EXP-28 left H5 untested: no symbol-vs-NL labelling, and it could not be added after the run (freeze-before-look). This experiment freezes the labelling rule first (snake_case | CamelCase | pathy | single token -> symbolic 15 / NL 18, sha256 7639ba1b), then runs both arms twice. A dense ON : NL 3/18 (16.7%), symbolic 8/15 (53.3%), 0 errors, mean 2559/2555ms B dense OFF: NL 5/18 (27.8%), symbolic 7/15 (46.7%), 0 errors, mean 2375/2354ms NL 2 better / 0 worse; symbolic 0 better / 1 worse (TR-P2). Speedup stable at 7.2% / 7.9%. Verdicts: H1 CONFIRMED (symbolic loses exactly 1 of 15, within the declared tolerance), H2 REFUTED as expected (NL is better), H3 REFUTED as expected (7.5%, below 15%), H4 CONFIRMED (0 errors), H5 CONFIRMED against expectation (two NL cases improved). CORRECTED: EXP-28 published an 18.0% speedup; two runs here show 7.2% and 7.9%. The old number was inflated by one noisy run (21% spread vs 0.2% here). Logged as a new entry with the supersede reference, not a silent edit. Harness incident: the first run from the worktree printed hit1=0/33 with rc=0 and 167ms - a silent zero on an empty index (index is per-project-path). Guard added: all-empty exits 2, and the project root is an explicit manifest parameter. That run is not counted. Phase 3 decision unchanged: no observed harm from dense, no proven gain (1-2 of 33 cases is noise), LanceDB stays as chunk primary store, the flag stays a kill-switch.
feat(search): MSCODEBASE_DENSE_OFF kill-switch + EXP-28 dense-off A/B
feat(search): on-demand cosine rerank path behind flag (Phase 2)
# Conflicts: # AGENT_DIARY.md
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What
Phases 1–3 of the search-stack migration, all three behind default-OFF flags, so the production path is byte-identical without any env var:
src/core/indexing/chunk_augment.py): FILE/SYMBOLS/DOC prefix on chunks, parity 27/27 with the frozen E26 template, flagMSCODEBASE_AUGMENT_CHUNKS.src/core/search/ondemand_rerank.py): top-N fused candidates re-scored with live embeddings instead of stored vectors + BGE, flagMSCODEBASE_ONDEMAND_RERANK.MSCODEBASE_DENSE_OFFskips the dense stage, plus EXP-28/EXP-29 which decided the question.Phase 3 conclusion: LanceDB stays. Quality without dense is unchanged (net +1 of 33 = noise) and the dependency analysis showed LanceDB is the chunk primary store, not a vector index —
bm25.py:93builds the BM25 index fromtable.to_pandas(), and only 5 of 52 table-API call sites in 13src/files are vector search. Removing it is a store migration, not cleanup.Measured results
Two corrections worth reading:
Verification
2052 passed / 0 failed.main): ubuntu, windows, clean-state, manifest-parity, transport-equiv-e2e, CodeQL and the Analyze matrix — all green.origin/mainwas merged in to resolve the conflict; the E24 fast-fallbacktry/exceptand theMSCODEBASE_DENSE_OFFcondition were checked together by hand (auto-merge kept both correctly).Safety
Every flag defaults to OFF and reloads from env through the same
field(default_factory=...)pattern as the existinglate_enrichmentflag. No reindex is required.KNOWN_ISSUES.mdwas deliberately not touched (another session has uncommitted edits) — proposals live inexperiments/e28_denseoff/PENDING_LEDGER.md.Notes
pull_request → main. This PR is the single landing point.experiments/e2{5,6,7,8,9}_*/(frozen panels with sha256, raw results, verdict tables).