feat(pipeline): report when a re-ingest disagrees with the page on disk (#104) #129

Merged

mathias merged 2 commits from fix/104-surface-merge-conflicts-v2 into main 2026-09-23 05:22:28 +00:00

Conversation 1 Commits 2 Files Changed 5

+190 -6

2 Commits

Author	SHA1	Message	Date
mathiasandClaude Sonnet 5	b05e9c48f4	fix(wiki): canonicalize whitespace across a whole section, not just its ends ... Review of PR #129 found MergeReport's disagreement check compared strings.TrimSpace(sA.content) against strings.TrimSpace(sB.content) -- only leading/trailing whitespace is tolerated. Every other content comparison in this file (appendNew, mergeBullets) already canonicalizes internal whitespace runs too, because a re-extraction's exact spacing isn't byte-stable (#103). A mid-line double-space would false-positive as a "disagreement" here even though the file's own established tolerance says it shouldn't. canonicalBlock applies the existing canonicalLine per line of a multi-line section body. A genuine content difference is still caught (TestMerge_ReportsSectionsWhereTheFreshVersionWasDiscarded stays green); only internal-whitespace-only variants stop false-positiving. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>	2026-09-23 07:20:40 +02:00
mathias	719eb8c2d6	feat(pipeline): report when a re-ingest disagrees with the page on disk (#104) ... The asymmetry that makes unattended ingest safe (#103) also discarded a real signal. When a fresh extraction produced a different version of a section already on disk, the on-disk version won and the model's version went in the bin without a word — so "the source is unchanged" and "the source changed its mind and we ignored it" were indistinguishable. wiki.MergeReport now returns the headings where the incoming version disagreed and lost, and the ingest run reports them as warnings, which AppendLog already writes into log.md. This is the simplified answer to contradiction flagging at ingest. The issue proposed calling brain-gardener's Judge from the ingest path, which needs either a shared Go module across two repos or an HTTP service that does not exist — and puts an LLM call on the write path, where a timeout must neither fail the ingest nor silently mark nothing as contested. None of that is necessary to get the signal. The merge already computes the disagreement; it was throwing it away. Surfacing it costs nothing, needs no model, no cue-word matcher, no cross-repo dependency, and is deterministic. Deliberately textual disagreement, not semantic contradiction. Deciding whether a disagreement is a genuine contradiction stays with the Judge-gated enrich pass that already runs hourly and is approval-gated by design (brain-gardener#175). Detection here, adjudication there. Merge is kept as a thin wrapper over MergeReport so existing callers are untouched. Refs #104, #103, brain-gardener#175, brain#21.	2026-09-23 07:11:53 +02:00