Repository navigation
Conversation
jeremydixon22
added a commit
that referenced
this pull request
Oct 20, 2026
## Brief Captured GitHub Actions failure logs live outside the normal `checks` snapshot component, so existing cache searches could not find their error text. ## Changes This adds an offline `ci_logs` source with frozen capture selection and verified citations. - Discover saved captures for selected PR snapshots within explicit limits. - Freeze immutable capture references when indexing. Query and retrieval verify source content, PR head, check/run/attempt/job identity and original ranges. - Preserve raw citation bytes while neutralizing ANSI escapes for search. Keep partial captures, omitted regions and unsupported content explicit. - Support the `pr-checks` corpus profile. - Return snapshot counts by default in `cache-info.v2`, with bounded ID pages and checksum-bound continuations. This includes the metadata pagination work from #4 and changes the previous default full-list response. - Remove obsolete proposal bridge/patch copies and point documentation to triage-o-mator’s maintained bridge. Replace compatibility-script rewriting with an explicit pre-index expectation. ## Validation Reposition’s 89-test suite completed with five optional-extra skips; Ruff passed. All 77 parent tests passed with the exact wheel, including capture-to-ranked-retrieval coverage. Independent offline wheel rebuild and compatibility checks passed. A disposable saved-CI trial verified exact citations, frozen selection and corruption refusal with zero GitHub requests. > The current version identifies the downstream build vendored by triage-o-mator. Local-verification notes and reports remain outside this change. --------- Co-authored-by: Jeremy Dixon <jeremydixon22@users.noreply.github.com>
Member
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
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.
Why
General performance improvement.
On the 4,932-item Omarchy corpus, search-info returned about 332 KB because cache-info.v1 includes every indexed snapshot ID. Most metadata inspections need the count; callers that need IDs could retrieve them in pages.
What changed
• cache-info.v2 returns indexed_snapshot_count and omits indexed_snapshots by default.
• --snapshot-ids-limit and --snapshot-ids-offset return sorted ID pages, capped at 100 IDs each.
• The index manifest still retains the complete, checksummed ID list.
• Documentation and a synthetic 4,932-ID regression test cover the new response. A one-line Ruff cleanup is included.
Compatibility
This changes the response contract. Consumers that read the full indexed_snapshots list from cache-info.v1 must request pages under v2. No checked-in consumer relies on that default list, but external consumers may. The list comes from Reposition’s index manifest.
Validation
The full test suite passed (80 tests; 5 optional tests skipped), and
ruff check .as well.