chore: re-vendor the canonical baseline from the current central bundle - #156
Merged
Conversation
The vendored copy had drifted eight migrations behind the assembled bundle, and its grants file predated three grant fixes. CanonicalBaselineContractTest was failing on any branch here, which blocked this repository's CI regardless of what the change was. Re-vendored per db/canonical-baseline/README.md: the assembled bundle copied verbatim, baseline.manifest and baseline.sha256 regenerated. Nothing was hand-edited. The result is checkable rather than asserted: the regenerated baseline.sha256 is 50ff03a9…, which equals canonical_baseline_sha256 already recorded in the root bundle's source-revisions.json at root develop 8be256a. Both sides now agree on the same digest, computed independently. What actually differed, after normalising CRLF the way the manifest does: - eight migrations absent here entirely, from backtest run outcome detail through the production RSI timeframe seeds - R__database_runtime_grants.sql, which had none of backend #241, #246 or #249 — the grants for the batch role's writes and the backtest role's operations and bot reads Worth recording because it misled me first: a raw byte comparison reported all 45 shared files as different. That was line endings alone. The manifest normalises CRLF to LF for exactly this reason, and comparing the same way reduced the real difference to one file plus the eight absent ones. I had told the trading-engine issue that a single file needed copying; that was wrong, and this commit is the correction. Verified: CanonicalBaselineContractTest passes, and the full ./gradlew test suite passes in 7m22s. This repository moved to kcrmin in the 2026-08-09 two-owner split, which is why the re-vendor no longer waits on another owner. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
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.
trading-engine #155 를 닫습니다.
무엇이 어긋나 있었나
벤더 사본이 조립된 번들보다 마이그레이션 8개 뒤처져 있었고, grants 파일은 grant 수정 3건 이전 상태였습니다.
그래서 이 저장소의
CanonicalBaselineContractTest가 어떤 브랜치에서든 실패해 CI 가 막혀 있었습니다.db/canonical-baseline/README.md절차대로 재벤더링했습니다 — 조립된 번들을 그대로 복사하고baseline.manifest·baseline.sha256을 재생성했습니다. 손으로 편집한 파일은 없습니다.결과는 주장이 아니라 검증됩니다
재생성한
baseline.sha256=50ff03a9…이고, 이는 루트develop8be256a의source-revisions.json에 이미 기록된canonical_baseline_sha256과 같습니다. 양쪽이 독립적으로계산해 같은 digest 에 도달합니다.
실제 차이 (manifest 방식대로 CRLF 정규화 후)
timeframe seed 까지
R__database_runtime_grants.sql— backend #241·#246·#249 가 하나도 반영되지 않은 상태였습니다(배치 역할의 쓰기 권한, backtest 역할의
operations·bot읽기)저를 속인 것도 기록해 둡니다
바이트 단위 비교는 공통 45개 전부가 다르다고 보고했습니다. 줄바꿈 차이뿐이었습니다. manifest 가
CRLF→LF 를 정규화하는 이유가 정확히 이것이고, 같은 방식으로 비교하니 실제 차이는 1개 + 없는 8개로
줄었습니다. 이슈에 "복사할 파일은 하나뿐" 이라고 적었던 것은 틀렸고 이 커밋이 그 정정입니다.
검증
이 저장소는 2026-08-09 2인 재분배로
kcrmin소유가 되었고, 그래서 이 재벤더링이 더 이상 다른 소유자를기다리지 않습니다.
🤖 Generated with Claude Code