Skip to content

pullrequests Search Results · language:Edge language:TypeScript language:HTML language:Python language:Python language:Ruby

Filter by

28.4M results  (2 s)

28.4M results

This PR contains the following updates: | Package | Change | Age | Confidence | | --- | --- | --- | --- | | eslint-plugin-prettier | 5.5.5 → 5.5.6 | age | confidence | Release Notes details summary ...

Introduce a SimplePhaseChecker debug utility in python/sglang/srt/utils/phase_checker.py: a Triton-kernel-based assertion helper that verifies an execution phase counter matches an expected value at runtime, ...

Summary - Bug 1.1 + 10.1 — addUnit was completely absent from the doPost switch, so the Add Unit form always returned a 404-style error. Added the case addUnit dispatch and the full addUnit() function ...

What Resolves the charting decision (recharts, per design review). Adds token-themed line / bar / area charts for arbitrary in-memory business data — the gap the audit flagged (charts were Prometheus-only). ...

phase-13: Service Foundations II · API Data → main API 데이터 ergonomics(Tier 2). pagination·idempotency·typed client·object storage·outbox·factory. 보일러플레이트 품질 트랙 2번째 phase. 📦 Scope: 계획 vs 실제 | 구분 | ...

요약 round6 eligible 풀 재사용(재스크레이프 0). 성인/YA 성숙 필터 유지 + 내용 점검으로 American Born Chinese(Printz=YA, 인종 캐리커처) 제외. - 원서 680 → 767권 (신규 87, 덮어쓰기 0), 총 books 5200 → 5287 - 저밴드 보강: band0:18 / band1~5:14 (ABC ...

This PR contains the following updates: | Package | Type | Update | Change | | --- | --- | --- | --- | | rubocop-hq/rubocop | repository | minor | v1.86.2 → v1.87.0 | Note: The pre-commit manager in ...

!-- I want to review in Japanese. -- 概要 OAuthのプロバイダー情報が不整合を引き起こさないよう、oauth_providers テーブルに (provider, uid) の一意制約(ユニークインデックス)を追加しました。 詳細な変更点 - [x] マイグレーションファイルを作成し、oauth_providers テーブルの provider, ...

…nger button class - Warehouse Bin Prep was rendering ALL active warehouse rows because renderWHTab s dispatcher had no whTab=== binprep branch -- it fell through to the default rows = d.all. The ...

为什么 Polish guard 评测体系有一道承重的循环论证:gold 金标准里 53% (1118/2096) 的行是由 guard 自己的函数 + 同一份 lexicon 判定的(_is_destutter_only、ascii 去空格豁免、vocab 白名单)。记分牌随后又拿同一个 guard 去对这份 gold 打分——在这些行上 guard == gold 是构造出来的恒等式,量的是 ...