refactor(providers): share one lease table across cloud device providers
维护者通常 1 天内回复
还没有人认领这个 Issue。
评估
- 难度
- 4/5
- 预计耗时
- 3-5 天
- 新手友好度
- 48/100
- Issue 类型
- 重构
- 描述清晰度
- 基本清楚
- 活跃度
- 活跃
- 技术栈
- typescript
- 领域
- backend
调研方向
Start by reading the lease bookkeeping in provider-limrun and the review findings in #4092 to understand the existing rules. Extract the shared lease table, migrate limrun without behavior changes, then move provider-revyl to it and remove duplicate bookkeeping. Run the relevant provider tests; done means both providers share the table and the typed cleanup-pending outcome is used when a stop is unconfirmed.
由索引模型根据 Issue 内容生成。
描述
provider-limrun and the new provider-revyl (#4092) each keep their own lease state: allocate once per lease, track in-flight allocations, release only after the provider confirmed the stop, drain on shutdown, bound work by the caller's deadline, and recover expired leases. Each provider re-implements these rules, so each one can get a different rule wrong. In #4092, most review findings were about this bookkeeping, not about Revyl itself.
Proposal: extract one provider lease table that owns these rules, with a typed provider-cleanup-pending outcome when a stop is not confirmed. A provider would supply only start, stop by lease, find by lease, and its interactor.
Order:
- Extract the table from
provider-limrunand migrate limrun onto it with no behavior change. - Move
provider-revylonto the same table. - Delete the per-provider lease bookkeeping.
Decided on #4092: Revyl lands first, and this refactor follows.
- 主要语言
- TypeScript
- 星标
- 5.3k
- 派生
- 377
- 平均合并
- 13 小时 49 分钟
- 30 天内合并 PR
- 650
环境准备
- 没有 Dockerfile 或 Docker Compose 文件
- 没有 Pull Request 模板
- 阅读贡献指南
从这里开始
- 先读完整个 Issue，再读项目的贡献指南。
- 在 Issue 下留言说明你要接手 —— 这能避免两个人做同样的事。
- Fork 仓库，在一个分支上完成修改。
- 提交 Pull Request，并在描述里引用这个 Issue 编号。
callstack/agent-device 的其他 Issue
-
bug ready-for-agent
难度 2/5 1-3 小时 新手友好度 90/100
callstack/agent-device#4212 · 1 条评论 ·
维护者通常 1 天内回复
-
iOS alert wait: the last poll can get ~30 ms and turns "no alert" into a failed read可能已有人在做 @pcbeingused333 于 5 天前认领。 未关闭bug ready-for-agent
难度 2/5 1-3 小时 新手友好度 77/100
callstack/agent-device#4018 ·
维护者通常 1 天内回复
-
难度 2/5 1-3 小时 新手友好度 68/100
callstack/agent-device#1869 ·
维护者通常 1 天内回复
-
iOS 27: HealthKit's authorization sheet reads as one node, so label and ref presses can't reach it未关闭
难度 4/5 3-5 天 新手友好度 48/100
callstack/agent-device#4226 ·
维护者通常 1 天内回复
-
难度 4/5 3-5 天 新手友好度 70/100
callstack/agent-device#4216 ·
维护者通常 1 天内回复
查看 callstack/agent-device 的全部 Issue
相似的 Issue
-
unwrapDatabaseError masks real DB errors with TypeError: Cannot convert object to primitive value未关闭
难度 2/5 1-3 小时 新手友好度 74/100
NEARBuilders/everything-dev#798 ·
维护者通常 1 天内回复
-
难度 2/5 1-3 小时 新手友好度 78/100
vercel-labs/durable-harness#53 ·
维护者通常 1 天内回复
-
[Bug]: Redesign tab bars ignore clicks on Linux: hidden chat layers keep window drag regions可能已有人在做 关联的 PR 仍在进行中或已合并。 未关闭
难度 2/5 1-3 小时 新手友好度 82/100
维护者通常 1 天内回复
-
experiência: a doc manda subir para o Micro do Supabase, mas não diz que exige plano pago nem quanto custa可能已有人在做 关联的 PR 仍在进行中或已合并。 未关闭area/docs area/ui enhancement
难度 1/5 1-3 小时 新手友好度 88/100
melgarafael/DeskcommCRM#4098 ·
维护者通常 1 天内回复
-
dependencies platform: ios
难度 1/5 1-3 小时 新手友好度 78/100
software-mansion/react-native-executorch#1617 ·
维护者通常 1 天内回复