Drive pdm.lock and poetry.lock rewrites through one lock_fragments driver instead of two mirrored triads
维护者通常 1 天内回复
已经有一个关联 PR 被合并了。
- #1898 来自 @mikolalysenko —— 已合并
评估
调研方向
Start by reading crates/socket-patch-core/src/utils/lock_fragments.rs, pdm_lock.rs, and poetry_lock.rs, especially the driver, pairing, and equivalence test sections cited in the issue. Then trace the listed call sites in redirect/pdm.rs, redirect/poetry.rs, vendor/pypi_pdm.rs, and vendor/pypi_poetry.rs. Done means one shared driver and pairing helper, equivalence coverage for both formats, and all exact byte tests passing unchanged.
由索引模型根据 Issue 内容生成。
描述
[agent] Filed by the scheduled architecture audit (audit-core). Register: register.
Kind: refactor. Register ID: A05.
Problem
utils/pdm_lock.rs and utils/poetry_lock.rs each carry the same driver around utils::lock_fragments:
- Poetry has
rewrite_poetry_lock/_with_edits/_in/_all/_all_with_version/_batch/_steps: poetry_lock.rs L426-L731. - PDM has the mirror: pdm_lock.rs L136-L353.
- The test modules are mirrored too.
parse_reuse_equivalence_testsandbatch_equivalence_testsexist in both files, and the "fragments changed shape" test exists three times (lock_fragmentsL595,pdm_lockL921,poetry_lock).
The _steps loops have already drifted:
- PDM calls
rewrite.edits()?before its Unchanged check: pdm_lock.rs L333-L340. - Poetry compares the text first: poetry_lock.rs L710-L719.
- This looks unreachable today, but nothing stops it from becoming reachable.
Inside lock_fragments itself, fragment pairing is written three times, and the copies disagree on how uniqueness is counted:
pair_fragmentsL191-L210 counts non-overlapping matches on both sides.- the
finishloop L316-L336 checksoldin the text andnewin the spliced result. rewrite_batchL473-L495 counts overlappingsole_occurrences.- The shape-change length check is repeated at L198, L322 and L477.
Proposed change
- Move the
_steps,_alland_with_editsdrivers intolock_fragments, generic over a small per-format value:plan,mutate,fragments_in. - Delete the per-format wrappers and the
PoetryLockRewrite/PoetryLockParsealiases. - Have
finishandrewrite_batchcall one pairing helper. - Keep one copy of each equivalence test, parameterized over both formats.
Size
Three files (lock_fragments.rs, pdm_lock.rs, poetry_lock.rs), plus call-site renames in redirect/pdm.rs, redirect/poetry.rs, vendor/pypi_pdm.rs and vendor/pypi_poetry.rs. About −250 net lines.
Acceptance criteria
- Every pdm/poetry exact-bytes test passes unchanged, including the batch and parse-reuse equivalence tests (kept once, run for both formats).
- The duplicated shape-change test is kept once, in
lock_fragments. - No user-visible change.
- 主要语言
- Rust
- 星标
- 9
- 派生
- 2
- 平均合并
- 20 小时 47 分钟
- 30 天内合并 PR
- 681
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。
SocketDev/socket-patch 的其他 Issue
-
agent:triaged arch-audit priority:p3 refactor
难度 2/5 1-3 小时 新手友好度 83/100
SocketDev/socket-patch#1860 ·
维护者通常 1 天内回复
-
agent:triaged arch-audit priority:p3 refactor
难度 2/5 1-3 小时 新手友好度 78/100
SocketDev/socket-patch#1838 ·
维护者通常 1 天内回复
-
arch-audit refactor
难度 4/5 3-5 天 新手友好度 54/100
SocketDev/socket-patch#1993 ·
维护者通常 1 天内回复
-
arch-audit refactor
难度 3/5 1-2 天 新手友好度 66/100
SocketDev/socket-patch#1992 ·
维护者通常 1 天内回复
-
agent:triaged arch-audit priority:p3 refactor
难度 4/5 3-5 天 新手友好度 35/100
SocketDev/socket-patch#1987 · 1 条评论 ·
维护者通常 1 天内回复
查看 SocketDev/socket-patch 的全部 Issue
相似的 Issue
-
难度 2/5 1-3 小时 新手友好度 65/100
维护者通常 1 天内回复
-
难度 2/5 1-3 小时 新手友好度 62/100
维护者通常 2 天内回复
-
难度 2/5 1-3 小时 新手友好度 65/100
GiganticMinecraft/seichi-portal-backend#1651 ·
维护者通常 1 天内回复
-
Android visitors are offered the Linux .deb on cap.so's homepage and download page可能已有人在做 @patrickballard 今天认领。 未关闭
难度 2/5 1-3 小时 新手友好度 78/100
CapSoftware/Cap#3677 · 1 条评论 ·
维护者通常 1 天内回复
-
难度 2/5 1-3 小时 新手友好度 72/100
维护者通常 1 天内回复