Repository navigation
Route the remaining hand-rolled child-process deadlines through utils::process #1067
Description
Activity
- addedarch-auditFiled by a scheduled architecture audit routine (see the architecture review discussion)Filed by a scheduled architecture audit routine (see the architecture review discussion)refactorStructural change: duplicated code or logic, missing abstraction, layering, dead codeStructural change: duplicated code or logic, missing abstraction, layering, dead code
on Oct 7, 2026 mikolalysenko commented
on Oct 7, 2026 CollaboratorAuthorMore actions[agent] Triage: priority:p3 (cross-cutting refactor). No duplicate or open PR found.
Generated by Claude Code
mikolalysenko commented
on Oct 8, 2026 CollaboratorAuthorMore actions[agent] Claiming this issue for the architecture refactor routine (highest leverage: the only refactor slice whose files no open PR changes; routes the PDM site probe through
utils::process::output_withinand adds the guard). Branch: arch-refactor/1067-pdm-probe-deadline. Claim-ID: 2026-10-08T02:56:01Z-9ee976Slice taken:
crawlers/python_crawler/pdm_site.rsplus a guard test that rejects newkill_on_dropspawns outsideutils::process, withvendor/npm_dir.rson its pending list. Thenpm_dirgit exchange remains for a later slice, because #1026 changesnpm_dir.rsandutils/process.rs.
Generated by Claude Code
mikolalysenko commented
on Oct 8, 2026 CollaboratorAuthorMore actionsmikolalysenko commented
on Oct 11, 2026 CollaboratorAuthorMore actions[agent] Janitor: releasing the stale claim
Claim-ID: 2026-10-08T02:56:01Z-9ee976. Its PR #1106 merged on 2026-10-08 as thepdm_siteslice only; thevendor/npm_dir.rsgit exchange still has a hand-rolled deadline. No follow-up PR exists and the claimer has been silent since 2026-10-08 03:01Z. The issue stays open and any fixer can claim it.
Generated by Claude Code
[agent] Filed by the October 7 architecture audit campaign (core). Register: arch-audit register.
Kind: refactor. Source: C48 remainder (#845 closed by #886 slice 1), register C48.
Problem: #886 added
utils::process::output_within(10 s probe deadline) and closed #845, but hand-rolled spawn deadlines remain:pdm_site.rs#L19keeps its own 5 sPROBE_TIMEOUTwithtokio::time::timeoutandkill_on_drop(added by #611), andnpm_dir.rs#L426and#L1220-L1225keep 30 s and 60 s git exchanges. The asyncCommandRunnerand the architecture guard from #886's plan were never tracked.Impact: probe deadlines and kill policy drift again; the refactor notes forbid new
timeout+kill_on_dropsites but nothing enforces it.Proposed change: route
pdm_siteand thenpm_dirgit exchange through the shared runner (an async variant if needed) and delete the local deadlines; add an architecture test that rejects newkill_on_dropspawns outsideutils::process.Acceptance criteria:
tokio::time::timeoutaround a child process outsideutils::process.Dependencies: none.
Generated by Claude Code