Phase 5: the 56 lint findings Phase 4 left behind #911

Closed
opened 2026-10-27 13:38:15 +00:00 by sam · 0 comments
Owner

Discovered as a side effect of the 6d4e0be audit (archie-core-ifi), not as its
goal: running golangci-lint over the two packages I had touched showed findings
in files nobody had flagged, so I ran it repo-wide.

All four children of archie-core-qa4 are closed and the epic reports "4/4
complete (100%) -- eligible for close", but golangci-lint run ./... still
reports 56 issues. The epic must not be closed on that prompt; its own
description says the gate is RED until it closes, so closing it early declares
a green gate that does not exist.

Breakdown: noctx 11, gocognit 11, gocritic 6, errcheck 6, nilerr 4, recvcheck 3,
nestif 3, godox 3, sloglint 2, predeclared 2, funlen 2, gofumpt 1, cyclop 1,
staticcheck 0 (the one hit, configtemplate.go:6, is fixed).

The gocognit remainder is concentrated in exactly the functions Phase 4 named as
deferred -- cmd/archied/main.go run (106, the description predicted "108->30
remaining" and it did not land), internal/workflow AgentStage.Stage (68),
Feasibility and TDD (funlen), cmd/archie-agent startMCPProviders (45) -- plus
the whole of internal/tools/builtin: grepFallback 58, runFindGoFallback 57,
makeReadExecutor 56, makeGrepExecutor 41, makeShellExecutor 40, makeWriteExecutor
38, suggestPaths 36. tomlwrite.Apply (82) is new, uncommitted-until-7fab597 code.

READ archie-core-qa4's notes before starting. Both known behaviour regressions
from this epic came out of structural extraction, and every remaining gocognit
site is a structural extraction waiting to happen. The nilerr findings in
grep.go are not noise either -- see archie-core-ts9, where the same swallow in
find.go is a live false-negative bug. Fixing nilerr by "returning the error"
without deciding what the caller does with it will change tool behaviour.

Acceptance Criteria

golangci-lint run ./... reports 0 issues, or every remaining finding has an explicit nolint with a reason. Each structural extraction lands with a behavioural test for the function it splits, per the discipline recorded in archie-core-qa4's notes.

Close reason

Stale as of 2026-08-08's 56-issue count; verified today golangci-lint run ./... reports 0 issues (same command Taskfile's lint target runs), thresholds unchanged (cyclop max-complexity 15, gocognit min-complexity 30 in .golangci.yml), no threshold loosening. All 21 remaining //nolint suppressions (nilerr x14, contextcheck x4, others) carry a documented reason in an adjacent comment. Acceptance criteria met.


Imported from beads archie-core-1786637499814-382-b27575d6 · GitHub #403 · type task · priority P2 · created 2026-08-08 · closed 2026-09-02

Discovered as a side effect of the 6d4e0be audit (archie-core-ifi), not as its goal: running golangci-lint over the two packages I had touched showed findings in files nobody had flagged, so I ran it repo-wide. All four children of archie-core-qa4 are closed and the epic reports "4/4 complete (100%) -- eligible for close", but `golangci-lint run ./...` still reports 56 issues. The epic must not be closed on that prompt; its own description says the gate is RED until it closes, so closing it early declares a green gate that does not exist. Breakdown: noctx 11, gocognit 11, gocritic 6, errcheck 6, nilerr 4, recvcheck 3, nestif 3, godox 3, sloglint 2, predeclared 2, funlen 2, gofumpt 1, cyclop 1, staticcheck 0 (the one hit, configtemplate.go:6, is fixed). The gocognit remainder is concentrated in exactly the functions Phase 4 named as deferred -- cmd/archied/main.go run (106, the description predicted "108->30 remaining" and it did not land), internal/workflow AgentStage.Stage (68), Feasibility and TDD (funlen), cmd/archie-agent startMCPProviders (45) -- plus the whole of internal/tools/builtin: grepFallback 58, runFindGoFallback 57, makeReadExecutor 56, makeGrepExecutor 41, makeShellExecutor 40, makeWriteExecutor 38, suggestPaths 36. tomlwrite.Apply (82) is new, uncommitted-until-7fab597 code. READ archie-core-qa4's notes before starting. Both known behaviour regressions from this epic came out of structural extraction, and every remaining gocognit site is a structural extraction waiting to happen. The nilerr findings in grep.go are not noise either -- see archie-core-ts9, where the same swallow in find.go is a live false-negative bug. Fixing nilerr by "returning the error" without deciding what the caller does with it will change tool behaviour. ### Acceptance Criteria golangci-lint run ./... reports 0 issues, or every remaining finding has an explicit nolint with a reason. Each structural extraction lands with a behavioural test for the function it splits, per the discipline recorded in archie-core-qa4's notes. ### Close reason Stale as of 2026-08-08's 56-issue count; verified today golangci-lint run ./... reports 0 issues (same command Taskfile's lint target runs), thresholds unchanged (cyclop max-complexity 15, gocognit min-complexity 30 in .golangci.yml), no threshold loosening. All 21 remaining //nolint suppressions (nilerr x14, contextcheck x4, others) carry a documented reason in an adjacent comment. Acceptance criteria met. --- _Imported from beads `archie-core-1786637499814-382-b27575d6` · GitHub #403 · type task · priority P2 · created 2026-08-08 · closed 2026-09-02_
sam added the mediumtask labels 2026-10-27 13:38:15 +00:00
sam self-assigned this 2026-10-27 13:38:15 +00:00
sam closed this issue 2026-10-27 13:38:15 +00:00
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: sam/archie#911