toasterson/akh-medu
Watch
1
0
Fork
You've already forked akh-medu
 0
Code Issues 10 Pull requests 22 Projects Releases 2 Packages Wiki Activity Actions 7

fix(lint): clear workspace clippy warnings in grug-data + harness-core #274

Merged
toasterson merged 1 commit from claude/fix-workspace-clippy-lint into main 2026-07-28 10:35:42 +00:00
Conversation 0 Commits 1 Files changed 6 +48 -43
toasterson commented 2026-07-28 10:19:32 +00:00
Owner
Copy link

Why this is its own PR

CI runs cargo check -p seshat — one crate. The workflow says broader
test/clippy/release-build OOM-kills rustc on the Solstice runner. So workspace-wide
clippy has been rotting unchecked while main stays green.

The akhs do not have that blind spot: compile_feedback covers the wider surface.
The result, measured on 2026-07-28, is that every akh-medu session opens by
re-fixing the same warnings in crates unrelated to its work item, then commits
them under that work item's title. WI-230 (Wave R1 — Config spine) burned four
rounds this way and pushed PR #246 containing nothing but these 48 lines, which the
reviewer correctly rejected for containing "work from MANY work items".

Landing it once, here, so the waves stop paying the tax.

Contents

  • crates/grug-data (ground.rs, eval.rs, generator.rs, retriever.rs) —
    collapsible if let → let-chains.
  • benchmarks/harness-core (site.rs, store.rs) — write! with a trailing
    newline → writeln!.

No behaviour change. The content was produced by the tecton session on WI-230 and
passed Solstice CI on that branch (tasks 15763/15768/15771/15774/15775, all green
on 18bed87).

Follow-ups filed

  • Anima 019fa7c6-ec4b-73e1-8b43-dc4deeca8373 — a turn that only repairs the base
    repo must not be committed, pushed and PR'd as the work item's round-1 work.
  • akh-medu — scope compile_feedback to the crates the work item touches, so an
    unrelated lint sweep never becomes the akh's turn.

Note that this will rot again until CI can lint the workspace (currently blocked by
the runner OOM).

## Why this is its own PR CI runs `cargo check -p seshat` — one crate. The workflow says broader test/clippy/release-build OOM-kills rustc on the Solstice runner. So workspace-wide clippy has been rotting unchecked while main stays green. The akhs do not have that blind spot: `compile_feedback` covers the wider surface. The result, measured on 2026-07-28, is that **every akh-medu session opens by re-fixing the same warnings in crates unrelated to its work item**, then commits them under that work item's title. WI-230 (Wave R1 — Config spine) burned four rounds this way and pushed PR #246 containing nothing but these 48 lines, which the reviewer correctly rejected for containing "work from MANY work items". Landing it once, here, so the waves stop paying the tax. ## Contents - `crates/grug-data` (`ground.rs`, `eval.rs`, `generator.rs`, `retriever.rs`) — collapsible `if let` → let-chains. - `benchmarks/harness-core` (`site.rs`, `store.rs`) — `write!` with a trailing newline → `writeln!`. No behaviour change. The content was produced by the tecton session on WI-230 and passed Solstice CI on that branch (tasks 15763/15768/15771/15774/15775, all green on `18bed87`). ## Follow-ups filed - Anima `019fa7c6-ec4b-73e1-8b43-dc4deeca8373` — a turn that only repairs the base repo must not be committed, pushed and PR'd as the work item's round-1 work. - akh-medu — scope `compile_feedback` to the crates the work item touches, so an unrelated lint sweep never becomes the akh's turn. Note that this will rot again until CI can lint the workspace (currently blocked by the runner OOM).
fix(lint): clear workspace clippy warnings in grug-data + harness-core
Some checks failed
CI / check-seshat (push) Successful in 13m1s
Details
CI / publish-chart (push) Successful in 19m8s
Details
CI / publish-chart (pull_request) Successful in 18m47s
Details
CI / docker-seshd (push) Failing after 21m13s
Details
CI / check-seshat (pull_request) Successful in 28m1s
Details
CI / docker-seshd (pull_request) Successful in 35m7s
Details
839078159c
 
CI only runs `cargo check -p seshat` (broader test/clippy/release-build OOM-kills
rustc on the Solstice runner), so workspace-wide clippy has been rotting unchecked.
The akhs' `compile_feedback` covers the wider surface, so every akh-medu session
opens by re-fixing the same warnings in crates unrelated to its work item — and
then commits them under that work item's title (Anima WI 019fa7c6-ec4b).

Content produced by the tecton session on WI-230 (akh-medu Wave R1); extracted
here so it lands once, on its own, instead of being misattributed to a wave.

- collapsible if-let → let-chains (grug-data: ground.rs, eval.rs, generator.rs, retriever.rs)
- write! with a trailing newline → writeln! (harness-core: site.rs, store.rs)
Sign in to join this conversation.
Reviewers
No reviewers
Labels
Clear labels
No items
No labels
Milestone
Clear milestone
No items
No milestone
Projects
Clear projects
No items
No project
Assignees
Clear assignees
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set

Reference
toasterson/akh-medu!274
No description provided.