Skip to content

claw: make pr_tools.t27 compile and pass (24/24), reseal (Closes #12493) - #12580

Merged
gHashTag merged 4 commits into
masterfrom
claude/claw-pr-tools-fix-12493
Oct 26, 2026
Merged

gHashTag merged 4 commits into
masterfrom
claude/claw-pr-tools-fix-12493

Conversation

@gHashTag

Copy link
Copy Markdown
Collaborator

What

#12520 merged specs/claw/pr_tools.t27 while it did not compile. Its checks were red for inherited reasons, so the BLOCKED test report was hidden. This PR makes the spec compile, pass and sealed.

Fixes

  • merge_plan is declared in both claw/merge_judge and claw/merge_lane; the splicer resolves bare names globally, so the call is now merge_judge::merge_plan. Same for text::has_prefix and text::str_eq.
  • run_gt takes five arguments; one call passed six.
  • t27c keeps string-typed names (struct fields) in a single set for the whole spliced spec. claw/evidence declares a field path : str, so path == 9 in merge_judge's t_named test helper was lowered to a string compare and Zig refused it. The two helper parameters are renamed pidx (merge_judge stays 67/67).

Evidence

  • Test report for pr_tools: 24 passed, 0 failed, 0 vacuous.
  • Test report for merge_judge: 67 passed, 0 failed.
  • Seals saved with the master-built t27c from the worktree root: claw_ClawPrTools.json (new) and claw_ClawMergeJudge.json; both spec hashes match the sha256 of the spec files.
  • The duplicate-body scan reports only the inherited same group (t27#12498), nothing from these files.

Lesson

Read the test report of a new spec before merging; red checks can hide a compile error.

Closes #12493

🤖 Generated with Claude Code

…loses #12493)

#12520 merged specs/claw/pr_tools.t27 while it did not compile; red checks
hid the BLOCKED test-report. Fixes:
- merge_plan is declared in merge_judge and merge_lane: call it as
  merge_judge::merge_plan; likewise text::has_prefix, text::str_eq
- run_gt takes five arguments, not six
- t27c keeps string-typed names (struct fields) in one set for the whole
  spliced spec, so evidence's `path : str` field turned merge_judge's
  `path == 9` test helper into a string compare; rename those two test
  helper parameters to pidx (merge_judge 67/67 still green)

Seals saved for ClawPrTools (new) and ClawMergeJudge, hashes checked
against the spec files.

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
@github-actions

Copy link
Copy Markdown
Contributor

PR Dashboard

Generated at: 2026-10-26 01:18:33 UTC

Summary

Status Count
Total Open PRs 50
PRs with Failing Checks 35
PRs with All Checks Green 15
READY 4
FAILING 35
PENDING 0
NO CHECKS YET 0

These columns do not partition: 4 + 35 + 0 + 0 = 39, and there are 50 open PRs. A PR is being counted twice or not at all.

Seal Status

  • ⚠️ STALE -- sha256(compiler.rs)=75249fe1efad != manifest seal=87e5cbd3ad94.
    The committed NMSE numbers were certified against an older compiler.rs.
    Run scripts/reseal-check.sh locally for the two-step reseal command (advisory; not a merge gate).

…12493)

The spec-guards check lowers specs/claw/merge_judge.t27 and diffs it against the
committed C file; the t_file and t_named parameter rename path -> pidx was not
carried into the generated file. Seven lines changed, all of them that rename.

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
@github-actions

Copy link
Copy Markdown
Contributor

PR Dashboard

Generated at: 2026-10-26 01:35:36 UTC

Summary

Status Count
Total Open PRs 50
PRs with Failing Checks 36
PRs with All Checks Green 14
READY 2
FAILING 36
PENDING 0
NO CHECKS YET 0

These columns do not partition: 2 + 36 + 0 + 0 = 38, and there are 50 open PRs. A PR is being counted twice or not at all.

Seal Status

  • ⚠️ STALE -- sha256(compiler.rs)=75249fe1efad != manifest seal=87e5cbd3ad94.
    The committed NMSE numbers were certified against an older compiler.rs.
    Run scripts/reseal-check.sh locally for the two-step reseal command (advisory; not a merge gate).

@github-actions

Copy link
Copy Markdown
Contributor

PR Dashboard

Generated at: 2026-10-26 02:26:05 UTC

Summary

Status Count
Total Open PRs 32
PRs with Failing Checks 18
PRs with All Checks Green 14
READY 1
FAILING 18
PENDING 0
NO CHECKS YET 0

These columns do not partition: 1 + 18 + 0 + 0 = 19, and there are 32 open PRs. A PR is being counted twice or not at all.

Seal Status

  • ⚠️ STALE -- sha256(compiler.rs)=8ddf64a0178e != manifest seal=87e5cbd3ad94.
    The committed NMSE numbers were certified against an older compiler.rs.
    Run scripts/reseal-check.sh locally for the two-step reseal command (advisory; not a merge gate).

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

claw: MCP tools that help with pull requests (status, red origin, merge plan, reseal plan, dupe group)

1 participant