Skip to content

seals: reseal 16 specs whose seals lagged master (Refs #13584, #5980) - #13585

Merged
gHashTag merged 1 commit into
masterfrom
lane/reseal-stale-seals
Nov 10, 2026
Merged

gHashTag merged 1 commit into
masterfrom
lane/reseal-stale-seals

Conversation

@gHashTag

@gHashTag gHashTag commented Nov 10, 2026 •

Copy link
Copy Markdown
Collaborator

Summary

Method note

  • Local Homebrew zig 0.17.1 rejects the emitted spaced ** repeat that official zig 0.15/0.16 accept; seals here were minted with zig 0.16.0. No compiler source was changed.

Refs #13584, Refs #5980


Note

Low Risk
Diff is limited to seal metadata; behavior changes are whatever the underlying specs/tests already encode, except recording one known forced test failure on vf card fit.

Overview
Refreshes 16 Trinity seal JSON files under .trinity/seals/ so they match current specs, generated artifacts, and test results. Each entry is reminted with bootstrap t27c-bootstrap@0.5.2+3994e9d43, new sealed_at timestamps, and updated spec_hash / gen_hash_* values.

Toolchain: Most seals now record zig 0.17.1 (up from 0.16.0) and the current test_runner hash where it had been stale.

Notable seal content updates: BlogEpicProgress gains a closure on specs/tri/blog/blog.t27; TriShootoutCodegenAudit’s closure switches from bench.t27 to measured_aot.t27; shared dependency hashes move for petal_menu, own_language, ui_kit, and t27os/install. Test totals shift on several modules (e.g. automation blog/meta_connect, blog specs, policy). T27fVfCardFit is resealed with forced: true, documenting 1 failing test (fit_card_at_375_changes_only_the_text_widths).

No compiler or spec source changes in this diff—only seal metadata to restore seal-coverage alignment.

Reviewed by Cursor Bugbot for commit 48679cd. Bugbot is set up for automated code reviews on this repo. Configure here.

@cursor

cursor Bot commented Nov 10, 2026

Copy link
Copy Markdown
Contributor

Bugbot couldn't run - usage limit reached

Bugbot is counted against Cursor usage for this user or team, and this run hit a usage or spend limit.

A user or team admin can review and increase usage limits in the Cursor dashboard.

(requestId: c59a4449-2963-4d5a-949c-7337f9e82889)

@github-actions

Copy link
Copy Markdown
Contributor

PR Dashboard

Generated at: 2026-11-10 15:25:20 UTC

Summary

Status Count
Total Open PRs 34
PRs with Failing Checks 29
PRs with All Checks Green 5
READY 0
FAILING 29
PENDING 0
NO CHECKS YET 0

These columns do not partition: 0 + 29 + 0 + 0 = 29, and there are 34 open PRs. A PR is being counted twice or not at all.

Seal Status

  • ⚠️ STALE -- sha256(compiler.rs)=b882be1e4933 != 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).

@gHashTag
gHashTag merged commit 3d3b435 into master Nov 10, 2026
34 of 40 checks passed
@gHashTag
gHashTag deleted the lane/reseal-stale-seals branch November 10, 2026 15:27
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.

1 participant