diag: does oxfmt 0.65.0 fix the native-binding install bug? (draft, throwaway) - #4
Closed
catomean wants to merge 4 commits into
Closed
diag: does oxfmt 0.65.0 fix the native-binding install bug? (draft, throwaway)#4catomean wants to merge 4 commits into
catomean wants to merge 4 commits into
Conversation
The repo already had oxfmt wired as format/format:check, but nothing in scripts/check.mjs ever called it — verify passing was never evidence the tree was formatted. These 11 files had drifted; fixed mechanically with 'oxfmt --write'.
format/format:check have existed since oxfmt was adopted, but scripts/check.mjs never called format:check — a passing 'pnpm check' was never evidence the tree was formatted, only that it built and typechecked. Prettier was never the right formatter here for a 17.6k-file repo; this repo already made the right call with a faster Rust-based one, it just never got wired in. Added as the last entry in the existing 'preflight guards' stage, alongside the other cheap npm-script guards it already runs in parallel with — one line, same shape as every other entry there. Proved by mutation: appended deliberately mis-formatted code to a tracked file and ran 'node scripts/check.mjs' end to end. It failed at the format step, naming exactly that file, and nothing else changed. Reverted before committing. Unrelated pre-existing finding while running the full check, noted for whoever owns it: 'npm shrinkwrap guard' also failed — 'npm-shrinkwrap.json is stale. Run pnpm deps:shrinkwrap:generate.' — with a clean git status on both npm-shrinkwrap.json and pnpm-lock.yaml beforehand, so it is not something this change caused.
The scripts/check.mjs wiring in the previous commit is what 'pnpm check' and any local dev running the aggregate check picks up. It turns out CI itself does not call check.mjs at all for this task: .github/workflows/ci.yml's check-shard job hand-lists the same commands a second time, independently, in a 'case "$TASK" in guards) ... esac' block — and that list had drifted from check.mjs's own array even before this PR (it's missing several entries check.mjs has, e.g. media-download-helpers, runtime-sidecar-loaders, opengrep-rule-metadata). Confirmed by reading check-guards' actual CI log: the workflow runs each 'pnpm check:*'/'lint:*' line directly, never invoking node scripts/check.mjs. So the previous commit alone, however correct, would never have been enforced by CI — only by a human or agent remembering to run 'pnpm check' locally, which is the exact kind of unenforced-by-default gap this PR exists to close. Added as one more line in the 'guards)' case, matching every other entry's shape. The two lists (check.mjs's array, and this case block) are still two independently-maintained copies of the same guard list after this change — that duplication is pre-existing and out of scope here, but is worth someone eventually collapsing to one source of truth.
Throwaway diagnostic branch, not for merging as-is. 0.52.0 (the pinned version) fails 'Cannot find native binding' for @oxfmt/binding-linux-x64-gnu on every CI attempt (4/4), 0/N locally, independent of cache state and with a verified-correct lockfile entry. 13 minor versions have shipped since 0.52.0; testing whether one of them fixed the binding resolution bug before ruling this out as unfixable from this side.
catomean
marked this pull request as ready for review
September 1, 2026 00:15
Dependency GuardThis PR changes dependency-related files. Maintainers should confirm these changes are intentional. Changed files:
Maintainer follow-up:
|
Dependency graph changes notedThis PR includes dependency graph changes. The dependency guard is informational because the PR author is a repository admin or a member of
Security review is still recommended before merge when the dependency graph change is intentional. |
Collaborator
Author
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Testing whether a version bump past 0.52.0 resolves the 'Cannot find native binding' failure seen on every CI attempt of #3. Draft on purpose — not for merging. Will close regardless of outcome.