[steipete]

What does this PR do?

Registers the listener passed to fs.watch() as a change listener on the returned FSWatcher. Listener inspection, manual emit(), removal, and registration order now include that callback. The separate callback field and its duplicate dispatch sites are removed. A rejected listener releases the native watch rather than leaving an open handle.

Ports the existing open proposal oven-sh/bun#43765, by robobun. No competing upstream proposal is needed.

How did you verify your code works?

• Node.js 24.22.0 passes the portable listener oracle; unpatched Bun fails at listenerCount("change").
• On a native Linux release build, the new listener group changes from 5 passes / 15 failures to 20 passes. Coverage includes overloads, callback receiver, synthetic and native events, ordering, removal, invalid listeners, and native-handle cleanup.
• The surrounding test/js/node/watch and test/js/node/fs run records 1,030 passes and 122 skips. Its four load errors came from omitting the internal-testing flag; rerunning only those four files with the standard BUN_FEATURE_FLAG_INTERNAL_FOR_TESTING=1 --expose-internals setup gives 18 passes, 5 platform skips, and no failures on both baseline and patched builds.
• The affected OpenClaw consumer file changes from 7 passes / 3 failures to 10 passes with the patched runtime and isolated temporary state.
• JavaScript lint, formatting, and independent P2 reviews pass, including the refresh onto current main. The refreshed head passes Linux and macOS CI.

[chatgpt-codex-connector]

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review	Status	Commit	Review trigger
📝 Code Review	✅ Completed 2026-10-16T11:33:01.163806Z	4061abf	Draft marked ready

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.