chore: migrate to forked steady - #3795
Conversation
Codex Review SummaryThis comment shows the latest Codex review activity on this pull request.
ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings. |
Castiron custom code✅ No new custom-code files detected. 36 mixed files remain; 0 existing customizations changed. Compared 36 existing customizations unchanged
A changed generated baseline means this report cannot reliably identify which handwritten lines changed. Inspect the custom-code diffDownload the exact patch produced by this run (requires repository access): gh run download 33822556742 --repo openai/openai-python \
--name castiron-custom-code-33822556742-1 --dir /tmp/castiron-custom-code-33822556742-1
git apply --stat /tmp/castiron-custom-code-33822556742-1/custom-code.patch
cat /tmp/castiron-custom-code-33822556742-1/custom-code.patchOr reproduce it from an SDK checkout containing the vendored reporter: git fetch --no-tags origin 88391abf981df3ea395ca1b5bf55ec6a4011ea93 fa24ffbf5060fa14f6fc1f1d0ed50ff3e81dd0d0
python3 scripts/castiron/custom_code_report.py report \
--base 88391abf981df3ea395ca1b5bf55ec6a4011ea93 \
--head fa24ffbf5060fa14f6fc1f1d0ed50ff3e81dd0d0 --fetch --require-head-hash --public \
--out /tmp/castiron-custom-code-fa24ffbf5060
cat /tmp/castiron-custom-code-fa24ffbf5060/custom-code.patchThis is the current full custom patch for mixed files, not an attribution of only the handwritten lines changed by this PR. |
jbeckwith-oai
left a comment
There was a problem hiding this comment.
Blocking: the atomic publisher is not safe under concurrent Windows/Git Bash bootstraps. only treats and as contention, but Node's directory rename can report or when another process wins. Because this change explicitly advertises Windows support, please handle those codes as contention only when the destination now exists, then rely on to authenticate the winner. Please add focused concurrent-publication coverage.
Superseded by corrected review text.
jbeckwith-oai
left a comment
There was a problem hiding this comment.
Blocking: the atomic publisher is not safe under concurrent Windows and Git Bash bootstraps. The publish_directory helper only treats EEXIST and ENOTEMPTY as contention, but Node directory rename can report EPERM or EACCES when another process wins. Because this change explicitly advertises Windows support, please handle those codes as contention only when the destination now exists, then rely on steady_verify to authenticate the winner. Please add focused concurrent-publication coverage.
jbeckwith-oai
left a comment
There was a problem hiding this comment.
Re-reviewed the requested fix. Concurrent publication now accepts Windows loser-race errors only when a real non-symlink destination exists, and the caller still authenticates the published winner. The focused publication test is wired into CI, HTTPS-only redirects are enforced, and exact-head checks are green.
migrate to forked steady