Repository navigation
fix(api): preserve Responses WebSocket connection behavior - #4221
Conversation
Castiron-Internal-PR: openai/openai-python-internal#389 Castiron-Source-SHA: be619e819f9bdb9e8904931387f5fed8388cf757 Castiron-Public-Base-SHA: 3622a97
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-Internal-PR: openai/openai-python-internal#389
jbeckwith-oai
left a comment
jbeckwith-oai
left a comment
There was a problem hiding this comment.
Reviewed cd9a9db with an independent static pass. No actionable findings. Shared helpers preserve stable/beta sync/async URL and auth preparation, case-insensitive header precedence and removal, connection defaults, retry-budget progress, iterator overrides, and entry/exit exception cleanup. Generated-file hashes match the changed resources.
Hosted checks: {'SUCCESS': 17, 'SKIPPED': 9}. No candidate code or tests executed locally.
Automated Release PR --- ## [3.42.1](openai/openai-python@v3.42.0...v3.42.1) (2026-11-01) ### Bug Fixes * **api:** preserve Responses WebSocket connection behavior ([openai#4221](openai#4221)) ([38075d8](openai@38075d8)) * **realtime:** validate beta WebSocket authentication profiles ([openai#4218](openai#4218)) ([3622a97](openai@3622a97)) ### Refactors * **websocket:** add SDK-owned Responses policy helpers ([openai#4219](openai#4219)) ([83ed269](openai@83ed269)) --- This PR was generated with [Release Please](https://github.com/googleapis/release-please). See [documentation](https://github.com/googleapis/release-please#release-please). Co-authored-by: openai-sdks[bot] <284451331+openai-sdks[bot]@users.noreply.github.com>
Use the SDK-owned connection helpers for stable and beta Responses WebSockets. This keeps header overrides, connection defaults, cleanup and recovery behavior in place when the clients are regenerated. It removes duplicate connection policy code from both generated resources.
Validation: 617 targeted tests and 10 retained exit tests pass against a fresh, noneditable wheel installation. Two normal generation runs produce identical code. The three SDK-owned test updates are included in this PR.