Skip to content

[pull] main from openai:main - #175

Merged
pull[bot] merged 5 commits into
Mattlk13:mainfrom
openai:main
Sep 24, 2026
Merged

pull[bot] merged 5 commits into
Mattlk13:mainfrom
openai:main

Conversation

@pull

@pull pull Bot commented Sep 24, 2026 •

Copy link
Copy Markdown

See Commits and Changes for more details.


Created by pull[bot] (v2.0.0-alpha.4)

Can you help keep this open source service alive? 💖 Please sponsor : )

HaokaiDing and others added 5 commits September 23, 2026 21:28
- [x] I understand that this repository is auto-generated and my pull
request may not be merged.

## Changes being requested

Fixes #3874.

When the Skills `files` parameter is a single file, the array extraction
path consumes it before the fallback single-file path can run. This
produces an empty JSON request instead of uploading the file.

Keep non-list values available when a path still expects an array.
Direct file extraction and list uploads retain their existing behavior.
Add bytes/tuple extraction regressions and synchronous/asynchronous
public-client tests that construct a real in-memory zip and inspect its
multipart field, filename and content through MockTransport.

## Additional context & links

- [Skills create accepts a single zip or multiple
files](https://developers.openai.com/api/reference/python/resources/skills/methods/create).
- Before the fix: 4 failed, 33 passed. After: 37 passed with
`PYTHONPATH=src .venv/bin/python -m pytest -o addopts=''
tests/test_extract_files.py tests/test_files.py -q`. Tests run serially
and make no live API requests.
- Ruff lint and format checks pass on all three changed files; `git diff
--check` passes.
- Python 3.13.15, httpx2 2.12.0, Pydantic 2.13.5 on macOS ARM64. The
full SDK suite, Pyright and mypy were not run locally.
- The custom-code policy, budget and generated ownership metadata are
unchanged. Budget validation remains with the trusted upstream CI; SDK
CODEOWNER review is required for this file-handling change.

Co-authored-by: Marcus Wood <marcuswood@openai.com>
#3949)

## Summary

Corrects fine-tuning numeric bounds and Realtime client-secret response
alternatives in the Python API reference.

## Changes

- Use OpenAPI 3.1 numeric exclusive bounds for fine-tuning parameters.
- Identify realtime and transcription session alternatives in
client-secret responses.






## Why

The API reference used boolean `exclusiveMinimum` values, which are
invalid in OpenAPI 3.1. This corrects seven numeric lower bounds and
restores their existing `auto` defaults. Explicit discriminator mappings
also identify the existing `realtime` and `transcription` alternatives
in client-secret responses.

This is a reference-only correction: SDK source, public types,
dependencies, and request/response serialization are unchanged.

Source: [published OpenAPI
correction](openai/openai-openapi@946e365).

## Compatibility review

Reviewed head `add92714d353df33e1ee98a604800ce8b6e6d42d` against base
`4d1274681db7f34f671ff9e71adc1461d22a927b`.

- The complete tree diff contains only `.castiron.stats.yml` and
`api_reference/openapi.transformed.yml`. All SDK source, public
signatures, related model types, serializers, dependencies, and
package/build configuration are unchanged. Neither changed file is
consumed by the SDK runtime or enters its normal distributed code.
- Paired baseline/candidate Python client and model probes pass (183
cases per revision, identical results).
- The transformed reference passes 464 focused schema cases, including
actual enclosing requests/responses, lower/upper boundaries, `auto`,
omission, and both Realtime response variants.
- [Current-head
CI](https://github.com/openai/openai-python/actions/runs/35922654398)
passes.

Scope: this establishes no identified SDK-package compile-time or
runtime compatibility regression from this patch. Tools that directly
read the OpenAPI reference (validators, mock servers, or third-party
generators) can observe the corrected bounds, defaults, and explicit
mappings; those integrations are a separate compatibility surface. No
deployed-service test or package release is claimed.

Co-authored-by: markstuart-oai <323302876+markstuart-oai@users.noreply.github.com>
## Summary

Clarifies how approximate user locations affect web search results.

## Changes

- Document that omitted or null user locations default to the United
States.
- Explain how to avoid the fallback with an approximate location
containing no location fields, or localize results with location fields.

Co-authored-by: marcuswood-oai <323319819+marcuswood-oai@users.noreply.github.com>
## Summary

Clarifies retained Realtime modality arrays in the API reference without
changing SDK types or method signatures.

## Changes

- Explicitly describe retained Realtime modality fields as arrays of
text or audio values.
- Preserve existing optionality and SDK behavior.

Co-authored-by: markstuart-oai <323302876+markstuart-oai@users.noreply.github.com>
Automated Release PR
---


##
[3.19.2](v3.19.1...v3.19.2)
(2026-09-23)


### Bug Fixes

* preserve single files for fallback extraction paths
([#3875](#3875))
([bfd3680](bfd3680))


### Chores

* **api:** clarify approximate web search location defaults
([#3953](#3953))
([a95c95e](a95c95e))
* **api:** clarify Realtime modality array definitions
([#3954](#3954))
([e79cf53](e79cf53))
* **api:** correct fine-tuning bounds and Realtime response reference
([#3949](#3949))
([325a948](325a948))

---
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>
@pull pull Bot locked and limited conversation to collaborators Sep 24, 2026
@pull pull Bot added the ⤵️ pull label Sep 24, 2026
@pull
pull Bot merged commit d9f7b7a into Mattlk13:main Sep 24, 2026
3 of 6 checks passed
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants