[retrry]

The httpx and httpx2 rest.py sends timeout=_request_timeout or 5 * 60 on every request. In HTTPX a per-request timeout replaces the client's, so a client set as pool_manager with its own timeout still waits up to 5 minutes unless every call passes _request_timeout. Configuration has no timeout setting, so there is no other way to change it.

This change moves the 5 minute default from the request to the client:

• A call without _request_timeout uses {{httpxModule}}.USE_CLIENT_DEFAULT, so the client's own timeout applies.
• _create_pool_manager builds its AsyncClient with timeout=5 * 60, so clients the generator creates keep the current 300 s default.

The urllib3 rest.py already behaves this way: without _request_timeout it leaves the timeout unset and uses the pool manager's. The aiohttp template has the same 5 * 60 line but a different timeout API, so it is left out of this PR.

Validation: regenerated the four httpx samples (python-httpx, python-httpx-sync, python-httpx2, python-httpx2-sync); only rest.py changed. Added two tests to the httpx2 samples' test_httpx2_transport.py: a supplied client's timeout applies when the call sets none, and the generated client defaults to 300 s. Both fail on master and pass with this change. The sample suites pass except test_pet_api.py, which needs a local Petstore server.

PR checklist

• Read the contribution guidelines.
• Run the following to build the project and update samples:

  ./mvnw clean package || exit
  ./bin/generate-samples.sh ./bin/configs/*.yaml || exit
  ./bin/utils/export_docs_generators.sh || exit
  

  Regenerated with ./bin/generate-samples.sh bin/configs/python-httpx*.yaml, the only configs using this template.
• If your PR is targeting a particular programming language, @mention the technical committee members, so they are more likely to review the pull request.

Python technical committee: @wing328

---

Summary by cubic

Fixes the httpx Python templates so calls without _request_timeout honor the pool manager's own timeout instead of always forcing a 5-minute default. The 5-minute default is now set on the generated AsyncClient in _create_pool_manager, so generated clients keep the current 300-second timeout.

• Requests without _request_timeout now send httpx.USE_CLIENT_DEFAULT so the client's timeout applies.
• Added tests verifying a supplied client's timeout is used and the generated client defaults to 300 seconds.
• Regenerated the four httpx samples; only rest.py changed. The aiohttp template is unchanged.

^{Written for commit e322f84. Summary will update on new commits.}

[cubic-dev-ai]

All reported issues were addressed across 7 files

<sub>Reply to a comment to ask cubic a question or push back. It learns from your replies.

View guided diff | Re-trigger cubic</sub>

[wing328]

thanks for the PR

cc @cbornet (2017/09) @tomplus (2018/10) @krjakbrjak (2023/02) @fa0311 (2023/10)