[Python] Support large pipeline options via file

[Mathdee]

Issue:
As described in #37370, running pipelines with large options on Dataflow causes fork/exec /usr/local/bin/python: argument list too long.
This occurs because the bootloader passes the complete JSON config via PIPELINE_OPTIONS environment variable --> exceeds the OS ARG_MAX limit.

The Fix:
This change uses an identical pattern from the Go SDK (Issue #27839, Commit e31e885) to Python.

1. boot.go(file): Writing the pipeline options to a temp file (pipeline_options.json) and sets the PIPELINE_OPTIONS_FILE environment variable.
2. sdk_worker_main.py(file): This then checks for PIPELINE_OPTIONS_FILE and loads the configs from the disk if present.

Outcome:

• Fixes [Feature Request]: Support large pipeline options in Python SDK #37370
• Feature Parity with Java and Go SDK's for handling large pipeline options
• Verified with unit tests that ensured priority of file-based loading.

---

• Mention the appropriate issue in your description (for example: addresses #123), if applicable. This will automatically add a link to the pull request in the issue. If you would like the issue to automatically close on merging the pull request, comment fixes #<ISSUE NUMBER> instead.

• Update CHANGES.md with noteworthy changes.

• If this contribution is large, please file an Apache Individual Contributor License Agreement.

[gemini-code-assist]

Summary of Changes

Hello @Mathdee, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request introduces a robust solution for handling large pipeline options in the Python SDK, which previously led to an 'argument list too long' error due to operating system limitations on environment variable size. By transitioning from direct environment variable passing to a file-based approach, where options are written to a temporary file and its path is referenced, the system can now accommodate extensive configurations without encountering runtime failures. This enhancement ensures greater stability and consistency across SDKs.

Highlights

• Addresses Argument List Too Long Error: This pull request resolves the fork/exec /usr/local/bin/python: argument list too long error that occurs when running Dataflow pipelines with excessively large pipeline options, by preventing the complete JSON config from being passed directly via the PIPELINE_OPTIONS environment variable.
• File-based Pipeline Options: Pipeline options are now written to a temporary file named pipeline_options.json by the Go bootloader. The path to this file is then passed to the Python SDK worker via the PIPELINE_OPTIONS_FILE environment variable.
• Python SDK Worker Adaptation: The Python SDK worker (sdk_worker_main.py) has been updated to check for the PIPELINE_OPTIONS_FILE environment variable and load the pipeline configuration from the specified file if present, ensuring compatibility with the new mechanism.
• Feature Parity: This change brings the Python SDK into feature parity with the Java and Go SDKs for handling large pipeline options, using an identical pattern previously implemented in the Go SDK.

🧠 New Feature in Public Preview: You can now enable Memory to help Gemini Code Assist learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. Click here to enable Memory in your admin console.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature	Command	Description
Code Review	/gemini review	Performs a code review for the current pull request in its current state.
Pull Request Summary	/gemini summary	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help	/gemini help	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

[Mathdee]

All linting and integration checks are passing. Ready for review

[github-actions]

Assigning reviewers:

R: @claudevdm for label python.

Note: If you would like to opt out of this review, comment assign to next reviewer.

Available commands:

• stop reviewer notifications - opt out of the automated review tooling
• remind me after tests pass - tag the comment author after tests pass
• waiting on author - shift the attention set back to the author (any comment or push by the author will return the attention set to the reviewers)

The PR bot will only process comments in the main thread (not review comments).

[claudevdm]

@shunping can you take a pass at this since you have more context?

[shunping]

Ack! Will do

[shunping]

Thanks for submitting the PR. Please check out my inline comments.

[Mathdee]

Hey @shunping. Just checking in on this one. Let me know if there's anything else you need from my end.

[shunping]

Hey @shunping. Just checking in on this one. Let me know if there's anything else you need from my end.

Sorry for the delay. I would want to build a custom container based on this and test before approving it. Will try to do that tomorrow.

[shunping]

/gemini review

[shunping]

Could you merge to the latest master and push again? @Mathdee

[gemini-code-assist]

Code Review

This pull request introduces a valuable enhancement to support large pipeline options by passing them through a file, thus avoiding issues with argument length limits. The implementation in both Go and Python appears correct and follows the intended logic. I have a couple of suggestions to improve the Python code: one for exception handling to make it more robust, and another to add unit tests for the new functionality.

[shunping]

Both failed test workflow PreCommit YAML Xlang Direct and PreCommit Python ML tests with ML deps installed were unrelated to the test change here.

[shunping]

I have verified the changes with a new container and a test pipeline on Dataflow. The pipeline options were successfully loaded via the file and the job completed as expected.

During testing, I found that the INFO log message for loading options from the file was being suppressed. This occurs because the default log level is WARNING until the pipeline options are fully parsed.

Therefore, I have added a fallback to INFO at the start of create_harness.

Below is a screenshot of the log in Dataflow.

[shunping]

@Mathdee: Could you also update CHANGES.md to document this change?

I suggest adding a note to the 'New Features/Improvements' section regarding the support for large pipeline options via a file. In the 'Breaking Changes' section, we should also mention that the Python SDK container's boot.go now passes pipeline options through a file instead of the PIPELINE_OPTIONS environment variable.

Please note that if a user pairs a new Python SDK container with an older SDK version (which does not support the file-based approach), the pipeline options will not be recognized and the pipeline will fail. In that case, we should advise users to ensure their SDK and container versions are synchronized.

[shunping]

LGTM! Thanks a lot! Will merge when the tests are complete.

[Mathdee]

@shunping Awesome, thanks for taking the time to test this and for adding the log level fallback.
I just updated the CHANGES.md with the notes exactly as requested. Let me know if it looks good!

[shunping]

Loosk like PreCommit Whitespace failed.

Could you fix the notes to include the issue link, i.e. #37370 (not the PR link)?

The error message is shown below:

Execution failed for task ':validateChanges'.
> CHANGES.md validation failed with the following errors:
  Line 71: Missing or malformed issue link. Each entry should end with ([#X](https://github.com/apache/beam/issues/X)): * Added support for large pipeline options via a file (Python) ([#37379](https://github.com/apache/beam/pull/37379)).
  Line 75: Missing or malformed issue link. Each entry should end with ([#X](https://github.com/apache/beam/issues/X)): * The Python SDK container's `boot.go` now passes pipeline options through a file instead of the `PIPELINE_OPTIONS` environment variable. If a user pairs a new Python SDK container with an older SDK version (which does not support the file-based approach), the pipeline options will not be recognized and the pipeline will fail. Users must ensure their SDK and container versions are synchronized ([#37379](https://github.com/apache/beam/pull/37379)).

[Mathdee]

I fixed the notes to include the issue link instead of the PR link, just waiting on the tests to be completed now.

[shunping]

Verified the failed workflow PreCommit Python Integration / beam_PreCommit_Python_Integration is irrelevant as well.

Good job! Thank you again! @Mathdee.

[Mathdee]

Thanks so much for the review and all your help getting this across the finish line @shunping. So glad we could get it officially merged today. Have a great rest of your week!