🔄 daily merge: master → main 2025-12-01

[antfin-oss]

This Pull Request was created automatically to merge the latest changes from master into main branch.

📅 Created: 2025-12-01
🔀 Merge direction: master → main
🤖 Triggered by: Scheduled

Please review and merge if everything looks good.

[sourcery-ai]

The pull request #692 has too many files changed.

The GitHub API will only let us fetch up to 300 changed files, and this pull request has 6194.

[gemini-code-assist]

Summary of Changes

Hello @antfin-oss, 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 automated daily merge pull request synchronizes the master branch with main, bringing in a comprehensive set of updates to the project's build and continuous integration infrastructure. The changes span from fundamental build system configurations and dependency management to detailed refinements in testing pipelines across various Ray components, ensuring consistency and improved efficiency.

Highlights

• Bazel Configuration Updates: The .bazelrc file has been significantly updated to enable incompatible_strict_action_env by default, add UTF-8 support for Windows builds, ignore warnings for third-party files, and introduce a .user.bazelrc for custom user options. Python 3.9 runtime and toolchain definitions have been removed, with hermetic_python now aliasing to //bazel:py310.
• CI Pipeline Refactoring: Buildkite CI configurations have undergone major restructuring. Image build steps are now centralized in a new _images.rayci.yml file, and documentation-related steps have been moved to a new doc.rayci.yml. Python 3.10 support has been expanded across various CI matrices, and several oss-ci-base build steps now include a cibase tag.
• Python Dependency Management with uv: The CI now leverages uv for Python dependency management, replacing miniconda with miniforge in Dockerfiles and updating related scripts. New raydepsets configurations have been introduced for compiling dependencies, and the pip-compile dependencies step has been moved to a dedicated dependencies.rayci.yml file.
• C++ and Java Build/Test Improvements: C++ worker tests have been updated, RAYCI_DISABLE_TEST_DB=1 is now added to sanitizer builds, and new ray-core-build and ray-java-build steps have been introduced. Java worker tests are now explicitly defined, and the C++ API has been updated to use more granular common libraries.
• macOS Build/Test Enhancements: macOS bisect tests now utilize macos-arm64 instance types, and the macOS wheel/jar build process has been streamlined. The build_x86_64 function has been removed in favor of a unified build function.
• Ray Serve CI Updates: Ray Serve CI configurations have been updated to use Python 3.10, include new tracing tests, and refine HA integration tests. The servebuild step has been removed, and minbuild-serve now supports Python 3.10.
• Ray Train/ML CI Updates: Ray Train tests are now split into v1 and v2, with new GPU-specific v2 tests. Python 3.10 has been added to ML build matrices, and except-tags for ml: {{matrix.python}} tests and flaky tests now include train_v2_gpu.
• Ray LLM CI Updates: Ray LLM GPU tests have been refined, including specific steps for 4-GPU configurations. The llmgpubuild step now uses oss-ci-base_cu128-multipy and cu128.
• Codeowners Refinement: The .github/CODEOWNERS file has been extensively updated to consolidate ownership for core components (C++, Python, Java) and add specific ownership for new areas such as Azure autoscaler and LLM-related dashboard metrics.
• Pre-commit Hooks and Linting: Pre-commit hooks have been updated to leverage semgrep, vale, cython-lint, and eslint. The format.sh script has been removed in favor of pre-commit, and the check_import_order.py script has been adjusted to reflect changes in bundled libraries.

Ignored Files

• Ignored by pattern: .gemini/** (1)
  • .gemini/config.yaml
• Ignored by pattern: .github/workflows/** (1)
  • .github/workflows/stale_pull_request.yaml

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. ↩

[gemini-code-assist]

Code Review

This pull request is an automated daily merge from master to main. It includes a very large number of changes, primarily focused on refactoring and improving the CI/CD pipeline and the Bazel build system. Key changes include modularizing build steps, updating dependencies and toolchains (e.g., moving to miniforge and uv), a shift to arm64 for macOS builds, and significant cleanups in various configuration files and scripts. I've found one issue in the buildkite annotation logic. Otherwise, the changes appear to be consistent and part of a large-scale improvement effort.

[gemini-code-assist]

There is a bug in the buildkite annotation logic. The aws_alias_image variable is used outside the for loop, so it will only contain the value from the last iteration. This means only the last created alias will be annotated. To fix this, you should collect all aliases and annotate them together after the loop.

            images_to_annotate = []
            for alias in self._get_image_tags():
                aws_alias_image = f"{aws_registry}/anyscale/{self.image_type}:{alias}"
                gcp_alias_image = f"{gcp_registry}/anyscale/{self.image_type}:{alias}"
                azure_alias_image = (
                    f"{azure_registry}/anyscale/{self.image_type}:{alias}"
                )
                cmds += [
                    f"docker tag {anyscale_image} {aws_alias_image}",
                    f"docker push {aws_alias_image}",
                    f"docker tag {anyscale_image} {gcp_alias_image}",
                    f"docker push {gcp_alias_image}",
                    f"docker tag {anyscale_image} {azure_alias_image}",
                    f"docker push {azure_alias_image}",
                ]
                images_to_annotate.append(aws_alias_image)

            if os.environ.get("BUILDKITE") and images_to_annotate:
                annotation_body = "<br/>".join(images_to_annotate)
                subprocess.run(
                    [
                        "buildkite-agent",
                        "annotate",
                        "--style=info",
                        f"--context={self.image_type}-images",
                        "--append",
                        annotation_body,
                    ]
                )