[Bugfix] Consistent ascii handling in tool parsers (#17704)

[abadea-ionos]

Because of Cuda drivers dependencies, we can't upgrade our vllm to 0.9.0 yet, so we would like to include this bugfix (which is already included in vllm:0.9.0) into the 0.8.5 version, which we are currently running on.

[gemini-code-assist]

Summary of Changes

Hello @abadea-ionos, 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 integrates a crucial bugfix from a newer vLLM release into our current 0.8.5 deployment. The change specifically targets the serialization of tool call arguments, ensuring that all characters, including non-ASCII ones, are correctly represented in JSON outputs. This resolves an underlying issue that could lead to incorrect tool call processing due to character encoding discrepancies.

Highlights

• Bugfix Backport: This pull request backports a bugfix from vLLM version 0.9.0 to the currently utilized 0.8.5, addressing an issue related to consistent ASCII handling in tool parsers.
• Consistent ASCII Handling: The core change involves modifying json.dumps calls across various tool parser files to include the ensure_ascii=False parameter. This ensures that non-ASCII characters within tool arguments are preserved directly in the JSON output rather than being escaped, preventing potential data corruption or misinterpretation.
• Affected Components: The fix is applied to multiple tool parser implementations, including those for Granite, InternLM2, Jamba, Llama, Phi4Mini, and Pythonic tools, ensuring broad coverage for tool call serialization.

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 in your pull request via creating an issue comment (i.e. comment on the pull request page) using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands.

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 issue 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 is currently in preview and 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 to provide feedback.

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 introduces a bug fix to ensure consistent ASCII handling in tool parsers by adding the ensure_ascii=False parameter to json.dumps calls. This change prevents non-ASCII characters from being escaped, which is essential for proper data representation when dealing with tool calls that may contain Unicode characters. It's important to verify that the receiving end supports UTF-8 encoding to correctly interpret these characters.

[github-actions]

👋 Hi! Thank you for contributing to the vLLM project.

💬 Join our developer Slack at https://slack.vllm.ai to discuss your PR in #pr-reviews, coordinate on features in #feat- channels, or join special interest groups in #sig- channels.

Just a reminder: PRs would not trigger full CI run by default. Instead, it would only run fastcheck CI which starts running only a small and essential subset of CI tests to quickly catch errors. You can run other CI tests on top of those by going to your fastcheck build on Buildkite UI (linked in the PR checks section) and unblock them. If you do not have permission to unblock, ping simon-mo or khluu to add you in our Buildkite org.

Once the PR is approved and ready to go, your PR reviewer(s) can run CI to test the changes comprehensively before merging.

To run CI, PR reviewers can either: Add ready label to the PR or enable auto-merge.

🚀