Adapt to the Ollama reasoning platform and use the "reasoning" JSON k…

[zhujunling-nj]

…ey to return the reasoning content

Pull Request Checklist

Thank you for your contribution! Before submitting your PR, please make sure you have completed the following checks:

Compatibility Check

• I have checked whether this change affects the backward compatibility of the plugin declared in README.md
• I have checked whether this change affects the forward compatibility of the plugin declared in README.md
• If this change introduces a breaking change, I have discussed it with the project maintainer and specified the release version in the README.md
• I have described the compatibility impact and the corresponding version number in the PR description
• I have checked whether the plugin version is updated in the README.md

Available Checks

• Code has passed local tests
• Relevant documentation has been updated (if necessary)

[gemini-code-assist]

Summary of Changes

Hello @zhujunling-nj, 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 enhances the system's compatibility with the Ollama reasoning platform by modifying how reasoning content is parsed from model outputs. It introduces a more robust mechanism to retrieve reasoning information, allowing the system to correctly process responses that might use an alternative key for this data. This ensures smoother integration and improved functionality when interacting with Ollama.

Highlights

• Ollama Platform Adaptation: The system has been updated to better integrate with the Ollama reasoning platform.
• Flexible Reasoning Content Extraction: The logic for extracting reasoning content from model responses now checks for both 'reasoning_content' and 'reasoning' keys, providing greater flexibility.

🧠 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. ↩

[gemini-code-assist]

Code Review

This pull request adapts the code to support the reasoning key from the Ollama reasoning platform, in addition to the existing reasoning_content key. The change is straightforward and correct. My review includes two points of feedback: one to improve the robustness of the code against potential type errors, and another to update the documentation to reflect this change, ensuring better maintainability.

[gemini-code-assist]

This change, combined with the existing logic, could lead to a TypeError if delta.get('reasoning_content') or delta.get('reasoning') returns a truthy non-string value (e.g., a dictionary or a list). The string concatenation on line 542 would fail. To make the code more robust and prevent runtime errors, it's best to explicitly convert the result to a string. Using str() and providing an empty string as a final fallback will handle None and other falsy values correctly while safely converting any other types.

Suggested change

	reasoning_content = delta.get("reasoning_content") or delta.get("reasoning")
	reasoning_content = str(delta.get("reasoning_content") or delta.get("reasoning") or "")

[Yeuoly]

Please also add some tests to ensure your code works, of course, this enforces others will not break your fixes.

[zhujunling-nj]

I use an OpenAI-API-compatible plugin to access the OLLAMA Qwen3:32B model.
Previously, the reasoning content could not be displayed:

Now the reasoning content can be displayed:

[Yeuoly]

I mean you can add some unittests to the repo.

[zhujunling-nj]

Added 3 test cases:

1. Includes the reasoning_content field
2. Includes the reasoning field
3. Does not include reasoning_content or reasoning fields

Test execution results:
DEBUG Output: '\nThinking started. Still thinking.\nContent started.'
.DEBUG Output: '\nThinking started. Still thinking.\nContent started.'
.DEBUG Output: 'Content started. Still content.'
.

Ran 3 tests in 0.002s

OK