Fix/refine deepresearch

[alcholiclg]

Change Summary

1. Fix error for generating empty follow-up questions
2. Support Ray resource cleanup to prevent errors at atexit

Related issue number

Checklist

• The pull request title is a good summary of the changes - it will be used in the changelog
• Unit tests for the changes exist
• Run pre-commit install and pre-commit run --all-files before git commit, and passed lint check.
• Documentation reflects the changes where applicable

[gemini-code-assist]

Summary of Changes

Hello @alcholiclg, 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 refines the deep research workflow by addressing two critical areas: improving the robustness of follow-up question generation and enhancing resource management for the Ray framework. The changes ensure smoother operation by preventing errors related to empty follow-up questions and guaranteeing proper cleanup of Ray resources, thereby avoiding potential issues during application shutdown.

Highlights

• Follow-up Question Generation: Resolved an issue where the system would encounter an error when attempting to generate empty follow-up questions, ensuring a fallback to the initial query.
• Ray Resource Management: Implemented robust Ray resource cleanup to prevent atexit errors by ensuring Ray is properly shut down after the workflow completes.

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

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 introduces two main changes. First, it fixes a bug where an empty list of follow-up questions would cause an error by explicitly handling this case. Second, it adds resource cleanup for Ray in a finally block to prevent atexit errors, wrapping the core logic in a new private _run method. The changes improve the robustness and reliability of the research workflow. I've suggested a small refactoring to the Ray cleanup logic to make it more concise.

[gemini-code-assist]

The logic for Ray cleanup in the finally block can be simplified to be more concise and robust. By combining the import and shutdown logic within a single try...except block, you can reduce nesting and eliminate the intermediate ray_available variable. This makes the code easier to read and maintain.

Suggested change

	# Clean up Ray resources to prevent atexit callback errors
	try:
	import ray
	ray_available = True
	except ImportError:
	ray_available = False
	if self._use_ray and ray_available:
	try:
	if ray.is_initialized():
	ray.shutdown()
	if self._verbose:
	logger.info('Ray shutdown completed successfully')
	except Exception as e:
	# Suppress Ray shutdown errors to avoid atexit callback issues
	if self._verbose:
	logger.warning(f'Ray shutdown warning (can be safely ignored): {e}')
	# Clean up Ray resources to prevent atexit callback errors
	if self._use_ray:
	try:
	import ray
	if ray.is_initialized():
	ray.shutdown()
	if self._verbose:
	logger.info('Ray shutdown completed successfully')
	except ImportError:
	# Ray is not installed, so no cleanup is needed.
	pass
	except Exception as e:
	# Suppress Ray shutdown errors to avoid atexit callback issues
	if self._verbose:
	logger.warning(f'Ray shutdown warning (can be safely ignored): {e}')