Fix crash dump replay script for image data replay

[ConnorLi96]

Motivation

When replaying crash dumps containing multimodal requests (images/videos/audio), asdict() converts ImageData objects to {'url': '...'} dicts, but the /generate API expects plain URL strings. This causes replay failures for multimodal requests.

Modifications

Added normalize_mm_data_item(): Converts {'url': '...'} dict back to plain URL string
Added normalize_mm_data(): Handles nested list structures in multimodal data fields
Added normalize_request_data(): Normalizes image_data, video_data, audio_data fields after asdict() conversion
Updated run_one_request_internal() to use the normalization pipeline

Accuracy Tests

N/A
Just script update and have verified locally, no need to do Accuracy and Benchmarking test.

Benchmarking and Profiling

N/A

Checklist

• Format your code according to the Format code with pre-commit.
• Add unit tests according to the Run and add unit tests.
• Update documentation according to Write documentations.
• Provide accuracy and speed benchmark results according to Test the accuracy and Benchmark the speed.

[gemini-code-assist]

Summary of Changes

Hello @ConnorLi96, 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 resolves an issue where replaying crash dumps containing multimodal requests would fail because asdict() would convert ImageData objects into dictionaries {'url': '...'} instead of the plain URL strings expected by the /generate API. The changes introduce a robust normalization process to convert these dictionary representations back into simple URL strings, ensuring successful and accurate replay of multimodal requests.

Highlights

• Multimodal Data Normalization: Introduced normalize_mm_data_item(), normalize_mm_data(), and normalize_request_data() functions to correctly handle multimodal data (images, videos, audio) during replay by converting dictionary representations to plain URL strings.
• Replay Compatibility Fix: Modified run_one_request_internal() to integrate the new normalization pipeline, ensuring that asdict() converted ImageData objects are properly transformed into the expected plain URL strings for the /generate API, thereby preventing replay failures for multimodal requests.

🧠 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 effectively addresses a bug in replaying crash dumps with multimodal data by normalizing the data structure before it's sent to the API. The logic is sound and the changes are well-targeted. I have one suggestion to refactor a new helper function to be more robust and readable by using recursion, which will allow it to handle arbitrarily nested data structures.

[gemini-code-assist]

The normalize_mm_data function can be simplified and made more robust. The current implementation is somewhat complex and only handles a single level of list nesting. A recursive approach would be cleaner, more concise, and capable of handling arbitrarily nested lists of multimodal data. This also removes the need for the explicit if data is None check, as the recursive call handles it gracefully.

def normalize_mm_data(data):
    if isinstance(data, list):
        return [normalize_mm_data(item) for item in data]
    return normalize_mm_data_item(data)