Add mock responses for invalid safety ratings during image gen

[andrewheard]

Added mock responses for image generation using gemini-2.0-flash-exp where the backend returns additional empty SafetyRating objects in the safetyRatings array (i.e., {}, {}, {}, {}).

Note: This issue only impacts Vertex AI.

[gemini-code-assist]

Hello @andrewheard, 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!

Summary of Changes

This pull request adds mock responses for image generation using gemini-2.0-flash-exp where the backend returns additional empty SafetyRating objects in the safetyRatings array. Specifically, I've added a mock response to handle cases where the backend returns empty SafetyRating objects, ensuring the system gracefully handles such scenarios.

Highlights

• Mock Response: Added a mock response for streaming image generation with invalid safety ratings in mock-responses/vertexai/streaming-success-image-invalid-safety-ratings.txt. The original response contained an image and several safety ratings, followed by multiple empty safety rating objects. The new mock response returns empty text content and a 'STOP' finish reason.
• Mock Response: Added a mock response for unary image generation with invalid safety ratings in mock-responses/vertexai/unary-success-image-invalid-safety-ratings.json. The original response contained an image and several safety ratings, followed by multiple empty safety rating objects.

Changelog

• mock-responses/vertexai/streaming-success-image-invalid-safety-ratings.txt
  • Replaced the original mock response with a new one that returns empty text content and a 'STOP' finish reason, instead of an image and a list of safety ratings that includes empty objects.
• mock-responses/vertexai/unary-success-image-invalid-safety-ratings.json
  • Added a new mock response file containing a response with an image and several safety ratings, followed by multiple empty safety rating objects.

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.

---

In mocks we trust, when real tests fail,
To catch the bugs, and then prevail.
With safety's check,
No rating's a wreck,
Our image gen will never bail.

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. ↩

[github-actions]

Coverage Diff

GenerateContentResponse: 46 -> 48 🔵
| candidates: 41 -> 43 🔵
| | content: 40 -> 42 🔵
| | | parts: 36 -> 38 🔵
| | | | inlineData: 0 -> 2 ✅
| | | | | data: 0 -> 2 ✅
| | | | | mimeType: 0 -> 2 ✅
| | | | text: 28 -> 29 🔵
| | | role: 29 -> 31 🔵
| | finishReason: 35 -> 37 🔵
| | | STOP: 31 -> 33 🔵
| | safetyRatings: 29 -> 31 🔵
| | | category: 29 -> 31 🔵
| | | | HARM_CATEGORY_HATE_SPEECH: 27 -> 29 🔵
| | | | HARM_CATEGORY_DANGEROUS_CONTENT: 28 -> 30 🔵
| | | | HARM_CATEGORY_HARASSMENT: 29 -> 31 🔵
| | | | HARM_CATEGORY_SEXUALLY_EXPLICIT: 27 -> 29 🔵
| | | probability: 29 -> 31 🔵
| | | | NEGLIGIBLE: 28 -> 30 🔵
| | | probabilityScore: 10 -> 12 🔵
| | | severity: 10 -> 12 🔵
| | | | HARM_SEVERITY_NEGLIGIBLE: 10 -> 12 🔵
| | | severityScore: 10 -> 12 🔵
| createTime: 0 -> 2 ✅
| modelVersion: 4 -> 6 🔵
| responseId: 0 -> 2 ✅
| usageMetadata: 18 -> 20 🔵
| | candidatesTokenCount: 14 -> 16 🔵
| | candidatesTokensDetails: 1 -> 3 🔵
| | | modality: 1 -> 3 🔵
| | | | IMAGE: 0 -> 2 ✅
| | | tokenCount: 1 -> 3 🔵
| | promptTokenCount: 18 -> 20 🔵
| | promptTokensDetails: 1 -> 3 🔵
| | | modality: 1 -> 3 🔵
| | | | TEXT: 1 -> 3 🔵
| | | tokenCount: 1 -> 3 🔵
| | totalTokenCount: 17 -> 19 🔵
| | trafficType: 0 -> 2 ✅
| | | ON_DEMAND: 0 -> 2 ✅
Total Coverage: 35.02% -> 38.71% ✅

Legend:
✅ : total coverage increase
🔵 : files number increase, total coverage unaffected
🟡 : files number decrease, total coverage unaffected
❌ : total coverage decrease

[gemini-code-assist]

Warning

Gemini encountered an error creating the review. You can try again by commenting /gemini review.