fix(patch): cherry-pick 56f9688 to release/v0.54.0-pr-28700 to patch version v0.54.0 and create version 0.54.1 - #28710
Conversation
…ponse (#28700) Co-authored-by: David Pierce <davidapierce@google.com>
Summary of ChangesHello, 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 addresses a regression where user messages were being incorrectly coalesced into previous turns if the conversation history ended on an unanswered tool response. By introducing a closing model turn when such a state is detected, the system ensures that new user inputs are treated as distinct turns, maintaining proper conversational context for the model. Highlights
Using Gemini Code AssistThe 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
Customization To customize the 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 Limitations & Feedback Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counterproductive. 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. Footnotes
|
|
📊 PR Size: size/L
|
There was a problem hiding this comment.
Code Review
This pull request addresses an issue where subsequent user messages were being incorrectly coalesced into preceding unanswered tool-response turns (such as after a stream failure or cancellation). It introduces a mechanism to close these dangling turns with an interrupted response placeholder, along with comprehensive unit tests and a test-timer mock to prevent timeouts. The review feedback correctly points out that the synthetic model turn appended to the history should also be recorded via this.chatRecordingService to ensure it is properly persisted to disk.
|
Size Change: +3.62 kB (+0.01%) Total Size: 35.2 MB
ℹ️ View Unchanged
|
This PR automatically cherry-picks commit 56f9688 to patch version v0.54.0 in the stable release to create version 0.54.1.
Contains the following additional cherry picks:
#28607
#28672