fix: Fix flaky TestGenerateContentTracing

[pawel-maciejczek]

This PR fixes flaky TestGenerateContentTracing.
The main fix is to reset the testExporter before the test, so traces from other runs don't interfere with assertions in the test.

Before:

go test -test.fullpath=true -count=2 google.golang.org/adk/internal/llminternal
--- FAIL: TestGenerateContentTracing (0.00s)
    /usr/local/google/home/pmaciejczek/git-repos/adk-go/internal/llminternal/base_flow_tracing_test.go:76: expected 0 spans after partial response, got 2
    /usr/local/google/home/pmaciejczek/git-repos/adk-go/internal/llminternal/base_flow_tracing_test.go:91: expected 1 span after final response, got 3
    /usr/local/google/home/pmaciejczek/git-repos/adk-go/internal/llminternal/base_flow_tracing_test.go:106: expected 1 span after final response, got 3
    /usr/local/google/home/pmaciejczek/git-repos/adk-go/internal/llminternal/base_flow_tracing_test.go:120: expected 1 span, got 3
FAIL
FAIL    google.golang.org/adk/internal/llminternal      0.041s
FAIL

After:

go test -test.fullpath=true -count=100 google.golang.org/adk/internal/llminternal
ok      google.golang.org/adk/internal/llminternal      0.745s

[gemini-code-assist]

Summary of Changes

Hello @pawel-maciejczek, 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 a persistent flakiness issue in the TestGenerateContentTracing suite. The core problem stemmed from shared state in the test environment, where tracing data from prior test executions would contaminate subsequent runs. By implementing a mechanism to reset the tracing exporter before each test, the changes ensure a clean and isolated testing context, thereby stabilizing the test suite and improving its reliability.

Highlights

• Flaky Test Resolution: Addressed the flakiness in TestGenerateContentTracing by ensuring that testExporter is reset before each test run.
• Test Exporter Management: Modified setupTestTracer to explicitly reset the testExporter at the beginning of each test, preventing interference from previous test runs.
• Test Assertion Updates: Updated test assertions to directly use the global testExporter variable instead of a local exporter variable, aligning with the new setup.

🧠 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.

Changelog

• internal/llminternal/base_flow_tracing_test.go
  • Removed the return value from setupTestTracer as the testExporter is now a global variable.
  • Introduced a testExporter.Reset() call at the beginning of setupTestTracer to clear traces from previous runs.
  • Updated all calls to exporter.GetSpans() to testExporter.GetSpans() and span to gotSpan to reflect the use of the global exporter.
  • Added a comment explaining why otel.SetTracerProvider is called only once.

Activity

• The pull request was created by pawel-maciejczek to address a flaky test.

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 addresses a flaky test, TestGenerateContentTracing, by ensuring the trace exporter is reset before each test run. This prevents state from one test from interfering with another. The changes involve modifying the setupTestTracer helper function to reset a global testExporter and removing its return value. Consequently, all calls to this function and usages of the exporter have been updated throughout the test file. The changes are correct, effectively fix the flakiness, and also include minor code quality improvements like avoiding redundant function calls and using more idiomatic variable names in tests. The pull request is well-executed and ready for merging.