test: fix Windows CI execution and resolve exposed platform failures#24476
test: fix Windows CI execution and resolve exposed platform failures#24476
Conversation
|
Note Gemini is unable to generate a summary for this pull request due to the file types involved not being currently supported. |
|
Hi @ehedlund, thank you so much for your contribution to Gemini CLI! We really appreciate the time and effort you've put into this. We're making some updates to our contribution process to improve how we track and review changes. Please take a moment to review our recent discussion post: Improving Our Contribution Process & Introducing New Guidelines. Key Update: Starting January 26, 2026, the Gemini CLI project will require all pull requests to be associated with an existing issue. Any pull requests not linked to an issue by that date will be automatically closed. Thank you for your understanding and for being a part of our community! |
|
Size Change: -174 B (0%) Total Size: 34 MB
ℹ️ View Unchanged
|
…prevent silent skips
Summary
This PR enables core package tests on Windows CI (fixing a PowerShell bug that was silently skipping them) and addresses the cross-platform test failures that were consequently uncovered, ensuring the test suite passes consistently on macOS, Linux, and Windows.
Details
1. CI Workflow Fixes (Windows)
npm run test:ci --workspacearguments in.github/workflows/ci.yml. Without quotes, PowerShell interpreted@(e.g.,@google/gemini-cli-core) as a splatting operator and dropped the argument, causing tests to be silently skipped.if ($LASTEXITCODE -ne 0) { exit $LASTEXITCODE }checks afternpmcommands. PowerShell continues execution by default when an external command fails, meaning subsequent successful steps were previously overwriting the failure code and falsely reporting a green build.2. Cross-Platform Test Fixes
To get the newly unskipped tests passing across all platforms, several types of adjustments were made:
/tmpas a symlink to/private/tmp. Assertions were updated to useresolveToRealPath()so that expected output accurately mirrors runtime path evaluations.fsandvalue-resolver) to behave reliably across OSes without relying on hardcoded, platform-specific path structures.Note: The Windows Sandbox tests have been temporarily disabled in this PR and will be properly addressed and re-enabled in a separate effort (#24480).