Skip to content

test(#9426): make repos and scaffold tests pass on macOS/arm64 - #9436

Open
fullsend-ai-coder[bot] wants to merge 2 commits into
mainfrom
agent/9426-macos-test-portability
Open

fullsend-ai-coder[bot] wants to merge 2 commits into
mainfrom
agent/9426-macos-test-portability

Conversation

@fullsend-ai-coder

Copy link
Copy Markdown
Contributor

Summary

Fixes two test-only portability failures that stopped make go-test from finishing on macOS/arm64. Production code is unchanged. CI on Linux/amd64 was not affected.

Changes

  • internal/repos/manifest_test.go: TestValidate_LocalConfigSourceIsContainedAndResolvedRelativeToManifest built its expected path from the raw t.TempDir(). Validate deliberately resolves local config_base sources against the symlink-canonicalized manifest directory. On macOS the temp dir is under /var, which is a symlink to /private/var. The expected value now uses filepath.EvalSymlinks(base). The escape and symlink-escape assertions are unchanged.
  • internal/scaffold/gitlab_job_scripts_test.go: runInstallFullsendCLIScript now also stubs uname. The stub prints $UNAME_M_STUB, defaulting to x86_64. Before this, the release-pin path ran the host's real uname -m. macOS reports arm64, which install-fullsend-cli.sh rejects before the stubbed curl runs, because the script targets Linux runners. The script itself is unchanged. The new TestInstallFullsendCLIScript_ReleasePinArchitecture test covers the aarch64 mapping and the unsupported-architecture error.

Testing

  • Reproduced the manifest failure on Linux by pointing TMPDIR at a symlinked directory. The test fails without the fix and passes with it.
  • The new architecture test feeds arm64 through the stub and confirms the exact error from the issue.
  • go test ./internal/repos/ ./internal/scaffold/ passes, as do go vet and gofmt.
  • pre-commit could not fetch remote hook repos in the sandbox (TLS verification failure). I ran the applicable hooks directly instead: gofmt, go vet, git diff --check, lint-broken-symlinks, plus a gitleaks secret scan.
  • Not run on real macOS hardware.

🤖 Generated with Claude Code


Closes #9426

Post-script verification

  • Branch is not main/master (agent/9426-macos-test-portability)
  • Secret scan passed (gitleaks — dda4623665ec7f034c0e8cd503e2aa8e0220565a..HEAD)
  • PR body secret scan passed (gitleaks — no-git)
  • Final PR description secret scan passed (gitleaks — no-git)

@fullsend-ai-review

fullsend-ai-review Bot commented Nov 13, 2026 •

Copy link
Copy Markdown

🤖 Finished Review · ✅ Success · Started 4:48 PM UTC · Completed 4:54 PM UTC

Commit: bbaf16e · View workflow run →

Runtime: pi · Model: sonnet → claude-sonnet-5-5 · Effort: high · Cost: $0.88

@fullsend-ai-review fullsend-ai-review Bot added the risk/moderate PR risk: moderate label Nov 13, 2026
@fullsend-ai-review

fullsend-ai-review Bot commented Nov 13, 2026 •

Copy link
Copy Markdown

Risk Assessment: moderate (2/5)

Details

Small test-only portability fix by a bot with exact issue alignment, elevated to moderate risk by recent regression history and high change coupling in the touched test files.

Previous run

Risk Assessment: moderate (2/5)

Details

Small test-only portability fix by a bot with exact issue alignment, elevated to moderate risk by recent regression history and high change coupling in the touched test files.

Previous run (2)

Risk Assessment: moderate (2/5)

Details

Small test-only portability fix by a bot with exact issue alignment, elevated to moderate risk by recent regression history and high change coupling in the touched test files.

@fullsend-ai-review

fullsend-ai-review Bot commented Nov 13, 2026 •

Copy link
Copy Markdown

Looks good to me

Previous run

<!-- fullsend:review-findings-v2:eyJ2ZXJzaW9uIjoyLCJhY3Rpb24iOiJhcHByb3ZlIiwiZmluZGluZ3MiOltdfQ== -->

Looks good to me

Previous run (2)

<!-- fullsend:review-diff-hash:6f419e6d86eaf0ad0ffacffeadc7fee1370e95be282f5656b50c7b9f33dc32da -->
<!-- fullsend:review-head:d73413fd38a3949aea24de9e7d66e56170918d8c -->
<!-- fullsend:review-findings-v2:eyJ2ZXJzaW9uIjoyLCJhY3Rpb24iOiJhcHByb3ZlIiwiZmluZGluZ3MiOltdfQ== -->

Looks good to me

Previous run (3)

<!-- fullsend:review-findings-v2:eyJ2ZXJzaW9uIjoyLCJhY3Rpb24iOiJhcHByb3ZlIiwiZmluZGluZ3MiOltdfQ== -->

Looks good to me

Previous run (4)

<!-- fullsend:review-findings-v2:eyJ2ZXJzaW9uIjoyLCJhY3Rpb24iOiJhcHByb3ZlIiwiZmluZGluZ3MiOltdfQ== -->

Looks good to me


Labels: Test-only Go change touching test code.

fullsend-ai-review[bot]

This comment was marked as outdated.

@fullsend-ai-review fullsend-ai-review Bot added ready-for-merge All reviewers approved — ready to merge go Pull requests that update go code testing labels Nov 13, 2026
@codecov

codecov Bot commented Nov 13, 2026

Copy link
Copy Markdown

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!

@ravelley

Copy link
Copy Markdown
Contributor

/fs-fix rebase and resolve conflicts

@fullsend-ai-coder

fullsend-ai-coder Bot commented Nov 16, 2026 •

Copy link
Copy Markdown
Contributor Author

🤖 Finished Fix · ✅ Success · Started 3:22 PM UTC · Completed 3:27 PM UTC

Commit: bbaf16e · View workflow run →

Runtime: claude · Model: sonnet → claude-sonnet-5-5 · Effort: high · Cost: $0.36

Two tests failed on macOS/arm64 because of host-dependent behavior in
the tests, not in production code:

- TestValidate_LocalConfigSourceIsContainedAndResolvedRelativeToManifest
  built its expected config source from the raw t.TempDir(). Validate
  intentionally resolves against the symlink-canonicalized manifest
  directory, and on macOS the temp dir lives under /var, a symlink to
  /private/var. The expected value now uses filepath.EvalSymlinks(base).
  The escape and symlink-escape assertions are unchanged.
- runInstallFullsendCLIScript stubbed git, curl, fullsend and go but
  not uname, so the release path ran the host's `uname -m`. macOS
  reports arm64, which install-fullsend-cli.sh (targeting Linux
  runners) rejects before the stubbed curl runs. The helper now stubs
  uname to print $UNAME_M_STUB (default x86_64). A new test covers the
  aarch64 mapping and the unsupported-architecture error.

Verified on Linux with TMPDIR pointed through a symlink (fails before,
passes after) and with the uname stub reporting arm64.

Note: pre-commit could not fetch remote hook repos (TLS verification
failure in the sandbox). Ran the applicable hooks directly instead:
gofmt, go vet, git diff --check, lint-broken-symlinks; gitleaks via
scan-secrets.

Closes #9426

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
@fullsend-ai-coder

Copy link
Copy Markdown
Contributor Author

🔧 Fix agent — iteration 1 (human-triggered)

Rebased the PR branch onto main and resolved the conflict in gitlab_job_scripts_test.go by combining upstream stub changes with the uname stub. go vet, gofmt, and go test for internal/repos and internal/scaffold pass.

Fixed (1):

  1. Rebase onto main and resolve merge conflicts (internal/scaffold/gitlab_job_scripts_test.go): Rebased the single PR commit onto origin/main. The forge reported a conflict in the install-fullsend-cli test helper, where main had added FULLSEND_STUB_VERSION and CURL_STUB_LATEST_TAG stub behavior and new image-skip tests. Resolved by keeping both sides: the new stub options and tests from main, plus the uname stub and the architecture test from this PR. Merged the helper doc comment to describe both.

Tests: passed

Conflict update: rebased; strategy rebase; forge CONFLICTING; target main@c095e16f2c54b49a9a86c93b07005ff91a5fb588

Next steps:

  • /fs-review — request a re-review of the changes
  • /fs-fix <your instruction> — run another fix pass with specific guidance
  • Push commits directly — review re-runs automatically on push

Updated by fullsend fix agent

@fullsend-ai-review

fullsend-ai-review Bot commented Nov 16, 2026 •

Copy link
Copy Markdown

🤖 Finished Review · ✅ Success · Started 3:30 PM UTC · Completed 3:35 PM UTC

Commit: d73413f · View workflow run →

Runtime: pi · Model: sonnet → claude-sonnet-5-5 · Effort: high · Cost: $0.89

@fullsend-ai-review fullsend-ai-review Bot removed the ready-for-merge All reviewers approved — ready to merge label Nov 16, 2026
@fullsend-ai-review
fullsend-ai-review Bot dismissed their stale review November 16, 2026 15:34

Superseded by updated review

fullsend-ai-review[bot]

This comment was marked as outdated.

@fullsend-ai-review fullsend-ai-review Bot added the ready-for-merge All reviewers approved — ready to merge label Nov 16, 2026
@ravelley
ravelley enabled auto-merge November 16, 2026 16:10
@fullsend-ai-review

fullsend-ai-review Bot commented Nov 17, 2026 •

Copy link
Copy Markdown

🤖 Finished Review · ✅ Success · Started 5:47 PM UTC · Completed 5:53 PM UTC

Commit: 37389c7 · View workflow run →

Runtime: pi · Model: sonnet → claude-sonnet-5-5 · Effort: high · Cost: $0.91

@fullsend-ai-review fullsend-ai-review Bot removed the ready-for-merge All reviewers approved — ready to merge label Nov 17, 2026
@fullsend-ai-review
fullsend-ai-review Bot dismissed their stale review November 17, 2026 17:52

Superseded by updated review

@fullsend-ai-review fullsend-ai-review Bot added the ready-for-merge All reviewers approved — ready to merge label Nov 17, 2026

This branch was previously deployed

1 inactive deployment
dev — 37389c7b Deployed Nov 17, 2026 by ravelley via behaviour #20515
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

go Pull requests that update go code ready-for-merge All reviewers approved — ready to merge ready-for-review Triggers review agent dispatch risk/moderate PR risk: moderate testing

Projects

None yet

Development

Successfully merging this pull request may close these issues.

make go-test fails on macOS/arm64 in repos and scaffold tests

1 participant