Handle deleted PR branches in push-to-pull-request-branch with opt-in skip mode

[Copilot]

design-decision-gate was failing when push_to_pull_request_branch ran after the target PR branch had been deleted. This change adds an opt-in mode to treat missing/deleted branch push failures as skipped instead of terminal, and enables it for that workflow.

• Safe-output behavior

  • Added ignore-missing-branch-failure to push-to-pull-request-branch.
  • When enabled, missing/deleted remote branch conditions return skipped: true instead of a hard failure.
  • Applies across branch existence checks (ls-remote, fetch/rev-parse failure paths) for consistent handling.

• Compiler + schema wiring

  • Added schema support for safe-outputs.push-to-pull-request-branch.ignore-missing-branch-failure in main_workflow_schema.json.
  • Extended workflow parsing/config structs to read ignore-missing-branch-failure.
  • Propagated the parsed value into runtime handler config (ignore_missing_branch_failure).

• Workflow enablement

  • Enabled in .github/workflows/design-decision-gate.md:

    safe-outputs:
      push-to-pull-request-branch:
        allowed-files:
          - docs/adr/**
        patch-format: bundle
        ignore-missing-branch-failure: true

  • Recompiled lock workflow to include the new handler config.

• Maintainability adjustments

  • Centralized missing-branch error text/pattern matching in the JS handler to avoid duplicated logic and keep behavior uniform.

[pelikhan]

@copilot rename to ignore-missing-branch

[Copilot]

Pull request overview

Adds an opt-in mode for push-to-pull-request-branch to treat “PR branch missing/deleted” push failures as a skipped outcome (instead of failing the workflow), and enables it for design-decision-gate.

Changes:

• Introduces ignore-missing-branch-failure in workflow YAML schema + Go compiler parsing, and emits ignore_missing_branch_failure into the runtime handler config.
• Updates the JS handler to detect missing-branch conditions across ls-remote/fetch/rev-parse paths and return skipped: true when configured, with accompanying JS tests.
• Enables the option in .github/workflows/design-decision-gate.md and recompiles the lock workflow.

Show a summary per file

File	Description
pkg/workflow/push_to_pull_request_branch.go	Adds config field and parsing for ignore-missing-branch-failure.
pkg/workflow/compiler_safe_outputs_handlers.go	Wires the parsed flag into handler-manager JSON config (ignore_missing_branch_failure).
pkg/parser/schemas/main_workflow_schema.json	Extends schema to allow ignore-missing-branch-failure under safe outputs config.
pkg/workflow/push_to_pull_request_branch_test.go	Adds compiler/lockfile regression test for the new config flag.
actions/setup/js/push_to_pull_request_branch.cjs	Implements opt-in missing-branch skip behavior and centralizes error matching.
actions/setup/js/push_to_pull_request_branch.test.cjs	Adds tests covering skip behavior for deleted/missing branches.
.github/workflows/design-decision-gate.md	Enables ignore-missing-branch-failure for the workflow.
.github/workflows/design-decision-gate.lock.yml	Updates compiled workflow config to include the new handler flag.

Copilot's findings

Tip

Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Comments suppressed due to low confidence (1)

actions/setup/js/push_to_pull_request_branch.cjs:538

• The git rev-parse --verify failure path is skipped for any error when ignore_missing_branch_failure is enabled. That can hide unrelated failures (e.g., unexpected git errors) as a missing/deleted branch. Consider checking the error text with looksLikeMissingRemoteBranchError(...) (or a tighter rev-parse-specific pattern set like "needed a single revision"/"unknown revision") before returning skipped: true; otherwise return the current hard failure.

      const missingBranchError = MISSING_BRANCH_ERROR_TEMPLATE(branchName);
      if (ignoreMissingBranchFailure) {
        core.warning(`${missingBranchError} Skipping as configured by ignore-missing-branch-failure.`);
        return { success: false, error: missingBranchError, skipped: true };
      }

• Files reviewed: 8/8 changed files
• Comments generated: 1

[Copilot]

looksLikeMissingRemoteBranchError treats any error text containing "exit code 128" as a missing-branch signal. Exit code 128 is a generic git failure (auth, network, repo state, etc.), so with ignore_missing_branch_failure enabled this can incorrectly mark real failures as skipped. Remove this pattern (or replace it with more specific stderr substrings tied to missing refs) so only true missing-branch cases are skipped.

This issue also appears on line 534 of the same file.

Suggested change

"exit code 128",

[github-actions]

🧪 Test Quality Sentinel Report

Test Quality Score: 90/100

✅ Excellent test quality

Metric	Value
New/modified tests analyzed	3
✅ Design tests (behavioral contracts)	3 (100%)
⚠️ Implementation tests (low value)	0 (0%)
Tests with error/edge cases	3 (100%)
Duplicate test clusters	0
Test inflation detected	⚠️ Yes — Go test file (4.6:1 ratio, see note)
🚨 Coding-guideline violations	0

---

Test Classification Details

Test	File	Classification	Issues Detected
TestPushToPullRequestBranchIgnoreMissingBranchFailureConfig	pkg/workflow/push_to_pull_request_branch_test.go:96	✅ Design	None — verifies new config option compiles correctly into the lock file
should skip deleted branch failure when ignore_missing_branch_failure is enabled	actions/setup/js/push_to_pull_request_branch.test.cjs:648	✅ Design	None — tests skip mode + warning emission for the pre-fetch deleted-branch path
should skip rev-parse missing branch failure when ignore_missing_branch_failure is enabled	actions/setup/js/push_to_pull_request_branch.test.cjs:708	✅ Design	None — tests skip mode + warning emission for the rev-parse deleted-branch path

---

Test Inflation Note

The Go test file gained 37 lines against 8 new production lines in push_to_pull_request_branch.go (4.6:1 ratio). This triggers the inflation penalty mechanically, but the high ratio is expected here: the test follows the standard codebase pattern that requires creating a temp directory, writing a markdown file, compiling it, reading the lock file, and asserting the output — substantial boilerplate for a single new configuration field. The 10-point deduction reflects the rubric; it is not a concern in practice.

---

Language Support

Tests analyzed:

• 🐹 Go (*_test.go): 1 test — unit (//go:build !integration) ✅
• 🟨 JavaScript (*.test.cjs): 2 tests (vitest)

JavaScript mocking note: vi.fn() is used to stub @actions/core, @actions/exec, and github.rest.*. All mocked targets are external I/O interfaces (GitHub Actions runtime, git subprocess, GitHub API) — this is the appropriate pattern for this codebase.

---

Verdict

✅ Check passed. 0% of new tests are implementation tests (threshold: 30%). All three new tests directly verify the observable behavioral contract of the ignore-missing-branch-failure opt-in feature: the Go test confirms the config is compiled correctly into the workflow's handler JSON, and the two JavaScript tests confirm the runtime skip-with-warning behavior for both deleted-branch code paths.

---

📖 Understanding Test Classifications

Design Tests (High Value) verify what the system does:

• Assert on observable outputs, return values, or state changes
• Cover error paths and boundary conditions
• Would catch a behavioral regression if deleted
• Remain valid even after internal refactoring

Implementation Tests (Low Value) verify how the system does it:

• Assert on internal function calls (mocking internals)
• Only test the happy path with typical inputs
• Break during legitimate refactoring even when behavior is correct
• Give false assurance: they pass even when the system is wrong

Goal: Shift toward tests that describe the system's behavioral contract — the promises it makes to its users and collaborators.

References: §24635898531

🧪 Test quality analysis by Test Quality Sentinel · ● 995.2K · ◷

[github-actions]

✅ Test Quality Sentinel: 90/100. Test quality is excellent — 0% of new tests are implementation tests (threshold: 30%). All 3 new tests enforce observable behavioral contracts for the ignore-missing-branch-failure feature.