.github/scripts/verify-labeler.cjs (5)

27-66: ⚡ Quick win

Remove unused createMockContext() function.

This function is defined but never called. The E2E testing approach appears to have been abandoned in favor of the manual label simulation starting at line 271.

🧹 Proposed cleanup

-function createMockContext(title, files, additions = 50, deletions = 30) {
-  const labelsAdded = [];
-
-  const mockGithub = {
-    rest: {
-      pulls: {
-        get: async () => ({
-          data: { additions, deletions, changed_files: files.length },
-        }),
-        listFiles: async () => ({
-          data: files.map(f => ({ filename: f })),
-        }),
-      },
-    },
-  };
-
-  const mockContext = {
-    payload: {
-      pull_request: { title },
-    },
-  };
-
-  // Intercept addLabels to track what would be added
-  const originalLabelPR = labelPR;
-  // We patch by wrapping the bot context build
-  const patchedGithub = {
-    ...mockGithub,
-    rest: {
-      ...mockGithub.rest,
-      issues: {
-        addLabels: async (params) => {
-          labelsAdded.push(...params.labels);
-          return { data: {} };
-        },
-      },
-    },
-  };
-
-  return { github: patchedGithub, context: mockContext, labelsAdded };
-}
-

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In @.github/scripts/verify-labeler.cjs around lines 27 - 66, Remove the unused
testing helper createMockContext (and its inner bindings labelsAdded,
mockGithub, mockContext, originalLabelPR, patchedGithub) from
.github/scripts/verify-labeler.cjs; locate the function declaration named
createMockContext and delete it and any dangling references, then run lint/tests
to ensure no remaining references to createMockContext or its variables remain.

---

227-266: ⚡ Quick win

Remove unused runE2ETest() function.

This async function is defined but never called. The E2E testing section (lines 224-225) prints a header but doesn't execute any tests. Remove this dead code or implement the E2E test suite.

🧹 Proposed cleanup

-// ── End-to-end: simulate labelPR with a mock payload ──
-console.log('\n── E2E: labelPR (full flow) ──');
-
-async function runE2ETest(name, title, files, additions, deletions) {
-  const labelsAdded = [];
-  const totalChanges = additions + deletions;
-
-  const github = {
-    rest: {
-      pulls: {
-        get: async () => ({ data: { additions, deletions, changed_files: files.length } }),
-        listFiles: async () => ({ data: files.map(f => ({ filename: f })) }),
-      },
-      issues: {
-        addLabels: async (params) => { labelsAdded.push(...params.labels); return { data: {} }; },
-      },
-    },
-  };
-  const context = {
-    payload: { pull_request: { title, number: 42 } },
-    repo: { owner: 'test', repo: 'test' },
-  };
-
-  // Manually inline the addLabels to capture labels
-  // The real labelPR uses buildBotContext({github, context}) which returns {owner,repo,number,github,...}
-  // So we provide the right shape
-  const botContext = { github, owner: 'test', repo: 'test', number: 42, _labels: labelsAdded };
-  github.rest.issues.addLabels = async (params) => { labelsAdded.push(...params.labels); return { data: {} }; };
-
-  // Override buildBotContext to inject our _labels tracker
-  const capturedLabels = [];
-  const buildBotContextOrig = require('./helpers/api.cjs').buildBotContext;
-  // Instead, just run labelPR with a wrapped github
-  try {
-    await labelPR({ github, context });
-  } catch (e) {
-    // The labelPR function uses buildBotContext({github, context}) and expects specific shape
-    // If the mock isn't perfect, we still get console output
-    console.log(`    ${name}: ${e.message}`);
-    console.log(`    (E2E mocking requires exact API shape — see pure function tests above for coverage)`);
-    return;
-  }
-}
-
-// Fall back to pure function verification + manual label assembly test

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In @.github/scripts/verify-labeler.cjs around lines 227 - 266, The async helper
runE2ETest is dead code (defined but never invoked); remove the entire
runE2ETest function declaration from the script to eliminate the unused
function, or if E2E coverage is intended, replace the removal by invoking
runE2ETest appropriately from the E2E section and ensure the mocked
github/context shape matches what labelPR and buildBotContext expect
(references: runE2ETest, labelPR, buildBotContext).

---

72-86: ⚡ Quick win

Remove unused vm.Script setup.

The vm.Script approach to extract internal functions from pr-labeler.cjs is set up but never executed. The script later duplicates the functions directly (lines 90-146) instead.

🧹 Proposed cleanup

-const fs = require('fs');
-const vm = require('vm');
-
-const source = fs.readFileSync(require.resolve('./pr-labeler.cjs'), 'utf8');
-// Wrap in an IIFE that returns the internal functions via module scope
-const script = new vm.Script(`
-  const module = { exports: {} };
-  const require = (m) => {
-    if (m === './helpers/api.cjs') return { buildBotContext: (x) => x, addLabels: async (ctx, labels) => { ctx._labels.push(...labels); return {success:true}; } };
-    if (m === './helpers/config-loader.cjs') return { loadAutomationConfig: () => (${JSON.stringify(config)}) };
-    return {};
-  };
-  ${source}
-  module.exports;
-`);
-
-// We need to eval in a sandbox. Simpler: just duplicate the pure functions here.
-

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In @.github/scripts/verify-labeler.cjs around lines 72 - 86, The vm.Script setup
(variables source, script, and the vm.Script wrapper that defines
module/require) is unused and duplicates functionality later; remove the unused
vm.Script-related code (the const source = fs.readFileSync(...), the const
script = new vm.Script(`...`) block and any unused vm/require stubs) so only the
direct function duplicates remain, and ensure any required mocks referenced
later (e.g., the inline require replacement for './helpers/api.cjs' and
'./helpers/config-loader.cjs') are preserved where actually used; target the
vm.Script, source, script and unused require shim declarations in
.github/scripts/verify-labeler.cjs to clean up the dead setup.

---

171-181: ⚡ Quick win

Size threshold tests are tightly coupled to config values.

The assertions hardcode specific boundaries (10, 50, 200, 500) that assume exact config values for prLabels.size. If the config changes, these tests will break even though the logic may be correct. Consider reading actual thresholds from the loaded config to make tests more robust.

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In @.github/scripts/verify-labeler.cjs around lines 171 - 181, The tests
hardcode size boundaries causing brittleness; update the assertions for
determineSize and prLabels.size to derive threshold values from the same config
used by the labeler instead of literal numbers — load the config (the
module/variable that provides size thresholds), extract the boundary values
(e.g., smallThreshold, mediumThreshold, largeThreshold, xlargeThreshold or an
array of thresholds) and use those values in the assert calls so determineSize(
threshold-1, prLabels.size ) and determineSize( threshold, prLabels.size ) tests
reflect the configured boundaries dynamically.

---

197-201: ⚡ Quick win

Module detection test assumes exact count.

The assertion mods1.length === 2 is brittle because it depends on the exact modulePaths configuration. If the config adds overlapping patterns (e.g., src/**/*.ts mapping to a new module), this test will fail even though the detection logic is correct.

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In @.github/scripts/verify-labeler.cjs around lines 197 - 201, Test is brittle
because it asserts exact module count; update the checks around
files1/mods1/detectModules so we only verify the expected modules are present
rather than enforcing mods1.length === 2. Concretely, remove or replace the
assert('No spurious modules', mods1.length === 2) with an assertion that the
expected modules are included (e.g., ensure mods1.includes('cli') and
mods1.includes('config') or check an expectedModules array and assert every
expectedModules.every(m => mods1.includes(m))). This keeps the test focused on
presence of 'cli' and 'config' without failing when modulePaths adds overlapping
patterns.

docs/README.md (1)

6-6: ⚡ Quick win

Fix markdown heading levels in release documentation files.

Three documentation files contain the same markdown heading hierarchy violation: the "Commits" section (h3) immediately follows the main heading (h1) without an intermediate h2 level. Per CommonMark specification, heading levels should increment by one level at a time.

Affected files:

• docs/README.md line 6: ### Commits should be ## Commits
• docs/v2.0.0.md line 4: ### Commits should be ## Commits
• docs/v2.0.1.md line 4: ### Commits should be ## Commits

Fix for all three files

Change ### Commits to ## Commits in each file.

-### Commits
+## Commits

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@docs/README.md` at line 6, Replace the incorrect third-level heading "###
Commits" with a second-level heading "## Commits" in each affected release
documentation file so the heading hierarchy increments correctly from the main
H1; search for the exact string "### Commits" and change it to "## Commits" in
the files containing release notes (e.g., where the main title is followed
immediately by the "Commits" section).

Source: Linters/SAST tools

Fix all unresolved CodeRabbit comments on this PR:

• Push a commit to this branch (recommended)
• Create a new PR with the fixes