Update content linting report to only include errors and expired content warnings (#56479)

Author: heiskr

.github/workflows/lint-entire-content-data-markdown.yml

@@ -41,7 +41,7 @@ jobs:
           REPORT_AUTHOR: docs-bot
           REPORT_LABEL: broken content markdown report
           REPORT_REPOSITORY: github/docs-content
-        run: npm run post-lints -- --path /tmp/lint-results.json
+        run: npm run lint-report -- --path /tmp/lint-results.json
 
       - uses: ./.github/actions/slack-alert
         if: ${{ failure() && github.event_name != 'workflow_dispatch' }}

package.json

@@ -66,7 +66,7 @@
     "move-content": "tsx src/content-render/scripts/move-content.js",
     "openapi-docs": "tsx src/rest/docs.js",
     "playwright-test": "playwright test --config src/fixtures/playwright.config.ts --project=\"Google Chrome\"",
-    "post-lints": "tsx src/content-linter/scripts/post-lints.js",
+    "lint-report": "tsx src/content-linter/scripts/lint-report.js",
     "postinstall": "cp package-lock.json .installed.package-lock.json && echo \"Updated .installed.package-lock.json\" # see husky/post-checkout and husky/post-merge",
     "precompute-pageinfo": "tsx src/article-api/scripts/precompute-pageinfo.ts",
     "prepare": "husky src/workflows/husky",

src/content-linter/scripts/post-lints.js src/content-linter/scripts/lint-report.jssrc/content-linter/scripts/post-lints.js renamed to src/content-linter/scripts/lint-report.js

@@ -5,10 +5,47 @@ import coreLib from '@actions/core'
 import github from '#src/workflows/github.ts'
 import { getEnvInputs } from '#src/workflows/get-env-inputs.ts'
 import { createReportIssue, linkReports } from '#src/workflows/issue-report.js'
+import { reportingConfig } from '#src/content-linter/style/github-docs.js'
 
 // GitHub issue body size limit is ~65k characters, so we'll use 60k as a safe limit
 const MAX_ISSUE_BODY_SIZE = 60000
 
+/**
+ * Determines if a lint result should be included in the automated report
+ * @param {Object} flaw - The lint result object
+ * @param {string} flaw.severity - 'error' or 'warning'
+ * @param {string[]} flaw.ruleNames - Array of rule names for this flaw
+ * @returns {boolean} - True if this flaw should be included in the report
+ */
+function shouldIncludeInReport(flaw) {
+  if (!flaw.ruleNames || !Array.isArray(flaw.ruleNames)) {
+    return false
+  }
+
+  // Check if any rule name is in the exclude list
+  const hasExcludedRule = flaw.ruleNames.some((ruleName) =>
+    reportingConfig.excludeRules.includes(ruleName),
+  )
+  if (hasExcludedRule) {
+    return false
+  }
+
+  // Check if severity should be included
+  if (reportingConfig.includeSeverities.includes(flaw.severity)) {
+    return true
+  }
+
+  // Check if any rule name is in the include list
+  const hasIncludedRule = flaw.ruleNames.some((ruleName) =>
+    reportingConfig.includeRules.includes(ruleName),
+  )
+  if (hasIncludedRule) {
+    return true
+  }
+
+  return false
+}
+
 // [start-readme]
 //
 // This script runs once a week via a scheduled GitHub Action to lint
@@ -46,15 +83,26 @@ async function main() {
   // or open an issue report, you might get cryptic error messages from Octokit.
   getEnvInputs(['GITHUB_TOKEN'])
 
-  core.info(`Creating issue for errors and warnings...`)
+  core.info(`Creating issue for configured lint rules...`)
 
   const parsedResults = JSON.parse(lintResults)
-  const totalFiles = Object.keys(parsedResults).length
-  let reportBody = 'The following files have markdown lint warnings/errors:\n\n'
+
+  // Filter results based on reporting configuration
+  const filteredResults = {}
+  for (const [file, flaws] of Object.entries(parsedResults)) {
+    const filteredFlaws = flaws.filter(shouldIncludeInReport)
+
+    // Only include files that have remaining flaws after filtering
+    if (filteredFlaws.length > 0) {
+      filteredResults[file] = filteredFlaws
+    }
+  }
+  const totalFiles = Object.keys(filteredResults).length
+  let reportBody = 'The following files have markdown lint issues that require attention:\n\n'
   let filesIncluded = 0
   let truncated = false
 
-  for (const [file, flaws] of Object.entries(parsedResults)) {
+  for (const [file, flaws] of Object.entries(filteredResults)) {
     const fileEntry = `File: \`${file}\`:\n\`\`\`json\n${JSON.stringify(flaws, null, 2)}\n\`\`\`\n`
 
     // Check if adding this file would exceed the size limit
@@ -77,7 +125,7 @@ async function main() {
   const reportProps = {
     core,
     octokit,
-    reportTitle: `Error(s) and warning(s) in content markdown file(s)`,
+    reportTitle: `Content linting issues requiring attention`,
     reportBody,
     reportRepository: REPORT_REPOSITORY,
     reportLabel: REPORT_LABEL,

src/content-linter/style/github-docs.js

@@ -1,3 +1,23 @@
+// Configuration for which rules should be included in automated weekly reports
+export const reportingConfig = {
+  // Always include all rules with these severities
+  includeSeverities: ['error'],
+
+  // Specific rules to include regardless of severity
+  // Add rule names (short or long form) that should always be reported
+  includeRules: [
+    'GHD038', // expired-content - Content that has passed its expiration date
+    'expired-content',
+  ],
+
+  // Specific rules to exclude from reports (overrides severity-based inclusion)
+  // Add rule names here if you want to suppress them from reports
+  excludeRules: [
+    // Example: 'GHD030' // Uncomment to exclude code-fence-line-length warnings
+    // Example: 'british-english-quotes' // Uncomment to exclude punctuation warnings
+  ],
+}
+
 const githubDocsConfig = {
   'link-punctuation': {
     // GHD001