Fix link check workflow failures (#59963)

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

Author: heiskr

.github/workflows/link-check-internal.yml

@@ -108,6 +108,7 @@ jobs:
     needs: [setup-matrix, check-internal-links]
     runs-on: ubuntu-latest
     permissions:
+      contents: read
       issues: write
     steps:
       - name: Checkout

content/copilot/responsible-use/copilot-in-github-desktop.md

@@ -39,7 +39,7 @@ The feature is intended to supplement rather than replace a human's work to draf
 
 ### Provide feedback
 
-If you encounter any issues or limitations with {% data variables.copilot.copilot_desktop_short %}, you can provide feedback by creating an issue in the [{% data variables.product.prodname_desktop %} open source repository](https://github.com/desktop/desktop/issues/new?template=bug_report.yaml ). This can help the developers to improve the tool and address any concerns or limitations.
+If you encounter any issues or limitations with {% data variables.copilot.copilot_desktop_short %}, you can provide feedback by creating an issue in the [{% data variables.product.prodname_desktop %} open source repository](https://github.com/desktop/desktop/issues/new?template=bug_report.yaml). This can help the developers to improve the tool and address any concerns or limitations.
 
 ## Limitations of {% data variables.copilot.copilot_desktop_short %}
 

src/links/lib/extract-links.ts

@@ -16,8 +16,9 @@ import type { Context, Page } from '@/types'
 // Link patterns for Markdown
 const INTERNAL_LINK_PATTERN = /\]\(\/[^)]+\)/g
 const AUTOTITLE_LINK_PATTERN = /\[AUTOTITLE\]\(([^)]+)\)/g
-// Handles one level of balanced parentheses in URLs (e.g., Wikipedia links)
-const EXTERNAL_LINK_PATTERN = /\]\((https?:\/\/(?:[^()\s]+|\([^()]*\))*)\)/g
+// Handles one level of balanced parentheses in URLs (e.g., Wikipedia links).
+// Uses an unrolled loop to avoid catastrophic backtracking on malformed URLs.
+const EXTERNAL_LINK_PATTERN = /\]\((https?:\/\/[^()\s]*(?:\([^()]*\)[^()\s]*)*)\)/g
 const IMAGE_LINK_PATTERN = /!\[[^\]]*\]\(([^)]+)\)/g
 
 // Anchor link patterns (for same-page links)

src/links/scripts/check-links-external.ts

@@ -155,10 +155,20 @@ async function extractAllExternalLinks(): Promise<Map<string, { file: string; li
 
   // Find all Markdown files
   const files = await glob('content/**/*.md')
+  console.log(`Found ${files.length} Markdown files to scan`)
 
-  for (const file of files) {
+  const extractStart = Date.now()
+  for (let i = 0; i < files.length; i++) {
+    const file = files[i]
+    const fileStart = Date.now()
     const content = fs.readFileSync(file, 'utf-8')
     const result = extractLinksFromMarkdown(content)
+    const fileMs = Date.now() - fileStart
+
+    // Warn if a single file takes longer than 1 second (possible regex issue)
+    if (fileMs > 1000) {
+      console.warn(`  ⚠️  Slow extraction: ${file} took ${(fileMs / 1000).toFixed(1)}s`)
+    }
 
     for (const link of result.externalLinks) {
       // Only check HTTPS links
@@ -173,8 +183,16 @@ async function extractAllExternalLinks(): Promise<Map<string, { file: string; li
       }
       links.get(url)!.push({ file, line: link.line })
     }
+
+    if ((i + 1) % 500 === 0) {
+      const elapsed = ((Date.now() - extractStart) / 1000).toFixed(0)
+      console.log(`  Scanned ${i + 1}/${files.length} files (${elapsed}s elapsed)`)
+    }
   }
 
+  const totalElapsed = ((Date.now() - extractStart) / 1000).toFixed(1)
+  console.log(`Extraction complete in ${totalElapsed}s`)
+
   return links
 }
 
@@ -275,7 +293,13 @@ async function main() {
 
     // Progress update every 100 URLs
     if (checkedCount % 100 === 0) {
-      console.log(`  Checked ${checkedCount}/${maxUrls} URLs...`)
+      const elapsed = ((Date.now() - startTime) / 1000).toFixed(0)
+      const rate = (checkedCount / (Date.now() - startTime)) * 1000 * 60
+      const remaining = maxUrls - checkedCount
+      const etaMin = (remaining / rate).toFixed(0)
+      console.log(
+        `  Checked ${checkedCount}/${maxUrls} URLs (${cachedCount} cached) — ${elapsed}s elapsed, ~${etaMin}m remaining`,
+      )
     }
 
     // Small delay between non-cached requests to avoid rate limiting