Merge branch 'main' into codespaces-prebuilds-ga

Author: saritai

.github/actions-scripts/content-changes-table-comment.js

@@ -20,6 +20,11 @@ if (!APP_URL) {
   throw new Error(`APP_URL environment variable not set`)
 }
 
+// the max size of the comment (in bytes)
+// the action we use to post the comment caps out at about 144kb
+// see docs-engineering#1849 for more info
+const MAX_COMMENT_SIZE = 125000
+
 const PROD_URL = 'https://docs.github.com'
 const octokit = github.getOctokit(GITHUB_TOKEN)
 
@@ -131,5 +136,25 @@ const lines = await Promise.all(
     return `| ${contentCell} | ${previewCell} | ${prodCell} | ${note} |`
   })
 )
-markdownTable += lines.join('\n')
+
+// this section limits the size of the comment
+const cappedLines = []
+let underMax = true
+
+lines.reduce((previous, current, index, array) => {
+  if (underMax) {
+    if (previous + current.length > MAX_COMMENT_SIZE) {
+      underMax = false
+      cappedLines.push('**Note** There are more changes in this PR than we can show.')
+      return previous
+    }
+
+    cappedLines.push(array[index])
+    return previous + current.length
+  }
+  return previous
+}, markdownTable.length)
+
+markdownTable += cappedLines.join('\n')
+
 setOutput('changesTable', markdownTable)

.github/workflows/check-all-english-links.yml

@@ -17,7 +17,7 @@ jobs:
   check_all_english_links:
     name: Check all links
     if: github.repository == 'github/docs-internal'
-    runs-on: ubuntu-latest
+    runs-on: self-hosted
     env:
       GITHUB_TOKEN: ${{ secrets.DOCUBOT_READORG_REPO_WORKFLOW_SCOPES }}
       FIRST_RESPONDER_PROJECT: Docs content first responder
@@ -50,18 +50,32 @@ jobs:
           NODE_ENV: production
           PORT: 4000
           DISABLE_OVERLOAD_PROTECTION: true
-          DISABLE_RENDER_CACHING: true
+          DISABLE_RENDERING_CACHE: true
           # We don't want or need the changelog entries in this context.
           CHANGELOG_DISABLED: true
           # The default is 10s. But because this runs overnight, we can
           # be a lot more patient.
           REQUEST_TIMEOUT: 20000
+          # Don't care about CDN caching image URLs
+          DISABLE_REWRITE_ASSET_URLS: true
         run: |
-          node server.mjs &
+          node server.mjs > /tmp/stdout.log 2> /tmp/stderr.log &
           sleep 5
-          curl --retry-connrefused --retry 3 -I http://localhost:4000/
+          curl --retry-connrefused --retry 4 -I http://localhost:4000/
+
+      - if: ${{ failure() }}
+        name: Debug server outputs on errors
+        run: |
+          echo "____STDOUT____"
+          cat /tmp/stdout.log
+          echo "____STDERR____"
+          cat /tmp/stderr.log
 
       - name: Run script
+        env:
+          # The default is 300 which works OK on a fast macbook pro
+          # but not so well in Actions.
+          LINKINATOR_CONCURRENCY: 100
         run: |
           script/check-english-links.js > broken_links.md
 
@@ -73,11 +87,6 @@ jobs:
       #
       # https://docs.github.com/actions/reference/context-and-expression-syntax-for-github-actions#job-status-check-functions
 
-      - if: ${{ failure() }}
-        name: Debug broken_links.md
-        run: |
-          ls -lh broken_links.md
-          wc -l broken_links.md
       - uses: actions/upload-artifact@6673cd052c4cd6fcf4b4e6e60ea986c889389535
         if: ${{ failure() }}
         with:

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

@@ -49,6 +49,9 @@ jobs:
         run: cat $HOME/files.json
 
       - name: Link check (warnings, changed files)
+        env:
+          # Don't care about CDN caching image URLs
+          DISABLE_REWRITE_ASSET_URLS: true
         run: |
           ./script/rendered-content-link-checker.mjs \
             --language en \
@@ -59,6 +62,9 @@ jobs:
             --list $HOME/files.json
 
       - name: Link check (critical, all files)
+        env:
+          # Don't care about CDN caching image URLs
+          DISABLE_REWRITE_ASSET_URLS: true
         run: |
           ./script/rendered-content-link-checker.mjs \
             --language en \