✅ Automate URL check

.github/workflows/check-urls.yml

@@ -0,0 +1,88 @@
+#
+# check-urls.yml
+# Check (most) distinct URLs in the project for reachability
+#
+
+name: Check URLs
+
+on:
+  schedule:
+    - cron: '0 4 * * 2' # Tues 4:00 AM UTC
+
+jobs:
+  check_urls:
+    name: Check All URLs
+    if: github.repository == 'MarlinFirmware/Marlin'
+
+    runs-on: ubuntu-latest
+
+    permissions:
+      issues: write
+      pull-requests: write
+
+    steps:
+
+    - name: Check out bugfix-2.1.x
+      uses: actions/checkout@v4
+      with:
+        ref: bugfix-2.1.x
+
+    - name: Check All URLs
+      env:
+        GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+        GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+      # Exit with an error to notify fail on URL tests
+      run: |
+        UA="Mozilla/5.0 (Linux; Android 10; SM-G996U Build/QP1A.190711.020; wv) AppleWebKit/537.36 (KHTML, like Gecko) Version/4.0 Mobile Safari/537.36"
+        UTMP=$(mktemp)
+        grep -R -E "https?:\/\/[^ \"''\(\)\<\>]+" . 2>/dev/null \
+          | grep -v "Binary file" \
+          | sed -E "s/\/https?:\/\//\//" \
+          | sed -E 's/.*\((https?:\/\/[^ ]+)\).*$/\1/' \
+          | sed -E 's/.*\[(https?:\/\/[^ ]+)\].*$/\1/' \
+          | sed -E 's/.*(https?:\/\/[^ \"''()<>]+).*/\1/' \
+          | grep -vE "(127\.0\.0\.1|localhost|myserver|doc\.qt\.io|docs\.google\.com|raw\.githubusercontent\.com|[\${}])" \
+          | sed -E 's/]$//' | sed -E "s/'$//" | sed -E "s/[#.',]+$//" \
+          | sed -E 's/youtu\.be\/(.+)/www.youtube.com\/watch?v=\1/' \
+          | sort -u -R \
+          >"$UTMP"
+
+          ISERR=
+          declare -a BADURLS
+          while IFS= read -r URL
+          do
+            echo -n "Checking ${URL} ... "
+            HEAD=$(curl -s -I -A "${UA}" --request GET "${URL}" 2>/dev/null) ; HERR=$?
+            if [[ $HERR > 0 ]]; then
+              # Error 92 may be domain blocking curl / wget
+              [[ $HERR == 92 ]] || { ISERR=1 ; BADURLS+=($URL) ; }
+              echo "[FAIL ($HERR)]"
+            else
+              HEAD1=$(echo $HEAD | head -n1)
+              EMSG=
+              case "$HEAD1" in
+                *" 301"*) EMSG="[Moved Permanently]" ;;
+                *" 302"*) EMSG="[Moved Temporarily]" ;;
+                *" 303"*) echo "[See Other]" ;;
+                *" 400"*) EMSG="[Invalid Request]" ;;
+                *" 403"*) EMSG="[Forbidden]" ;;
+                *" 404"*) EMSG="[Not Found]" ;;
+                *" 503"*) EMSG="[Unavailable]" ;;
+                *" 200"*) echo "[ OK ]" ;;
+                       *) EMSG="[Other Err]" ;;
+              esac
+              [[ -n $EMSG ]] && { ISERR=1 ; BADURLS+=($URL) ; echo $EMSG ; }
+            fi
+          done <"$UTMP"
+
+          if [[ -n $ISERR ]]; then
+            # Join bad URLs into a bulleted markdown list
+            printf -v BADSTR -- "- %s\n" "${BADURLS[@]}"
+            BODY=$(echo -e "URL Checker reports one or more URLs could not be reached:\n${BADSTR}")
+            echo -e "\n$BODY"
+            #gh issue comment 26975 --repo $GITHUB_REPOSITORY --body "${BODY}"
+            exit 1
+          fi
+
+          echo -e "\nURL Check Passed."
+          exit 0

buildroot/share/PlatformIO/scripts/signature.py

@@ -31,7 +31,6 @@ def enabled_defines(filepath):
     our crude scraping method and the actual compiler output.
     We end up with the actual configured state,
     better than what the config files say. You can then use the
-    a decent reflection of all enabled options that (probably) came from
     resulting config.ini to produce more exact configuration files.
     '''
     outdict = {}
@@ -188,7 +187,7 @@ def tryint(key):
 
         ini_fmt = '{0:40} = {1}'
         ext_fmt = '{0:40}   {1}'
-        ignore = ('CONFIGURATION_H_VERSION', 'CONFIGURATION_ADV_H_VERSION', 'CONFIG_EXPORT')
+        ignore = ('CONFIGURATION_H_VERSION', 'CONFIGURATION_ADV_H_VERSION', 'CONFIG_EXAMPLES_DIR', 'CONFIG_EXPORT')
 
         if extended_dump:
             # Extended export will dump config options by section