👷 Local URL checking script #39
Workflow file for this run
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
# | |
# check-urls.yml | |
# Check (most) distinct URLs in the project for reachability | |
# | |
name: Check URLs | |
on: | |
pull_request: | |
branches: | |
- bugfix-2.1.x | |
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) | |
echo "[debug 1] UTMP = ${UTMP}" | |
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" | |
echo "[debug 2] link count = $(wc -l $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" | |
echo "[debug 3]" | |
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 | |
else | |
gh issue comment 26975 --body "Comment generated by action." | |
fi | |
echo -e "\nURL Check Passed." | |
exit 0 |