generated from osc/skins-template
148 lines
4.9 KiB
YAML
148 lines
4.9 KiB
YAML
name: Test Skins
|
|
|
|
on:
|
|
push:
|
|
pull_request:
|
|
|
|
jobs:
|
|
link-check:
|
|
runs-on: ubuntu-latest
|
|
|
|
steps:
|
|
- uses: actions/checkout@v4
|
|
|
|
- name: Validate links and assets
|
|
shell: bash
|
|
run: |
|
|
set -uo pipefail
|
|
|
|
RED="\033[31m"
|
|
GREEN="\033[32m"
|
|
RESET="\033[0m"
|
|
|
|
ERRORS=()
|
|
|
|
############################################################
|
|
# Safe unicode URL decode (never produces NULL bytes)
|
|
############################################################
|
|
urldecode() {
|
|
local url="$1"
|
|
url="${url//+/ }"
|
|
printf '%s' "$url" | perl -pe 's/%([A-Fa-f0-9]{2})/chr(hex($1))/eg'
|
|
}
|
|
|
|
############################################################
|
|
# Detect Gitea "soft 404" even when HTTP=200 OK
|
|
############################################################
|
|
check_http() {
|
|
local url="$1"
|
|
echo " → Checking external: $url"
|
|
|
|
local status body
|
|
|
|
status=$(curl -Is --max-time 10 "$url" | head -n1 | awk '{print $2}')
|
|
if [[ "$status" =~ ^2|3 ]]; then
|
|
body=$(curl -Ls --max-time 10 "$url")
|
|
if echo "$body" | grep -qiE "404 Not Found|doesn't exist|File not found|Not Found"; then
|
|
return 1
|
|
fi
|
|
return 0
|
|
fi
|
|
|
|
status=$(curl -Is --max-time 10 -X GET "$url" | head -n1 | awk '{print $2}')
|
|
if [[ "$status" =~ ^2|3 ]]; then
|
|
body=$(curl -Ls --max-time 10 "$url")
|
|
if echo "$body" | grep -qiE "404 Not Found|doesn't exist|File not found|Not Found"; then
|
|
return 1
|
|
fi
|
|
return 0
|
|
fi
|
|
|
|
return 1
|
|
}
|
|
|
|
############################################################
|
|
# Local file check
|
|
############################################################
|
|
check_local() {
|
|
local path="$1"
|
|
path="${path#/}" # strip leading slash
|
|
local decoded
|
|
decoded=$(urldecode "$path")
|
|
|
|
echo " → Checking local: $decoded"
|
|
|
|
[[ -e "$decoded" ]]
|
|
}
|
|
|
|
############################################################
|
|
# Extract links from markdown + HTML
|
|
############################################################
|
|
extract_links() {
|
|
local f="$1"
|
|
|
|
grep -oE '\[[^]]*\]\([^)]*\)' "$f" \
|
|
| sed -E 's/.*\((.*)\).*/\1/'
|
|
|
|
grep -oE '!\[[^]]*\]\([^)]*\)' "$f" \
|
|
| sed -E 's/.*\((.*)\).*/\1/'
|
|
|
|
grep -oE 'https?://[^ )"]+' "$f"
|
|
|
|
grep -oE '<img[^>]*src="[^"]+"' "$f" \
|
|
| sed -E 's/.*src="([^"]*)".*/\1/'
|
|
|
|
grep -oE '<video[^>]*src="[^"]+"' "$f" \
|
|
| sed -E 's/.*src="([^"]*)".*/\1/'
|
|
}
|
|
|
|
echo "🔍 Scanning Markdown files..."
|
|
echo
|
|
|
|
############################################################
|
|
# Process all markdown files
|
|
############################################################
|
|
find . -type f -name '*.md' | while IFS= read -r mdfile; do
|
|
echo "📄 Checking: $mdfile"
|
|
|
|
while IFS= read -r url; do
|
|
[[ -z "$url" ]] && continue
|
|
[[ "$url" == mailto:* ]] && continue
|
|
|
|
# Skip versioned docs tag links
|
|
if [[ "$mdfile" == ./docs/* ]] && [[ "$url" == *"/src/tag/"* ]]; then
|
|
continue
|
|
fi
|
|
|
|
# External links
|
|
if [[ "$url" == http* ]]; then
|
|
if ! check_http "$url"; then
|
|
ERRORS+=("❌ Broken external link: $url (in $mdfile)")
|
|
fi
|
|
else
|
|
# Local links
|
|
if ! check_local "$url"; then
|
|
ERRORS+=("❌ Missing local file: $url (in $mdfile)")
|
|
fi
|
|
fi
|
|
|
|
# 🌟 FULL FIX: sanitize extract_links output BEFORE loop
|
|
done < <(extract_links "$mdfile" | tr -d '\000' | tr -d '\r' | sed 's/[[:cntrl:]]//g')
|
|
|
|
echo
|
|
done
|
|
|
|
############################################################
|
|
# Final report
|
|
############################################################
|
|
echo
|
|
if (( ${#ERRORS[@]} > 0 )); then
|
|
echo -e "${RED}✖ Errors found:${RESET}"
|
|
printf "%s\n" "${ERRORS[@]}"
|
|
echo
|
|
echo -e "${RED}❌ Failing job because broken links were found.${RESET}"
|
|
exit 1
|
|
else
|
|
echo -e "${GREEN}✔ All links OK!${RESET}"
|
|
fi
|