wordpress static export using wget and cloudflare pages

01-export.sh

#!/usr/bin/env bash

set -eo pipefail

SITE_URL="http://localhost:8080"

cd "$(dirname "$0")"
mkdir -p export

# Check if the site is up
if ! curl -s -o /dev/null $SITE_URL; then
  echo "Site is not up"
  exit 1
fi

# Copy favicon
cp wordpress/uploads/2017/06/favicon.png export/favicon.png
cp wordpress/uploads/2017/06/favicon.png export/favicon.ico

# Add a 404 page for CloudFlare Pages
cat > export/404.html << 'EOF'
<h1>Not Found</h1>
EOF

# Copying wp-content from the container
rsync -a \
    --exclude='*.php' \
    --exclude='*.tmp*' \
    --exclude='cache/' \
    --exclude='upgrade/' \
    --exclude='upgrade-temp-backup/' \
    --exclude='w3tc/' \
    --exclude='w3tc-config/' \
    --exclude='smush/' \
    --exclude='debug.log' \
    --exclude='*.log' \
    wordpress/ export/wp-content/

# Use wget to crawl the site and save it to export/
# adjust-extension will dirs vs. files and CloudFlare Pages will handle this
wget \
  --trust-server-names \
  --recursive \
  --adjust-extension \
  --page-requisites \
  --convert-links \
  --no-parent \
  --no-host-directories \
  --directory-prefix=export \
  --tries=1 \
  --no-verbose \
  --exclude-directories=wp-content,wp-admin,wp-json \
  $SITE_URL

# Done
echo "Done."

02-post-process.sh

#!/usr/bin/env bash

set -eo pipefail

ORIGINAL_URL="http://localhost:8080"
NEW_URL="https://mysite.com"
#NEW_URL="https://mysite.pages.dev"

cd "$(dirname "$0")"

# Replace the original URL with the new URL
find export -type f \( -name "*.html" -o -name "*.css" -o -name "*.js" \) -exec sed -i '' \
    -e "s|$ORIGINAL_URL|$NEW_URL|g" \
    {} \;

# Remove old stats script
find export -type f \( -name "*.html" -o -name "*.css" -o -name "*.js" \) -exec sed -i '' \
    -e "s|<script src=\"https://some-stats-service/track.js\" data-token=\"ef06b08b\" defer></script>||g" \
    {} \;

# Copy quote of the day script
cp qotd.js export/qotd.js

# Modify export/index.html to include the quote of the day script
#sed -i '' \
#    -e "s|<\/body>|<script src=\"/qotd.js\"></script></body>|" \
#    export/index.html

# Fix %3Fver= in URLs (replace %3F with ? to decode URL-encoded query strings)
# Use perl for safer handling of special characters
find export -type f \( -name "*.html" -o -name "*.css" -o -name "*.js" \) -exec perl -i -pe 's/%3Fver=/?ver=/g' {} \;

# Some filenames get saved with `?ver=...` in the actual filename. Fix that to remove it.
find export -type f -name "*\?ver=*" | while IFS= read -r file; do
    newname="${file%%\?*}"
    if [ "$file" != "$newname" ]; then
        mv "$file" "$newname"
    fi
done

03-deploy.sh

#!/usr/bin/env bash

set -eo pipefail

cd "$(dirname "$0")"

wrangler pages deploy ./export --project-name=mysite

apache-config.conf

<VirtualHost *:80>
        ServerAdmin webmaster@localhost
        DocumentRoot /var/www/html

        <Directory /var/www/html>
                Options Indexes FollowSymLinks
                AllowOverride All
                Require all granted
        </Directory>

        ErrorLog ${APACHE_LOG_DIR}/error.log
        CustomLog ${APACHE_LOG_DIR}/access.log combined
</VirtualHost>

docker-compose.yml

services:

  mysite-db:
    container_name: mysite-db
    image: mysql:9
    user: "1000:1000"
    volumes:
      - /Users/ian/mysite/wordpress/db/data:/var/lib/mysql
      - /Users/ian/mysite/wordpress/db/custom.cnf:/etc/mysql/conf.d/custom.cnf:ro
    restart: always
    environment:
      MYSQL_ROOT_PASSWORD: 'sekrit'
      MYSQL_DATABASE: wordpress
      MYSQL_USER: wordpress
      MYSQL_PASSWORD: wordpress
    security_opt:
      - seccomp=unconfined
    healthcheck:
      test: ["CMD", "mysqladmin", "ping", "-h", "localhost"]
      timeout: 5s
      retries: 10

  mysite-wordpress:
    container_name: mysite-wordpress
    depends_on:
      - mysite-db
    ports:
      - '8080:80'
    image: wordpress:latest
    user: "1000:1000"
    restart: always
    volumes:
      - /Users/ian/mysite/wordpress/wordpress:/var/www/html/wp-content
      - /Users/ian/mysite/wordpress/apache-config.conf:/etc/apache2/sites-available/000-default.conf
    environment:
      WORDPRESS_DB_HOST: mysite-db
      WORDPRESS_DB_USER: wordpress
      WORDPRESS_DB_PASSWORD: wordpress
      WORDPRESS_DB_NAME: wordpress
      WORDPRESS_CONFIG_EXTRA: |
        define('ALTERNATE_WP_CRON', true);
        define('SS_CRON', true);