Wordpress downloader and unpacker

Question

Wrote a small bash-script today that downloads the latest swedish wordpress archive , extracts it and writes wp-config.php

Would appreciate any feedback that could improve it / prettify it.

The script assumes that the database name & the username is always the same.

#!/bin/bash
#Download wordpress tar.gz
echo "Laddar ner senaste wordpress på svenska"
tarfile=$(curl -s https://sv.wordpress.org/ | grep -ioE "https:\/\/sv.wordpress.org\/wordpress-[0-9].[0-9].[0-9]-sv_SE.tar.gz")
wget $tarfile -O senaste_wp.tar.gz
tar xvfz senaste_wp.tar.gz -C ./
rm senaste_wp.tar.gz
# Move the wordpress files to this dir
mv wordpress/* .
rm -r wordpress
# Get wordpress salts
salts=$(curl https://api.wordpress.org/secret-key/1.1/salt/)
# Read userinput
read -p "Ange databasanvändare : " dbuser
read -p "Ange databaslösenord : " dbpass
# Write wp-config.php
cat > wp-config.php << EOF
define('DB_NAME', '$dbuser');
define('DB_USER', '$dbuser');
define('DB_PASSWORD', '$dbpass');
define('DB_HOST', 'localhost');
define('DB_CHARSET', 'utf8');
define('DB_COLLATE', '');
$salts
\$table_prefix = 'wp_';
define('WP_DEBUG', false);
if ( !defined('ABSPATH') )
 define('ABSPATH', dirname(__FILE__) . '/');
require_once(ABSPATH . 'wp-settings.php');
EOF

Answer 1

You don't need to escape slashes for grep, you can simplify to:

tarfile=$(curl -s https://sv.wordpress.org/ | grep -ioE "https://sv.wordpress.org/wordpress-[0-9].[0-9].[0-9]-sv_SE.tar.gz")

curl and wget are too similar tools. Consider using only one of them, to reduce the number of dependencies of your script.

Instead of this:

wget $tarfile -O senaste_wp.tar.gz
tar xvfz senaste_wp.tar.gz -C ./
rm senaste_wp.tar.gz

You can pipe directly to tar, without needing a temporary file, like this:

curl $tarfile | tar xvz

Lastly, your script is not using anything Bash specific, so you could change the shebang line to !#/bin/sh.