Some TextPattern .htaccess Goodness

No-www

One of my key things for SEO is to reduce the number of URLs that point to the same content. My personal preference is to remove www from URLs, which is a simple rule seen all over the internet.

RewriteCond %{HTTP_HOST} ^www\\.(.*) [NC]
RewriteRule ^(.*)[\\/]?$ http://%1/$1 [R=301,NC,L]

If you prefer, have the ‘www’ I’m no fascist, it’s more of a tidiness thing. But pick one.

Removing the trailing slash

When using a CMS such as textpattern, that uses URL rewriting, you can often get the same url with a trailing slash, I’ve found a way to remove that. Note the two lines that single out the mint (analytics) and rpc (connection for desktop blogging software) directories, if these aren’t in there the respective service will not function. This works for any directory, just copy the line and replace the name.

RewriteCond %{REQUEST_FILENAME} !-f
RewriteCond %{REQUEST_URI} !rpc/(.*)$
RewriteCond %{REQUEST_URI} !mint/(.*)$
RewriteCond %{REQUEST_URI} (.*)[\\/]$
RewriteRule ^(.*)[\\/]$ http://%{HTTP_HOST}/$1 [R=301,L]

This also means Mint and Google Analytics do not record two separate URLs, one with a trailing slash and one without.

Standard redirects

When importing the old articles I realised there was a couple of out of date pieces so I simply removed the articles and added standard redirects to the latest information.

Redirect 301 /writing/setting-up-a-ruby-on-rails-development-mac http://andycroll.com/writing/the-big-reinstall-setting-up-rails-on-leopard
Redirect 301 /writing/setup-a-ruby-on-rails-development-mac-on-os-x-leopard http://andycroll.com/writing/the-big-reinstall-setting-up-rails-on-leopard