.htaccess - translation to layman terms needed - regex

Question

My .htaccess is running with the Zend framework and one of our developers previously added this code:
RewriteCond %{THE_REQUEST} ^.*/index.php
RewriteRule ^(.*)/index.php$ http://www.mydomain.com/$1 [R=301,L]
Can someone please translate this to English for me? I believe it is saying, if someone tries to access index.php, then do a 301 redirect to the main domain (i.e. don't show the world "index.php" exists). I don't know why the RewriteCond is there when the RewriteRule has the same rule it seems, perhaps this was a mistake. Anyway I don't know if it even works as there are other rules and my site can actually access the index.php file (no redirection) so perhaps the above can be removed entirely. Looking for a proper translation of this to understand.
Many thanks!

Answer 1

Actually both RewriteCond and RewriteRule are required here. But first thing first this rule isn't going to work way it is.
Correct working code will be this:
RewriteCond %{THE_REQUEST} /index\.php [NC]
RewriteRule ^(.*?)index\.php$ /$1 [L,R=301,NC,NE]
Main difference is presence of / in your RewriteRule which will fail it for http://domain.com/index.php
Explanation:
THE_REQUEST variable represents original request received by Apache from your browser
You need to match URI with THE_REQUEST to make sure /index.php is there in the original URI and not a result of some internal rewrite to /index.php
If you remove RewriteCond it might still work but you have to make sure that none of rules are rewriting your URI to /index.php anywhere.

Answer 2

The RewriteCond test string contains the server-variable %{THE_REQUEST} which takes the form:
GET /index.php HTTP/1.1
or
POST /login.php HTTP/1.1
The condition pattern ^.*/index.php contains a regular expression that matches anything with /index.php in it...
The ^ marks the start of the string, the .* matches zero-or-more instances of any character and the /index.php is self-explanatory.
So, any URL which contains /index.php will be matched. If the condition pattern ended with a $ symbol (the end-of-string symbol), then it wouldn't match any instance of %{THE_REQUEST} as all instances of %{THE_REQUEST} contain the HTTP request type after the URL.
The RewriteRule is then used to do the 301 redirect as you describe, but importantly, the RewriteRule directive matches it's pattern against the current URL (and not against the whole request string like the RewriteCond does in this example).
The RewriteRule pattern ^(.*)/index.php captures (denoted by the brackets) everything in front of /index.php and then creates a 301 redirect that removes the /index.php.
#anubhava has pointed out that if you visit website.com/index.php then the RewriteRule won't work, because the URL it matches against does not include the leading /.
Those two lines can be rewritten as:
RewriteRule ^(.*)index.php$ http://www.mydomain.com/$1 [R=301,L,NC]
Then, if you visit http://website.com/index.php then the (.*) part of the pattern will match an empty string.
Or if you visit http://website.com/sub/folder/index.php then the (.*) part of the pattern will match sub/folder/, which is then referenced in the substitution string by $1.
Remember that every redirect works by telling the browser to request a new URL, and each request gets re-processed by the htaccess, so if there are other rules that allow viewing index.php then that could explain why you can see some URLs containing index.php.