Match the path of a URL, minus the filename extension

Try this:

preg_match("/net(.*)\.php$/","http://php.net/manual/en/function.preg-match.php", $matches);
echo $matches[1];
// prints /manual/en/function.preg-match

There's no need to use a regular expression to dissect a URL. PHP has built-in functions for this, pathinfo() and parse_url().

Just for the fun of it, here are two ways that have not been explored:

substr($url, strpos($s, '/', 8), -4)

Or:

substr($s, strpos($s, '/', 8), -strlen($s) + strrpos($s, '.'))

Based on the idea that HTTP schemes http:// and https:// are at most 8 characters, so typically it suffices to find the first slash from the 9th position onwards. If the extension is always .php the first code will work, otherwise the other one is required.

For a pure regular expression solution you can break the string down like this:

~^(?:[^:/?#]+:)?(?://[^/?#]*)?([^?#]*)~
                              ^

The path portion would be inside the first memory group (i.e. index 1), indicated by the ^ in the line underneath the expression. Removing the extension can be done using pathinfo():

$parts = pathinfo($matches[1]);
echo $parts['dirname'] . '/' . $parts['filename'];

You can also tweak the expression to this:

([^?#]*?)(?:\.[^?#]*)?(?:\?|$)

This expression is not very optimal though, because it has some back tracking in it. In the end I would go for something less custom:

$parts = pathinfo(parse_url($url, PHP_URL_PATH));
echo $parts['dirname'] . '/' . $parts['filename'];

This general URL match allows you to select parts of a URL:

if (preg_match('/\\b(?P<protocol>https?|ftp):\/\/(?P<domain>[-A-Z0-9.]+)(?P<file>\/[-A-Z0-9+&@#\/%=~_|!:,.;]*)?(?P<parameters>\\?[-A-Z0-9+&@#\/%=~_|!:,.;]*)?/i', $subject, $regs)) {
    $result = $regs['file'];
    //or you can append the $regs['parameters'] too
} else {
    $result = "";
}

|(?<=\w)/.+(?=\.\w+$)|

• select everything from the first literal '/' preceded by
• look behind a Word(\w) character
• until followed by a look ahead
  • literal '.' appended by
  • one or more Word(\w) characters
  • before the end $

  re> |(?<=\w)/.+(?=\.\w+$)|
Compile time 0.0011 milliseconds
Memory allocation (code space): 32
  Study time 0.0002 milliseconds
Capturing subpattern count = 0
No options
First char = '/'
No need char
Max lookbehind = 1
Subject length lower bound = 2
No set of starting bytes
data> http://php.net/manual/en/function.preg-match.php
Execute time 0.0007 milliseconds
 0: /manual/en/function.preg-match

|//[^/]*(.*)\.\w+$|

• find two literal '//' followed by anything but a literal '/'
• select everything until
• find literal '.' followed by only Word \w characters before the end $

  re> |//[^/]*(.*)\.\w+$|
Compile time 0.0010 milliseconds
Memory allocation (code space): 28
  Study time 0.0002 milliseconds
Capturing subpattern count = 1
No options
First char = '/'
Need char = '.'
Subject length lower bound = 4
No set of starting bytes
data> http://php.net/manual/en/function.preg-match.php
Execute time 0.0005 milliseconds
 0: //php.net/manual/en/function.preg-match.php
 1: /manual/en/function.preg-match

|/[^/]+(.*)\.|

• find literal '/' followed by at least 1 or more non literal '/'
• aggressive select everything before the last literal '.'

  re> |/[^/]+(.*)\.|
Compile time 0.0008 milliseconds
Memory allocation (code space): 23
  Study time 0.0002 milliseconds
Capturing subpattern count = 1
No options
First char = '/'
Need char = '.'
Subject length lower bound = 3
No set of starting bytes
data> http://php.net/manual/en/function.preg-match.php
Execute time 0.0005 milliseconds
 0: /php.net/manual/en/function.preg-match.
 1: /manual/en/function.preg-match

|/[^/]+\K.*(?=\.)|

• find literal '/' followed by at least 1 or more non literal '/'
• Reset select start \K
• aggressive select everything before
• look ahead last literal '.'

  re> |/[^/]+\K.*(?=\.)|
Compile time 0.0009 milliseconds
Memory allocation (code space): 22
  Study time 0.0002 milliseconds
Capturing subpattern count = 0
No options
First char = '/'
No need char
Subject length lower bound = 2
No set of starting bytes
data> http://php.net/manual/en/function.preg-match.php
Execute time 0.0005 milliseconds
 0: /manual/en/function.preg-match

|\w+\K/.*(?=\.)|

• find one or more Word(\w) characters before a literal '/'
• reset select start \K
• select literal '/' followed by
• anything before
• look ahead last literal '.'

  re> |\w+\K/.*(?=\.)|
Compile time 0.0009 milliseconds
Memory allocation (code space): 22
  Study time 0.0003 milliseconds
Capturing subpattern count = 0
No options
No first char
Need char = '/'
Subject length lower bound = 2
Starting byte set: 0 1 2 3 4 5 6 7 8 9 A B C D E F G H I J K L M N O P 
  Q R S T U V W X Y Z _ a b c d e f g h i j k l m n o p q r s t u v w x y z 
data> http://php.net/manual/en/function.preg-match.php
Execute time 0.0011 milliseconds
 0: /manual/en/function.preg-match