Regular expression for extracting certain URLs?

I've already tried my best but regular expressions aren't really my thing. :(

I need to extract certain URLs that end in a certain file extension. For example, I want to be able to parse a large paragraph and extract all URLs that end with *.txt. So for example,

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Nulla hendrerit aliquet erat at ultrices. Donec eu nunc nec nibh http://www.somesite.com/somefolder/blahblah/etc/something.txt iaculis dictum. Quisque nisi neque, vulputate quis pellentesque blandit, faucibus eget nisl.

I need to be able to take http开发者_StackOverflow社区://www.somesite.com/somefolder/blahblah/etc/something.txt out of the above paragraph but the number of URLs to extract will vary. It will be dynamic based on what the user inputs. It can have 3 links that end with *.txt and 3 links that don't end with *.txt. I only need to extract those that does end in *.txt. Can anyone possibly give me the code I need for this?

You can find what you need with /(?<=\s)http:\/\/\S+\.txt(?=\s)/

Which means:

• A space/tab/new line before.
• http://
• more that one non-space character.
• .txt
• A space/tab/new line after.

Assuming these are all proper URLs, then they won't have any spaces in them. We can take advantage of that fact to make the regular expression really simple:

preg_match_all("/([^ ]+\.(txt|doc))/i", $text, $matches);
//   ([^ ]+     Match anything, except for a space.
//   \.         A normal period.
//   (txt|doc)  The word "txt" or "doc".
//   )/i        Case insensitive (so TXT and TxT also work)

If you don't need to match multiple file extensions, then you can change "(txt|doc)" to "txt".

$matches will contain a number of arrays, you'll want key number 0 or 1. To make the array easier to read, you can use:

preg_match_all("/(?P<matched_urls>[^ ]+\.(txt|doc))/i", $text, $matches);

This will make $matches look something like this:

array([0] => array(), [1] => array(), [2] => array(), ["matched_urls"] => array());

Should be obvious which key you need.

How about:

$str = 'Lorem ipsum dolor sit amet. Donec eu nunc nec nibh http://www.somesite.com/somefolder/blahblah/etc/something.txt. Lorem ipsum dolor sit amet. Donec eu nunc nec nibh http://www.somesite.com/somefolder/blahblah/etc/something.doc.';
preg_match_all('#\b(http://\S+\.txt)\b#', $str, $m);

explanation:

#             : regex delimiter
\b            : word boundary
(             : begin capture group
http://       : litteral http://
\S+           : one or more non space
\.            : a dot
txt           : litteral txt
)             : end capture group
\b            : word boundary
#             : regex delimiter