preg_match doesn't recognize \\2?

[Amailer]

Hey,
I have a place that does in one function

$text = preg_match("bla", "bla".valid_function("\\2")."bla");

In the valid function, it does another preg match, searching for a specific text; But, for some odd reason, its not taking the matched value for \\2, insted its taking \\2 as a string. So if I do a preg_match for "\\2", it works

Any idea how to fix this?

[Tycoon]

I may have an idea, it may be simular to putting $array['item'] in a select statement - like so "SELECT * FROM `a` WHERE `id` = '$array['item']'";

I would try:

$valid_text = valid_function("\\2");

$text = preg_match("bla", "bla".$valid_text."bla");

Unsure if that will work but give it a try.

[Amailer]

Nope, didn't work. I don't see how it would have made a difference though :/

Is there something I have to do in preg_match() to like..recognize \\2, $1, $2 and etc?

[octopi]

I'm not exactly sure what your trying to do....but.....

preg_match returns an int, and not the acctual matched text. to access this, you need to add a 3rd parameter '

preg_match(/www\.(.*?).ca/,'www.compsci.ca',$matches);

now....$matches[0] should contain 'www.compsci.ca'
and $matches[1] should contain compsci

based on what your attempting to do (from what i can tell) you'll probally have to do a general match first, store it in matches, then feed that to your sub/function then do another match


I'm not exactly sure what your trying to do though....


\\2 from what I can tell your refering to a back reference, I don't believe you can place one of these into your haystack, as its meant to be part of your pattern.

[Amailer]

K sorry, let me clear it up a bit.
This function searches for urls

php:

function make_clickable($text) {         $text = preg_replace('#(script|about|applet|activex|chrome):#is', "\\1:", $text);         // pad it with a space so we can match things at the start of the 1st line.         $ret = ' ' . $text;         // matches an "xxxx://yyyy" URL at the start of a line, or after a space.         // xxxx can only be alpha characters.         // yyyy is anything up to the first space, newline, comma, double quote or <     $ret = preg_replace("#(^|[\n ])([\w]+?://[\w\#$%&~/.\-;:=,?@\[\]+]*)#is", "\\1<a href=\"\\2\" target=\"_blank\">".validate_url('$2')."</a>", $ret);         // Remove our padding..         $ret = substr($ret, 1);         return($ret); }

As you can see, it searches for the text http://blablabla in the post and makes it into a url. What I want to do is, if the urls domain matches that of the current site, strip the extra part, just show the page

So:
http://wii.com/index.php?t=12 -> index.php?t=12

That was suppose to have been done in the validate_url() function which has:

php:

if( preg_match('#'.$forum_url.'#i', $url) ) {                 $valid = true;         }

And some other stuff- but the preg_match there does not recognize the $2 given :/. Meaning, if I do

php:

if( preg_match('#2#i', $url) ) {                 $valid = true;         }

$valid will be true :/ Even if there is no 2 to match!

[octopi]

You need to use preg_replace_callback() to do what you want, heres an example:
http://www.magnify.ca/devel/compsci/adm/preg.php

code:

<?php   function validate_url($matches) {     if($matches[2] == "www.magnify.ca") {       return "<a href='{$matches[3]}'>{$matches[3]}</a> ";     }     else {       return "<a href='{$matches[1]}{$matches[2]}{$matches[3]}'>{$matches[1]}{$matches[2]}{$matches[3]}</a> ";     }   }   function make_clickable($subject)  {    // $patterns[]="/http(s?):\/\/www.magnfiy.ca(:\d){1,5}\/(.*?)\w?/";  //our site     $pattern="!(https?://)(.*?)(/.*?)*(\s|$)!";   //another site     return preg_replace_callback($pattern,"validate_url",$subject);   }   print make_clickable("goto http://www.magnify.ca/index.php?test=hello&this=that now") . "<br>";   print make_clickable("goto https://www.magnify.ca/index.php now") . "<br>";   print make_clickable("goto https://www.compsci.ca/index.php now") . "<br>";   print make_clickable("goto http://www.compsci.ca") . "<br>";   print make_clickable("goto http://www.compsci.ca:8080") . "<br>"; ?>

[Amailer]

Excellent! That worked perfectly, thanks.

Though its, odd, I had seen what I was trying to do someone where else, and... it worked :/ Maybe I over looked something.

[Tycoon]

Lol I looked stupid

Question though: How do you get the code to appear in its own box? Just use [code][/code] ?

[Amailer]

Yeah, you can use [code][/code] or if you want to highlight it for a specific language use [syntax="langname"][/syntax]

A question: I got it to all work out, this is how it shows it now:
http://www.compsci.ca/v3/posting.php?t=123 posting.php?t=123

But I if I have just the domain, this happnes:
http://www.compsci.ca/v3/ [blank]

Is there a way to make the preg_replace() only remove the text if there is some text after the match?

***EDIT***
Nevermind, I got it

[Tycoon]

Amailer @ Sat Jan 13, 2007 1:57 pm wrote:

Nevermind, I got it

Glad we could help Don't be afraid to ask questions. haha

[Tycoon]

php:

<?php                 function replace_reply($message) {                         $expression = "#\[reply=(.*?)](.*?)\[/reply]#si";                                                 $count = preg_match_all($expression, $message, $replace);                                                 if (count($count) > 0)                         {                                 for ($a = 0; $a < count($count); $a++)                                 {                                         $quote_user = $replace[1][$a];                                         $replace_string = $replace[2][$a];                                 }                                 $str_replace = $this->str_replace_nonphp("Reply", $quote_user, $replace_string);                                 $message = preg_replace($expression, $str_replace, $message);                         }                                                 return $message;                 } ?>

As we can see, this code SHOULD replace all the [reply] and [/reply] tags in the message BUT what it does is grab the first [reply=username] code and then it grabs, not the last [/reply] tag that matches, but the first one it fines.