Looking for information.. preg_replace

[Scooby08]

I basically have a function like so:

 

<?php
function AdjustForEdit($block)
{
$search[] = "/(['\"])?(templ\/)?images\//m"; 
$replace[] = "\\1".$SITEURL."/\\2images/";

$ret = preg_replace($search, $replace, $block);

return $ret;
}
?>

 

I was wondering if anybody could tell me what the numbers 1 and 2 in this line are doing exactly:

 

$replace[] = "\\1".$SITEURL."/\\2images/";

 

I can't seem to find any documentation on that either because I'm not too familiar with regex and do not know what exactly to search for.. Can anybody recommend any good resources??

 

 

[.josh]

they represent the captured data.  (['\"])  is the first capture, and (templ\/) is the 2nd capture.  The parenthesis mark the capture start and end.

[nrg_alpha]

Can anybody recommend any good resources??

 

My personal favorite reference is this book. You can also view phpfreak's resources, as well as their regex tutorial, or sites like this.

[Scooby08]

Thank you for the info guys..

 

I have read through some of the documentation and am having troubles figuring out how to do a  search for everything except specific words.. For example, I have this line:

 

/href=\"(.*)?\"/

 

which replaces all href's with whatever I like.. The thing I'm trying to figure out is how would I do the same, except not replace any href's like so:

 

<a href="javascript://">

 

I'm thinking something like this:

 

[^javascript\:\/\/]

 

but Im not sure how to incorporate that into this line:

 

/href=\"(.*)?\"/

 

Thanks again..

 

[nrg_alpha]

/href=\"(.*)?\"/

 

A few things wrong here as is... you don't need to escape " characters in this case.. and avoid .* If you are going to use dot all matching, its generally safer to make it into a lazy quantifier. See this discussion to see why (read from reply 10 onwards). In this case, I would use a negated character class instead.

/href="[^"]+"/

 

The thing I'm trying to figure out is how would I do the same, except not replace any href's like so:

 

<a href="javascript://">

 

I'm thinking something like this:

 

[^javascript\:\/\/]

 

you are using a negated character class to tell the regex engine not to match either a j, or an a, or a v, etc.. what you need to understand is that a character class just represents one position in the string.. not a sequence such as a capture or non capturing set (...) or (?:...).

 

So if I understand you correctly, you want to replace any href content that is NOT javascript://? You can use negative look ahead assertions.

 

$arr = array('<a href="javascript://" id = "clueless_ID">', '<a href="whatever" class = "I_dunno">', '<a href="javascript://" id = "Rock_Me_Amadeus">');
foreach($arr as &$val){
    $val = preg_replace('#(<a.+?href=")(?!javascript://)[^"]+("[^>]*>)#', '\\1\\2', $val);
}
echo "<pre>".print_r($arr, true);

 

When you run the above script, you'll have to right-click to see the results... any href that has the content javascript:// doesn't get wiped out. Keep hacking away at the material you were reading.. in time, it really starts to fit together.