Question

I wrote a regex in Rad Soft to pull out a div tag from some html. It tested well in the editor and php regex tester. Now I am trying to get it to implement with php on my current project but just doesn’t work. Usually its some simple solution that is failing to register since I have been staring at the problem too long.

here is the test sample being used

$temp = '<div>blah blah blah.
    You ordered these items: 
    <div id="product_list">

 - 1  Large -  Target
<span class="pricedisplay">R$ 1,00</span>
- 1  Large - Black
<span class="pricedisplay">R$ 1,00</span>
- 1  Large - Leather
<span class="pricedisplay">R$ 1,00</span>
- 1  Large - Preto
<span class="pricedisplay">R$ 1,00</span>

</div><span class="total-shipping"></span><span class="total-price">Total: <span class="pricedisplay">R$ 1,00</span>
</span></div>';

and for the regex:

<div\sid="product_list">([\s\w\d<=>/\\\$:",\.-]?)*[^(</div>)]</div>

which I tried using here:

$outputHTML = preg_replace_callback(preg_quote('<div\sid="product_list">([\s\w\d<=>/\\\$:",\.-]?)*[^(</div>)]</div>'), 'findProducts', $temp);

I have tried with/without the preg_quote. And if you know of any software similar to Rad that better simulates php’s regex, it would help alot. Thanks in advance!

Answer 1

Regardless of whether this regex happened to work with some tester there are many mistakes in it.

First :

[\s\w\d<=>/\\\$:",\.-]

Equals too :

[\s\w\d<=>/$:",.\\-] 

You need not unescape characters inside character class. Additionally you use preg_quote which escapes : . \ + * ? [ ^ ] $ ( ) { } = ! < > | : -

Secondly :

[^(</div>)]

Means :

"
[^(</div>)]    # Match a single character NOT present in the list “(</div>)”
"

While what you probably wanted to write is :

(?<!</div>)

"
(?<!         # Assert that it is impossible to match the regex below with the match ending at this position (negative lookbehind)
   </div>       # Match the characters “</div>” literally
)
"

Lastly :

([\s\w\d<=>/\\\$:",\.-]?)* equals : [\s\w\d<=>/$:",.\\-]*

Oh and yeah, what you are doing is potentially dangerous. You shouldn’t try to extract/parse html/xml/xhtml with regex when you have ready to use tools in your language.