preg_match for nested XML Tags

[Hawk2005]

Hi,

i'm trying to parse an XML document. It looks something like this:

<block name="var3">
<tr>
	<td width="20"><variable name="id"/></td>
	<td width="180"><variable name="name"/></td>
                <block name="test">
                </tr>
                        <td>-</td>
                        <td>-</td>
                <tr>
                </block>
</tr>
</block>
<block name="var4">
<tr>
	<td width="20"><variable name="id"/></td>
	<td width="180"><variable name="name"/></td>
</tr>
</block>

I used this

preg_match ("/<block name=\"".$block."\">(.*)<\/block>/s", $tpl, $out);

for templates with only one block and it worked fine. But how can i solve it with nested blocks ?

I want an array something like this:

Array
(
    [var3] => <tr>
				<td width="20"><variable name="id"/></td>
				<td width="180"><variable name="name"/></td>
                <block name="test"/>
              </tr>
    [var4] => <tr>
 	    		<td width="20"><variable name="id"/></td>
		 	    <td width="180"><variable name="name"/></td>
              </tr>
    [test] => </tr>
              	<td>-</td>
                <td>-</td>
              <tr>
)

But i don't know how to solve this.

[sfullman]

Think about this, with white space, nested tags, etc., regex's are complex. Try using this function I wrote; it's got a few notes in it but will do what you want.

Basically, what you're looking for requires you break it down into smaller chunks for people here on the forum; to do what you want to do is some work.

to see how the code works use it this way:

$string=implode('',file('myfile.htm')); //file you want to parse
include('function_xml_read_tags_v131.php');
$a=xml_read_tags($string, 'tr');

echo "<pre>";
print_r($a);
echo "</pre>";

This should give you starting and ending points, you can use this to get substr() calls (which is very fast).

Sam

[Hawk2005]

Great work.

This is what i needed.

Am i allowed to modify the code and use it in my programs ?

cu...

[sfullman]

To be honest, that script is worth $20.00 US to me; took many hours to program right. PM me if this seems fair to you. However I obviously gave you the script cause I wanted you to learn how, so it's up to you.

One way the function could be modified is to store the actual matching string in the _special node of the array, perhaps you can modify it appropriately.

Also, in case you didn't figured out, single attributes like nowrap get placed in the array as [nowrap]=>1