PHP Forum

**Brian07002** · 09-15-2009, 08:39 PM

Can someone show me an example of how I would go about extracting just the text contained inside a div tag on a site and save to a text file?

Example to extract the content of the div.text1 tag throughout a particular website.

**bunchjesse** · 09-15-2009, 08:59 PM

You can't using PHP, but you can with JavaScript and then pass that info to PHP using AJAX. IN JavaScript, you just use the .InnerHTML property of the DIV.

**NullPointer** · 09-15-2009, 09:35 PM

Using preg_match:

PHP Code:


			
$content; //your html content

//this may be wrong someone a little better

//with regular expressions will probably know the correct way

$expression = '/<div(.*)>(.*)</div>/';



$matches = array(); //this will store all of the matches

preg_match($expression, $content, $matches);



$divContent = array();



foreach($matches as $match)

{

     $match = preg_replace('/<div(.*)>/', '', $match);

     $divContent[] = str_replace('</div>', '', $match);

}



//$divContent will now contain an array of strings extracted 

//from the html

You'll need to double check the regular expressions, but that should do it.

**JeremyMiller** · 01:55 PM

Your question doesn't make sense to me unless you're talking about crawling a site and extracting content from a div on the page. RegEx's are powerful, but can be challenging to match in a lot of situations. Since ID's are supposed to be unique, we can use this to our advantage by searching for it (this WILL NOT WORK if the id is not unique):

PHP Code:


			
<?php
$content = '<!DOCTYPE html>
            <html lang="en">
              <head>
                <meta charset="UTF-8">
                <title></title>
                <script src="http://html5shiv.googlecode.com/svn/trunk/html5.js"><!-- Makes HTML 5 elements available for styling --></script>
              </head>
              <body>
                <nav id="header_nav">
                  <ul>
                    <li><a href=""></a></li>
                    <li><a href=""></a></li>
                    <li><a href=""></a></li>
                  </ul>
                </nav>
                <div id="text1">My content</div>
              </body>
            </html>';
//Find id
if (($id_pos = strpos($content, 'text1')) !== false) {
  $end_of_tag = strpos($content, '>', $id_pos);
  $close_tag = strpos($content, '</div>', $end_of_tag);
  echo substr($content, $end_of_tag + 1, $close_tag - $end_of_tag - 1);
}
?>

@Matt: When writing a reg ex to apply to HTML tags, I usually use [^>]* for matching the inner contents as the > character should not show up (if done correctly) in the rest of a tag and will give non-greedy matches. When I use the dot character, I always end up playing with the ? greedy toggler to figure out if I have it right. Others more versed in RegEx's may have a different philosophy or easier way of understanding it, but I thought you may appreciate knowing someone else's way (with reasoning).

**chrishirst** · 09-16-2009, 02:10 PM

Quote:

Originally Posted by bunchjesse

You can't using PHP, but you can with JavaScript and then pass that info to PHP using AJAX. IN JavaScript, you just use the .InnerHTML property of the DIV.

Nope, you can't as that would be blocked as cross site scripting.

**Brian07002** · 09-17-2009, 01:25 AM

Thanks for all the advice, will try it eventually.

**weker** · 09-18-2009, 10:12 AM

You can try http://simplehtmldom.sourceforge.net/ for manipulation of HTML elements, it is very powerful DOM analyser.

**dweebsonduty** · 09-20-2009, 07:21 PM

Quote:

Originally Posted by chrishirst

Nope, you can't as that would be blocked as cross site scripting.

But what if you used php to read and print it then javascript to do the rest...

Inefficient, I know but possible.

The preg match is the best way to go.