Help me. parsing html documents with php

Hello,
I need to parse html documents to get "values" of all html tags and their attributes like if I have:
<a href="url_here" title="title_here"> text_here </a>
Then I need 3 values of:
array(
'href' => 'url_here',
'title' => 'title_here',
'tag_value' => 'text_here'
);
How to do this?
Please let me know. If possible please give me some code.
regards

 

I found some code here:
http://www.weberdev.com/get_example-1817.html
and modified it to return the value between the anchor tags too (it did only return the attributes).

Here is the code:

<?php 
           /* 
            * parseHtml.php 
            * My notation of variables: 
            * i_ = integer, ex: i_count 
            * a_ = array,       a_html 
            * b_ = boolean, 
            * s_ = string 
            * 
            * What it does: 
            * - parses a html string and get the tags 
            *   - exceptions: html tags like <br> <hr> </a>, etc 
            * - At the end, the array will look like this: 
            *      ["IMG"][0]["SRC"] = "xxx" 
            *      ["IMG"][1]["SRC"] = "xxx" 
            *      ["IMG"][1]["ALT"] = "xxx" 
            *      ["A"][0]["HREF"] = "xxx" 
            * 
            */ 

$html = '<a href="url_here" title="title_here"> text_here </a>';
$tags = parseHtml($html);
print_r ($tags);

           function parseHtml( $s_str ) 
           { 
            $i_indicatorL = 0; 
            $i_indicatorR = 0; 
            $s_tagOption = ""; 
            $i_arrayCounter = 0; 
            $a_html = array(); 
            // Search for a tag in string 
            while( is_int(($i_indicatorL=strpos($s_str,"<",$i_indicatorR))) ) { 
                    // Get everything into tag... 
                    $i_indicatorL++; 
                    $i_indicatorR = strpos($s_str,">", $i_indicatorL); 
                    $s_temp = substr($s_str, $i_indicatorL, ($i_indicatorR-$i_indicatorL) ); 
                    $a_tag = explode( ' ', $s_temp ); 
                    // Here we get the tag's name 
                    list( ,$s_tagName,, ) = each($a_tag); 
                    $s_tagName = strtoupper($s_tagName); 
                    // Well, I am not interesting in <br>, </font> or anything else like that... 
                    // So, this is false for tags without options. 
                    $b_boolOptions = is_array(($s_tagOption=each($a_tag))) && $s_tagOption[1]; 
                    if( $b_boolOptions ) { 
                            // Without this, we will mess up the array 
                            $i_arrayCounter = (int)count($a_html[$s_tagName]); 
                            if ($s_tagName == "A") {
                              $a_value = substr($s_str, $i_indicatorR+1, strpos($s_str,"</a>", $i_indicatorR) - $i_indicatorR - 1); 
                              $a_html[$s_tagName][$i_arrayCounter]["VALUE"] = $a_value; 
                            }
                            // get the tag options, like src="htt://". Here, s_tagTokOption is 'src' and s_tagTokValue is '"http://"' 
                            do { 
                              $s_tagTokOption = strtoupper(strtok($s_tagOption[1], "=")); 
                              $s_tagTokValue  = trim(strtok("=")); 
                              $a_html[$s_tagName][$i_arrayCounter][$s_tagTokOption] = $s_tagTokValue; 
                              $b_boolOptions = is_array(($s_tagOption=each($a_tag))) && $s_tagOption[1]; 
                            } while( $b_boolOptions ); 
                    } 
            } 
            return $a_html; 
           } 

           ?>

PHP:

There are probably better ways of doing this depending on exactly what it is you are trying to achieve.

 

<?php
if( !defined( 'MODE_HTML' ) )	define( 'MODE_HTML', 1 );	# Work in HTML source mode
if( !defined( 'MODE_FILE' ) )	define( 'MODE_FILE', 2 );	# Work in FILE/URL source mode

if( !defined( 'ERROR_NONE' ) )	define( 'ERROR_NONE', 0 );	# No errors occured
if( !defined( 'ERROR_DOM' ) )	define( 'ERROR_DOM', -1 );	# DOMDocument class not available
if( !defined( 'ERROR_MODE' ) )	define( 'ERROR_MODE', -2 );	# No valid mode passed to getLinks
if( !defined( 'ERROR_LOAD') )	define( 'ERROR_LOAD', -3 );	# Not able to load source
if( !defined( 'ERROR_LINK' ) )	define( 'ERROR_LINK', -4 );	# No links found in document

/**
 * getLinks - search for links in document
 *
 * @param string $source
 * @param int $mode
 * @param array $links
 * @param DOMDocument $dom
 * @return int
 * 
 * <code>
 * #search for links in source of http://interviolet.com
 * getLinks( "http://interviolet.com", MODE_FILE, $links, $dom );
 * #search for links in the source '<a href="http://interviolet.com" title="TheTitle">Interviolet</a>'
 * getLinks( "<a href=\"http://interviolet.com\" title=\"TheTitle\">Interviolet</a>", MODE_HTML, $links, $dom );
 * </code>
 */
function getLinks( $source, $mode, &$links = null, &$dom = null )
{
	$links = array( );	# Create link array for storage
	
	if( class_exists( 'DOMDocument' ) )	# Check for DOMDocument class
	{
		$dom = new DOMDocument( );	# Create new DOMDocument

		switch( $mode )	# Load HTML into DOM object
		{
			case MODE_HTML: if( !@$dom->loadHTML( $source ) ) return ERROR_LOAD; break;
			case MODE_FILE: if( !@$dom->loadHTMLFile( $source ) ) return ERROR_LOAD; break;
			default: return ERROR_MODE;
		}
		
		if( ( $tags = $dom->getElementsByTagName( 'a' ) ) )	# Find links in document
		{
			foreach( $tags as $tag )
			{	
				if( $tag->hasAttribute( 'href' ) or $tag->hasAttribute( 'title' ) )	# If the current link is valid
				{
					$links[ ] = array	# Store the current link
					(
						'href' => $tag->getAttribute( 'href' ),
						'title' => $tag->getAttribute( 'title' ),
						'text' => $tag->nodeValue,
						'node' => $tag	# Store DOMNode in results too ...
					);	
				}
			}
			
			return ERROR_NONE;
		}
		else return ERROR_LINK;
	}
	else return ERROR_DOM;
}

switch( getLinks( 'http://forums.digitalpoint.com/showthread.php?t=863580', MODE_FILE, $links, $dom ) )
{
	case ERROR_NONE: foreach( $links as $index => $link )
	{
		printf
		( 
			"--------------------#%03d--------------------\r\n", 
			$index
		);
		foreach( $link as $key => $value ) if( $key != 'node' ) printf
		(
			"%s\t\t%s\r\n", 
			$key, 
			$value ? $value : '(null)'
		);
		printf
		(
			"--------------------#%03d--------------------\r\n\r\n", 
			$index
		);
	}
	break;
	
	case ERROR_DOM: print( "PHP DOM extension not loaded\r\n" ); break;
	case ERROR_MODE: printf( "No valid mode passed to function\r\n"); break;
	case ERROR_LOAD: print( "Failed to load document\r\n" ); break;
	case ERROR_LINK: print( "No links found in document\r\n" ); break;
}
?>

PHP: