find the position of HTML characters in a string

[nozom]

I have a string that contains HTML tags and i want to record where each tag starts and ends:

eg. for the text

 hello <b>this is some text</b>, This is a <font size=2>tag</font> 

for the above HTML I want to record the a <b> tag starts at character 7 and end as character 9, the </b> is character 27-29 and so on.

Any ideas on how to go about this? I've started with building an array of the tags i need to check for (15 or so) but i'm not sure about where to go from there. I'm looking at using the strrpos function but not quite sure what to do with it.

Thanks.
Hugh Newsome

[Astor]

<?
$string="hello <b>this is some text</b>, This is a <font size=2>tag</font>";
$start_pos=1;
while ($start_pos>0) {
 $start_pos=strpos ($string, "<", $offset);
 $end_pos=strpos ($string, ">", $start_pos);
 $offset=$end_pos;
 if ($start_pos>0) {
	 echo ($start_pos+1)." : ".($end_pos+1)."<br />\n";
	 // record $start_pos+1 & $end_pos+1
 }
}
?>

7 : 9
27 : 30
43 : 55
59 : 65

[nozom]

thats along the lines that i've been working, although it gets triped up if a rogue < tag gets in there.

I've come up with the solution to the problem using a similar system to that, thanks partially to a script on the php.net site.

Appreciate the help though, if you want me to post the solution on here i will do.

Cheers,
Hugh

[Weedpacket]

Originally posted by nozom
Appreciate the help though, if you want me to post the solution on here i will do.

It will help other people with a similar problem in the future when they search here if you did.

[nozom]

here we go, rather messy but it does the job of returning the character locations of designated tags within a string.

Called with the two lines below. 1 to set up which tags we want to search for, 2nd is the call from within the class.

 $badtags=array('<b','</b','<i','</i','<P','</p','<STRONG','</STRONG','<EM','</EM','<U','</U','<FONT','</FONT','<OL','</OL','<LI','</LI','<A','</A','<BLOCKQUOTE','</BLOCKQUOTE','<HR');
$nogo_locations=$this->stripos_words($thetext,$badtags);
 

The function

 
	function stripos_words($haystack,$needles_array,$pos_as_key=true)
		{
		for ($iz=0;$iz<sizeof($needles_array);$iz++)
			{
			$needles=$needles_array[$iz];
			$idx=0; // Used if pos_as_key is false

		// Convert full text to lower case to make this case insensitive
		$haystack = strtolower($haystack);

		// Split keywords and lowercase them
		$needle=strtolower($needles);


		// Get all occurences of this keyword
		$i=0; $pos_cur=0; $pos_found=0;
		while (  $pos_found !== false )
			{
			// Get the strpos of this keyword (if thereis one)
			$pos_found = strpos(substr($haystack,$pos_cur),$needle);
			if ( $pos_found !== false )
				{
		        // Set up key for main array
				$index = $pos_as_key ? $pos_found+$pos_cur : $idx++;

				// Populate main array with this keywords positional data
		        $positions[$index]['start'] = $pos_found+$pos_cur;
		        $pos_cur += ($pos_found+strlen($needle));
				$positions[$index]['endish']  = $pos_cur;
				// find the close of the >
				$pos_cur = strpos($haystack, '>', $pos_cur); 
				$positions[$index]['end']  = $pos_cur;
				//
		        $positions[$index]['word'] = $needle;
		        $i++;

				$thebadchars[]=$positions[$index]['start'];
				$num=$positions[$index]['start'];
				// create a new array filling in the gaps between the starts and ends of the tag
				while($num!=$positions[$index]['end'])
					{
					$num+=1;
					$thebadchars[]=$num;
					}

				}

			}

		}


	   // If we found anything then sort the array and return it
	if ( isset($positions) )
		{
		sort($thebadchars);
		return $thebadchars;

		}
	// If nothign was found then return false
	return false;
	}  

The only issue with it currently is it falls over on things like &nbsp; or similar. It will be a simple fix to get sort that out i think.
ta,
Hugh