Description:
------------
preg_replace has a bug

//CASE STUDY: remove multiple instances of &nbsp; inside <table> <div id="Paginator">&nbsp; <span></span>&nbsp;<a href="#"></a> &nbsp; </div></table> 
$m = "?";//Any marker that isn't currently in the strHTML string. Note, this is an example, but you should use a function that gets a safeMarker.
// replaces all instances of </table> with </table>?
$strHTML = preg_replace('/(<\/table>)/','$1'.$m, $strHTML);
//looks for all &nbsp; within <table><div id="Paginator"> and removes them
//This would be the code to use but php seems to be greedy and doesn't match all instances and only gets the last &nbsp;, where as perl, .NET or javascript regular expressions would
$strHTML = preg_replace('/(\<div\sid\=\"Paginator\"\>[^\?]*)\&nbps\;/','$1', $strHTML);

Reproduce code:
---------------
---
From manual page: function.preg-replace
---

//CASE STUDY: remove multiple instances of &nbsp; inside <table> <div id="Paginator"></div></table> 
$m = "?";//Any marker that isn't currently in the strHTML string. Note, this is an example, but you should use a function that gets a safeMarker.
// replaces all instances of </table> with </table>?
$strHTML = preg_replace('/(<\/table>)/','$1'.$m, $strHTML);
//looks for all &nbsp; within <div id="Paginator"> and removes them
//This would be the code to use but php seems to be greedy and doesn't match all instances and only gets the last &nbsp;, where as perl, .NET or javascript regular expressions would
$strHTML = preg_replace('/(\<div id\=\"Paginator\"\>[^\?]*)\&nbps\;/','$1', $strHTML);

Expected result:
----------------
Would remove all instances of &nbsp; inside <div id="Paginator"

Actual result:
--------------
Only removes the last &nbps; inside <div id="Paginator"