Question

I want to make a function that gets a string with html code (as a string) including the script tag – for example: “<div>dkjdg</div><script>blabla</script>fgfgh<span>hey</span>” and returns all the script inside the script tag including the open and close tags.

I tried this so far:

var s;
function(string)
{
    var a = string.toLowerCase().match("/(.*?)<script>(.*?)<//script>(.*?)/");
    s = a[2];
    return a[1]+a[3];
}

s is containing what is between the script tags and the return is every thing else.

But it is not working…

Answer 1

And what happens if the block looks like this?

<script type="text/javascript"><![CDATA[
  alert('hello </script>');
]]></script>

Never parse HTML with regular expressions. Instead, you can do something like this (with a little help from jQuery):

function GetScript(str) {
  var div = $('<div>'), tmpdiv = $('<div>'), scr, ret = '<script';
  div[0].innerHTML = str;
  scr = div.find('script');

  for ( var i = 0; i < scr[0].attributes.length; i++ ) {
    var attr = scr[0].attributes[i];

    if ( attr.value && attr.value != '' )
      ret += ' ' + attr.name + '="' + 
             tmpdiv.text(attr.value).html().replace(/"/g, '&quot;') +
             '"';
  }

  return ret + '>' + scr.text() + '</script>';
}

That will do exactly what you want, doesn’t break unless the input is totally screwed and is more or less 100% cross-browser compatible. Note that we don’t want to use jQuery’s .html() to set the div content, because that will eval() the blocks!

This:

GetScript("<div>dkjdg</div><script type='yadda\"yadda'>blabla</script>fgfgh<span>hey</span>")

will return

'<script type="yadda&quot;yadda">blabla</script>'