Regex challenge - find "foobar" in HTML document

Question

I have a fairly long and complex HTML document, and I need to find all occurences of a given string, e.g. "foobar", unless it's between <a> and </a> anchor tags.

The trouble is: it could be inside some text between the anchor tags, e.g.

<a>this is a foobar test</a>

and even in this case, I should not find the match.

How can I do that with a regex?? I would have no trouble finding <a>foobar</a> and so on - but finding every "foobar" except when it's between the anchor tags and surrounded by possible a lot of other text seems a bit tricky.....

Any ideas??

ANSWER:
We ended up using this Regex to solve this problem - just in case anyone is a) curious, or b) in the same place :-)

(?<!\<A.*(?=\<\/A))Test(?!\<\/A.*(?=\<A))

Answer 1

You should be able to do with negative lookahead and loohbehind patterns. Here is a good tutorial:

http://www.regular-expressions.info/lookaround.html

Answer 2

'foobar(?![^<]*</a>)'

works for me in the simplest case. it's obviously not resistant to having other tags within a tag.

Answer 3

Try this:

$str = 'foobar <a>this is a foobar <span>foobar</span> test</a> foobar';

$pattern = '<a(?:[^"\'>]+|"[^"]*"|\'[^\']*\')*>(?:[^<]+|(?!<\/a\s*>)<)*<\/a\s*>';
$parts = preg_split('/('.$pattern.')/', $str, -1, PREG_SPLIT_DELIM_CAPTURE);
$isLink = (bool) preg_match('/^'.$pattern.'$/', $parts[0]);
foreach ($parts as &$part) {
    if (!$isLink) {
        $part = str_replace('foobar', '!!!found!!!', $part);
    }
    $isLink = !$isLink;
}
$str = implode('', $parts);

echo htmlspecialchars($str);