> Apart from the problem of having to specify all the possible tag names up front in an array, this approach
> will fail to pick up on certain problems that are quite common, such as:
>
> <foo><bar> blah blah </foo></bar>

True, but again all he said he was looking for was to see if all open tags had a close tag, not if the ordered syntax was correct. Also he stated that he was only searching for a few tags and NOT the entire range of possible HTML tags so the array approach would be sufficient for his implementation, though a bad idea on a large scale parser.

All html parser modules I am aware of would not do what he is looking for since they basically parse the text inside the tags and do nothing with the actual tags themselves except demilit on them.