The author's stated stance is that you can, but usually shouldn't.

Personally, I think to myself before using regexes to parse HTML:

1. Am extracting more than the contents of a single element?

2. Is the input HTML prone to change?

3. Will there be issues if the parsing completely fails?

4. Do I plan to use this code for more than a few months?

If the answer to any of the questions is "Yes", then don't use a regex :)