lxml all the way if you care about speed! It’s a beast for parser html python stuff, but like you said, it’s picky.
If the HTML’s a disaster, maybe pre-clean it with something like `tidy` or `bleach` before throwing it at lxml.
Also, check out `selectolax`—it’s a newer parser html python lib that’s *blazing* fast and a bit more forgiving than lxml.
If the HTML’s a disaster, maybe pre-clean it with something like `tidy` or `bleach` before throwing it at lxml.
Also, check out `selectolax`—it’s a newer parser html python lib that’s *blazing* fast and a bit more forgiving than lxml.