In opposite to most other CSS parsers, no regular expressions are used and thus CSSTidy has full CSS2 support and a higher reliability.
I wonder why CSSTidy built their own parser and not utilized the official SAC API (which also includes support for the new CSS3)? Is there any technical or practical reason for not using SAC and writing your own parser? Infact, there're NO css tools right now that use SAC, I wonder why?