When creating String objects, it is for the most part free or extremely cheap to make the distinction between ascii and latin1 where the detection of latin1 strings is already taking place. Moreover, there are bits already available in StringImpl to hold this data, making storage of it cheap/free, depending on which bit is used.
By knowing when a latin1 string is actually ascii, several optimizations become available such as memcpy() conversion to utf8 and faster JavaScript regex engines, which for the most part perform their regexing on ascii strings. A meta tracking bug has been created here for comments, etc: https://bugs.webkit.org/show_bug.cgi?id=92677. Please let me know if anyone has questions or concerns. Thanks, Dan _______________________________________________ webkit-dev mailing list [email protected] http://lists.webkit.org/mailman/listinfo/webkit-dev
