That means not only foregoing treating numbers as LTR but also foregoing ignoring the pieces bracketed by LRE|RLE|LRO|RLO and PDF. Thus, "[LRE]mamma mia![PDF] HAS LOTS OF SONGS" will be considered LTR and displayed as "mamma mia! SGNOS FO STOL SAH". I think that both of the "improvements" are indeed improvements, and we will be giving them just to fit in better with the plaintext feature, which has so many problems that I have a hard time believing it will get anywhere. Speaking of which - where do we stand on the issues in the unicode-bidi:plaintext spec (alignment and dealing with inline elements)? Aharon On Mon, Sep 27, 2010 at 12:32 PM, fantasai <fantasai.lists@inkedblade.net>wrote: > I think first-strong should match the UBA. Otherwise we need two > first-strongs: one to be compatible with the UBA, and one that's > "smarter" than UBA. I'm not sure that's a good idea. > > ~fantasai >
Received on Monday, 27 September 2010 14:00:39 UTC