On Wed, 2006-05-31 at 21:41 +0700, Christian Perrier wrote: > > wikipedia languages whose names galeon could not render: > > at all: Gutisk, Inuktitut, Yi > > problems: Kannada, Sichuan Yi > > We should notify the indlinux gorup about this. I'll try to talk about > this with Guntupalli Karunakar, as we will meet (in real life!) tomorrow.... Excellent, thanks :) > > Thai: 16 chars draw on U+200D > > Here, Thep (the Thai translator) could be involved.) Hmm, seems U+200D is ZWJ - zero-width-joiner, not a printable character. So, perhaps it needs to be defined somewhere with an empty zero-width glyph? > Great work, indeed, Paul. You've just volunteered to setup a wiki page > to summarize all this..:-) EEEK... I'll try this on the weekend. > And hello from Thailand, by the way... :) nice! On Wed, 2006-05-31 at 22:58 +0800, Arne Götje wrote. > For the CJK glyphs: It is currently not necessary to have all of them > <snip> Good to hear :) > All blocks mentioned above will be supported by the ttf-arphic-{ukai| > uming} fonts in the future, but the missing characters are not so > important as that life depends on them... ;) Thanks for the summary, good to hear you are working on them. On Wed, 2006-05-31 at 17:08 +0200, Mohammed Adnène Trojette wrote: > On Wed, May 31, 2006, Paul Wise wrote: > > Arabic presentation forms A: about 2 thirds to 3 quarters empty > > Please could you precise which unicode range you are referring to? In gucharmap in sid, the block is U+FB50 to U+FDFF. That is split up into several smaller blocks surrounded by characters not defined in Unicode. U+FBB1 to U+FBB1 is fully filled. The last block is small and fully filled (U+FDF0 to U+FDFD). The middle two blocks are the ones that are mainly empty (second one has one char): U+FBD3 to U+FD3F and U+FD50 to U+FDC7. As a side note, U+FDFD (ARABIC LIGATURE BISMILLAH AR-RAHMAN AR-RAHEEM) is an amazingly detailed character, and quite beautiful, does anyone know the meaning? adn? > May 31 21:52:20 +0800 <adn> tell me what fonts to package :) As a start, I think we should ensure that all the DFSG-free fonts mentioned in the guide below are packaged, then move on to searching for fonts that have characters in the other empty blocks. Some of them we will have to take care with - Myazedi and some of the other Burmese fonts intrude on undefined parts of Unicode, or use not-yet approved parts of blocks. A warning in the package description is probably enough. I'm probably remembering wrong, but I think MPH Yangon renders incorrectly under Pango, that may have changed with recent releases. Unicode Font Guide For Free/Libre Open Source Operating Systems http://www.unifont.org/fontguide/ -- bye, pabs http://wiki.debian.org/PaulWise
Attachment:
signature.asc
Description: This is a digitally signed message part