On Sun, Dec 18, 2005 at 07:23:03AM +0100, Christian Perrier wrote: > > ko: \300\314\270\247 > > tr: \335[Ss][\335i][Mm] > > zh_CN: \303\373\263\306 > > zh_TW: \246\127\272\331 > > > > You can find these codes very easily, copy man pages into another file, > > delete everything but the wanted word, and run 'od -to1' on it. > > Thanks. > > About cs and hu, I think the problem comes from what's currently in > lexgrog.l: > > cs_name J[Mm][ÉéEe][Nn][Oo] > hu_name N[ÉéEe][Vv] > > > The "e acute" chould probably be encoded in the above way.....So, > should this be: > > hu_name N[\303\211\303\251Ee][Vv] > > ?
This sequence is wrong because you took UTF-8 encoded characters, it should be hu_name N[\311\351Ee][Vv] But that's not the point, because in lexgrog.l, those accented characters are seen as bytes by the compiler, and their value is of course similar to their octal value. Anyway encoding these characters would certainly be a good idea to avoid this encoding mismatch. > This is the last bit I'm missing for sending a patch to man-db. What is the error message, and on which man page? Denis -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
