Hi, Lucho, El sáb, 11-09-2004 a las 19:56, Luchezar Georgiev escribió: > Thanks! But earlier, I changed the format of the CTYINFO table back to the > old, much more readable format, by using a macro (idea by Arkady). So it > won't apply. Would it be too difficult for you to modify it for the new > COUNTRY.ASM? You can update it from unstable CVS branch. When I can apply
OK, done again. New patch in: http://perso.wanadoo.es/samelborp/country3.zip It also adds complete NLS info for (de,at)/(850,858,437), and keeps the entry for tr/850, but makes tr/857 the default. > it successfully, I can try to write a code to read the tables from > COUNTRY.SYS into the kernel tables (unless you want to do this yourself, > of course ;-) Please, do it yourself. However, you should keep in mind the following limitations if you overwrite the kernel hardcoded tables: * The hardcoded NLS package points the UCASE and FUCASE pointers to the same table. This is OK for the most of the country/codepage pairs, but will not allow loading info for codepage 852. * The harcoded package does not have enough space for a non-empty DBCS table, so it won't be possible to load the info for Japanese, Chinese or any other language with DBCS. * There is no room for the LCASE table, so it won't be possible to load the ru/866 pair. I see three options: - Overwrite the hardcoded tables for "compatible" packages and just throw a message like "NLS package too complex. Use NLSFUNC" if the package has FUCASE, LCASE or non-empty DBCS (I don't remember who suggested that) - Leave the hardcoded pages untouched and allocate memory for the new package, then chain to nlsInfo.chain and make it the active package. This is what FD NLSFUNC does and what I think that Steffen had in mind when designing the nls support. - This one will not make me very popular :-) : Make enough room in the hardcoded tables. This will mean 258 (LCASE) + 130 (FUCASE) + 132 (DBCS) = 520 more bytes. Probably 132 bytes for DBCS is way too much and 16 or 32 should be enough in practice, but this is the theoretical maximum, AFAIK. Eduardo. ------------------------------------------------------- This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170 Project Admins to receive an Apple iPod Mini FREE for your judgement on who ports your project to Linux PPC the best. Sponsored by IBM. Deadline: Sept. 13. Go here: http://sf.net/ppc_contest.php _______________________________________________ Freedos-kernel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/freedos-kernel
