Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20130808053107.GL221@brightrain.aerifal.cx>
Date: Thu, 8 Aug 2013 01:31:07 -0400
From: Rich Felker <dalias@...ifal.cx>
To: musl@...ts.openwall.com
Subject: Re: Re: Status of Big5 and extensions

On Thu, Aug 08, 2013 at 06:50:57AM +0200, Szabolcs Nagy wrote:
> * Rich Felker <dalias@...ifal.cx> [2013-08-08 00:30:35 -0400]:
> > On Thu, Aug 08, 2013 at 05:53:21AM +0200, Szabolcs Nagy wrote:
> > > note that
> > > C87A, C87C, C8A4 are mapped to 2xxxx in hkscs
> > > (old hkscs pua codes agree with uao)
> > 
> > OK, so is this non-conflicting?
> > 
> 
> i just wanted to note that these are the only codes
> in the 89x157 table that map to non-16bit unicode
> codepoints
> 
> there is no conflict

OK, great. I think rather than using any fancy tables for the 89x157
grid and HKSCS, we can just special-case these two in the code.

It also looks to me like HKSCS and UAO are essentially non-conflicting
in the 89x157 grid once you remove PUA junk; the only real conflict is
the half-width kana UAO mapped over part of HKSCS. So it may make
sense to just add all the extended mappings in this range except the
kana to the base Big5 table. This would definitely simplify HKSCS
support. If we later want UAO support, this range could just be
special-cased algorithmically since it seems to be direct range
mappings to Unicode.

Rich

Powered by blists - more mailing lists

Confused about mailing lists and their use? Read about mailing lists on Wikipedia and check out these guidelines on proper formatting of your messages.