It's going to take me some time to track down that table. I may even have to do as you suggested :) Unfortunately I don't have time to look at that for a while, though I'm trying to find someone who has the table.
-Shawn -----Original Message----- From: Anne van Kesteren [mailto:[email protected]] Sent: Wednesday, March 28, 2012 11:27 AM To: Shawn Steele Cc: [email protected] Subject: Re: big5 and big5-hkscs On Wed, 28 Mar 2012 19:56:26 +0200, Shawn Steele <[email protected]> wrote: > http://blogs.msdn.com/b/shawnste/archive/2007/03/12/cp-951-hkscs.aspx > > By default, no, HK and other machines are the same. We used to provide > a hacked code page as an optional file for HKSCS support, which used a > hacked CP950 (aka CP951) and installed that as CP950. That provided > HKSCS glyphs by reusing some of the PUA code points. > > We recommend people convert from that mechanism to Unicode, using the > tool you found to convert from the UTF-16 PUA to UTF-16 HKSCS. > > What is actually found on the web, I have no clue. Pretty much any > CP951 encoded data isn't going to be very portable. Our recommendation > would be for people having such content to convert to UTF-16 (or maybe > UTF-8). Thanks, this is awesome. Is there public documentation on the PUA code point to Unicode code point mapping? I guess I can install the utility and write some software around it to extract the mapping table, but I'm not very familiar with writing software on Windows. -- Anne van Kesteren http://annevankesteren.nl/
