nanogui: UniCode
Subject:
Re: Unicode
From:
"Greg Haerr" ####@####.####
Date:
29 Mar 2000 15:48:37 -0000
Message-Id: <0d3a01bf9994$af698180$15320cd0@gregh>
: > So this could speak for a 16-Bit representation. I think, it would
: > be more easy to convert a given program to 16 bit, because you have
: > a single 16 bit value for a character, not a floating scheme. So it
: > could be easier to use existing routines for example to get the
: > length of a string etc. (This could be different: Is a length the
: > lenght in bytes or the number of characters -what now could be
: > different. But in old programs its always the same.)
:
: I couldn't disagree with you more. I loathe the idea of not using char for
: strings, and so does most of GNU/Linux I think.
Well, the good news is that although Microwindows does support
16 and 32-bit unicode out of the box, we also have UTF-8, which,
as Alan posted, has the dual advantage of working with most all the
C string manipulations functions, as well as being identical to ascii-7
in the low half. Also, UTF-8 can express all unicode 16 and 32 characters.
So a program written specifying UTF-8 text display will work well
with ascii and can be used without modification to display any unicode
character. Note this only works now with truetype fonts, as the T1LIB
adobe T1 font library still wants a char-sized index for the glyph.
Regards,
Greg
