<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=Content-Type content="text/html; charset=iso-8859-1">
<META content="MSHTML 6.00.2722.900" name=GENERATOR></HEAD>
<BODY bottomMargin=0 leftMargin=3 topMargin=0 rightMargin=3>
<DIV></DIV>
<DIV>>It may end up being a lot of work to use old stacks <BR>>containing
Japanese, since it looks like we've got to recode the scrips <BR>>as well,
but I haven't really played with them much yet.</DIV>
<DIV>Well if we gave you guys what you wanted, which is to deal with shift-JIS,
multibyte, it wouldn't be cross platform, and you would be unsure if a character
was indeed one or two bytes. We're also in the process of modifying the
uniencode and unidecode property so you will be able to easily convert form
shift-JIS, utf8 to unicode and vice versa.</DIV>
<DIV> </DIV>
<DIV>When typing plain text (ASCII character 0 to 255) the engine actually saves
that as one byte per character. If you want<BR>to know for sure, it's always
good to check the textfont of a selection. Even when setting the unicodetext,
the engine optimizes. It also always saves<BR>*carriage* returns, tabs,
as one byte per character. This is to save memory, hard disk space, and
speed up processing. Rather than bloat everything, and go with everything is a
unicode character like Java, we decided to find a way for unicode to fit in
comfortably with the ASCII text.<BR><BR>If you plan to manipulate unicode text
in your variables, and want to ensure that it is two bytes per character you
should always get the unicodetext of<BR>a field or selection first..this will
convert all character in a field (even if they are not using a unicode font) to
2 byte per character. Then when<BR>your done manipulating the text, set the
unicodetext of that field or selection.<BR></DIV>
<DIV>For those who may be wishing to write some unicode processing algorithms,
they can check out the useunicode property, and numtochar and chartonum
routines. I would be interested in hearing what if any special challenges people
have run into when modifying their programs to deal with unicode.</DIV>
<DIV><BR>Tuviah Snyder <<A
href="mailto:tuviah@runrev.com">tuviah@runrev.com</A>> <<A
href="http://www.runrev.com/">http://www.runrev.com/</A>><BR>Runtime
Revolution Limited - The Solution for Software Development</DIV></BODY></HTML>