FAQ
Cover
This is the archive Discourse for the Processing (ALPHA) software.
Please visit the new Processing forum for current information.

   Processing 1.0 _ALPHA_
   Bugs
   Bug Fixes, Implemented Suggestions
(Moderator: fry)
   Multibyte characters
« Previous topic | Next topic »

Pages: 1 
   Author  Topic: Multibyte characters  (Read 2623 times)
saito

tatsuyas1979 WWW Email
Multibyte characters
« on: May 3rd, 2004, 5:40pm »

I would like to draw multibyte characters. I tried to make a .vlw file from japanese true type font file and text()ed to draw texts as usual. but, it didn't work.
I guess multibyte characters can not be displayed as it is.
 
Is it difficult or fundamentally impossible to show multibyte characters??
 
-T
 

-SAITO / tatsuyas@ucla.edu
fry


WWW
Re: Multibyte characters
« Reply #1 on: May 4th, 2004, 4:24am »

only a matter of spending some time to work on it, so not impossible and not particularly difficult. i just recently got support for other (roman-based) languages working, and it would be possible to do the same for japanese, i probably need to add something to the font builder window to allow you to select the character set that's produced.  
 
we don't want to be biased in language support because we see processing as an international project. so it's certainly a priority to get it working.  
 
in some ways, it's mostly an interface issue.. i don't want to make people select a "charset" or a "code page"  or learn any of the other terminology for this stuff.
 
just a thought, do you (or anyone else reading this) know what subsets of unicode characters would be needed for japanese? maybe we can start adding languages one-by-one. maybe we can just call the current charset "Roman" and then add Japanese (as another option) if we know what to include. then we can add other languages as people show interest (and tell us what needs to be included for that language).
 
TomC

WWW
Re: Multibyte characters
« Reply #2 on: May 4th, 2004, 10:48am »

At the end of his fairly lucid article on Unicode, Joel Spolsky says that all 29 translations of Joel On Software are published as UTF-8, but his editor uses UCS-2 (2-byte encoding) internally. If I understand him correctly this should mean that UTF-8 is adequate for almost everything.
 
Great article anyway... The Absolute Minimum Every Software Developer Absolutely, Positively Must Know About Unicode and Character Sets (No Excuses!) http://www.joelonsoftware.com/articles/Unicode.html  
« Last Edit: May 4th, 2004, 10:51am by TomC »  
fry


WWW
Re: Multibyte characters
« Reply #3 on: May 4th, 2004, 5:38pm »

right, utf-8 will do just about everything, it's just that i can't embed the entire utf8 charset (up to 65535 characters!) into a single font, when all that is needed is, say, the roman set, or the japanese set (with a handful of roman stuff).  
 
so what needs to be done is to figure out proper subsets for different languages, so we have something called "japanese" for instance, that includes the unicode subsets for the characters most likely to be used.
 
alternatively, i suppose we could just include all the characters available in the font, and hope that that does the right thing.. hmm.. maybe that's the way to go. maybe i'll include that as the default in 70, and if it causes trouble, it can be disabled.
 
fry


WWW
Re: Multibyte characters
« Reply #4 on: Jul 17th, 2004, 5:40am »

now implemented for megabucket. there's an option in the font builder for "all characters" which will include the entire (sometimes enormous) charset.
 
also lots of bug fixes to the font code that showed up as a result of being stretched that way. oy.
 
Pages: 1 

« Previous topic | Next topic »