Re: emacs 22 on win XP, selection-coding-system, and copy & paste outside emacs

[Peter Dyballa <Peter_Dyballa@Web.DE>]

Am 03.08.2006 um 12:18 schrieb std:

> By the way, is latin-9 a smart choice? (compared to latin-1, or other
> utf-8 or -16...). I have no idea on these things.

Depends – I use ISO Latin-9/ISO 8859-15 in LaTeX, because then I can  
see € (ISO Latin-1 does not have the Euro currency), my usual  
environment is UTF-8 based. GNU Emacs 22.0.50 still is not perfect in  
UTF-8 use (a month name like Mär in dired is displayed as 'Mär',  
file names appear as 'RGB aÌ\210oÌ\210uÌ\210æÃ\206UÌ\210OÌ 
\210AÌ210.txt' or 'Perl_und_BuÌ\210cher'), the Unicode Emacs 23.0.0  
performs better, so I'd recommend so stay with some 8 bit encoding  
like ISO Latin-9/ISO 8859-15. And use UTF-8 or UTF-16 for data  
exchange with MS Windows XP applications. Both encodings are  
"representations" of a file's contents. UTF-8 uses 8 bit words to  
code this contents, usually using three or two of these words for  
each character, only in case of ASCII or ISO Latin-1 use one such  
word suffices. UTF-16 uses 16 bit words – so it can address the first  
64 K characters in Unicode uniformly, although it makes a difference  
in which sequence/direction the two octets of bits are read, BE or  
LE. Behind the Basic Multilingual Plane (BMP), UTF-8 and UTF-16 need  
to use 32 bits to address the Unicode characters – just as UTF-32  
uses from the start. But of course each of these three systems uses  
its own codes in this range(s)/plane(s) ...

--
Greetings

   Pete

Bake Pizza not war!