different codings of a char in the same litteral

different codings of a char in the same litteral

[Sébastien Kirche]

Hi,

In Gnus, to display the recipients  names or the newsgroups names instead of
my  name   in  the  sent   messages  lists,  I  eventually   configured  the
gnus-ignored-from-addresses that is a regexp for the From: names to rid off.

However, in the past i used different From: in different coding systems :
- Sebastien Kirche
- Sébastien Kirche encoded in latin-1 then latin-9

My problem was to handle the both versions of the accentuated form. 

I tried to  copy/paste the chars from each version, but  as my coding system
is latin-9, the  latin-1 was re-encoded. Thanks to C-x =  i managed to build
the following regexp :

(setq gnus-ignored-from-addresses "S\\(e\\|\x8e9\\|\xf69\\)bastien Kirche")

Where \x8e9 is the é in latin-1 and \xf69 is the latin-9.

But i wonder if it exists a proper manner less tricky to build such a string
(without simply replacing the possibly accentuated char by a .+) ?


-- 
Sébastien Kirche

Re: different codings of a char in the same litteral

[Peter Dyballa]

Am 11.04.2005 um 14:49 schrieb Sébastien Kirche:

> But i wonder if it exists a proper manner less tricky to build such a 
> string
> (without simply replacing the possibly accentuated char by a .+) ?
>

These should work:

   unify-8859-on-decoding-mode: t
   unify-8859-on-encoding-mode: t

--
Greetings

   Pete

"They're putting dimes in the hole in my head to see the change in me."

Re: different codings of a char in the same litteral

[Sébastien Kirche]

Le 11 Apr 2005, Peter Dyballa a formulé :

> These should work:
> 
> unify-8859-on-decoding-mode: t
> unify-8859-on-encoding-mode: t

I used  to define this  until some recent  tests where I determined  that it
causes some  mess with  OSX and  fontsets for displaying  at least  the euro
char.

I keep only unify-8859-on-encoding-mode.

Maybe it is only relative to the current cvs version ?

-- 
Sébastien Kirche