[MPlayer-cvslog] CVS: main ChangeLog,1.197,1.198

[Diego Biurrun]

On Fri, Apr 22, 2005 at 10:55:55AM -0400, Rich Felker wrote:
> On Fri, Apr 22, 2005 at 12:36:35PM +0200, Diego Biurrun wrote:
> > On Thu, Apr 21, 2005 at 03:30:22PM -0400, Rich Felker wrote:
> > > On Sun, Apr 17, 2005 at 05:29:53PM +0200, Diego Biurrun wrote:
> > > > On Sun, Apr 17, 2005 at 10:14:20AM -0400, Rich Felker wrote:
> > > > > On Sat, Apr 16, 2005 at 10:32:56PM +0200, Diego Biurrun CVS wrote:
> > > > > > 
> > > > > > +    * put changed notices into all imported files for GPL §2a compliance
> > > > > 
> > > > > may i object to this senseless use of latin1 encoding in ascii files?
> > > > > writing out the word "section" would not be difficult, or just using
> > > > > the ascii character S...
> > > > 
> > > > You may, although a bit of an explanation why this is a problem would be
> > > > appreciated.  § is valid in all charsets I am familiar with...
> > > 
> > > It's not valid ascii or utf8. The character exists in unicode of
> > > course but it's not encoded as you have written it in the docs, which
> > > is a latin1 (maybe latin*?) encoding.
> > 
> > I'm quite confident it should be valid in all latin encodings.  Anyway,
> > I've removed it.
> 
> The point is that all text files should be assumed to be ascii or
> utf8, unless they're localized, and even then they should be utf8.
> Otherwise we're making it more difficult for people to abandon legacy
> encodings and switch to utf8.

OK, I'll keep it in mind.  It's a problem we will have to attack sooner
or later.  I'm already running ugly sed hacks in the documentation
generation cronjob on mphq because the outdated groff version there does
not support the proper encoding for the east european languages...

Diego