bring speed back into KConfig

[David Faure]

On Monday 21 April 2008, Juergen Pfennig wrote:
> On Thu 17.04.2008, Dirk Mueller wrote:
> > On Wednesday 16 April 2008, Juergen Pfennig wrote:
> > > I also found (at that time) that the code was not perfectly optimized and
> > > it was not capable of handling different line ending styles CR vs LF vs
> > > CR LF and it did not handle various unicode encodings correctly.
> >
> > there is only one encoding: utf8, so this is lesser of a concern. the line
> > ending style might be a new problem, but it certainly wasn't an isseu with
> > kde3 :)
> 
> NO NO NO !!!!
> 
> Your assumption is that only kconfig writes to config files.
Well it's called KConfig for a reason :)

It's used for config files and desktop files, both of which are specified to contain utf8.

> But any user 
> or user program could do. What about running on windows? 

What about windows? All KDE-created config files and desktop files are utf8, on Windows like anywhere else.

Did you mean *.ini files? The KConfig parser is not meant for those, since KConfig has different
escaping rules - and indeed different encoding rules. If someone ever wants to parse *.ini files
he'll have to write an actual ini file parser, not try to reuse KConfig for the job.

> What about users that 
> store config files in some sort of repository. Just an example: cvs on linux 
> dislikes a directory structure created with the windows version of cvs.

I don't see the relation with KConfig.

> Any program that reads a text file could be capable of handling various line 
> styles and encodings.

KConfig doesn't read any text file, it reads files with a very specific format.

-- 
David Faure, faure at kde.org, sponsored by Trolltech to work on KDE,
Konqueror (http://www.konqueror.org), and KOffice (http://www.koffice.org).