speeding up KFind

[Stephan Kulow]

On Monday 26 May 2003 04:12, Andras Mantia wrote:
> On Sunday 25 May 2003 23:07, Maks Orlovich wrote:
> > On Monday 26 May 2003 04:06 pm, Daniel Naber wrote:
> > > Hi,
> > >
> > > KFind can be very slow if ones searches for file content because it
> > > searches any file, even large binary files that usually don't even
> > > contain text. As a fix I suggest to ignore all files that have a mime
> > > type unlike "text/*". Of course there are exceptions like KOffice and
> > > OpenOffice.org files (these are the only exceptions currently). Searches
> > > that don't use the "contains text" field won't be affected.
> > >
> > > Is that change okay or will some important files be missed this way?
> >
> > Beware application/x-python, x-perl, x-ruby, etc., and similar mimetypes.
> > Perhaps it can use the new "is it text" functionality that was added
> > recently?
> 
> Yes, I think the X-KDE-Text should be used. Otherwise you will miss some pure
> text documents, which are not under text/*.

I would like to have this replaced by a X-KDE-IsAlso hierarchy. This X-KDE-Text should
be removed for a general solution ;(

Greetings, Stephan