headers
Piotr Szymanski | 1 Apr 2003 16:15
Picon

problem with 030329 snapshot.

Hi,
I keep getting these errors whentrying to start koffice apps such as krita or  
kexi:
koffice (lib kofficecore): ERROR: Couldn't find the native MimeType in kexi's 
desktop file. Check your installation !
KWord and stuff work ok.
Im running qt-copy from 030317 and kde from 030329.
--

-- 
Piotr Szymanski
djurban <at> redfox.pl
Permalink | Reply | 1 comment |
headers
David Faure | 1 Apr 2003 17:02
Picon

Re: problem with 030329 snapshot.


On Tuesday 01 April 2003 16:15, Piotr Szymanski wrote:
> Hi,
> I keep getting these errors whentrying to start koffice apps such as krita or  
> kexi:
> koffice (lib kofficecore): ERROR: Couldn't find the native MimeType in kexi's 
> desktop file. Check your installation !
> KWord and stuff work ok.

Please use koffice/tools/scripts/check_koffice_install.sh to debug this problem
after changing instance=kword to instance=krita or instance=kexi.

--

-- 
David Faure -- faure <at> kde.org, dfaure <at> klaralvdalens-datakonsult.se
Qt/KDE/KOffice developer
Klarälvdalens Datakonsult AB, Platform-independent software solutions
Permalink | Reply | 0 comments |
headers
Matthias Kretz | 2 Apr 2003 13:47
Picon
Favicon
Gravatar

KSpreads COUNTBLANK function

Hi,

yesterday I commited a fix to the koffice branch for COUNTBLANK. I'd fix the 
function for HEAD, too but it's commented out there. Should I just fix the 
commented code or is the function supposed to disappear?

Please CC me, I'm not on this list.

--

-- 
C'ya
        Matthias
________________________________________________________
Matthias Kretz (Germany)                          <><
http://Vir.homeip.net/
MatthiasKretz <at> gmx.net, kretz <at> kde.org,
Matthias.Kretz <at> urz.uni-heidelberg.de
Permalink | Reply | 2 comments |
headers
Norbert Andres | 2 Apr 2003 19:45
Picon

Re: KSpreads COUNTBLANK function

Hi,

the function is not commented out: it is not in the information group anymore 
but in math. So you need to look here:
kspread_functions_math.cc

Regards
Norbert

On Wednesday 02 April 2003 13:47, Matthias Kretz wrote:
> Hi,
>
> yesterday I commited a fix to the koffice branch for COUNTBLANK. I'd fix
> the function for HEAD, too but it's commented out there. Should I just fix
> the commented code or is the function supposed to disappear?
>
> Please CC me, I'm not on this list.
Permalink | Reply | 1 comment |
headers
Matthias Kretz | 2 Apr 2003 21:40
Picon
Favicon
Gravatar

Re: KSpreads COUNTBLANK function

Ok, I found it. Looks like the fix won't work there ;-)
Thanks.

On Wednesday April 2 2003 19:45, Norbert Andres wrote:
> Hi,
>
> the function is not commented out: it is not in the information group
> anymore but in math. So you need to look here:
> kspread_functions_math.cc

--

-- 
C'ya
        Matthias
________________________________________________________
Matthias Kretz (Germany)                          <><
http://Vir.homeip.net/
MatthiasKretz <at> gmx.net, kretz <at> kde.org,
Matthias.Kretz <at> urz.uni-heidelberg.de
Permalink | Reply | 0 comments |
headers
thierry lorthiois | 5 Apr 2003 00:31
Picon

patch for wmf library

well,
http://www.malistebb.com/kde/libkowmf

this patch include the new wmf library : libkowmf
tested with the current cvs and doesn't conflict with 
others wmf libraries.

the configuration isn't complete because I have to edit
/configure.in by hand. But don't know how to do more ?

It's an evolution of qwmf and the goal is to cover
all the need with wmf format (read/write).

regards
Permalink | Reply | 2 comments |
headers
Rob Buis | 4 Apr 2003 22:34
Picon
Picon
Favicon

Re: patch for wmf library

Hi,

On Saturday 05 April 2003 00:31, thierry lorthiois wrote:
> well,
> http://www.malistebb.com/kde/libkowmf
>
> this patch include the new wmf library : libkowmf
> tested with the current cvs and doesn't conflict with
> others wmf libraries.

Looking at the code reminds me, karbon can handle ellipses, arcs, pies and 
chords now, so it shouldnt be too hard to add that functionality to the wmf 
import filter.
Just a thought.
Cheers,

Rob.
Permalink | Reply | 0 comments |
headers
Nicolas Goutte | 4 Apr 2003 23:50
Picon

Re: patch for wmf library

On Saturday 05 April 2003 00:31, thierry lorthiois wrote:
> well,
> http://www.malistebb.com/kde/libkowmf
>
> this patch include the new wmf library : libkowmf
> tested with the current cvs and doesn't conflict with
> others wmf libraries.
>
> the configuration isn't complete because I have to edit
> /configure.in by hand. But don't know how to do more ?

Doing it in configure.in.in is better or you will lose your change at the 
first "make -f Makefile.cvs".

What do you want to check?

A few examples:
- the ImageMagick check in koffice/configure.in.in
- the KDE 3.1.x check in koffice/mimetypes/configure.in.in

(Autoconf's doc is at info:autoconf )

>
> It's an evolution of qwmf and the goal is to cover
> all the need with wmf format (read/write).
>
> regards

Have a nice day/evening/night!
(Continue reading)

Permalink | Reply | 0 comments |
headers
thierry lorthiois | 5 Apr 2003 02:22
Picon

Re: patch for wmf library

> > the configuration isn't complete because I have to edit
> > /configure.in by hand. But don't know how to do more ?
>
> Doing it in configure.in.in is better or you will lose your change at the 
> first "make -f Makefile.cvs".
>

thanks,
the new patch include the change in /lib/configure.in.in

http://www.malistebb.com/kde/libkowmf

bye
Permalink | Reply | 501 comments |
headers
Nicolas Goutte | 5 Apr 2003 17:06
Picon

Re: patch for wmf library

Looks good for me. (I have taken more time to look at your code.)

I have just a few small critics.

In KoWmf::dibToBmp, you could avoid one casting in:
  bmpHeader = (BMPFILEHEADER*)((const char*)pattern);
by using:
  bmpHeader = (BMPFILEHEADER*)(pattern.data());

In KoWmf::StateWmf, why are using unsigned instead of bool, if each data is 
only 1 bit long? I also wonder if the memory gain of :1 is worth the extra 
machine code for accessing single bits.

Also perhaps for binary compatibility problems you should move all private 
variables of KoWmf in an extra class not defined in kowmf.h in the style
  class KoWmfPrivate;
  class KoWmf
 {
 ...
    KoWmfPrivate* d;
 }
like it is done in many KDE classes whose headers are installed.

The last one is in KoWmf:.qtColor. I think that the variables should be const 
or the compiler willl not be able to optimize. Also perhaps it is a candidate 
for inline (as the function will not change much in future.)

Otherwise, I think that you can commit. (You can even make the small changes 
later.)
(Continue reading)

Permalink | Reply | 501 comments |

Gmane