[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-commits
Subject:    Re: koffice/libs/flake/tests
From:       jaham <jaham () gmx ! net>
Date:       2007-07-20 20:07:00
Message-ID: 200707202207.00497.jaham () gmx ! net
[Download RAW message or body]

On Friday 20 July 2007 18:40:19 Thomas Zander wrote:
> On Friday 20 July 2007 17:48:47 Jan Hambrecht wrote:
> > Thomas Zander wrote:
> > > On Friday 20 July 2007 17:34:52 Jan Hambrecht wrote:
> > >>> I just noted; a change that is a bit of a surprise to me :)
> > >>>
> > >>> I'm seeing that the names chosen for this (resize(), rotate(),
> > >>> scale()) are not the most useful for the absolute manner.  But all
> > >>> of KOffice depended on them being absolute...
> > >>>
> > >>> I'll rename the methods to avoid further confusion and make the
> > >>> rotate be absolute again.
> > >>
> > >> No you can't. Because there is no reliable way of retrieving the
> > >> actual rotation with arbitrary transformations applied to a shape.
> > >> Remember that we had that discussion already?
> > >
> > > But you can't just change the behavior of a method without altering
> > > all the code that calls it either... :(
> > >
> > > I'll see what I can do.
> >
> > Then please have my apologies. If you can point me to the now broken
> > code i will fix it for you.
>
> No worries :)
> I found one problem and fixed it (karbon/dockers/vtransformdocker.cc:346)
> I found another problem thats a bit bigger, its that the
> KoShapeRotateCommand still assumes absolute rotation.
>
> Of interrest there is that its being used in kivio and karbon;
> http://lxr.kde.org/ident?i=KoShapeRotateCommand
>
> Not sure if its better to remove the command and port those two usages, or
> fix the command to store the relative rotation per shape.
>
> Thanks for looking in to this!

The transform docker is going to be reworked anyway. So i think the best 
solution would be to remove these old commands, except the move command which 
still works.

Ciao Jan
[prev in list] [next in list] [prev in thread] [next in thread]