Review Request: Added support for drawing the shadow for a shape group, fixed shadow blur zooming bug.

[Jan Hambrecht]

> On Aug. 3, 2011, 4:55 p.m., Jan Hambrecht wrote:
> > libs/flake/KoShapeGroup.cpp, lines 63-76
> > <http://git.reviewboard.kde.org/r/102072/diff/4/?file=30652#file30652line63>
> >
> >     A group shape can also have filter effects applied, so that must be taken into account too.
> >     And I think that now the group shape has a proper boundingRect() function, the size() function should return a correct size too.
> 
> Yue Liu wrote:
>     I don't know if the non-zero size would break other things.
> 
> Jan Hambrecht wrote:
>     I don't know either. But having both functions return conflicting values is not something we should settle on.
> 
> Yue Liu wrote:
>     I suggest write a TODO here but not change size unless one day someone has to use a non-zero size.

Or in other words: let someone else deal with that inconsistency. ;-)


> On Aug. 3, 2011, 4:55 p.m., Jan Hambrecht wrote:
> > libs/flake/KoShapeShadow.cpp, lines 358-361
> > <http://git.reviewboard.kde.org/r/102072/diff/4/?file=30654#file30654line358>
> >
> >     Where is the number 3 comming from. Shouldn't d->blur be the exact blur radius?
> 
> Yue Liu wrote:
>     I forgot to move the comment here, the author of the shadow blur algorithm told me 3 times of the blur radius match the shadow region expansion well.
> 
> Jan Hambrecht wrote:
>     As d->blur is the actual number showing up in the gui, i think it is really confusing. How should the user know that when setting the blur radius to 10, it actually blurs 30 pixel wide? I think that is not a great thing to do.
> 
> Yue Liu wrote:
>     I'll ask Ariya what does the blur radius results to the shadow, it seems not simply expand a fixed value.

Yeah it would be great to have a better understanding of the blur radius' effect on the result.


- Jan


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
http://git.reviewboard.kde.org/r/102072/#review5355
-----------------------------------------------------------


On Aug. 3, 2011, 7:20 a.m., Yue Liu wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> http://git.reviewboard.kde.org/r/102072/
> -----------------------------------------------------------
> 
> (Updated Aug. 3, 2011, 7:20 a.m.)
> 
> 
> Review request for Calligra.
> 
> 
> Summary
> -------
> 
> Added support for drawing the shadow for a shape group, fixed shadow blur zooming bug.
> 
> 
> Diffs
> -----
> 
>   libs/flake/KoShapeContainer.cpp e7585ec 
>   libs/flake/KoShapeGroup.h 5ccd574 
>   libs/flake/KoShapeGroup.cpp aaa58b4 
>   libs/flake/KoShapeShadow.h 07c760c 
>   libs/flake/KoShapeShadow.cpp 5485b19 
>   libs/flake/tests/TestShapeGroupCommand.cpp 397c17b 
>   libs/widgets/KoShadowConfigWidget.cpp 1a41042 
>   plugins/dockers/shadowdocker/ShadowDocker.cpp ef53fae 
> 
> Diff: http://git.reviewboard.kde.org/r/102072/diff
> 
> 
> Testing
> -------
> 
> http://my.cdash.org/viewTest.php?onlydelta&buildid=212010
> 
> 
> Screenshots
> -----------
> 
> groupshadow / shape shadow
>   http://git.reviewboard.kde.org/r/102072/s/210/
> 
> 
> Thanks,
> 
> Yue
> 
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/calligra-devel/attachments/20110804/2d1c183a/attachment.htm>