Review Request 121336: Fix vector sapes resolution in Krita

[Dmitry Kazakov]

> On Дек. 4, 2014, 11:29 д.п., Camilla Boemann wrote:
> > libs/odf/KoUnit.cpp, line 382
> > <https://git.reviewboard.kde.org/r/121336/diff/1/?file=331818#file331818line382>
> >
> >     since you call this at all times in the widget, and the equal operator checks this value even if type is not PIXEL, then this shouldn't be done if type is != PIXEL
> >     
> >     OR the equal oper should be changed

Thanks for the catch! I have fixed the issue in the branch!


- Dmitry


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/121336/#review71339
-----------------------------------------------------------


On Дек. 3, 2014, 10:53 д.п., Dmitry Kazakov wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/121336/
> -----------------------------------------------------------
> 
> (Updated Дек. 3, 2014, 10:53 д.п.)
> 
> 
> Review request for Calligra and Boudewijn Rempt.
> 
> 
> Bugs: 313600 and 341107
>     http://bugs.kde.org/show_bug.cgi?id=313600
>     http://bugs.kde.org/show_bug.cgi?id=341107
> 
> 
> Repository: calligra
> 
> 
> Description
> -------
> 
> This is a cumulative patch for the work done in krita-chili-kazakov branch in git.
> 
> The main idea of the changes is that KoUnit returned by the canvas now contains proper pt->px conversion coefficient and the main shape widgets can handle it properly. They are now connected to the resource manager's signal and update the unit on every change.
> 
> There is one "not-ideal" point. The shapes can have internal transformations, but their size and position member functions do not take it into account. So in some of the widgets I had to take this "transformation factor" into account using KoUnit's pt->px factor. This looks a bit ugly, but I'm not sure we can do anything else here.
> 
> 
> Here is the full list of changes.
> 
> commit 5efa5b4e9205632e1945d5fb467bc37836a582c4
> Author: Dmitry Kazakov <dimula73 at gmail.com>
> Date:   Tue Dec 2 14:07:35 2014 +0300
> 
>     Fix width of the Pencil Tool stroke after changing resolution
>     
>     There were two things needed:
>     
>     1) Deselect shapes when a shape-based tool is activated.
>        That is not needed when twe creating a new shape and it causes
>        the shape stroke widget edit the options of the lastly
>        selected stroke, which is not expected.
>     2) Reset unit on every selection change in KoStrokeConfigWidget.
>        Otherwise deselection is not handled properly.
> 
> commit ef5e1d2544beb30a027f874f7aedd8add951fedd
> Author: Dmitry Kazakov <dimula73 at gmail.com>
> Date:   Tue Dec 2 13:05:30 2014 +0300
> 
>     Fix units in the shadow config docker
>     
>     1) Added correct unit updates to the widget
>     2) Moved approxTransformScale() calculation to KoUnit since it
>        is too global
> 
> commit 38eea366b6477c4fcb5f0b46ced9be4fb9b93c1f
> Author: Dmitry Kazakov <dimula73 at gmail.com>
> Date:   Fri Nov 28 15:23:30 2014 +0400
> 
>     Fix a nice infinite loop in KoShadowConfigWidget
>     
>     How to reproduce:
>     1) Add a shape in Krita
>     2) Enable Shadow
>     3) Zoom canvas with mouse wheel
> 
> commit fddc5e9cc648c4d0eceb67b793faa3c0c14b8d62
> Author: Dmitry Kazakov <dimula73 at gmail.com>
> Date:   Fri Nov 28 15:14:33 2014 +0400
> 
>     Fix Default tool widgets to show values in real pixels, not points
>     
>     Now the option widgets use correct KoUnit object to convert their
>     options in correct user-visible pixels. Some note on the implementation:
>     
>     1) Affects KoStrokeConfigWidget and DefaultToolWidget
>     2) The correct conversion value is stored in KoUnit, which is reported
>        by KisCanvas2 and retransmitted by KisView2.
>     3) There is a hack in KoStrokeConfigWidget. KoShapeStroke knows nothing
>        about the absoluteTransformation() of the shape, which doesn't stop
>        the shape from doing the transformation of the outline (loaded into
>        QPainter directly). So we take it into account manually, by adding
>        a multiplier into KoUnit.
>     
>     BUG:313600
>     BUG:341107
> 
> 
> Diffs
> -----
> 
>   krita/ui/kis_view2.cpp aa894e8 
>   krita/ui/tool/kis_delegated_tool.h c63c6f9 
>   krita/ui/tool/kis_delegated_tool_policies.h PRE-CREATION 
>   krita/ui/tool/kis_delegated_tool_policies.cpp PRE-CREATION 
>   libs/flake/KoShapeShadow.h ccb3a3f 
>   libs/flake/KoShapeShadow.cpp ead0b80 
>   libs/odf/KoUnit.h ae26734 
>   libs/odf/KoUnit.cpp dc22265 
>   libs/widgets/KoShadowConfigWidget.h b3140dd 
>   libs/widgets/KoShadowConfigWidget.cpp d08f232 
>   libs/widgets/KoStrokeConfigWidget.cpp 445ea9f 
>   libs/widgets/KoUnitDoubleSpinBox.cpp ff94913 
>   plugins/defaultTools/defaulttool/DefaultToolWidget.cpp 55885d2 
>   krita/ui/CMakeLists.txt e28134a 
>   krita/ui/canvas/kis_canvas2.cpp 60ade0b 
>   krita/ui/flake/kis_shape_controller.h 39ae1eb 
>   krita/ui/flake/kis_shape_controller.cpp a0db22d 
>   krita/ui/flake/kis_shape_layer.cc c0fe4ac 
>   krita/plugins/tools/defaulttools/kis_tool_path.h 36db16a 
>   krita/plugins/tools/defaulttools/kis_tool_pencil.h 8a69428 
>   krita/plugins/tools/selectiontools/kis_tool_select_path.h 0bb252b 
>   krita/plugins/tools/tool_transform2/kis_transform_utils.cpp 9fd7e29 
> 
> Diff: https://git.reviewboard.kde.org/r/121336/diff/
> 
> 
> Testing
> -------
> 
> Tested in Krita
> 
> 
> Thanks,
> 
> Dmitry Kazakov
> 
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/calligra-devel/attachments/20141204/f51f1402/attachment.htm>