| Summary: | Mirror on layer with oddly shaped selection corrups layer data | ||
|---|---|---|---|
| Product: | [Applications] krita | Reporter: | Sven Langkamp <sven.langkamp> |
| Component: | Tools/Transform | Assignee: | Krita Bugs <krita-bugs-null> |
| Status: | RESOLVED FIXED | ||
| Severity: | normal | CC: | halla |
| Priority: | NOR | ||
| Version: | git master (please specify the git hash!) | ||
| Target Milestone: | --- | ||
| Platform: | unspecified | ||
| OS: | Linux | ||
| Latest Commit: | Version Fixed In: | ||
| Sentry Crash Report: | |||
| Attachments: | Screenshot | ||
|
Description
Sven Langkamp
2012-07-11 17:01:08 UTC
Created attachment 72457 [details]
Screenshot
Should mirroring even take the selection into account? If the selection is used to select the source and limit the destination, I think it's almost unpredictable by the user how the result would look like. I also looked at Gimp and it does always mirror the whole layer, even if there is a selection. Hm.... I think at one point we wanted to have the selection determine which part was going to be mirrored, but not limit the pixels we'd write to; and the selection would also determine the midpoint of the mirrored part. It's a bit confusing, I think, as the other actions in that group like shear, scale and rotate all work directly on the layer without using the selection. Beside that it might also make more sense if the layer has a local selection. If we mirror the layer and the mask there is a situation where the layer would be mirrored using the selection and the selection itself woud be mirrored as part of the local selection. Git commit 6e06c169b14b903f32280669369d83408474afaf by Sven Langkamp. Committed on 12/07/2012 at 11:30. Pushed by langkamp into branch 'master'. moved mirroring code to a node visitor. This fixes mirroring of group layers and local selections. Also removes duplicated code from layer and mask manager and puts it into the node manager. Using selections is disabled for now as that produces some problems. Mirroring shape selections still not working. Related: bug 298719 M +1 -0 krita/ui/CMakeLists.txt M +1 -76 krita/ui/kis_layer_manager.cc M +0 -2 krita/ui/kis_layer_manager.h M +0 -44 krita/ui/kis_mask_manager.cc M +0 -10 krita/ui/kis_mask_manager.h A +156 -0 krita/ui/kis_mirror_visitor.cpp [License: GPL (v2+)] A +48 -0 krita/ui/kis_mirror_visitor.h [License: GPL (v2+)] M +29 -4 krita/ui/kis_node_manager.cpp M +2 -0 krita/ui/kis_node_manager.h http://commits.kde.org/calligra/6e06c169b14b903f32280669369d83408474afaf I have disabled the use of the selection until this bug is resolved. Downgrading to normal. Git commit a5bc1bd8100f24624f767fd4eb8d19a100bf5c83 by Thorsten Zachmann, on behalf of Sven Langkamp. Committed on 12/07/2012 at 11:30. Pushed by zachmann into branch 'calligra/2.5'. moved mirroring code to a node visitor. This fixes mirroring of group layers and local selections. Also removes duplicated code from layer and mask manager and puts it into the node manager. Using selections is disabled for now as that produces some problems. Mirroring shape selections still not working. Related: bug 298719 (cherry picked from commit 6e06c169b14b903f32280669369d83408474afaf) M +1 -0 krita/ui/CMakeLists.txt M +1 -76 krita/ui/kis_layer_manager.cc M +0 -2 krita/ui/kis_layer_manager.h M +0 -44 krita/ui/kis_mask_manager.cc M +0 -10 krita/ui/kis_mask_manager.h A +156 -0 krita/ui/kis_mirror_visitor.cpp [License: GPL (v2+)] A +48 -0 krita/ui/kis_mirror_visitor.h [License: GPL (v2+)] M +29 -4 krita/ui/kis_node_manager.cpp M +2 -0 krita/ui/kis_node_manager.h http://commits.kde.org/calligra/a5bc1bd8100f24624f767fd4eb8d19a100bf5c83 I think the current behaviour is actually fine :-). Let's close the bug. |