Advertisement
Guest User

Untitled

a guest
May 17th, 2017
714
0
Never
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
text 5.09 KB | None | 0 0
  1. commit 9d5df13605bae1329a8e6d2140cff6b134f99e09
  2. Author: Boudewijn Rempt <boud@valdyas.org>
  3. Date: Tue May 16 16:28:48 2017 +0200
  4.  
  5. Don't abort if the tile directory has the wrong size
  6.  
  7. This allows us to at least load the file referenced in
  8.  
  9. BUG:379889
  10.  
  11. which is incomaptible with xcf version v0003, but still carries
  12. that version number. Now it loads in krita... But no guarantees
  13. about 2.9 files. Mostly, they will still crash krita because
  14. the xcftools library aborts hard on encountering a problem, which
  15. we cannot prevent.
  16.  
  17. commit 0badd8a7e7ff0d5ad98446efa9cedb5da5eb101e
  18. Author: Boudewijn Rempt <boud@valdyas.org>
  19. Date: Tue May 16 16:09:46 2017 +0200
  20.  
  21. Revert "Don't try to load version 3 XCF files made with Gimp 2.9"
  22.  
  23. This reverts commit 62327b36653bc878acda87d0597b82b1c1f1b5bb.
  24.  
  25. commit 62327b36653bc878acda87d0597b82b1c1f1b5bb
  26. Author: Boudewijn Rempt <boud@valdyas.org>
  27. Date: Tue May 16 12:22:17 2017 +0200
  28.  
  29. Don't try to load version 3 XCF files made with Gimp 2.9
  30.  
  31. Finally Gimp 2.9's file format has changed so much that xcftools
  32. cannot load the tiles anymore. Until someone, somewhere updates
  33. xcftools to support the new file format, Krita won't load Gimp
  34. 2.9 files.
  35.  
  36. BUG:379889
  37.  
  38. commit 6b40888ab35df441e47d5e7a0b94ef616f4181e4
  39. Author: l10n daemon script <scripty@kde.org>
  40. Date: Tue May 16 05:32:16 2017 +0200
  41.  
  42. SVN_SILENT made messages (.desktop file) - always resolve ours
  43.  
  44. In case of conflict in i18n, keep the version of the branch "ours"
  45. To resolve a particular conflict, "git checkout --ours path/to/file.desktop"
  46.  
  47. commit 9908e7668cf268b63560e27f9a5412097682bf9e
  48. Author: l10n daemon script <scripty@kde.org>
  49. Date: Sun May 14 04:39:26 2017 +0200
  50.  
  51. GIT_SILENT made messages (after extraction)
  52.  
  53. commit c0ce7240c790fecc4bcdc27ba653a1e59119bc88
  54. Author: Kevin Funk <kfunk@kde.org>
  55. Date: Mon May 8 12:30:43 2017 +0200
  56.  
  57. clang-tidy: Use override where needed
  58.  
  59. This time for 3.1 branch
  60.  
  61.  
  62. commit 6d56639afe16f2e5ac9ee0e30861bdd9b4da2b36
  63. Author: Dmitry Kazakov <dimula73@gmail.com>
  64. Date: Tue May 2 21:21:24 2017 +0300
  65.  
  66. Fix crash when starting playback with openGL disabled
  67.  
  68. That is a critical bug in 3.1.3 :(
  69.  
  70. commit 4e41d829a363aaac2a7b555ec321d83dfe157738
  71. Author: Dmitry Kazakov <dimula73@gmail.com>
  72. Date: Mon May 1 17:15:44 2017 +0300
  73.  
  74. Fix crash when switching color-filtered layers too quickly
  75.  
  76. Sometimes, during the removal process, the model emits currentChanged()
  77. signal, which may cause a double-recursion, which causes a crash.
  78.  
  79. BUG:377732
  80.  
  81. commit 89641bfc428c268ae3fe42a0f4dd350b111ebc72
  82. Author: Dmitry Kazakov <dimula73@gmail.com>
  83. Date: Mon May 1 11:17:31 2017 +0300
  84.  
  85. Don't let the user close the document while saving is in progress
  86.  
  87. Ideally, we should postpone closing till the saving is finished,
  88. but it is impossible to do right now, because doc->setModified()
  89. is called outside the saving lock held. Later, when we refactor
  90. the saving code and the modified state will be cleared under
  91. the lock, we can easily use KisSignalCompressor to close the
  92. document.
  93.  
  94. The crash itself happened because the hack in KisPart::removeView()
  95. just skipped the deinitialization of the view and the input manager
  96. was still connected to the canvas that was destroyed.
  97.  
  98. BUG:378538
  99.  
  100. # Conflicts:
  101. # libs/ui/KisDocument.cpp
  102.  
  103. commit 79ab516d1ad5f8f427c4a6b279184b37dda0eedd
  104. Author: Dmitry Kazakov <dimula73@gmail.com>
  105. Date: Fri Apr 28 21:43:33 2017 +0300
  106.  
  107. Don't scale down too small previews for the transform tool
  108.  
  109. BUG:374388
  110.  
  111. commit e1fc8b8eda4f90177fd4b216d91eaef0114ffa8e
  112. Author: Dmitry Kazakov <dimula73@gmail.com>
  113. Date: Fri Apr 28 20:14:13 2017 +0300
  114.  
  115. Don't open the screen resolution settings dialog multiple times
  116.  
  117. BUG:375773
  118.  
  119. commit 58ab108605b0a59b456e5e407fdc36567b9d4525
  120. Author: Dmitry Kazakov <dimula73@gmail.com>
  121. Date: Fri Apr 28 15:54:33 2017 +0300
  122.  
  123. Fix rendering of animation frames if the directory not exist
  124.  
  125. Summary:
  126. This patch fixes four bugs:
  127.  
  128. 1) Recover warning dialog if the user tries to render animation
  129. frames into the directory with already existing frames. (QDir
  130. uses wildcard symbols instead of c-like patterns used by ffmpeg)
  131.  
  132. 2) Warn the user if rendering of frame files didn't complete
  133. successfully. Previously, Krita just tried to run ffmped on the
  134. non-existing files.
  135.  
  136. 3) Create the exporting directory if it doesn't exist.
  137.  
  138. 4) Create the rendering directory if it doesn't exist.
  139.  
  140. NOTE: the patch doesn't introduce any new strings, so it should be
  141. safe for merging into 3.1
  142.  
  143. Test Plan:
  144. Try rendering animation into unexisting/read-only directories. In the former
  145. case the new directory should be created. In the latter, error message should
  146. appear.
  147.  
  148. Reviewers: #krita, rempt
  149.  
  150. Differential Revision: https://phabricator.kde.org/D5637
Advertisement
Add Comment
Please, Sign In to add comment
Advertisement