10462 commits found for project KOffice.
Regular Commiters: mlaurent, nikolaus, zander, staniek, sebsauer, dfaure, rempt
Last update: 07:45, 01.09.2006

TimeRevisionAuthorRepliesMessageFriday, 01. September 2006
09:40579511zachmann13976o make tests compile


TimeRevisionAuthorRepliesMessageFriday, 01. September 2006
03:23579384staniek513update
03:17579382staniek0version -> 1.9.90 (make more space for multiple beta versions)
03:00579375staniek0SVN_SILENT fix height & possible crash
00:55579335sebsauer0revert r578171 since it does not solve the crash-on-file-close issue :-/
00:23579323sebsauer0* dox++ * kis_filter_configuration.h got a properties()-method to be able to know within scripts what props are defined. Hope that's ok?
22:44579292langkamp0reenable Krita flake shapes. Only the paintdevice shape can be inserted for now


TimeRevisionAuthorRepliesMessageThursday, 31. August 2006
02:11579051sebsauer0Improved script-package handling. Should work now (mostly :) again.
01:40579045jaham0do not compile encrypted store if qca is not there
00:27579017langkamp0fixed a bug where the shape where still selected after deleting them
23:57579011langkamp0removed old commands which allready ported to flake
23:57579011langkamp0removed old commands which allready ported to flake
23:57579011langkamp0removed old commands which allready ported to flake
23:57579011langkamp0removed old commands which allready ported to flake
23:21578999langkamp0removed the old toolbox
21:51578974dannya0fixed in trunk/ according to http://en.wikipedia.org/wiki/The_quick_brown_fox_jumps_over_the_lazy_dog BUG: 121352
19:00578915sebsauer0Disabled bytes() / setBytes() again. May a thing to look later at again.
18:49578910mlaurent0Move into kdelibs
18:18578894coolo0another URL port
17:47578873schaap0Implemented support for encrypted documents in KOffice. Opening documents is completely transparent. Note that saving documents is not implemented in the GUI yet, nor is it currently functioning fully correct: there are still interoperability problems. Also moved KoXmlReader from libs/kofficecore to libs/store.
16:41578855sebsauer0Ups. true != false :-/ A lot of thanks to Thomas!
04:35578719sebsauer0typo-- SVN_SILENT
04:35578719sebsauer0typo-- SVN_SILENT
04:24578714sebsauer0Added the bytes() and setBytes() methods to be able to manipulate the pixels direct. This is rock fast (near C++ speed), but very experimental stuff!


TimeRevisionAuthorRepliesMessageWednesday, 30. August 2006
00:48578646nikolaus0Functions EVEN, ODD EVEN rounds up to the next nearest even integer. ODD rounds up to the next nearest odd integer. The former behaviour was the same as ISEVEN or ISODD, respectively.
19:12578534sebsauer0Hmpf. Whatever optimization I try here, it doesn't provide more performance. So, at least add the channel/SetChannel method && dox++ :-/
18:30578522reed0I don't know why PIGMENT_INCLUDES isn't getting LCMS_INCLUDE_DIR itself :(
17:06578490staniek0Added unicode->latin1 mappings for Czech characters, thanks to: Michal Svec
15:56578459reed0rename Locale.* to Localization.* to avoid conflicts on case-insensitive FSes
12:23578392sebsauer0It assert's whyever... to remove the assert doesn't show any direct problems. So, does it make sense to have that assert in the dtor?
10:48578375zachmann0o Fix moving of points in rotated objects. Save the cursor pos in document and not in shape position as the shape position can change.
06:45578340reed0x11-specific
06:35578339reed0oh man that is ugly, use #include_next to work around case-insensitive FS issue
05:56578336reed0no x11 if we are using Qt/Mac, use the minimal implementation
00:46578329reed0fix an ordering issue with find_package, and add some missing LCMS includes
00:42578327reed0fix building with apple gcc 4.0.1


TimeRevisionAuthorRepliesMessageMonday, 28. August 2006
19:59578269coolo1replaced several URLs with Url and URI with Uri - the patch became a bit larger than expected, so I couldn't have the time to check if it still compiles with the changes done in between
18:16578227danders0More porting to do...
16:51578199sebsauer0hmmm... not needed to iter over the args to don't do anything if debugging is disabled :-/
16:09578171sebsauer0KoToolBox should always have a QWidget parent cause else a) the instance don't get destroyed cause addDockWidget() does not reparent and even more worse b) the application may crash (what happened here reproducable at least with karbon and krita on trying to file=>close)
13:46578134sebsauer0More performance. To paint to often slows things down. So, while the cursor is fast now, the painting itself is still slow. Seems the reason is located in kis_view.cc#2420 which calls e.g. KisToolFreehand::move which starts to repaint instantly rather then caching the painting/invalidating for some ms's...
12:24578103staniek0combo box widget icon: 16x16 version added
11:51578084sebsauer0port r578081 Fixed some possible deadlock-cases. Now Krita starts and exit's normal without any crashes :)
07:21578020sebsauer0Show the progressbar even a second time :-/
06:52578015zachmann0o Move changing of position into the normalize method. This fixes moving of points in rotated shapes. o First check if there is a control point at a position. This makes it possible to change a control point when it is at the same position as a point.
06:36578011sebsauer0in pilexport display only drivers that support writing
06:23578008sebsauer0less debugging at core-methods == better performance :) SVN_SILENT
05:59578002sebsauer0Krita is able to use the Python Image Library ( PIL, see http://www.pythonware.com/library/ ) now to import and export from additional 38 fileformats. * not much testet yet. * expect some probs with non-rgb images. * since PyQt does not currently run with qt 4.2 snapshot tkinter is used for the gui. FEATURE


TimeRevisionAuthorRepliesMessageSunday, 27. August 2006
23:13577951sebsauer0made the randompaint.rb ruby-example working again.
20:03577899sebsauer0made the invert.rb ruby-example working again.
19:07577873sebsauer0Qt's dox say that start() should be used only once...
19:02577872sebsauer0Now it's ideal and imho worth a backport since testing shows, that it's even much faster on Qt3 :)
18:22577859sebsauer0revert since while it's faster, the progressbar didn't got updated anymore :-/
17:55577847sebsauer0even faster.
17:47577845sebsauer0Fix for the fix.
17:14577839sebsauer0Improved speed of all filters by some hundred percent.
16:36577824sebsauer0* made scripting-progress work again. * fix possible div by zero.
16:30577822nikolaus0Formula Fix evalution of quoted named areas.
14:50577787sebsauer0display the progressbar again.
14:07577770nikolaus0Unit test Adjustment. revision 562957 [Select for revision A] [Select for revision B] date: 16.07.2006 12:53; author: mecir The ^ operator is now left-associative. This change made in accordance with the latest OpenFormula draft.
13:54577764nikolaus0OpenDocument Rewrite the formula/expression encoding, using the tokenizer. Formula Fix the quoted sheetnames.
13:02577748nikolaus0Unit tests Let 'em run.
11:35577735nikolaus0OpenDocument Fix loading of cell values. Don't overwrite loaded formulas!
09:41577644nikolaus0OpenDocument Fix loading of named ranges. Formula Allow whitespaces in named areas.


TimeRevisionAuthorRepliesMessageSaturday, 26. August 2006
18:27577450sebsauer0* made invert.py working again. We are 300% (yes, 3 times) faster with the new framework. * merged filterconfig into filter. * don't know why the progressbar isn't displayed. theoretical it should work :-/
15:53577401sebsauer0porting++
14:09577385sebsauer0krosskritacore does not depend on kross any longer.
13:55577384sebsauer0More porting work.


TimeRevisionAuthorRepliesMessageFriday, 25. August 2006
23:30577247sebsauer0Some more porting.
22:33577235sebsauer0Some initial porting work.
21:12577217sebsauer0publish it under "Krita" :-/ SVN_SILENT
21:05577216sebsauer0disable samples till ported too. SVN_SILENT
20:47577206sebsauer0EnabÃled parts ofKrita's scripting-plugin again. * merged things together * provide ScriptingModule as with KSpread to have a "Krita" main-/root-module. * Moved scripting-menuitems to Tools to be more consistent with KSpread and Kexi. * scripting/kritacore/* is disabled for now. Will be ported later.
18:42577164sebsauer0Since this is another case where something crashes now at exact the same place deep inside Qt4-snapshot what worked before, I suspect a prob in Qt4 itself... So, seems a Pixmap that isNull() could crash QX11PaintEngine. QPainter::begin: Cannot paint on a null pixmap Program received signal SIGSEGV, Segmentation fault. 0xb652e0e8 in QX11PaintEngine::updatePen (this=0x8a2ca20, pen=@0xbf8c62a4) at painting/qpaintengine_x11.cpp:1284 1284 } else if (d->pdev->devType() == QInternal::Pixmap && d->pdev_depth == 32 (gdb) bt #0 0xb652e0e8 in QX11PaintEngine::updatePen (this=0x8a2ca20, pen=@0xbf8c62a4) at painting/qpaintengine_x11.cpp:1284 #1 0xb652e517 in QX11PaintEngine::updateState (this=0x8a2ca20, state=@0x8a2d9b8) at painting/qpaintengine_x11.cpp:1123 #2 0xb64d25f7 in QPainterPrivate::updateState (this=0x8a2d610, newState=0x8a2d9b8) at painting/qpainter.cpp:552 #3 0xb64dab1a in QPainter::eraseRect (this=0xbf8c64d8, r=@0xbf8c6470) at painting/qpainter.cpp:5095 #4 0xb7ef1a9b in QPainter::eraseRect (this=0xbf8c64d8, x=0, y=0, w=0, h=0) at /opt/qt4-copy/include/QtGui/qpainter.h:632 #5 0xb7ef71e2 in KisRuler::drawRuler (this=0x8843ef0) at /home/kde4/koffice/krita/ui/kis_ruler.cc:199 #6 0xb7ef7b1d in KisRuler::recalculateSize (this=0x8843ef0) at /home/kde4/koffice/krita/ui/kis_ruler.cc:108 #7 0xb7ef7b9a in KisRuler::setZoom (this=0x8843ef0, zoom=1) at /home/kde4/koffice/krita/ui/kis_ruler.cc:128 #8 0xb7f27ab8 in KisView::zoomAroundPoint (this=0x8757ef0, x=0, y=0, zf=1) at /home/kde4/koffice/krita/ui/kis_view.cc:1480 #9 0xb7f24d50 in KisView::setCurrentImage (this=0x8757ef0, image=@0xbf8c66a8) at /home/kde4/koffice/krita/ui/kis_view.cc:3659 #10 0xb7f24e21 in KisView::slotLoadingFinished (this=0x8757ef0) at /home/kde4/koffice/krita/ui/kis_view.cc:3846 #11 0xb7f2a167 in KisView (this=0x8757ef0, doc=0x817b6b8, adapter=0x817b744, parent=0x867ddb8) at /home/kde4/koffice/krita/ui/kis_view.cc:356 #12 0xb7eb8f6e in KisDoc::createViewInstance (this=0x817b6b8, parent=0x867ddb8) at /home/kde4/koffice/krita/ui/kis_doc.cc:873 #13 0xb75f59a5 in KoDocument::createView (this=0x817b6b8, parent=0x867ddb8) at /home/kde4/koffice/libs/kofficecore/KoDocument.cpp:339 #14 0xb761cc31 in KoMainWindow::setRootDocument (this=0x84f8d28, doc=0x817b6b8) at /home/kde4/koffice/libs/kofficecore/KoMainWindow.cpp:387 CC_MAIL: kde-core-devel@kde.org
17:57577144sebsauer0Port of ruby-backend done. Everything works now again, more unittests later.
14:27577085sebsauer0More work on the ruby-backend.
09:47576951danders0Port later.


TimeRevisionAuthorRepliesMessageThursday, 24. August 2006
23:38576855jaham0- implemented KoPointPropertyCommand to change the KoPathPoint properties - introduced a KoPointBaseCommand which stores the path shape of a command and can repaint after altering the command (avoids code duplication) - derive KoPointPropertyCommand and KoPointMoveCommand from KoPointBaseCommand
22:18576833jaham0- implemented a KoPointMoveCommand for undo/redo path points moving - added a KoPointType enum to KoPathPoint to identify the KoPathPoints parts - fixed repainting after moving points
20:07576795jaham0Let the KoPathShape::normalize function return the offset by which the points are translated and use that to move the shape afterwards abd adjust the last mouse position. No need to calculate the offset twice.
19:54576785zander0Fix loader, the consequative counting was inverted. Maybe I should use a easier word ;) Make the counter us a proper layout so it will be painted correctly! Follow hint from david and rename member, I was just too lazy to come up with a better name when I first committed it.
18:48576751sebsauer0Be able to use kjs e.g. from python; import Kross kjsaction = Kross.action("MyKjsScript") kjsaction.setInterpreter("javascript") kjsaction.setCode( "println(\"Hello world from Kjs\");" ) kjsaction.trigger()
18:25576740zander0Add some more tests for lists and implement them.
16:12576689sebsauer0revert since dfaure's hint solved it. thx :)
16:05576686sebsauer0Disable cause does not compile yet with kdelibs<today.
16:00576684sebsauer0heh. small typo that sounds interessting :)
15:56576682sebsauer0Initial implementation of the kjs/kjsembed backend.
12:10576612zander0Actually paint the listitem. Still a text for now :) Fix the loader with a counter of type=0 to remove the list from the parag :)
11:30576602dfaure0One new/delete pair less. Not sure why ListItemsHelper needs a d pointer, too.
10:46576515zander0Some refactorings of the text-lists code and make the indenting work for single level lists by adding the relevant init code in the KWDLoader.
03:32576416sebsauer0Don't assert on "WARNING: The KOfficePart service type isn't installed!" (whyever that warning shows up now). At least KSpread runs again that way and I am able to continue coding rather then re-compiling again and again.
00:45576396sebsauer0* added ScriptingModule which provides access to KoApplicationAdaptor, KoDocumentAdaptor, KSpread::ViewAdaptor, KSpread::SheetAdaptor, map, etc. * made GUIManager GHNS working again. * unittests++


TimeRevisionAuthorRepliesMessageWednesday, 23. August 2006
23:51576388staniek0Table View = displaying cell's selection background fixed for nonstandard font 2.0: merged
23:34576383staniek0- StyleProxy ported - warnings--
22:26576361staniek0Forms - display default values in the new row; default text is displayed with a special style (italic blue) as in table view; cancelling changes in a widget reverts its contents to default value (if available); the special style is removed as soon as the editing starts - "redo" action removed from the data-aware widgets' context menu (to avoid problems with data handling) Table Designer = fixed setting subtype by user 2.0: merged
22:23576359staniek0Table View - do not display default values for autonumbered fields - do not fill default values for already filled fields = boolean editor in tristate mode: fixed switching between true/false/NULL values - display default boolean values in blue - display tooltip with row number when the vertical scrollbar is dragged = fixed displaying tooltips for default values in the "insert" row 2.0: merged
20:53576338zander0More work on Lists
20:28576333jaham0move the path point selection info into the tool as suggested by Thomas
20:27576332jaham0move the path point selection info into the tool as suggested by Thomas
17:50576287jaham0- the path points now paint themself - added an IsSelected property to the path point -> clicking an a path point selects it -> only selected path points draw their control points -> added function to select/deselect all points of a path - sync last mouse after moving shape - fixed some warnings in the path tool