|
| This Week... |
|
The "Web Browser" Plasmoid now uses the Konqueror bookmarks system, and gains support for favicons. Improvements to the D-Bus interface of Plasma, with continued developments in the Plasmoids-on-Screensaver project. More animations in Plasma, and custom colors support in the "Binary Clock" and "Notes" applets. First implementation of a browserbar, and wheel scrolling support in the "Previewer" applet. A new "Screen Management" Plasmoid for showing connected displays. Start of a set of 40 Oxygen weather icons, for use in various Plasma applets and across KDE. Basic implementation of javascript window operations in WebKit-KDE. More work in DVCS support in KDevplatform (the basis of KDevelop4). Initial work on a Kross backend and automatic Qt bindings for Lua. A QtScript module for scripting C# applications. The start of a flashcard mode in Parley. Return of spellchecking support in the KPhotoAlbum annotation dialog, and improved spellchecking support in Kopete chat windows. "SystemC" syntax highlighting in Kate. Start of using Akonadi to store contacts in KPilot. A new "Albums" context view applet, and a "Script Console" script (using Qt bindings) in Amarok 2. Playback of DVD ISO's in DragonPlayer. Support for EMule collection files in KMLDonkey. Support for the NX protocol in KRDC. More page transition effects in KPresenter. Start of work on an authorization framework in the Kexi "Web Forms" project. A PackageKit backend plugin in the Shaman package manager. KDiff3 moves from playground/devtools to extragear/utils. Start of porting KMathTool to KDE 4. The KHotKeys redesign is merged back into kdebase. A new KWin branch for window tiling experimentation. Initial import of DarkRoom, an experimental raw decoder based on top of libkdcraw.
|
Amarok provides an update on the progress of its 5 Summer of Code projects:
|
Daniel Caleb Jones is working on biased and fuzzy playlists. In Amarok 1.4.x we had dynamic playlists, which allowed a user to have Amarok dynamically choose upcoming songs and add them to the playlist. However, this was built upon code which was not designed to be extended in such a way. In Amarok 2, we were able to refactor our codebase and provide a much more powerful and flexible playlist framework, and are now able to do some awesome things.
Daniel's first work encompassed the creation of biased playlists - that is, the ability to provide playlist track selection based on a defined set of metadata criteria. Biased playlists allow users to choose how Amarok selects upcoming songs from chosen playlists. For example, Amarok could select songs from a "5-star" smart playlist 60% of the time, and from a "4-star" smart playlist 40% of the time. Furthermore, with the addition of fuzzy biases, Amarok will be able to select matching tracks that are not strictly matching or non-matching against a certain criteria. These fuzzy biases let amarok mix a small amount of tracks that are just slightly not matching: in other words, it can choose tracks from a probability distribution centered around the selection criteria.
Basically, this system allows a user (through a simple and intuitive UI) to easily create custom unique dynamic playlists. This is way more powerful than the kind of thing that you can find in any other media player, and gives Amarok a significant advantage over its competition.
Alejandro Daniel Wainzinger is working on porting and improving the support for media devices in Amarok 2.0. Amarok 1.4 had support for media devices, but there were a few rough edges with configuration and speed of media device operations such as connect/disconnect and file management. The issue where an iPod with a large amount of songs locked Amarok on connect has been dealt with, and loading times are down to less than a second, not counting album covers. Now iPod covers can be not just set on the iPod, but read from the iPod to show in the collection browser. Basic file management (copying, deleting, editing tags) for iPod and MTP are good to go, though the interface needs touching up. The inability to play songs directly off of MTP devices is being dealt with soon.
Coming Soon: To be consistent with Amarok 2.0's theme of having transparent operations between different types of collections, sync'ing between media devices and other collections will be implemented. This is possible to implement only recently thanks to Jeff Mitchell's work on AFT. Instead of the media device tab from Amarok 1.4, an applet will be used to handle connect/disconnect, battery level, percent free space, and so forth.
Casey Link is integrating the MP3tunes.com service with Amarok. MP3Tunes provides an “online music space” for its users to store their entire music collection so they can listen to their music anywhere. Using the feature-rich API and client libraries, applications can be developed for just about any device, so you can literally listen to your music anywhere. This summer, Casey implemented full MP3Tunes functionality into Amarok so that users can seamlessly synchronize their local music collections with MP3Tunes. Not only are you be able to upload your local music collection, but when you add music to your MP3Tunes Locker via other methods, such as buying music from an online store, the new tracks will automatically be downloaded to your local collection.
Peter Zhou is doing the scripting interface project for Amarok. Unlike the interface in Amarok 1.x, the scripting in Amarok2 will be mainly powered by QtScript.
Considering the planned usability and convenience of the users' scripts, we've decided not to use the Python and Ruby bindings to reduce the extra dependencies of every single script. Using the Qt bindings, the scripts can use not only the rich Amarok API's, but also the Qt API's. As one of the highlight points of Amarok2, services can be also created by the scripts.
Amarok2 also provides a D-Bus interface for external programs. The D-Bus API's are based on MPRIS (Media Player Remote Interface Specification), which is sufficient enough to provide most of the control of the media player.
William Viana Soares is continuing the development of the Context View. He has brought some of the concepts being used in the Plasma desktop, like the zooming user interface and activities to Amarok's Context View. He is trying to polish things up as he continues improving the navigation system through the contextual information.
|
|
Albert Cervera i Areny introduces Kandau (and puts out a call for assistance):
|
Kandau is a backend-independent object persistency framework for Qt/KDE. This means developers write their classes (almost) like they're used to and properties and relations among them can be stored into an XML, .desktop file or SQL database. It can even create the appropiate tables with a single function call. It's built with extensibility in mind and you can easily create new backends.
One of the most useful features is that it not only stores objects, but also relations among them. Kandau provides a simple way of defining many-to-many, one-to-many and one-to-one relations between two classes and stores them for you. It even loads objects on demand from the SQL backend just by browsing your objects. In fact, it's built around a transaction-based manager that handles a cache in which objects are allocated and destroyed dynamically as needed, depending on the number of objects you allow it to keep in memory.
However, as you're nobody in the KDE world if you can't create some screenshots, Kandau also provides a GUI library which uses introspection for automatically creating dialogs to browse your data. Also, UI files can be used to load and store data in your Kandau classes. Here you can see a screenshot of an example application that uses the widgets that create the interface from your classes.
You can find more screenshots here
There's some documentation explaining most of its features here, so take a look at it and see how Kandau could help you. Or you may take a look at its API.
I think there are lots of applications inside and outside KDE that could benefit from such a framework and developers would save quite some time of boring coding, for saving and loading data. It's also quite useful having the GUI auto-generated because allows developers to feed their applications with data without the need of actually creating a GUI, for example.
Unfortunately, I don't have the time to work on Kandau, indeed my work was discontinued some time ago and it's still using Qt/KDE 3. Recently, Percy Camilo Triveño Aucahuasi (from KColorEdit) made the core library build with Qt4 but there's still work to be done and he hasn't got the time either. It'd be great if somebody would like to pick up mantainership, update KandauUI to work with Qt4, and create a KandauModel to work with the model/view framework, for example.
So feel free to contact me at albert at nan-tic dot com if you're willing to work on it or simply want more information on its architecture.
|
|
|
| Statistics |
|
Commits: |
3534
by 263
developers, 11162
lines modified, 1649
new files. |
|
Open Bugs:
|
16605
|
|
Open Wishes:
|
14275
|
|
Bugs Opened:
|
555 in the last 7 days. |
|
Bugs Closed:
|
396 in the last 7 days. |
|
|
Commit Summary |
|
Module
|
Commits
|
|
/trunk/l10n-kde4 |
|
|
/trunk/KDE |
|
|
/branches/KDE |
|
|
/trunk/playground |
|
|
/branches/kdepim |
|
|
/trunk/extragear |
|
|
/trunk/koffice |
|
|
/branches/work |
|
|
/trunk/www |
|
|
/trunk/kdesupport |
|
|
|
Lines
|
Developer
|
Commits
|
|
|
Thomas McGuire
|
|
|
|
Allen Winter
|
|
|
|
Laurent Montel
|
|
|
|
Matt Rogers
|
|
|
|
Chusslove Illich
|
|
|
|
Volker Krause
|
|
|
|
Cyrille Berger
|
|
|
|
Matthias Kretz
|
|
|
|
Szymon Stefanek
|
|
|
|
Aaron J. Seigo
|
|
|
|
|
Internationalisation (i18n) Status
|
|
|
Bug Killers and Buzz |
|
Bug Killer
|
Number Of Bugs Closed
|
|
Aaron J. Seigo
|
|
|
Thomas McGuire
|
|
|
Peter Penz
|
|
|
Christophe Giboudeaux
|
|
|
A. Spehr
|
|
|
Pino Toscano
|
|
|
Mark Kretschmann
|
|
|
Robert Knight
|
|
|
Lukas Appelhans
|
|
|
Allen Winter
|
|
|
|
Program |
Buzz |
|
Amarok |
|
9815 |
|
|
K3B |
|
4875 |
|
|
KMail |
|
4840 |
|
|
Kopete |
|
3320 |
|
|
KDevelop |
|
|
Plasma |
|
|
Kaffeine |
|
|
Kate |
|
|
Solid |
|
|
Kontact |
|
|
|
Person |
Buzz |
|
David Faure
|
|
2110
|
|
|
Stephan Kulow
|
|
1749
|
|
|
Aaron Seigo
|
|
1390
|
|
|
Torsten Rahn
|
|
1367
|
|
|
Jonathan Riddell
|
|
1132
|
|
|
Laurent Montel
|
|
1030
|
|
|
Stephan Binner
|
|
782
|
|
|
Thiago Macieira
|
|
668
|
|
|
Zack Rusin
|
|
|
Adriaan de Groot
|
|
|
|
|
|
| Contents |
|
|
Bug Fixes |
Features |
Optimise |
Security |
Other |
|
Accessibility |
|
|
|
|
|
|
Development Tools |
|
|
|
|
|
|
Educational |
|
|
|
|
|
|
Graphics |
|
|
|
|
|
|
KDE-Base |
|
|
|
|
|
|
KDE-PIM |
|
|
|
|
|
|
Office |
|
|
|
|
|
|
Konqueror |
|
|
|
|
|
|
Multimedia |
|
|
|
|
|
|
Networking Tools |
|
|
|
|
|
|
User Interface |
|
|
|
|
|
|
Utilities |
|
|
|
|
|
|
Games |
|
|
|
|
|
|
Other |
|
|
|
|
|
|
There are 172 selections this week.
|
|
Bug Fixes |
|
Educational |
|
Jason Harris committed a change to /trunk/KDE/kdeedu/kstars/kstars/skymap.cpp:
|
Fixing the bug that was causing runaway CPU usage when drawn lines moved off-screen at high zoom. Akarsh submitted a patch which fixed the symptom, but had some side effects. This patch reverts his change and corrects a very simple bug that was causing the problem.
We have a function SkyMap::onScreenLine( QPointF &p1, QPointF &p2 ) that returns the o-screen segment of a line, given endpoints p1 and p2. The modified on-screen endpoints are returned through the same reference values. In the case that no segment of the line connecting p1 and p2 passes through the SkyMap widget, null values of -10000000 are returned.
At some point, IIRC, James created an alternative onscreenLine2() function which does essentially the same thing, except it returns a bool instead of void, and the return value is used to indicate whether an onscreen line segment was found or not. There are no null values in this version of the function.
However, in one of the places where onscreenLine2() should have returned false (indicating the line was totally off-screen with no onscreen segment), it actually returned true. This was the cause of the runaway CPU usage.
James, can you confirm this patch, and then I will backport it to the 4.1 branch.
I don't remember why James created a second version of this function rather than modifying onscreenLine(). |
|
|
|
|
|
|
Torsten Rahn committed changes in /branches/KDE/4.1/kdeedu/marble/src/lib:
|
- Addressing the symptoms of the rendering desync. This fixes the apparent visual desync problems.
However the way it gets done isn't satisfying and still has some less apparent bugs remaining, so I'll fix the issue entirely in a different way in my next commit. |
|
|
|
|
|
|
|
|
|
|
David Faure committed a change to /trunk/KDE/kdelibs/khtml/khtml_part.cpp:
|
Don't create the view* and save* actions when not in "browser" mode. Only khtml_browser.rc lists those actions anyway, and this fixes Ctrl+U triggering "show html source" in kmail unwantedly. |
|
|
|
|
|
|
Dominik Haumann committed changes in /trunk/KDE/kdelibs/kate:
|
Changes include - the undo cursor and undo selection are saved with the first KateDocument::editStart() - the redo cursor and redo selection are saved with the last KateDocument::editEnd()
For this to work, the following parts needed to be fixed - wrap doDeleteWordLeft/Right with editStart() and editEnd(), otherwise wordLeft() and wordRight() change the cursor position - drag & drop: while moving the cursor position is changed. To make it work, we need to fix the cursor position before we really do the drop operation. The undo selection is only set in the case of a move, otherwise it does not make sense (it feels really strange)
Please test the patch, as it should go into KDE 4.1 if it works. Right now, the KDE 4.1 branch crashes sometimes when doing undo/redo.
Port to KDE 4.1.0? Wait for KDE 4.1.1? dunno... |
|
|
|
|
|
|
KDE-PIM |
|
Thomas McGuire committed a change to /trunk/KDE/kdepim/kontact/src/mainwindow.cpp:
|
Also quit the event loop when a part still holds a reference to a KGlobal object, probably by a KJob which wasn't properly deleted.
This fixes that the Kontact process was still running (without a window) after viewing an article in Akregator.
The real fix would be to fix Akregator, it should take care of properly cleaning up each KJob.
@David: maybe KJob::allowQuitWhileJobsAreRunning(bool) wasn't such a bad idea after all? |
|
|
|
|
|
|
Multimedia |
|
Michael Pyne committed changes in /trunk/KDE/kdemultimedia/juk:
|
Be more careful in the event of a encrypted APIC frame (or compressed with no zlib in TagLib).
Code was put in to check for this earlier but was incomplete as although the pointer was correct it was still blindly dereferenced.
In addition add QImage header to coverinfo.h (was included indirectly before).
Neither of these need backported based on the rarity of encrypted/uncompressed picture frames but I'll probably backport the potential crash fix just in case. |
|
|
|
|
|
|
Daniel Caleb Jones committed changes in /trunk/extragear/multimedia/amarok/src:
|
Fixed crash when off is pushed while repopulating. Fixed bad things happening when repopulate is pushed a bunch of times. More reasonable results for infeasible sytems (i.e. you can put two artists at 100% and get about 50% of both). |
|
|
|
|
|
|
|
|
Features |
|
|
|
|
|
|
|
David Nolden committed a change to /trunk/KDE/kdevplatform/language/duchain/duchain.cpp:
|
Store and load all the repositories from/to disk. A unique directory for each kdevelop instance is used, so we don't get crashes from that.
What we store to disk now: - Identifiers - Macros - 99% of environment management information(macro sets, string sets, etc.) - Qualified identifiers - Strings - The complete type-system - Comments - C++ template instantiation information
The only missing part is the actual declarations/contexts, and the top-level environment management information. |
|
|
|
|
|
|
David Nolden committed changes in /trunk/KDE/kdevplatform/language/duchain:
|
Make the whole repository storing actually work, and add a lot of optional debugging code to itemrepository.h.
Pick the storage-directory from within itemrepository.cpp instead of duchain.cpp.
It is important that the repositories need to be loaded at the earlierst point possible. The problem: At that early point, neither KStandardDirs, nor KLockFile are functioncal, because KComponentData doesn't exist yet.
So for now we just store the data in ~/.kdevduchain, and don't support multiple running instances of kdevelop until we find another solution for the locking problem. |
|
|
|
|
|
|
Hamish Rodda committed changes in /trunk/KDE/kdevplatform/language/duchain/types:
|
Adding / comitting changes to files not yet added to the build system.
I was wrong earlier, c++ makes extensive use of ConstantIntegralType seperate from EnumeratorType, so separate them again.
The full patch looks like it's going to take a while to get right, and with the big changes happening in duchain persistance, I've decided to put them on the backburner for the moment. |
|
|
|
|
|
|
|
|
Educational |
|
Wouter Haffmans committed changes in /trunk/playground/edu/kiddraw/src:
|
Implement loading of KidDraw files. Drawings can now be saved and reloaded again! Previously saved files (within the past 24 hours or so) are incompatible due to a minor change in the drawingwriter. |
|
|
|
|
|
|
David Capel committed changes in /branches/work/soc-parley/parley/src:
|
Starting flashcard mode.
Some work on integrating parley and parleypractice: PPMainWindow now can have a KEduVocDocument in its ctor Parley listens for signalPracticeFinished to reopen -- crash bug partially fixed. I'll work on it more. New generic button class.
Small theme_reqs change |
|
|
|
|
|
|
|
|
|
|
|
|
Games |
|
Stefan Majewsky committed changes in /trunk/playground/games/palapeli/game:
|
Silently lay the foundations for pattern plugins with a complete redesign of the Pattern classes. * The old base class Pattern is split into two parts: PatternConfiguration specifies handles the whole user interaction part (values which have to be entered by the user or are read from and written to savegames). (This class will later be the one that is created by KPluginFactory.) When enough information has been given, the PatternConfiguration object creates a Pattern object which actually does the work of splitting the image into pieces with neighbor relations.
* The Pattern and PatternConfiguration base classes offer full abstraction of the features offered by Palapeli core. For example, the new implementation of the rectangular pattern in pattern-rectangular.[cpp|h] does only depend on pattern.h and pattern-configuration.h.
IMPORTANT: The new code is not yet included in the build. Actually, this is more or less a brain dump of this evening's work. (That means I do not even know if it compiles, but I'm quite sure that the logic is correct.) To get it to work, I need to do a new "New game" dialog and modify the Manager class a bit. Wow, I'm quite talkative today... |
|
|
|
|
|
|
Stefan Majewsky committed changes in /trunk/playground/games/palapeli/game:
|
Make new library libpalapelicore which includes all functionality of Palapeli. The application palapeli itself does now only consist of a single code file which creates a KApplication and passes control to libpalapelicore. |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Chani Armitage committed changes in /trunk/KDE/kdebase/workspace/krunner/lock:
|
a dbus interface for plasma, and proper stacking plasma can be shown and hidden and quit, we can get the winid, and we can make it stay above the screensaver. yay! :)
now if only I knew why I got kcheckpass working on my laptop but couldn't do the same on my desktop... |
|
|
|
|
|
|
Chani Armitage committed changes in /trunk/KDE/kdebase/workspace/krunner/lock:
|
make mouse & keyboard input work. the code for this is a bit ugly in places.
there's one bug: if a config dialog is closed and focus reverts to another config dialog, the only keyboard things that work are esc and enter. |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Marco Martin committed changes in /trunk/playground/base/plasma/widgets/tabbar:
|
support for mouseover animation for buttons (will be disabled in the tabbar)
new themes for the normal button and the mouseover pressed still missing new hints in the tabbar svg will be useful for experiments in PanelSvg |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Szymon Stefanek committed changes in /branches/kdepim/kmail-soc/kmail:
|
Allow switching the message sort order also by clicking on columns. The simple default skin is now 100% backward compatible. Promote some widely used functions to their own ComboBoxUtils namespace.
Also explicit a couple of message sorting options which weren't totally clear since depending on the ThreadLeader setting. |
|
|
|
|
|
|
|
|
|
|
Thomas McGuire committed a change to /trunk/KDE/kdepim/kmail/templateparser.cpp:
|
Also allow the german leading quote mark ( Unicode character 0x201C ) as a separator for template quotes. This fixes replies and forwards being empty when running under German locale.
Thanks to Friedrich Kröner for figuring out that this bug was locale-dependent (no wonder we developers couldn't reproduce it).
When translating the templates of KMail, please only use U+0022 or U+201C to translate the quotation mark, otherwise you'll trigger this bug. Simply using the ASCII " character is the best. |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Alejandro Daniel Wainzinger committed changes in /trunk/extragear/multimedia/amarok/src:
|
CollectionFactory classes should be in charge of dealing with Collection classes, and nothing else.
Introducing the MediaDeviceMonitor, another layer on top the MediaDeviceCache and Solid, which takes care of lower-level hardware detection, connection, and disconnection.
Media Device Collection classes plug into it and seamlessly get the needed information to build Collections, without doing dirty work. See IpodCollection for reference implementation. |
|
|
|
|
|
|
Ian Monroe committed changes in /trunk/extragear/multimedia/amarok/src/scripts:
|
a working start of a "Script Console" script using the Qt Bindings. Basically its speedcrunch/irb for the amarok scripting system. Requires you uncomment the generator at the top of amarok/CMakeLists.txt |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Matt Rogers committed changes in /branches/work/kopete/dev-4.2/kopete:
|
Add a contact list model class for the contact list. Make it handle the signals that are sent for the addition of groups and metacontacts. (It will handle a few more things in the future).
Provide the current tree view in use with an instance of the new model so we actually have some sort of contact list to work with. |
|
|
|
|
|
|
|
|
|
|
Matt Rogers committed changes in /branches/work/kopete/dev-4.2/kopete/contactlist:
|
Allow items to be dragged and to be drop targets.
This is the start of drag and drop support for the new contact list. In order to enable drag and drop, we need to allow the model to tell the view that we have draggable items. |
|
|
|
|
|
|
|
|
|
|
Sebastian Sauer committed changes in /trunk/extragear/network/kmldonkey/kmldonkey:
|
Added support for emulecollection files.
Gioacchino: Forgive me for not using the work done on Kollektion, but its imho much easier to just implement that functionality direct in KMLDonkey using C++.
Though for creating such *emulecollection-files Kollektion is still the better way (as in KMLDonkey doesn't support it ;) |
|
|
|
|
|
|
|
|
|
|
Thomas Braun committed changes in /trunk/extragear/office/kile/src:
|
reworked toolbar for the tools, now we got 4 buttons in the toolbar from these one can select all possible tools they are grouped as compile, view, convert, quick
the crash in the config dialog is not related to this commmit (I already checked) |
|
|
|
|
|
|
|
|
|
|
|
|
Pierre Ducroquet committed a change to /trunk/koffice/libs/kotext/KoTextShapeData.cpp:
|
Here is a first draft code to save lists. I don't think I can write any simpler code for this task. It supports nested lists, but I still have to test a few things...
And it doesn't save the list styles, what a shame for something that took less than 45 minutes to be written... :)
(BTW, personnal message to dipesh who will read this : as you can see, I'm not sleeping ;) ) |
|
|
|
|
|
|
Pierre Ducroquet committed changes in /trunk/koffice/libs/kotext:
|
Ok, now, we have ***basic*** support for saving list styles... I said basic.
Well, primitive would perhaps be more adequate... Anyway, it's working, that's what matters. |
|
|
|
|
|
|
|
|
|
|
Florian Merz committed changes in /trunk/koffice/plugins/textshape:
|
Implement support for paragraphs which span more than one shape (actually 1 or 2 shapes right now, but larger numbers should now be easy to do).
This introduced a few small bugs which need to be fixed, but the tool should still be stable and usable. |
|
|
|
|
|
|
Other |
|
Gael Guennebaud committed changes in /branches/work/eigen2/bench/btl:
|
enhancements of the plot generator: - removed the ugly X11 and PNG gnuplots terminals - use enhanced postscript terminal - use imagemagick to generate the png files (with compression) - disable the fortran impl by default since it is as meaningless as a "C impl" - update line settings |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Optimise |
|
Development Tools |
|
Hamish Rodda committed changes in /trunk/KDE/kdevplatform/plugins/standardoutputview:
|
I couldn't stand the terrible performance of the standard output view any longer.
Change the scrolling to bottom feature to use a delay (100ms chosen to start with)
Alleviates the worst of the lockups (eg, 20s lockup on my quad core2 machine) during output view spamming, but makes it look a bit strange.
Definitive solution still required. |
|
|
|
|
|
|
|
|
|
|
Other |
|
Gael Guennebaud committed changes in /branches/work/eigen2:
|
Optimization: added super efficient rowmajor * vector product (and vector * colmajor). It basically performs 4 dot products at once reducing loads of the vector and improving instructions scheduling.
With 3 cache friendly algorithms, we now handle all product configurations with outstanding perf for large matrices. |
|
|
|
|
|
|
Other |
|
Development Tools |
|
Joachim Eibl committed changes in /trunk:
|
|
Moved kdiff3 from playground/devtools to extragear/utils. |
|
|
|
|
|
|
|
|
Jakob Petsovits committed a change to /trunk/KDE/kdesdk/kate/plugins/CMakeLists.txt:
|
trunk is 4.2, feature freeze is over, which means I can now re-enable the snippets plugin that I ported some time ago.
Also include the helloworld plugin in the CMakeLists.txt (but don't build it), so that it won't be so easily forgotten when further interface changes happen. |
|
|
|
|
|
|
|
|
|
|
|
|
Akarsh Simha committed changes in /trunk/KDE/kdeedu/kstars/kstars:
|
Merging the summer branch into trunk.
This means that:
* KStars now has 2.5 million stars * There's the coordinate grid clipping trouble at high zoom (so it might be a good idea to disable the coordinate grid) |
|
|
|
|
|
|
|
|
|
|
|
|
Stefan Majewsky committed changes in /trunk/playground/games/palapeli:
|
Who needs atomic commits? Here comes the GIGANTIC commit!!! Hopefully, this is the last commit in the very long series of "Let's break the total internal infrastructure of Palapeli" commits. (To be honest, I doubt.)
Here are the changes:
* Rename libpalapelicore to libpalapelipattern because everything else is moved back to Palapeli application. This is possible because the Pattern class gets a proper interface to the Palapeli game engine. * The Palapeli::Pattern::slice function is renamed to Palapeli::Pattern::doSlice, because there is now another non-virtual slice function which does some initialisation in the base class. * The piece positions are now loaded or randomized in the Palapeli::Pattern class.* Move libpalapelipattern to a separate directory. Do only include this folder in API documentation. |
|
|
|
|
|
|
|
|
|
|
Graphics |
|
Jesper Pedersen committed changes in /trunk/extragear/graphics/kphotoalbum/MainWindow:
|
|
OK enough time wasted on this statistics dialog, it will likely go away again, but lets keep it around for when we work on the DB backend, as it might be useful for checking that categories and subcategories are converted correctly between backends. |
|
|
|
|
|
|
Jan Kundrát committed changes in /trunk/extragear/graphics/kphotoalbum:
|
Preload image thumbnails from images in the thumbnail viewer even if they aren't visible yet.
Preloading starts as soon as the ThumbnailWidget knows what images it contains and continues until another bunch of images is to be shown.
This behavior is open for discussion, perhaps it should stop when ThumbnailWidget goes out of (visible) place, perhaps it should only load images that aren't too far from current viewport etc. |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
André Wöbbeking committed changes in /branches/KDE/4.1/kdebase/workspace/kcontrol/keys/schemes:
|
adapt keys
Bar -> | Equal -> = Minus -> - Plus -> + Underscore -> _
and actions
Enable/Disable Clipboard Actions -> clipboard_action Logout -> Log Out Logout without Confirmation -> Log Out Without Confirmation Manually Invoke Action on Current Clipboard -> repeat_action Show Klipper Popup -> show_klipper_popup Window Iconify -> Window Minimize
The logout shortcuts also don't work in KDE 3.5. |
|
|
|
|
|
|
|
|
Chani Armitage committed changes in /trunk/KDE/kdebase/workspace/krunner/lock:
|
merge in groundwork branch this creates a plasma-overlay app with its own corona and view
it'll start up fine and load a default desktop containment + folderview however, once it's hidden you can't bring it back - you have to kquitapp plasma-overlay the lockprocess will run plasma-overlay when the screensaver starts, but also fails to stop it, and it gets created *under* the screensaver
I'm sure there are a bazillion other bugs - but it does run, and you can add widgets. :) |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Dirk Mueller committed changes in /trunk/KDE/kdebase/workspace:
|
introduce plugin versioning for plasma and ensure that only applets with a matching major.minor version are loaded.
needs porting to plugins other than applets. |
|
|
|
|
|
|
|
|
|
|
Pau Garcia i Quiles committed changes in /trunk/KDE/kdeplasma-addons/desktopthemes/heron:
|
Rename 'license' file to 'LICENSE2'. Please, never have files with the same filename but different case, that won't work in case-insensitive filesystems (for instance, on Windows).
BTW, I wonder if I can just remove LICENSE2, its contents are the same than those of LICENSE |
|
|
|
|
|
|
Aaron J. Seigo committed a change to /trunk/KDE/kdebase/workspace/plasma/applets/pager/pager.cpp:
|
two different update timeouts: a fast one for simple events, and a slower one for expensive ones. we use the latter for infrequent events and for ones that get repeated a lot (particularly moving windows around).
we use the fast timeout for new windows, closing windows, switching desktops ... results in a much, much more snappy feeling pager without compromising cpu usage for window movement (the original reason i introduced the delay ..) |
|
|
|
|
|
|
|
|
|
|
Chani Armitage committed changes in /trunk/KDE/kdebase/workspace:
|
only allow configuration while widgets are unlocked something went wrong, very wrong, when pulling from gitorious.
I don't know what or why. but this is a copy of the last commit I made on my laptop... |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Stefan Nikolaus committed changes in /trunk/koffice:
|
Interfaces KoChart::ChartInterface
Detach KSpread and the chart shape from the bridging lib 'koshape'. :)
Why the KChart KPart still needs this lib, I cannot explain. The embedding in other KOffice apps is done by using the flake library.
Maybe for embedding in external apps? Either way, that's up to the KChart developers. |
|
|
|
|
|
|
|
|
|
|
|
|
Jakob Petsovits committed changes in /trunk/KDE:
|
Move pinheiro's last-minute KJots app icon out of core Oxygen and into KJots itself.
Someone (i.e. not me) needs to backport this to 4.1, as we don't want application icons in core Oxygen for various reasons. |
|
|
|
|
|
|
|
|
|