|
| This Week... |
|
Further XMP tag support in Digikam. Beginnings of a Plasma lock/logout applet and a weather applet, to display data from the existing weather data engine. Continued work on the new Plasma-based KNewsTicker applet. Continued work and development ideas in Parley. More various developments and optimisations in KHTML. Jamendo album download support in Amarok 2.0. Interface usability work in Kopete. Shredding functionality removed from KGPG. Fixes in the Sonnet spellchecking solution for KDE 4. Work on the foundations of KChart 2 in KOffice. Lancelot and Raptor develop as alternatives to the Kickoff menu of KDE 4.0. Okteta, a planned successor to KHexEdit, uploaded to playground/utils. KOffice 1.9.95 and KDE 4.0 Beta 4/KDE Platform RC1 are tagged for future release.
|
Ivan Čukić introduces his work on Lancelot, a menu alternative based on Plasma:
|
Lancelot is an experimental launcher menu designed to provide a place from which all your jobs begin. It provides quick access to applications, places, documents, contacts and system information.
The word "experimental" stands for a couple of things:
- The (optional) no-click interface that allows you to navigate through the menu, and start any application just by moving the mouse - without making a single click.
- Arranging the items in the menu depending on the position of the menu button which activates it.
Existing "menu replacements" have, in my opinion, one common weakness - in order to make some things more accessible, sacrifices had to be made. One example would be Kickoff's way of browsing non-favorite applications, which is too slow and hides the "whole picture" from the user by not showing the list of all categories at all times. On the other hand, Tasty Menu does the job of application browsing very well, but it is too large and not well-thought through.
Lancelot is trying to make a compromise between the two approaches - to be relatively small, and to have unimpaired accessibility.
Currently, the vast majority of GUI work is finished. The only thing left to develop is the Passageway component, which will be the responsible for application browsing.
Apart from that, data models need to be made. Currently only models for the "Computer" category are made and can be tested. I hope to finish the basic functions by the time KDE 4.0 appears.
After that, special features will roll in - ability to add Plasma applets into the menu itself, adding parts of the menu to the Plasma desktop, composite effects, and more.
Like the most things related to Plasma, I intend to make Lancelot just-work(TM) for 4.0, and to make it Work-As-I-Think-It-Should(TM) for 4.1.
|
|
Jasem Mutlaq writes about his experiences integrating KStars as the control mechanism for a telescope at an observatory in Kuwait:
|
I've been working on Kuwait Science Club 0.5m telescope for the past few months and what an interesting project it is! It is one of these projects where you have to dwell deep in many disciplines to get things right.
The observatory houses a Swiss-made equatorial fork telescope with two OTA each 0.5m in diameter. It was used to do pretty interesting research up until a few years ago when new cities were build close by and now light pollution degraded its abilities drastically.
Since its commision, it's always been in manual control where the observer uses a keypad to move around in the sky. This is obviously a tedious process but for the few bright objects in the sky. Furthermore, the Swiss company who delivered the system with a computer never really got it to work, or so I'm told since the computer was thrown away long time ago.
What was particularly irritating is the complete lack of documentation. No one knew anything about the telescope save for the diameter of the mirrors and its commission date, it seems that everything else got lost somehow. So this is when I had to 'reverse engineer' the system starting from the keypad up to the servo motors..etc. I had to install new 12 bit absolute encoders for the RA and DEC axis, and design a PCB to act as a 'digital' keypad using optocouplers primary.
Then by using NI 6905 Digital I/O card, which is well supported under Linux, I controlled the new board, and received the encoder data. Then all I had to do was to write the INDI driver to achieve a closed-loop control.
Suffice to say, the hardware took the most time, mostly because I had no docs and I had to find how the electronics work out on a deep level. It was all new to me. Also, since this is a really heavy telescope (I was told 9 tons, but I really doubt it), safety is quite important.
There are minimum allowable altitude safety guards in the driver, auto and emergency park in case something goes wrong, and a watchdog that shuts the telescope even if the whole operating system crashes. But would you bet your life on software alone?
There are also two hardware limits switches for each axis that shuts the telescope down if it goes too far, and if all that fails, I installed an emergency stop button connected directly to mains. If THAT fails, then you probably deserve to die anyway.
The last tricky part of the whole deal was pointing performance. Since we're dealing with a really old mechanical system, the pointing accuracy is limited by numerous factors including index and collimation errors, fork flexture, and a lot more. We had to use TPoint to build a pointing model for the telescope to bring the pointing accuracy down. This is still in progress, but the overall control system is done.
Here is a video showing the telescope and KStars. I used my digital camera so I apologize if it's not clear enough.
|
|
|
| Statistics |
|
| Contents |
|
|
Bug Fixes |
Features |
Optimise |
Security |
Other |
|
Accessibility |
 |
|
|
|
|
|
Development Tools |
![[*]](/images/standard/bugfixes) |
![[*]](/images/standard/features) |
|
|
![[*]](/images/standard/other) |
|
Educational |
|
|
|
|
|
|
Graphics |
|
|
|
|
|
|
KDE-Base |
|
|
![[*]](/images/standard/optimise) |
|
|
|
KDE-PIM |
|
|
|
|
|
|
Office |
|
|
|
|
|
|
Konqueror |
|
|
|
|
|
|
Multimedia |
|
|
|
|
|
|
Networking Tools |
|
|
|
|
|
|
User Interface |
|
|
|
|
|
|
Utilities |
|
|
|
|
|
|
Games |
|
|
|
|
|
|
Other |
|
|
|
|
|
|
|
Bug Fixes |
|
Development Tools |
|
Oswald Buddenhagen committed changes in /branches/KDE/3.5/kdevelop/lib/widgets:
|
stop chopping up message lines from compiler, etc.
note that lines that were previously chopped up will now come in out of order. while this "regression" could be fixed, there is no guarantee at all that the complete lines arrive in order in the first place, so i see relatively little point in bothering with this. |
|
|
|
|
|
|
Andreas Pakulat committed a change to /branches/KDE/3.5/kdevelop/languages/cpp/cppsupportpart.cpp:
|
This should fix the lag that could be experienced, specially on larger projects. As David pointed out the updatedSourceInfo signal is more like a "the whole code-model changed" thing.
Everybody who experiences short delays when typing normal code, please svn up and test. I couldn't get the delay to work here the last days so I'm not 100% sure it fixes this problem (but I'm pretty confident) |
|
|
|
|
|
|
Educational |
|
Jason Harris committed a change to /trunk/KDE/kdeedu/kstars/kstars/skymapdraw.cpp:
|
Jure: thanks for finding a memory leak in KStars!
Unfortunately, your fix was not correct, because we set "pLast = pThis" prior to your "delete pThis". Therefore, pLast would be a null pointer when we still need it.
Instead, I have made pThis and pLast normal objects, rather than pointers. James, can you review this patch? I am in the midst of a large set of changes ATM, so my local copy is not compilable. |
|
|
|
|
|
|
|
|
Graphics |
|
Albert Astals Cid committed changes in /branches/KDE/3.5/kdegraphics/kpdf/xpdf:
|
Splash rework, check if font is inside clip area before rendering it to a temporary bitmap. Fixes bug 150693
This change is not trivial. What i did is:
It is getGlyph the one that does the intersection between clip area and rendering area of the font instead fillGlyph2
That means some clipRes = state->clip->testRect but we win more robustness against broken pdf that specify HUGE fonts |
|
|
|
|
|
|
|
|
KDE-Base |
|
Maksim Orlovich committed a change to /trunk/KDE/kdelibs/kparts/browserrun.cpp:
|
|
Unbreak redirections in konq, at least on form submit (affects rafb.net/paste), and who knows what else. Corrects clearly wrong code from a "cleanup" commit (and people wonder why I hate those?) |
|
|
|
|
|
|
|
|
|
|
Maksim Orlovich committed changes in /branches/KDE/3.5/kdelibs/khtml/ecma:
|
Provide body.focus, for comments on youtube.
Reported by a developer whose name is withheld for fear of the damage to their reputation that the revelation of them reading youtube comments may bring. |
|
|
|
|
|
|
Germain Garand committed changes in /trunk/KDE/kdelibs/khtml:
|
- fix widget event handling, then
- enable redirection of iframes so we get correct z-order
- propagate wheel event to widgets (151160)
yeah, I know it's late, I've spend the damn night on that thing.
Nevertheless, it's important.
Besides, its time to realize the king is naked!
Give us another beta! |
|
|
|
|
|
|
|
|
|
|
Peter Penz committed changes in /trunk/KDE/kdebase/apps/dolphin/src:
|
due to the recent tagging freeze this commit contains several fixes in one:
* the filterbar now also filters directories (works also in the column-view :-))
* The "Additional Information" menu entry now also works for showing/hiding the columns of the details view. This also implies that the columns for the details view can now be adjusted per directory -> we have now a consistent behavior between the icons view and details view. Still open: the view properties dialog must be fixed
* Don't show a "Nepomuk not available" error message when starting Dolphin and Nepomuk is not available.
* Fix issue that the information panel blocked the application because of parsing the full meta data of a huge file. |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Joris Guisson committed a change to /trunk/extragear/network/ktorrent/ktorrent/generalpref.ui:
|
|
Fix typo, and make sure tooltip for encryption option is displayed properly, pino's attempt to cleanup useless html formating, seems to result in tooltips wider then my screen. It seems that Qt won't break it up over multiple lines when there is no html stuff around it. |
|
|
|
|
|
|
|
|
|
|
Features |
|
Development Tools |
|
Anders Lund committed changes in /trunk/KDE/kdesdk/kate/app:
|
* Make the file list configurable, add a config dialog page.
- turn shading on/off
- select shading colors (defaults taken from the color schems, only saved when differing)
- select sorting (doesnt do anything yet, that is a bug)
* Indicate that the file list should be sortable (I know how to do this, but didn't finish it yet) |
|
|
|
|
|
|
|
|
Graphics |
|
Clarence Dang committed changes in /trunk/KDE/kdegraphics/kolourpaint:
|
Alphabetize and update mime information for KDE 4.0.
In terms of saving:
1. TIFF:
a) Saving a TIFF to a QByteArray does not work (bug reported to Qt).
In pratical terms, KolourPaint's saving preview does not work.
b) Saving a QImage as a TIFF file always outputs TIFFs that are read back in by QImage as 24-bit (hence this commit adds no special mime info for TIFF).
2. Saving to video/x-mng does not work. |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Marcel Wiesweg committed changes in /trunk/extragear/graphics/digikam/libs/database:
|
Restore full functionality of ImageInfo.
Consequently cache all fields (including rating, tags of course), use lazy loading everywhere.
Remove tags paths method, code outside libs/database may use the AlbumManager with its cache please, it's faster. Only supply (cached) tag ids.
Remove unused constructor.
Do not cache DatabaseUrl, but construct on demand from ids and image name (album name is cached in ImageInfoCache, TODO change notification, album root is retrieved from CollectionManager)
Add method to retrieve ImageComments object for sophisticated comments storage.
TODO: copy method, change notification |
|
|
|
|
|
|
|
|
KDE-Base |
|
Hamish Rodda committed a change to /trunk/KDE/kdelibs/kate/view/kateview.cpp:
|
All of these actions are specific to the view, and when you have two views open side-by-side, the shortcuts become ambiguous because they were window-level shortcuts. Make them widget level shortcuts.
This patch depends upon the patch I sent to kde-core-devel, because of the position of the call to setAssociatedWidget. Hopefully I will commit that patch soon if there are no objections. |
|
|
|
|
|
|
|
|
Fredrik Höglund committed changes in /trunk/KDE/kdebase/workspace/plasma/applets/tasks:
|
Refactor the task button text layout and rendering code to handle text that won't fit in the available space better.
* Elide lines that won't fit by fading the text out at the end.
* Don't extend the button labels above and below the buttons when there are more lines than will fit.
* Handle reversed layout correctly within the buttons. |
|
|
|
|
|
|
|
|
|
|
|
|
Kévin Ottens committed a change to /trunk/KDE/kdelibs/kfile/kfileplacesview.cpp:
|
Try another animation for appearing items. It doesn't generate lot of icon resizing as before, I find it much easier on the eyes (doesn't skip lots of frame like before).
Woot! I'm exploding my commit record from the last few weeks! Looks like I'm finally back on business. :-) |
|
|
|
|
|
|
|
|
Robert Knight committed a change to /trunk/KDE/kdebase/workspace/plasma/applets/kickoff/ui/itemdelegate.cpp:
|
Revert back to the previous hover/selected indicator for items: Draw the item background using the palette's highlight color and the text using the palette's highlighted text color.
The use of bold made it difficult for me to spot which item was selected, particular when manipulating items with the context menu. |
|
|
|
|
|
|
Matthew Woehlke committed changes in /trunk/KDE/kdebase/workspace/kcontrol/colors:
|
New class to hold windeco colors (like KColorScheme holds scheme colors)... this fixes a design flaw where loading WM colors didn't work.
TODO: All settings seem to have this problem, which is settings must be stored somewhere other than the UI, due to how we load from configs and update the UI. Thus contrast and 'shade sorted column' are probably broken the same way. Some placeholder comments have been added. |
|
|
|
|
|
|
Aaron J. Seigo committed changes in /trunk/KDE/kdebase/workspace/plasma/containments:
|
put the containments into the hidden plugin group so that, well, they don't show up =)
Hidden=true doesn't work as that screws with actually loading them the same way we do other applets, and i'd rather not special case that code. this is a small hack, but it works nicely and allows me to keep the rest of the code generic and uncomplicated |
|
|
|
|
|
|
Aaron J. Seigo committed changes in /trunk/KDE/kdebase/workspace/libs/plasma:
|
animate items being removed. whee. the only problem would be if the item is being removed but plasma is quit before the anim finishes... or if some top-level applet calls disappear on itself (though why they'd do that i don't?)
let's see if that ever really matters. if it does, then we can simply keep a list of applets being removed around. but let's play fast 'n loose for now. |
|
|
|
|
|
|
Maksim Orlovich committed changes in /trunk/KDE/kdebase/apps/nsplugins:
|
Make the nsplugin support actually do something:
- Actually build a kpart and install as is.
- Actually build a working kcminit module. Which might not be such a hot idea since KConfig seems to like to loop infinitely on this stuff
- Actually let nspluginviewer start
Right now it's lost someplace in d-bus communication...
And we didn't deal with the Xt issue yet. Oh boy. |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
KDE-PIM |
|
David Jarvie committed changes in /trunk/KDE/kdepim/kalarm:
|
|
Add D-Bus interfaces and command line options for new alarm edit dialog invocation interface, and for command display alarms. |
|
|
|
|
|
|
|
|
|
|
|
|
Networking Tools |
|
Will Stephenson committed changes in /trunk/KDE/kdenetwork/kopete:
|
Commit usability changes by Dennis Nienhüser and myself to the accounts and identity config:
*) change identityId to id for clarity
*) use an immutable id as an internal handle
*) introduce user-editable label field
*) use pointers for default identity
*) replace new identity source dialog with Add and Copy buttons
*) don't show identity selector dialog on switch when there are only 2
identities
*) adapt dbus interface to use ids, map these to labels
*) more string tweaks
*) order the KCM pages using X-KDE-Weight |
|
|
|
|
|
|
Urs Wolfer committed changes in /trunk/KDE/kdenetwork/krdc:
|
* Add clipboard sharing between host and client again in VNC. Mostly the same way as it has been done in the old KRDC.
* Less debug noise for floatingtoolbar |
|
|
|
|
|
|
|
|
|
|
Frerich Raabe committed changes in /trunk/KDE/kdenetwork/knewsticker:
|
- Renamed 'HeadlineItem' to 'HyperlinkItem' since the latter is more descriptive (to me at least) - and 'hyper' is a pretty word.
We should invent 'Ultralink items'; you can right-click on them to go to an alternative destination or so. |
|
|
|
|
|
|
|
|
Office |
|
Inge Wallin committed changes in /trunk/koffice/kchart:
|
* OpenDocument: Make the ChartShape load the chart type.
This is the first ODF loading code in ChartShape; from here on, it's only details, the infrastructure is DONE.
This makes all the subtasks under "Build the foundation" also DONE, which is a big milestone. I'll go grab a "Ballerina" cookie now.
They are great, with the nougat and stuff. |
|
|
|
|
|
|
Johannes Simon committed changes in /trunk/koffice/kchart/shape:
|
* Done with the interface for the legend tool
* Some code cleanup along the way
Some UI problems I noticed:
* The QFontComboBox is at least ~200 pixels wide. That's way too much for a tool! We might want to find an alternative, or use a dialog instead.
* Oxygen doesn't seem to fully support QComboBoxes yet. It paints reaaally slowly, and setting it to checkable doesn't quite work yet - it doesn't gray out its elements when unchecked. |
|
|
|
|
|
|
Marijn Kruisselbrink committed changes in /trunk/koffice/shapes/musicshape:
|
|
start implementing support for multiple linked music shapes that display portions of the same sheet. Lots of stuff doesn't work yet properly (and some stuff works with ugly hacks), but the start is there... |
|
|
|
|
|
|
|
|
|
|
John Layt committed changes in /trunk/koffice:
|
Port from KPrinter to QPrinter, remove dependency on KDE4_KDEPRINT_LIBS.
This change alters the api of KoView to cater for KSpread requiring a custom dialog. KSpread itself currently crashes on printing, but other apps are fine. I'll fix kspread tonight, but need to commit this now as KDEPrint is scheduled for removal from kdelibs today. |
|
|
|
|
|
|
Boudewijn Rempt committed changes in /trunk/koffice/krita/ui:
|
|
Implement the properties dialog for the filter mask and make it undoable. Note that we still need to fix the live preview update. |
|
|
|
|
|
|
|
|
Johannes Simon committed changes in /trunk/koffice/kchart/shape:
|
Finally: Now it's possible to change the position of the chart legend. I also set the default pen for the diagram to QPen( Qt::black, 0.4 ), which makes everything (and most importantly, the 3D mode), much prettier ;)
I admit, that was part of my previous patch, but I didn't mention it. |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Johannes Simon committed changes in /trunk/koffice:
|
* We now have a flag that allows us to tell the shape whether it is supposed to take over ownership of the model: ChartInterface::setModel( QAbstractItemModel *model, bool takeOwnershipOfModel = false )
* After moving initializing of default data of the chart shape back and forth between ChartShape and ChartShapeFactory like 3 times, we now decided to put it in the ChartShapeFactory, since that is necessary for implementing different chart templates in ChartShapeFactory.
* You can now hide the legend title more easily, in a more obvious way: in the legend tool, the "Title" group box is now checkable. Rock on, Qt.
* Added ChartShape::Private deconstructor and moved deleting of private elements to it.
* Removed KChartPart::createDefaultData()
* Basic support for bubble charts. Actually, no.. No bubble charts. It's more like a scatter chart with circular markers. Once KDChart supports real bubble charts, I'll integrate them. |
|
|
|
|
|
|
|
|
|
|
Matthew Woehlke committed changes in /trunk/KDE/kdebase/runtime/kstyles/oxygen:
|
integrate new hole code from playground, holes now use nicer colors for 3D and are more consistent with slabs
TODO: fix fill non-rounded corners, fix scrollbars (go back to slider not overlapping hole 1px) |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Optimise |
|
KDE-Base |
|
Maksim Orlovich committed changes in /trunk/KDE/kdelibs/kjs:
|
Decrease memory usage for JS objects/values by a third (48->32 bytes on 32-bit), and tune the allocator properly, giving a ~10+% speed up on 32-bit, and a huge speed up on 64-bit, since it was previously not configured properly at all.
This is heavily based on JSC revisions 21017, 21018, 21019, 21025, 21029, 21031, 21047, 21048 which are in turn heavily based on Chris Hyde's work in our kjs-tweaks branch; however, this does not introduce what I consider to be questionable GC semantics, which make the allocator abort/crash if the object size is too big for its liking (OK, OK. We still abort on objects ~64kb in size or more.
But surely no one has those. Right? Right!?)
Discussed with Harri, and regression tested twice, once w/normal allocator, and once with additional usage of oversize allocator forced, since it's barely used at all otherwise.
In more detail:
- Move lots of stuff in RegExp object into d-pointer
- Remove the two modes of marking in KJS::List, and just mark them globally, but purge them aggressively when possible.
This is the only SIC bit, but it also simplifies semantics and makes the use of KJS::List less error-prone
- Store the size of array objects with data, to shrink them
- Get rid of threadsafe-dtor bit, as we don't need it.
- Tuck localInjected bit with the property map
- Make property maps smaller by taking advantage of the fact that the single-entry and table modes are disjoint, and hence can shared some storage via a union
- Make the allocator work with well-aligned, fixed-size blocks, to make it possible to easily find heap metadata from the pointer.
This also simplifies checking of pointers at stack scan.
- Using the above, move the mark bits into heap metadata, finally fixing the 3-bits-eating-4-or-8-bytes insanity
- Implement a completely new oversize allocator that is capable of working with mark bits.
This is a simple bitmap'd allocator which is not very fast on its own, but is designed carefully to not intefere with the fast fixed-size allocator's operation and performance, and to do very little work when there are just a few oversize objects around, which is the common case for KHTML. |
|
|
|
|
|
|
|
|
|
|
Peter Penz committed a change to /trunk/KDE/kdelibs/kio/kio/kmimetyperesolver.cpp:
|
|
Fixed performance bottleneck when resolving the mimetypes for several thousand items: Only invoke KDirModel::itemChanged() for items that are visible within the view. This decreases the CPU-load from ~100 % down to ~7 % during MIME type resolving (reviewed by David Faure). |
|
|
|
|
|
|
Harri Porten committed changes in /trunk/KDE/kdelibs/kjs:
|
|
adapted jsc revision 26945 to our optimized structure. makes the diff smaller and gains ~1% time in a benchmark. |
|
|
|
|
|
|
|
|
|
|
Other |
|
Development Tools |
|
Andreas Pakulat committed changes in /trunk/KDE/kdevelop:
|
Remove the kate modelines, we have a .kateconfig file with the proper setup in the kdevelop root directory.
Everybody please check that you have set a depth of at least 5 under the Kate configuration page "Open/Save" |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Luboš Luňák committed a change to /trunk/KDE/kdebase/workspace/startkde.cmake:
|
I officially declare dbus autolaunch to be broken. I repeatedly get two (session) dbus daemons if I have none running before starting KDE.
Symptoms are things like Ctrl+Alt+Del not being the right shortcut to get out of KDE.
Make startkde use the starting code suggested in dbus-launch manpage (dbus-1.0.2), if somebody knows a better way of fixing it, feel free to. |
|
|
|
|
|
|
Stephen Leaf committed changes in /trunk/playground/base/autostart:
|
Update email address
caps all cmake commands
use kcmshell4 instead of kcmshell and set a System Settings category, makes it so we finally appear in systemsettings!
Use class Desktop anywhere appropriate |
|
|
|
|
|
|
|
|
Alex Merry committed changes in /trunk:
|
|
Goodbye, libkdeprint, how we loved thee. |
|
|
|
|
|
|
Aaron J. Seigo committed a change to /trunk/KDE/kdelibs/kdeui/icons/kiconloader.cpp:
|
the api docs don't mention you have to preface with emblem- and this way you can use any icon you want for your overlays. yes, this lets you shoot yourself in the foot, but if you pick a poor choice for the emblem icon that's your problem ;)
i looked in lxr to find overlay users, but couldn't find any ... but if i missed you and your emblems no longer show up, preface their names with emblem- |
|
|
|
|
|
|
|
|
|
|
Aaron J. Seigo committed a change to /trunk/KDE/kdebase/workspace/plasma/applets/kickoff/ui/flipscrollview.cpp:
|
make this keyboard nav friendly again .. now that left/right always switches active view, Enter/Escape (or alt/ctrl left/right) flips through the menu structure.
also fix a bug where when going up the menu hierarchy, the current index was still set to the submenu (so, press up or down and hit enter and a very surprising thing would happen: an app in the submenu would launch) |
|
|
|
|
|
|
Aaron J. Seigo committed changes in /trunk/KDE/kdebase/workspace/libs/plasma/widgets:
|
when an item is added to the layout, give it a chance to react. in the case of Widget, we're going to set the movable state of the QGraphicsItem. the result: items move, or don't, based on whether or not they are in a layout. finally, you can't just move shit around in the panel.
it's pretty hackish since we have to special case FreeLayout but hey, since when were layouts in plasma ever sane? |
|
|
|
|
|
|
|
|
|
|
Harri Porten committed changes in /trunk/KDE/kdelibs:
|
|
this doesn't do anything really yet but I want it in place now for developing dynamic HTML object extensions in KDE 4.x. |
|
|
|
|
|
|
Aaron J. Seigo committed a change to /trunk/KDE/kdebase/workspace/libs/plasma/containment.cpp:
|
* don't cache the painting; prevents blowing the pixmap cache and killing performance on large screens. we don't move this around anyways, and we only paint the exposed rect when using a bitmap wallpaper
* default to a bitmap wallpaper as we handle those a lot better (e.g. only painting the exposed rect)
the paper painting code really does need to move to DesktopContainment at some point =) |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Zack Rusin committed changes in /trunk/KDE/kdelibs:
|
Left hand - nothing, right hand - nothing! I just move this behind my back and... watch it... watch it...
Bam, not crashing anymore. Neat trick, eh?
Make Sonnet not crash with the Enchant engine which is the default. Enchant caches global references to dictionaries, so we need to keep a reference as well and only delete dictionaries when their reference count is 0. |
|
|
|
|
|
|
|
|
Dirk Mueller committed changes in /trunk/extragear/multimedia/amarok/src:
|
I don't know how this is supposed to compile, but in the last 8 years or so I've never seen a compiler that allows to call a non-const function from within a const function.
Dashbot also complained about this for a couple of days now, with noone caring? |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Utilities |
|
Rolf Eike Beer committed changes in /trunk/KDE/kdeutils/kgpg:
|
Remove shredder from KGpg
This feature was never related to GPG at all, is removed from KDElibs and had some security implications. Since it will not work at all from now on better completely remove it before users start to argue. |
|
|
|
|
|
|
Friedrich W. H. Kossebau committed changes in /trunk/playground/utils:
|
Uploading the Okteta program into KDE's playground, so the code isn't lost, after growing slowly only on my hdd for more than a year.
Okteta is a planned successor to KHexEdit, yet misses all of it's functionality.
With it's modular architecture, based on the co-developed lib kakao, it should soon offer more than would could be done with the monolithic KHexEdit. I hope. |
|
|
|
|
|
|
|
