Issue 2
16th April 2006 by Danny AllenThis Week...
But after months of work we ended up basically with a fork of scons.
So apparently scons wasn't the best tool for the job.
...
CMake needed some enhancements mainly for the windows stuff (we are probably the first project which uses mingw heavily). The cmake developers have been very supportive for KDE, they are on the buildsystem mailinglist and fixed the issues we found within days. They want cmake to become the buildsystem for KDE 4.
One of the reasons, at least, was to maintain compatibility with an old Python version. By dropping that compatibility, we made some progress, but created a fork.
...
between how much slower cmake is and the constant breakages that i have no idea how to fix (like the NOGUI in kdesu/kdesud/CMakeLists.txt not working, even though i have the latest cmake modules installed from kdelibs that seem to include something about it) i really, really, really hope that it improves dramatically over the next month. we don't need a build system holding us back. i understand the need for something better than autotools, but in my book the baseline for "better" is "actually builds reliably". meh.
With CMake, I get the feeling that there's only a very few gurus who know what is going on (Alex, David, and of course William Hoffmann on -core-devel). Whining often leads to fixes of a mysterious nature. But waiting for osmosis to somehow bring enough knowledge to developers is a slow process. The wiki page has some information, but it's mostly about using CMake under ideal conditions, as opposed to fixing it when it's broke.
...
I believe Aaron's point on CMake is that a build system shouldn't get in the way, and if CMake breaks right now and keeps him from working on whatever it is he does, those are valuable wasted hours. autofoo had progressed to a point that it didn't regularly get in the way of developers on KDE's UNIX-like platforms. CMake does get in the way, since it's still being broken. For the -- presumably many -- people who do not want to be build-system guru's, the system just has to work. That goes double for external tools that use the build system, like Coverity's static checking.
Now I'm done being annoyed for CMake getting in the way, though, I can get on with loving it for the things that it does right.
...i'm fairly confident that it'll work out in the end; i have great faith in the people who work on this project. we're not "in the end" yet, more in the middle. and along the way i think it is healthy to discuss what is working and what isn't working along the way.
Repetitive Strain Injury is an illness which can occur as a result of working with a mouse and keyboard. This utility can be used to remind you to take a break now and then. It will show you a random picture from a collection you can configure yourself for a configurable duration at a configurable interval.
You can use these breaks to do some stretch exercises for example, or as a reminder to walk away from the computer for a while.
RSIBreak will sit in your system tray and when it is time for a break it will show you the picture full screen. All timings can be set by clicking with the right mouse button on the icon in the system tray.
Version 0.6.0 Changelog:
- Extensive amount of statistics.
- Colors to indicate how far away you are from a break in the tooltip
- libxss for the idle detection is a no longer optional, it is required.
- RSIBreak remembers the timer states when you quit. It can restore the timers if you restart RSIBreak afterwards in a short timeframe. This is handy for rebooting (for those zealots who believe rebooting solves problems).
- Bug fixes:
- For Gnome there were two Quit entries in the context menu.
- Welcome message on first launch was incorrect.
- Compiles with KDE 3.3 and probably with gcc 2.9x.
- Memory leak while querying idle time.
- Some buddies in the setup were not set.
- Some i18n calls in the setup were not correct.
- Clear focus from buttons to prevent accidental closure of a break.
Statistics
Commits | 2636 by 212 developers, 5981 lines modified, 886 new files |
Open Bugs | 12107 |
Open Wishes | 10590 |
Bugs Opened | 349 in the last 7 days |
Bugs Closed | 313 in the last 7 days |
Commit Summary
Module | Commits |
/trunk/KDE |
716
|
/trunk/www |
421
|
/trunk/l10n |
308
|
/trunk/extragear |
264
|
/trunk/koffice |
185
|
/branches/stable |
150
|
/branches/KDE |
128
|
/branches/work |
121
|
/trunk/playground |
104
|
/branches/koffice |
84
|
Lines | Developer | Commits |
284
|
Laurent Montel |
143
|
269
|
David Faure |
111
|
177
|
Dirk Mueller |
83
|
157
|
Gilles Caulier |
70
|
217
|
Luboš Luňák |
58
|
107
|
Frans Englich |
41
|
85
|
Volker Krause |
40
|
79
|
Albert Astals Cid |
37
|
76
|
Mark Kretschmann |
37
|
83
|
Stefan Nikolaus |
36
|
Internationalization (i18n) Status
Language | Percentage Complete |
Portuguese (pt) |
99.99%
|
Danish (da) |
99.12%
|
Swedish (sv) |
99.04%
|
French (fr) |
95.1%
|
Spanish (es) |
94.78%
|
Italian (it) |
94.51%
|
Estonian (et) |
93.94%
|
Dutch (nl) |
93.55%
|
Greek (el) |
93.38%
|
British English (en_GB) |
92%
|
Bug Killers and Buzz
Person | Bugs Closed |
Thomas Zander |
91
|
Tommi Tervo |
31
|
Thiago Macieira |
17
|
Olivier Goffart |
14
|
Luboš Luňák |
12
|
Mark Kretschmann |
12
|
Seb Ruiz |
11
|
Julian Seward |
10
|
Charis Kouzinopoulos |
8
|
Jens Dagerbo |
8
|
Program | Buzz |
amaroK |
2215
|
K3B |
1079
|
Kopete |
1069
|
Kate |
759
|
KMail |
691
|
SuperKaramba |
666
|
Kontact |
589
|
Kicker |
566
|
KDevelop |
540
|
Quanta |
355
|
Person | Buzz |
zogje |
397
|
Aaron J. Seigo |
371
|
George Staikos |
352
|
pipitas |
291
|
Stephan Kulow |
290
|
Zack Rusin |
281
|
David Faure |
276
|
riddell |
236
|
telex |
210
|
boud |
190
|
Commit Countries
Commit Demographics
Sex
Age
Contents
Bug Fixes | Features | Optimization | Security | Other | |
---|---|---|---|---|---|
Accessibility | |||||
Development Tools | [] | [] | [] | ||
Educational | [] | [] | [] | [] | |
Graphics | [] | [] | |||
KDE Base | [] | [] | [] | [] | |
KDE-PIM | [] | [] | [] | [] | |
Office | [] | [] | [] | [] | |
Konqueror | [] | [] | [] | ||
Multimedia | [] | [] | [] | ||
Networking Tools | [] | [] | [] | [] | |
User Interface | |||||
Utilities | [] | [] | [] | ||
Games | [] | [] | |||
Other |
There are 131 selections this week
Bug Fixes
Development Tools
don't crash when deleting the designer window. Instead, crash in
kdelibs instead. :)
reviewed another 50 coverity reports
Educational
Graphics
digikam from trunk: Camera Rename Customizer bug fix: If 'Add Sequence Number' and 'Add Date Time' are unselected, the customize filename have non-sence because all target files to download will have all the _same_ filename. Also, camera GUI will ask to user is target must be overwritted!
This commit fix the problem to use a Combo Box options list instead separate Check Box. Like this at least one option will be used if user want a customized renaming.
ForwardingSlave simply invents mimetypes if there are none
(for whatever reason). So give them a mimetype so konqueror
is not confused between application/octet-stream and S_IFDIR
digikam from trunk : patch from Paco about screen ICC color profile rules.
The implemented have been fixed about a crash generated by in DImg::convertToPixmap() witch return a null Pixmap.
The code isn't optimized : ICC color profile files are loaded from disk at each pixmap computation. better way is to load these files at Image Editor startup, store like bytes array into memory, and use it when we want.
digikam from trunk : reverse RAW file generator source code about thumbnail rotation. I suspect a problem in this code witch broke CR2 file thumbs and don't rotate properly thumbs (certainly some camera tags orientation flags without respect TIFF/EP spec).
Alternative solutions for the future :
1 / using future libopenraw library from freedesktop.org (http://www.freedesktop.org/wiki/Software_2flibopenraw)
2 / using Exiv2 thumbs extractor when TIFF/EP will be finalized (my prefered way (:=)))
KDE Base
Missing file in no-namespace commit.
Default to no namespace in XML rather than to XHTML
Ok, I've had enough with Alt+Tab and mouse focus policies:
- use click to focus or focus follows mouse and I'll be fine
- with the remaining unreasonable focus policies only the CDE-style
tabbox works that raises and lowers windows and ignores
keep above/below windows
- if you don't like that, provide a patch for kwin/tabbox.cpp
- tabbox should be completely redesigned for KDE4
BUG: 123890
CCBUG: 105263
CCBUG: 84424
use the same icon as on the button (the last task in the group's icon)
for eagle eye bart.
don't use KAutoMount, it isn't available on win32
Lower-case tags so it actually works with XHTML.
Regressed because namespaces actually work now.
the floppy case is worse than expected. But now it really works:
we have to stress KMountPoint a bit as hal is completly useless
for floppies and we can't easily integrate the fstab backend.
Handle X timestamp wrapping correctly even on 64bit platforms.
Ok, I've had enough with Alt+Tab and mouse focus policies:
- use click to focus or focus follows mouse and I'll be fine
- with the remaining unreasonable focus policies only the CDE-style
tabbox works that raises and lowers windows and ignores
keep above/below windows
- if you don't like that, provide a patch for kwin/tabbox.cpp
- tabbox should be completely redesigned for KDE4
(#123890,#105263,#84424)
One more of "When a window resizes, first try to keep it inside its xinerama screen."
Separate focus chains are nice, but KDE-style Alt+Tab not limited
to the current desktop still needs the global one for MRU.
BUG: 124721
Revert previous two commits. For the second time.
hicolour is not a theme, it is a namespace for third party icons.
Fix commandline export.
BUG: 125165
Fix bug where 'fade out' setting was always used for fading between opacities, even if we were actually fading in. Also committed to KDE 4.
Handle even su that doesn't write "<whatever>:" - looks like in China
they have a different colon than we do.
guessing mimetypes just for the sake of it is wrong
(especially for directories that become octet-streams)
and slow.
The XML parser lets \t through. For now just rewrite them to spaces, because
otherwise Qt print them as boxes.
OK, it turns out that when you reset(), you have
to put the keys/certs back in. With that, it stops
asserting. Then it just fails to compare the messages.
If you encrypt the same data, with the same certs,
twice, you don't get the same result each time.
So I added the decrypt part, including re-use, and
checked that if you encrypt the same data twice, then
each decrypt produces the same result, and that result
is exactly what you started with. The test now passes.
Fix compilation on Mac OS X
Fix build on Mac OS X but not using ksocks.h in the kcontrol module
(the problem was kde_socklen_t but Thiago suggested this solution, to avoid using ksocks.h altogether)
forward port fix for fullscreen windows and kickertips showing anyways
with autohide panels
unbreak compilation with gcc-3.3. As stated on k-c-d gcc-3.3 should be supported for now.
BUG: Changing an user's password can change another password for a different
user (Malone bug #39444)
when checking the "don't show again" box and then pressing Yes,
remember "Yes", not "No". Everything else doesn't seem to make
sense to me.
KDE-PIM
Allow system tray icon to appear in non-KDE window managers (Bug 123651)
Fixing PDU decoding, checking if it is an incoming or an outgoing SMS.
This should fix a lot of problems, with outgoing sms, so please upgrade, test, and report.
This fixes the issues reported by the docbook sanitizer found at EBN for kdepim. I fixed all issues which were raised at the kde-pim mailinglist. One of the changes involved a new entity, so it will only 'compile' with the latest entities installed. (cp $svn/kdelibs/kdoctools/customization/entities/general.entities $prefix/share/appses/general.entities)
Office
Don't hide the font preview -- we want it!
Don't draw dashed line over solid line :)
Fix typing of accented characters with compositing-like keys for specific accents (like with azerty keyboards) and probably more things like that, in KWord tables.
This is triggered by the immodule patch Debian Unstable has applied to its Qt 3.3.6.
Thomas, David: please check if this is OK, and perhaps backport and forwardport if it is.
change "&&" to "and" in i18n strings to get the translators less confused and correct spelled tooltips
Enable the input methods _before_ setting the focus. Should fix the odd bug that some people had where they could not enter accented characters before manually focussing another widget.BUG: 124532
This was a tricky one: when I tried inputting some japanese text with skim, it wouldn't work in KWord. Turns out that when KWord gets a IMComposite event, it (indirectly) calls this function. That would then unconditionally update that KAction, that then would update a QCombobox to set the current text, and that then would reset the inputcontext on the poor old canvas, thus resulting in a loss of your poor hiragana each time skim would compose one.
I fix this by checking if the fontsize and font family actually changed. Since I assume it won't change in the middle of an IMStart-IMEnd sequence, this fixes the input with me.
From what I understand, this would basically mean that Japanese input would be impossible without this patch. However, I think I recall actually being able to do that in KWord, so I'm inclined to think that it's partially caused by qt-immodule. Anyway, the CCBUG seems related, but I'm going to check this with someone who actually has a real Japanese keyboard (Kiruwa) and maybe close it if he can confirm this a bit better.
CCBUG: 41380
CSV Import
- fixed problem with importing data with more than 100 columns (for default settings)
1.0/2.0: ported
- only the first 10KiB of data is loaded for the preview by default; can be adjusted
by setting ImportExport/MaximumBytesForPreviewInImportDialog option in kexirc
1.0/2.0: ported
o Fix Bug 125418: Resizing textarea resizes wrong textarea
The problem was that another obejcts was selected when trying to
resize the object. This has be fixed by checking if there is allready
an other object selected at this place.
Thanks for reporting the bug.
BUG: 125418
o Fix Bug 125424: When setting slide background to "Use slide master
background", preview is not shown correctly.
Now we show the background of the master in the preview when
"Use slide master background" is selected.
Thanks for reporting this bug.
BUG: 125424
BUG: 125433
Fix crash when trying to use tools on layers that cannot be painted on: now we
select the nearest layer that can be painted on (as was my original intention).
Fix the XMLGUI related hang, bringing back a lot of the ui. krita is now usable for painting and things.
Konqueror
* Crystal ball programming: assume Michael Rys' comment #1 in http://www.w3.org/Bugs/Public/show_bug.cgi?id=3097 gets accepted, and align. Basically, s/FORG0006/XPTY0004.
* Improve Doxygen in AggregateFNs.h
- use goto icon in bookmark dialog, bookmark icon can be misinterpreted
Multimedia
Disable "Unset Cover Image" for embedded images, since we can't remove them anyway. Patch by Shane King <kde@dontletsstart.com>.
BUG: 125372
fix display of free space for devices larger than 4gb
Fix a regression when downloading lugradio episodes.
If anyone notices podcasts with strange url's in the enclosures, please report them to me.
nice guess. The chances were just 50:50 to get it right ;)
The difference is 0 for IDE, but noticably for SATA
start porting the context browser to more properly display ratings
this is very much a work in progress (for one, the default theme should be made to not use stars for the scores, as two rows of them is a bit confusing), but I've been putting it off for far too long.
Networking Tools
Use the following method to determine the status message to
be displayed (the term "contact" should be read as
"contact of the metacontact" and "online" as "not offline"):
* Display the new status message if
- the new status message is not empty and
- the contact who set it is online or there are no contacts online at all.
* Otherwise display the first non-empty status message among all contacts
online---or offline if there are no contacts online at all.
* If no status message is displayed yet display no status message at all.
BUG: 116614
Fix Bug 118689: Incoming emoticons appear unpredictably
BUG: 118689
This was caused because on rich text messages, there is <span> or <p> arround the text, and the > or < was not considered as a space. now it is.
Utilities
Update the script that creates api documentation via robodoc.
Should be more full proof now and leave the existing source files
completely untouched.
Enjoy!
Games
it seems cmake is not able to deal with lib called kolf and a binary called kolf, so i renamed the lib to kolflib and i can play kolf4 now :-)
Features
Development Tools
Apply patches from http://www.geeksoc.org/~jr/umbrello/uml-devel/9396.html
implementing AutoCompletion for ComboBox dialog widgets. Many thanks Achim.
BUG:104477
* Added xml/saxbookmarks, xml/dombookmarks and widgets/tooltips examples
* Improved garbage collection marking, Children QTreeWidgets and
QListWidgets are now marked. QObject tree instances are only marked
starting at an instance which has no parent to improved efficiency,
otherwise the same sub trees would be marked several times as a
traversal is started for each node in the heirarchy.
* The 'type' method in QEvent subclasses now works correctly.
* Added inspect and pretty_print methods to the Qt::WidgetTreeItem class
Add code to use the kdelibs4_snapshot for kdelibs if the 'branch kde4-snapshot' option is
selected, instead of requiring weird hacks in the config file to use the snapshot.
e.g.
module kdelibs
branch kde4-snapshot # Use the branches/work/kdelibs4_snapshot stuff.
end module
Make the snapshot the default for kdelibs, which should help with the Coverity stuff.
Don't try to download a snapshot if override-url is in effect for a module.
Update the sample file to match, including a comment at the use-stable-kde option to warn
the user to also update the kdelibs options.
Touch up the pretend mode output a bit.
Educational
The Observing list tool is now functional. The table view is now a
QTableWidget, and the "Tiny" (name only) view is a QListWidget.What works? Adding objects to the list in the map (popup menu, or "O"
key), selecting them in the tool and using the action buttons. You can
also switch to "Tiny" mode, which shows just the name field and hides
the Log notes editor.TODO: Shrink the window when entering Tiny mode. Sort rows when
clicking a column. ObsListWizard is still a crash.
Added legendre polynomials up to n=6 (P_[0->6]).
Added code for drawing plots in a variety of styles (dashes, dots, etc) (there
is no way of configuring which style to use for plots yet, though).
Allow changing of plot line style (e.g. solid, dash, dot, etc). "Appearance"
group box needs improving though - atm it is slightly cluttered.
Graphics
digikam from trunk : GPS locator tool :
- Code polishing.
- Remember last web GPS locator service used.
Nota: the implementation is ready to use new GPS maps service from the web. If you have some url witch accept world latitude and longitude GPS coordinates, let's me hear...
111560
digikam from trunk : improving image properties restoration in database using image metadata:
- Fix DMetadata method to get image Exif/Iptc tags properlly. (With the old implementation, Comments from Exif and Iptc, Rating from Iptc and are never checked duing a wrong validity test - stupid bug)
- Now at startup, these informations are backported to database :
==> Comments from JFIF section, or Exif UserComments tag, or Iptc Caption tag.
==> Date & time stamp from Exif dateTime tag or Iptc date & time tags.
==> Rating from Iptc Urgency tag !!! If you add new files in your Album library, witch are rated using Mapivi for example, digiKam items rating will be appear in main interface.
Nota : these updates in database are only performed to new files :
==> when all albums are parsed during statup (or manually from Tool menu).
==> when new items are downloaded using camera interface.
==> when a folder is imported from main interface.
TODO :
- Performed a database update at startup when files are already in database and when metadata have been changed outside digiKam (using ExifTools for example)
- Do something with digiKam Tags, since they are stored in IPTC Keywords tags. This is most complicated to do because there is no hierarchy between IPTC Keywords like with digiKam tags. We store only Tags name in IPTC keywords. I propose :
==> to check if a digiKam tags name already exist in database and taging automaticly item using it.
==> do nothing if Tags name do not exist (no new digiKam Tags will be created in database).
==> If dupplicate Tags name exist in digiKam database (for ex. Travel/City and Travel/France/City), use only the first Tag name found in database.
digiKam Tags <==> IPTC Keywords rules is a complex problem. Please give me your viewpoints into B.K.O. Thanks in advance
91811
digikam from trunk:
The ImageEditor can now access the database independently
from main view, i.e. also if main moved to a different directory.
- adjust ImageInfo to be more like a database item wrapper class
- add a method to ImageWindow to start it with a list of
ImageInfo objects, use this in AlbumIconView
- adapt ImageWindow to (optional) ImageInfo list
- adjust behavior of saveAsIsComplete
- add itemChanged methods to ImagePropertiesSidebarDB:
- one takes only a URL (no database)
- one takes an ImageInfo object (used by ImageWindow, full DB support)
- one takes AlbumIconItem and -View (used by main view,
full DB support, optional future interaction with the view)
- ImageDescEditTab now works on an ImageInfo object
- add a copyItem method to ImageInfo as a wrapper for said function from AlbumDB
- modify AlbumDB::copyItem to return image id of inserted image
- add DISTINCT keyword to second SELECT statement of AlbumDB::copyItem.
With all of above changes, I constantly produced bug #113915.
I do not know why the SELECT statement produces (in my case) three times
the same result so that INSERT complains, and I do not know which of
my changes should ever trigger this problem, but the DISTINCT fixes this.
If anyone speaks SQL, you might want to check my comment in AlbumDB::copyItem.
Currently, the rating is not copied when database information is copied (in copyItem).
Should this be done?
CCBUG: 113915
KDE Base
QT_PLUGIN_PATH is useful so that the KDE widgets are available in designer.
Make it user friendly:
- Qt qmake not found!
+ The installed Qt version 4.1.0-rc1 is too old, at least version 4.1.1 is required
remove SConstruct as kdelibs doesn't have bksys support anymore (commit 523268)
FEATURE: --noignorebutton, based on patch by "Nicolas Lécureuil" <neoclust.kde gmail.com>
starting work on dataengines, datasource, datavisualizations, the plasma
app and applets ... doesn't compile atm as it's a work in progress off
the design concepts. but dinner becons and i don't want to lose progress
in case my laptop degrades further.
Added classinfo to the bindings, this is going to be _very_ helpful later
on.. also got rid of the stuff in the top level of autobind because its
not really useful anymore, we're going to have to rethink object/widget
bindings.. but variants first!
A replacement for the make parser target until someone figures out how to do that in CMake
KDE-PIM
Office
Add a patches directory to store all non-ported
1.6 patches so we won't forget them.
implemented cell reference to another sheet, like "Sheet2!A1"
this is done by partial support for EXTERNSHEET and SUPBOOK records
should work both on Excel 95 and Excel 97/later
Konqueror
use grayscale pixmap for unencrypted connections
Multimedia
Patch from Laszlo Pandy to let xine engine play .part files from KGet.
BUG: 110833
The collection browser now has tabs to switch between flat and tree view
instead of KToggleActions.
Looking for feedback, should the buttons common to both be moved into
the toolbar?
*Collection browser remembers what mode its in between sessions again.
*Tabs have wheel support (though its kind of funky, since my wheel
often sends two events at once so it skips right back to where it started)
Move custom icons option to the Appearance settings.
forwardport adridg's changes of this january
Random string function implemented. Note that this function restricts the possible characters to 0-9a-zA-Z, which is good
because we can print to stdout, but bad because there's less bits to choose from. Also there is a little favoritism
because of avoiding non alphanumeric characters, but in general that's alright. Even with these drawbacks, assuming
that the randomization sequence is good and a decent seed is found using the chosen method, there are still up to
51649738599652491790651549397174942699476
584262665539608782445962684816148
42987330263563657801857314603738370932736
possible IDs. I think it'll suffice for now.
more work on context browser + ratings
i've introduced some new css classes, and ported the default theme (because it's the default), and example (because I use it)
there's still some issues, like the gray line on the left of the score boxes isn't continuous now, and in the current track info, if you displayed both rating and score, it'd mysteriously move next to the 'last played' text instead of above it, whereas it works fine if you only display one of them.
i've hacked around this for now by sticking a bunch of <br /> at the end, but if someone who actually understands this shit could take a look at it, that'd be good.
Networking Tools
Adding 2 more validy tests for a style archive while keeping my current threshold. So if a package doesn't have Header or Footer, it will detected as valid
Compilation of Messenger plugin can now be optional (desatived via ccmake), can happen the plugin doesn't compile.
Support .jisp theme (as described in the JEP-0038)
The old emoticons theme is of course still supported.
There are no new feature tha JEP-0038 has over the current format.
(xml:lang and sounds are not supported)
Utilities
Games
new i18n framework
Removing ksmiletris, as discussed a while ago, see
http://lists.kde.org/?l=kde-games-devel&m=112536434505736&w=2
Optimization
Educational
KDE Base
Improved speed of valgrind unit tests.
Startup sequence reorder, see ksmserver README for how it works
now (in a nutshell: get kicker and kdesktop up as soon as possible,
start as much stuff as possible as late as possible).
remove silly code; doesn't really do anything and confuses covarity ;)
Forwardport of r455251, whatever that is exactly about. Penalty
points for missing a trunk commit for a 3.5 commit go to
Scott Wheeler this time.
Updated kjscmd for better command line support, now also supports exec without gui support leading to ~0-2ms startup time instead of the usual ~200ms for QApplication..
KDE-PIM
Avoid probing again devices is going back in the New Mobile Phone Wizard.
Office
Due to popular request: ditch the madvise stuff to 'swap' out tiles, and actually swap them out with munmap. This introduces some complications like memory areas becoming invalid and so. That, and the fact that I did have to change some stuff for this, might cause some fallout. I didn't notice anything strange with it anymore, but please yell if you see odd behaviour or crashes that you can directly link to this commit :-)(This deprecates some functions, I won't remove them because afaik we'd want at least sourcecode compatibility with Krita 1.5. They exist, but _please_ don't use them anymore. To the best of my knowledge, all current occurances are gone, keep it that way :P)
Konqueror
improve felt performance for preference dialog
we can enable the back button already when we start to load the next url
Multimedia
Change the rand function a bit. Why seed on every call? So do it once. Also call rand() directly after seeding, no point
in the extra function call 64 times.
Networking Tools
Security
KDE Base
the usual security fixes:
- add proper quoting
- add proper LD_LIBRARY_PATH export's
- don't make world executable. Ok, it would be better to
honor umask, but I don't know how to do that..
Other
Development Tools
first steps in porting
Educational
more porting
KDE-PIM
Office
Adapt to new kdelibs api