Re: Re: kwin: build failure with gcc-4.9.0

[Martin Gräßlin]

Martin GräßlinOn Tuesday 29 April 2014 07:39:44  wrote:
> On Monday 28 April 2014 21:41:16 Treeve Jelbert wrote:
> > I recently installed gcc-4.9.0
> 
> > kwin fails to compile  with the following error:
> thanks for letting us know. I just installed gcc-4.9 and can reproduce the
> build error.
> 
> Will investigate

and fixed: 
http://commits.kde.org/kwin/b18c672e368b0ff26367b6b5ea05e927227a7824

Cheers
Martin

signature.asc
Description: This is a digitally signed message part.
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: kwin: build failure with gcc-4.9.0

[Martin Gräßlin]

On Monday 28 April 2014 21:41:16 Treeve Jelbert wrote:
> I recently installed gcc-4.9.0
> 
> kwin fails to compile  with the following error:

thanks for letting us know. I just installed gcc-4.9 and can reproduce the 
build error.

Will investigate

Cheers
Martin

signature.asc
Description: This is a digitally signed message part.
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: Review Request 117814: Make examples optional (default to off)

[Kevin Ottens]

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/117814/#review56817
---


Wouldn't making the examples work without installation been a better fix?

As a sidenote, I wonder why the frameworks group wasn't added to this review, 
are we missing some bit of config somewhere? I found it thanks to kde-commits.

- Kevin Ottens


On April 28, 2014, 5:35 p.m., David Edmundson wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/117814/
> ---
> 
> (Updated April 28, 2014, 5:35 p.m.)
> 
> 
> Review request for Plasma.
> 
> 
> Repository: plasma-framework
> 
> 
> Description
> ---
> 
> Make examples optional (default to off)
> 
> 
> Most users won't want to have tech demos on their real system. 
> 
> 
> Diffs
> -
> 
>   CMakeLists.txt bc5d6d0 
> 
> Diff: https://git.reviewboard.kde.org/r/117814/diff/
> 
> 
> Testing
> ---
> 
> Build both variants editing flag with CCMake
> 
> 
> Thanks,
> 
> David Edmundson
> 
>

___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Build failed in Jenkins: kmenuedit_master_qt5 #12

[KDE CI System]

See 

Changes:

[scripty] SVN_SILENT made messages (.desktop file)

--
Started by remote host 127.0.0.1 with note: Triggered by commit
Building remotely on LinuxSlave - 3 (PACKAGER LINBUILDER) in workspace 

Running Prebuild steps
[kmenuedit_master_qt5] $ /bin/sh -xe /tmp/hudson4718387239349104181.sh
+ /home/jenkins/scripts/setup-env.sh

Preparing to perform KDE Continuous Integration build
== Setting Up Sources

>From git://anongit.kde.org/kmenuedit
   847502b..08a5114  master -> origin/master
Branch jenkins set up to track remote branch master from origin.

== Cleaning Source Tree

HEAD is now at 847502b SVN_SILENT made messages (.desktop file)
Removing build/
Removing install/
Success build forhudson.tasks.Shell@1b489b95
Fetching changes from the remote Git repository
Fetching upstream changes from git://anongit.kde.org/kmenuedit
Checking out Revision 08a5114e1b92b88bca63fa5061c693cff38ff99b 
(refs/heads/jenkins)
[kmenuedit_master_qt5] $ /bin/sh -xe /tmp/hudson8717629781243792169.sh
+ /home/jenkins/scripts/execute-job.sh

KDE Continuous Integration Build
== Building Project: kmenuedit - Branch master
== Build Dependencies:
 kwallet - Branch master
 kdnssd - Branch master
 sonnet - Branch master
 kcrash - Branch master
 kdeclarative - Branch master
 attica - Branch master
 threadweaver - Branch master
 kpty - Branch master
 kinit - Branch master
 knotifications - Branch master
 kio-extras - Branch master
 kcodecs - Branch master
 kparts - Branch master
 kded - Branch master
 plasma-workspace - Branch master
 kross - Branch master
 karchive - Branch master
 qt5 - Branch stable
 krunner - Branch master
 kconfigwidgets - Branch master
 kcmutils - Branch master
 baloo - Branch frameworks
 libksysguard - Branch master
 kf5umbrella - Branch master
 kemoticons - Branch master
 knotifyconfig - Branch master
 kiconthemes - Branch master
 libdbusmenu-qt - Branch master
 kdesignerplugin - Branch master
 kservice - Branch master
 kdbusaddons - Branch master
 kidletime - Branch master
 kjs - Branch master
 kcoreaddons - Branch master
 polkit-qt-1 - Branch qt5
 kdewebkit - Branch master
 kunitconversion - Branch master
 kdelibs4support - Branch master
 kde-cli-tools - Branch master
 kwidgetsaddons - Branch master
 kglobalaccel - Branch master
 kactivities - Branch master
 phonon - Branch master
 ktexteditor - Branch master
 kxmlgui - Branch master
 knewstuff - Branch master
 kwindowsystem - Branch master
 solid - Branch master
 kdesupport-svn - Branch master
 kfilemetadata - Branch frameworks
 kguiaddons - Branch master
 frameworkintegration - Branch master
 kplotting - Branch master
 extra-cmake-modules - Branch master
 kitemmodels - Branch master
 milou - Branch frameworks
 cmake - Branch master
 kjobwidgets - Branch master
 kitemviews - Branch master
 plasma-framework - Branch master
 kcompletion - Branch master
 ki18n - Branch master
 khtml - Branch master
 kconfig - Branch master
 kauth - Branch master
 kwin - Branch master
 kjsembed - Branch master
 khelpcenter - Branch master
 kbookmarks - Branch master
 kdoctools - Branch master
 ktextwidgets - Branch master
 kio - Branch master
 kdesu - Branch master

== Applying Patches
=== No patches to apply

== Syncing Dependencies from Master Server

file has vanished: 
"/srv/jenkins/install/linux/x86_64/g++/kf5-qt5/kde/workspace/kwin/inst/bin/.kwin.c35XNT"
file has vanished: 
"/srv/jenkins/install/linux/x86_64/g++/kf5-qt5/kde/workspace/kwin/inst/lib64/.libkdecorations.so.4.95.0.v88wFK"
rsync warning: some files vanished before they could be transferred (code 24) 
at main.c(1530) [generator=3.0.9]
Syncing dependencies from master server for project kmenuedit failed.
Build step 'Execute shell' marked build as failure
[WARNINGS] Skipping publisher since build result is FAILURE
Recording test results
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

kwin: build failure with gcc-4.9.0

[Treeve Jelbert]

I recently installed gcc-4.9.0

kwin fails to compile  with the following error:

FAILED: /var/lib/sorcery/build/c++   -DKCOREADDONS_LIB -DQT_CONCURRENT_LIB -
DQT_CORE_LIB -DQT_DBUS_LIB -DQ
T_DISABLE_DEPRECATED_BEFORE=0 -DQT_GUI_LIB -DQT_NETWORK_LIB -DQT_NO_DEBUG -
DQT_QML_LIB -DQT_QUICK_LIB -DQT
_WIDGETS_LIB -DQT_X11EXTRAS_LIB -DQT_XML_LIB -D_GNU_SOURCE -
D_LARGEFILE64_SOURCE -D_XOPEN_SOURCE=500 -Dkwi
n4_effect_builtins_EXPORTS -march=native -mtune=native -m64 -pipe -ffast-math -
funroll-loops -O3  -std=c++
0x -fno-exceptions -Wall -Wextra -Wcast-align -Wchar-subscripts -Wformat-
security -Wno-long-long -Wpointer
-arith -Wundef -Wnon-virtual-dtor -Woverloaded-virtual -Werror=return-type -O3 
-DNDEBUG -fPIC -fvisibility
=hidden -fvisibility-inlines-hidden -Ieffects -I/var/git/plasma-ws/kwin/effects 
-I/var/git/plasma-ws/kwin/
tabbox -I/var/git/plasma-ws/kwin/libkdecorations -I/var/git/plasma-
ws/kwin/libkwineffects -I. -Ilibkwineff
ects -Ilibkdecorations -I/var/git/plasma-ws/kwin -isystem /opt/qt5/include -
isystem /opt/qt5/include/QtDBu
s -isystem /opt/qt5/include/QtCore -isystem /opt/qt5/mkspecs/linux-g++ -
isystem /opt/qt5/include/QtWidgets
 -isystem /opt/qt5/include/QtGui -isystem /opt/qt5/include/QtX11Extras -
isystem /opt/qt5/include/KF5/KConf
igCore -isystem /opt/qt5/include/KF5 -isystem /opt/qt5/include/KF5/KService -
I/opt/qt5/include/KF5/KCoreAd
dons -isystem /opt/qt5/include/KF5/KWindowSystem -isystem 
/opt/qt5/include/KF5/KConfigGui -isystem /opt/qt
5/include/QtXml -isystem /opt/qt5/include/KF5/KConfigWidgets -isystem 
/opt/qt5/include/KF5/KCodecs -isyste
m /opt/qt5/include/KF5/KWidgetsAddons -isystem /opt/qt5/include/KF5/KAuth -
isystem /opt/qt5/include/KF5/KG
lobalAccel -isystem /opt/qt5/include/KF5/KI18n -isystem 
/opt/qt5/include/KF5/KF5 -isystem /opt/qt5/include/KF5/KF5/Plasma -isystem 
/opt/qt5/include/KF5/KIconThemes -isystem /opt/qt5/include/QtConcurrent -
isystem /opt/qt5/include/QtQuick -isystem /opt/qt5/include/QtQml -
I/opt/qt5/include/QtNetwork 

-MMD -MT effects/CMakeFiles/kwin4_effect_builtins.dir/effect_builtins.cpp.o -MF 
effects/CMakeFiles/kwin4_effect_builtins.dir/effect_builtins.cpp.o.d -o 
effects/CMakeFiles/kwin4_effect_builtins.dir/effect_builtins.cpp.o -c 
/var/git/plasma-ws/kwin/effects/effect_builtins.cpp
In file included from /var/git/plasma-ws/kwin/effects/effect_builtins.cpp:20:0:
/var/git/plasma-ws/kwin/effects/effect_builtins.h:89:10: error: 'function' in 
namespace 'std' does not name a template type
 std::function createFunction;
  ^
/var/git/plasma-ws/kwin/effects/effect_builtins.h:90:10: error: 'function' in 
namespace 'std' does not name a template type
 std::function supportedFunction;
  ^
/var/git/plasma-ws/kwin/effects/effect_builtins.h:91:10: error: 'function' in 
namespace 'std' does not name a template type
 std::function enabledFunction;
  ^

I tried adding '#include ' to toplevel.h.

 This fixes this problem but there are then others.

A full compile log is attached


kwin-5.x.bz2
Description: application/bzip
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: Review Request 117814: Make examples optional (default to off)

[David Edmundson]

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/117814/
---

(Updated April 28, 2014, 5:35 p.m.)


Status
--

This change has been marked as submitted.


Review request for Plasma.


Repository: plasma-framework


Description
---

Make examples optional (default to off)


Most users won't want to have tech demos on their real system. 


Diffs
-

  CMakeLists.txt bc5d6d0 

Diff: https://git.reviewboard.kde.org/r/117814/diff/


Testing
---

Build both variants editing flag with CCMake


Thanks,

David Edmundson

___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: Review Request 117773: make kwin-decoration co-installable, remove unused kconfig_update script and rename oxygen-shadow-demo to oxygen-shadow-demo5

[Aleix Pol Gonzalez]

> On April 28, 2014, 5 p.m., Aleix Pol Gonzalez wrote:
> > Does this demo really need to be installed?
> 
> Hugo Pereira Da Costa wrote:
> To be honest, I guess, no ... 
> Relic from debug days. (also I still uses every now and then, and right 
> now it exposes a 'bug' in Qt raster graphicssystem).
> Same would be true of oxygen/kstyles/demo/oxygen-demo5
>

But you don't need to install it for debugging, that's what I meant. It's just 
a test.


- Aleix


---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/117773/#review56778
---


On April 25, 2014, 2:40 p.m., Jonathan Riddell wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/117773/
> ---
> 
> (Updated April 25, 2014, 2:40 p.m.)
> 
> 
> Review request for Plasma, Martin Gräßlin and Hugo Pereira Da Costa.
> 
> 
> Repository: oxygen
> 
> 
> Description
> ---
> 
> make kwin-decoration co-installable with kdelibs4 equivalents, remove unused 
> kconfig_update script and rename oxygen-shadow-demo to oxygen-shadow-demo5
> kconfig upgrade script not used as KF5 apps will start with a fresh config
> 
> 
> Diffs
> -
> 
>   kwin-decoration/CMakeLists.txt ff99b256e94507b364965f3b8d4401dd07f55d9a 
>   kwin-decoration/data/CMakeLists.txt 
> 7a4a2f7959b086d10dac313d02b49d97860fa057 
>   kwin-decoration/data/oxygen.upd 2271c35e20f139625be419b2e9166aa35f9203a6 
>   kwin-decoration/data/update_oxygen.pl 
> da5c34e0a331ada6e0ef3be20838b3392a4ef051 
>   kwin-decoration/demo/CMakeLists.txt 
> 60c30180165a879ff3b3ab9980b47da186ce7ffc 
> 
> Diff: https://git.reviewboard.kde.org/r/117773/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Jonathan Riddell
> 
>

___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: Review Request 117773: make kwin-decoration co-installable, remove unused kconfig_update script and rename oxygen-shadow-demo to oxygen-shadow-demo5

[Hugo Pereira Da Costa]

> On April 28, 2014, 5 p.m., Aleix Pol Gonzalez wrote:
> > Does this demo really need to be installed?

To be honest, I guess, no ... 
Relic from debug days. (also I still uses every now and then, and right now it 
exposes a 'bug' in Qt raster graphicssystem).
Same would be true of oxygen/kstyles/demo/oxygen-demo5


- Hugo


---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/117773/#review56778
---


On April 25, 2014, 2:40 p.m., Jonathan Riddell wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/117773/
> ---
> 
> (Updated April 25, 2014, 2:40 p.m.)
> 
> 
> Review request for Plasma, Martin Gräßlin and Hugo Pereira Da Costa.
> 
> 
> Repository: oxygen
> 
> 
> Description
> ---
> 
> make kwin-decoration co-installable with kdelibs4 equivalents, remove unused 
> kconfig_update script and rename oxygen-shadow-demo to oxygen-shadow-demo5
> kconfig upgrade script not used as KF5 apps will start with a fresh config
> 
> 
> Diffs
> -
> 
>   kwin-decoration/CMakeLists.txt ff99b256e94507b364965f3b8d4401dd07f55d9a 
>   kwin-decoration/data/CMakeLists.txt 
> 7a4a2f7959b086d10dac313d02b49d97860fa057 
>   kwin-decoration/data/oxygen.upd 2271c35e20f139625be419b2e9166aa35f9203a6 
>   kwin-decoration/data/update_oxygen.pl 
> da5c34e0a331ada6e0ef3be20838b3392a4ef051 
>   kwin-decoration/demo/CMakeLists.txt 
> 60c30180165a879ff3b3ab9980b47da186ce7ffc 
> 
> Diff: https://git.reviewboard.kde.org/r/117773/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Jonathan Riddell
> 
>

___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: Review Request 117773: make kwin-decoration co-installable, remove unused kconfig_update script and rename oxygen-shadow-demo to oxygen-shadow-demo5

[Aleix Pol Gonzalez]

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/117773/#review56778
---


Does this demo really need to be installed?

- Aleix Pol Gonzalez


On April 25, 2014, 2:40 p.m., Jonathan Riddell wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/117773/
> ---
> 
> (Updated April 25, 2014, 2:40 p.m.)
> 
> 
> Review request for Plasma, Martin Gräßlin and Hugo Pereira Da Costa.
> 
> 
> Repository: oxygen
> 
> 
> Description
> ---
> 
> make kwin-decoration co-installable with kdelibs4 equivalents, remove unused 
> kconfig_update script and rename oxygen-shadow-demo to oxygen-shadow-demo5
> kconfig upgrade script not used as KF5 apps will start with a fresh config
> 
> 
> Diffs
> -
> 
>   kwin-decoration/CMakeLists.txt ff99b256e94507b364965f3b8d4401dd07f55d9a 
>   kwin-decoration/data/CMakeLists.txt 
> 7a4a2f7959b086d10dac313d02b49d97860fa057 
>   kwin-decoration/data/oxygen.upd 2271c35e20f139625be419b2e9166aa35f9203a6 
>   kwin-decoration/data/update_oxygen.pl 
> da5c34e0a331ada6e0ef3be20838b3392a4ef051 
>   kwin-decoration/demo/CMakeLists.txt 
> 60c30180165a879ff3b3ab9980b47da186ce7ffc 
> 
> Diff: https://git.reviewboard.kde.org/r/117773/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Jonathan Riddell
> 
>

___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: KDE VDG forums and "Ask a Dev" thread

[David Edmundson]

OK, sign me up.
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: KDE VDG forums and "Ask a Dev" thread

[Jens Reuterberg]

For starters to bridge the divide between community members and devs. To
make technical facts understood by the community and to, if that dev wants
to talk ideas.
Harassing him or her over past choices wouldn't help anyone so obviously
not that.
Den 28 apr 2014 16:32 skrev "David Edmundson" :

> What do you imagine the goal of it being?
>
> As a friendly getting to know some devs? To interrogate devs over
> unpopular decisions? For people to try and talk through new ideas and
> plans that people have?
> ___
> Plasma-devel mailing list
> Plasma-devel@kde.org
> https://mail.kde.org/mailman/listinfo/plasma-devel
>
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: Review Request 117773: make kwin-decoration co-installable, remove unused kconfig_update script and rename oxygen-shadow-demo to oxygen-shadow-demo5

[Hugo Pereira Da Costa]

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/117773/#review56777
---

Ship it!


looks good, as far as oxygen is concerned

- Hugo Pereira Da Costa


On April 25, 2014, 2:40 p.m., Jonathan Riddell wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/117773/
> ---
> 
> (Updated April 25, 2014, 2:40 p.m.)
> 
> 
> Review request for Plasma, Martin Gräßlin and Hugo Pereira Da Costa.
> 
> 
> Repository: oxygen
> 
> 
> Description
> ---
> 
> make kwin-decoration co-installable with kdelibs4 equivalents, remove unused 
> kconfig_update script and rename oxygen-shadow-demo to oxygen-shadow-demo5
> kconfig upgrade script not used as KF5 apps will start with a fresh config
> 
> 
> Diffs
> -
> 
>   kwin-decoration/CMakeLists.txt ff99b256e94507b364965f3b8d4401dd07f55d9a 
>   kwin-decoration/data/CMakeLists.txt 
> 7a4a2f7959b086d10dac313d02b49d97860fa057 
>   kwin-decoration/data/oxygen.upd 2271c35e20f139625be419b2e9166aa35f9203a6 
>   kwin-decoration/data/update_oxygen.pl 
> da5c34e0a331ada6e0ef3be20838b3392a4ef051 
>   kwin-decoration/demo/CMakeLists.txt 
> 60c30180165a879ff3b3ab9980b47da186ce7ffc 
> 
> Diff: https://git.reviewboard.kde.org/r/117773/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Jonathan Riddell
> 
>

___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: FYI: multi-threaded rendering is disabled in Qt for all of mesa

[Marco Martin]

On Monday 28 April 2014, Martin Graesslin wrote:
> 
> Given that we now have a setup used by most (Intel+Radeon) and a setup used
> by some (NVIDIA) users, it might make sense to switch to single-threaded
> by default. Investigating issues depending on whether single or multi
> threaded renderer will be used, sounds like lot of "fun" (works for me -
> oh are you using an NVIDIA?, Marco you are the only NVIDIA user, aren't
> you? Could you please look into it?)
> 

i can test:
nvidia
catalyst/radeonhd on a decent card
catalyst on a card from the middle ages

and yes, i would go single threaded for a couple of releases still (maybe with 
a runtime option for the brave?)

how can be disabled? only setting a variable? doesn't seem to exist a public 
api for that...

-- 
Marco Martin
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: FYI: multi-threaded rendering is disabled in Qt for all of mesa

[Martin Graesslin]

On Monday 28 April 2014 16:45:08 Marco Martin wrote:
> On Monday 28 April 2014, Martin Gräßlin wrote:
> > Hi all,
> > 
> > that's kind of a follow-up to my mail to try running KWin with
> > QSG_RENDER_LOOP=basic.
> > 
> > After I updated qtbase I noticed that my KWin became way more stable
> > without the env variable and looked at the blacklist and noticed the
> > commit copied below.
> > 
> > Looks like the asserts in xcb are history. It's not good that the threaded
> > renderer is disabled, but not crashing > multi threading.
> 
> interesting. (also, we have then to review our crashes to see what's
> reproduible now)
> 
> even tough, i'm not that optimist on nvidia and catalyst being less crashy
> towards multithreading :p

Given that we now have a setup used by most (Intel+Radeon) and a setup used by 
some (NVIDIA) users, it might make sense to switch to single-threaded by 
default. Investigating issues depending on whether single or multi threaded 
renderer will be used, sounds like lot of "fun" (works for me - oh are you 
using an NVIDIA?, Marco you are the only NVIDIA user, aren't you? Could you 
please look into it?)

At least for KWin I will prepare a change tomorrow.

Cheers
Martin

signature.asc
Description: This is a digitally signed message part.
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: Breeze repo

[Marco Martin]

On Monday 28 April 2014, Aleix Pol wrote:
> > ___
> > Plasma-devel mailing list
> > Plasma-devel@kde.org
> > https://mail.kde.org/mailman/listinfo/plasma-devel
> 
> They might tell you to go through kdereview, maybe we can put it in
> playground for now, so it can start to be worked on ASAP?

will see how they answer, as soon they created it i'll tell where it ended up, 
workspace or playground.

-- 
Marco Martin
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: Breeze repo

[Aleix Pol]

On Mon, Apr 28, 2014 at 6:07 PM, Marco Martin  wrote:

> On Monday 28 April 2014, Andrew Lake wrote:
> > Thanks so much Martin. I've been so swamped these past few days. The Qt
> > Quick Controls style can also be moved over (no real need to preserve the
> > history). I'm happy to do it later tonight if you're swamped right now.
> >
>
> ok.
> it will just take a while that the sysadmin request to create the repo is
> processed.
>
> --
> Marco Martin
> ___
> Plasma-devel mailing list
> Plasma-devel@kde.org
> https://mail.kde.org/mailman/listinfo/plasma-devel
>

They might tell you to go through kdereview, maybe we can put it in
playground for now, so it can start to be worked on ASAP?

Aleix
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: Review Request 117814: Make examples optional (default to off)

[David Edmundson]

> On April 28, 2014, 3:25 p.m., Marco Martin wrote:
> > good for me, is the same for me wether this is done explicitly here or in 
> > ecm if somebody else thinks they have the same usecase.

I'll ship it for now, I don't want examples installed on our release.


If ECM gets support, I can adjust pretty easily.


- David


---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/117814/#review56766
---


On April 28, 2014, 12:39 a.m., David Edmundson wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/117814/
> ---
> 
> (Updated April 28, 2014, 12:39 a.m.)
> 
> 
> Review request for Plasma.
> 
> 
> Repository: plasma-framework
> 
> 
> Description
> ---
> 
> Make examples optional (default to off)
> 
> 
> Most users won't want to have tech demos on their real system. 
> 
> 
> Diffs
> -
> 
>   CMakeLists.txt bc5d6d0 
> 
> Diff: https://git.reviewboard.kde.org/r/117814/diff/
> 
> 
> Testing
> ---
> 
> Build both variants editing flag with CCMake
> 
> 
> Thanks,
> 
> David Edmundson
> 
>

___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: Breeze repo

[Marco Martin]

On Monday 28 April 2014, Andrew Lake wrote:
> Thanks so much Martin. I've been so swamped these past few days. The Qt
> Quick Controls style can also be moved over (no real need to preserve the
> history). I'm happy to do it later tonight if you're swamped right now.
> 

ok.
it will just take a while that the sysadmin request to create the repo is 
processed.

-- 
Marco Martin
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: Breeze repo

[Andrew Lake]

Thanks so much Martin. I've been so swamped these past few days. The Qt
Quick Controls style can also be moved over (no real need to preserve the
history). I'm happy to do it later tonight if you're swamped right now.

Thanks again,
Andrew


On Mon, Apr 28, 2014 at 7:16 AM, Marco Martin  wrote:

> So, following the discussion last week, i made a scratch called breeze
> here:
> http://quickgit.kde.org/?p=scratch%2Fmart%2Fbreeze.git
>
> that at the moment has only the cursor theme.
>
> and opened a sysadmin ticket for the final repository location at
> kde/workspace/breeze.git
>
> Cheers,
> Marco Martin
> ___
> Plasma-devel mailing list
> Plasma-devel@kde.org
> https://mail.kde.org/mailman/listinfo/plasma-devel
>
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: Review Request 117814: Make examples optional (default to off)

[David Edmundson]

> On April 28, 2014, 6:44 a.m., Michael Palimaka wrote:
> > I wonder if it would be useful to make the option part of ECM, so we could 
> > propagate this to other frameworks too keeping it consistent.

Personally, I don't think so. It's only the one line, and I want to customise 
the message.

Plasma's usage is also a bit different from others; for kwidgetaddons (for 
example) wouldn't need to actually install the examples as everything can be 
run and seen from the build folder. In our case we need to install it for it to 
work.


- David


---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/117814/#review56705
---


On April 28, 2014, 12:39 a.m., David Edmundson wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/117814/
> ---
> 
> (Updated April 28, 2014, 12:39 a.m.)
> 
> 
> Review request for Plasma.
> 
> 
> Repository: plasma-framework
> 
> 
> Description
> ---
> 
> Make examples optional (default to off)
> 
> 
> Most users won't want to have tech demos on their real system. 
> 
> 
> Diffs
> -
> 
>   CMakeLists.txt bc5d6d0 
> 
> Diff: https://git.reviewboard.kde.org/r/117814/diff/
> 
> 
> Testing
> ---
> 
> Build both variants editing flag with CCMake
> 
> 
> Thanks,
> 
> David Edmundson
> 
>

___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: Review Request 117814: Make examples optional (default to off)

[Marco Martin]

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/117814/#review56766
---

Ship it!


good for me, is the same for me wether this is done explicitly here or in ecm 
if somebody else thinks they have the same usecase.

- Marco Martin


On April 28, 2014, 12:39 a.m., David Edmundson wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/117814/
> ---
> 
> (Updated April 28, 2014, 12:39 a.m.)
> 
> 
> Review request for Plasma.
> 
> 
> Repository: plasma-framework
> 
> 
> Description
> ---
> 
> Make examples optional (default to off)
> 
> 
> Most users won't want to have tech demos on their real system. 
> 
> 
> Diffs
> -
> 
>   CMakeLists.txt bc5d6d0 
> 
> Diff: https://git.reviewboard.kde.org/r/117814/diff/
> 
> 
> Testing
> ---
> 
> Build both variants editing flag with CCMake
> 
> 
> Thanks,
> 
> David Edmundson
> 
>

___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: stracing KWin

[Marco Martin]

On Monday 28 April 2014, Martin Gräßlin wrote:
> On Monday 28 April 2014 15:21:19 Aleix Pol wrote:
> > Hi,
> > I was looking something up and I realized it keeps looking for an
> > unexistant "plasma_shell" icon all the time, maybe you know what it could
> > be going on?
> 
> it's probably trying to find an icon for a plasma dialog. E.g. for kickoff
> or the systray. Relevant code should be in KWin::Client::getIcons()
> 
> Might make sense to set a proper icon in plasma.

now all Dialogs (in the end if works other window will have to have the same 
threatment) have a default icon, the usual cashew (irrelevant anyways since is 
never shown anywhere)

Aleix: can you see if it's complaining a bit less now?

-- 
Marco Martin
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: FYI: multi-threaded rendering is disabled in Qt for all of mesa

[Marco Martin]

On Monday 28 April 2014, Martin Gräßlin wrote:
> Hi all,
> 
> that's kind of a follow-up to my mail to try running KWin with
> QSG_RENDER_LOOP=basic.
> 
> After I updated qtbase I noticed that my KWin became way more stable
> without the env variable and looked at the blacklist and noticed the
> commit copied below.
> 
> Looks like the asserts in xcb are history. It's not good that the threaded
> renderer is disabled, but not crashing > multi threading.

interesting. (also, we have then to review our crashes to see what's 
reproduible now)

even tough, i'm not that optimist on nvidia and catalyst being less crashy 
towards multithreading :p


-- 
Marco Martin
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

FYI: multi-threaded rendering is disabled in Qt for all of mesa

[Martin Gräßlin]

Hi all,

that's kind of a follow-up to my mail to try running KWin with 
QSG_RENDER_LOOP=basic.

After I updated qtbase I noticed that my KWin became way more stable without 
the env variable and looked at the blacklist and noticed the commit copied 
below.

Looks like the asserts in xcb are history. It's not good that the threaded 
renderer is disabled, but not crashing > multi threading.

Cheers
Martin

commit 6977700bed929c7f1a20192d07a06abc2fdca278
Author: Ulf Hermann 
Date:   Thu Apr 10 15:51:56 2014 +0200

Disable threaded rendering if Mesa is used

Mesa and xcb show some bad interaction which leads to frequent crashed
on multithreaded access. Also, the selective approach to blacklisting
only specific chipsets isn't feasible, given the resources available.

The client glx vendor string is used to identify mesa instead of the
server GL vendor and/or renderer string as that is much more reliable.

Task-number: QTBUG-38221
Change-Id: I2d8c037aa4fd9c38eb9537452a5e7e62f72a081d
Reviewed-by: Gunnar Sletta 



signature.asc
Description: This is a digitally signed message part.
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: KDE VDG forums and "Ask a Dev" thread

[David Edmundson]

What do you imagine the goal of it being?

As a friendly getting to know some devs? To interrogate devs over
unpopular decisions? For people to try and talk through new ideas and
plans that people have?
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Breeze repo

[Marco Martin]

So, following the discussion last week, i made a scratch called breeze here:
http://quickgit.kde.org/?p=scratch%2Fmart%2Fbreeze.git

that at the moment has only the cursor theme.

and opened a sysadmin ticket for the final repository location at
kde/workspace/breeze.git

Cheers,
Marco Martin
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: KDE VDG forums and "Ask a Dev" thread

[Jens Reuterberg]

Good point but I still like the "one place to find old comments" 
PLUS I thought it would look more unified... Although your example 
makes it more tidy... Lets start the first one and then see where 
we go from there, shall we? One new each week if that feels ok?

(also another good point for your thing is that WHEN the week is 
over it can be a general Kwin thing? Like the Qt thing or the 
Inkscape thread? "No question to stupid" - like a sex advice column 
without the sex?)

(*Writing up Martin as the first dev to have an AMA with* ;) )

On Monday 28 April 2014 11.09.32 Martin Gräßlin wrote:
> On Monday 28 April 2014 10:54:24 Jens Reuterberg wrote:
> > Well either that or we did it a week event - I mean it's not like 
it's
> > gonna be ten thousand questions and most of them will be
> > considering design, layout and usability - so say if a dev 
promised to
> > pop in once a day and answer the questions that's cropped up?
> 
> wouldn't it make more sense to have a different thread then per 
dev who
> answers. E.g. having a long living thread for all general KWin things 
people
> want to know?

___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: stracing KWin

[Martin Gräßlin]

On Monday 28 April 2014 15:21:19 Aleix Pol wrote:
> Hi,
> I was looking something up and I realized it keeps looking for an
> unexistant "plasma_shell" icon all the time, maybe you know what it could
> be going on?

it's probably trying to find an icon for a plasma dialog. E.g. for kickoff or 
the systray. Relevant code should be in KWin::Client::getIcons()

Might make sense to set a proper icon in plasma.

Cheers
Martin

signature.asc
Description: This is a digitally signed message part.
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Review Request 117839: [kglobalaccel] Update X11 appTime from key press events

[Martin Gräßlin]

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/117839/
---

Review request for Plasma.


Repository: plasma-workspace


Description
---

[kglobalaccel] Update X11 appTime from key press events

KGlobalAccelD sends the current appTime through DBus to the application
and KF5::GlobalAccel updates the appTime from the timestamp. This is
needed for following actions to succeed. E.g. KWin needs the updated
appTime for grabbing the keyboard following the kill window shortcut.

To get the current timestamp we just use the timestamp of the key press
event delivered to kglobalacceld.


Diffs
-

  kglobalaccel/kglobalaccel_x11.cpp 454c62e17ac0fb188bbbf2bcefcf42241d7f4ec1 

Diff: https://git.reviewboard.kde.org/r/117839/diff/


Testing
---

Tested with ctrl+alt+esc


Thanks,

Martin Gräßlin

___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: stracing KWin

[Marco Martin]

On Monday 28 April 2014, Aleix Pol wrote:
> Hi,
> I was looking something up and I realized it keeps looking for an
> unexistant "plasma_shell" icon all the time, maybe you know what it could
> be going on?

maybe it looks for an icon with the name as the component of the app for all 
windows that don't have an icon?


-- 
Marco Martin
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

stracing KWin

[Aleix Pol]

Hi,
I was looking something up and I realized it keeps looking for an
unexistant "plasma_shell" icon all the time, maybe you know what it could
be going on?

To reproduce (the result is after ~10s of runtime):
$ strace -e file kwin --replace  2> cosa &
$ grep plasma_shell. cosa  | grep stat | wc -l
10464

Aleix
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: Review Request 117837: move argument parsing logic inside application from libs

[Bhushan Shah]

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/117837/
---

(Updated April 28, 2014, 6:47 p.m.)


Status
--

This change has been discarded.


Review request for Plasma, Shantanu Tushar, Sinny Kumari, and Sujith Haridasan.


Repository: plasma-mediacenter


Description
---

Argument parsing should not be done inside library.


Diffs
-

  libs/mediacenter/playlistmodel.h 0de7c56 
  libs/mediacenter/playlistmodel.cpp b233f18 
  shells/newshell/mainwindow.cpp c1be61a 

Diff: https://git.reviewboard.kde.org/r/117837/diff/


Testing
---

Opened single file with pmc and multiple files with it.


Thanks,

Bhushan Shah

___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: Review Request 117813: Make the system tray faster

[Mark Gaiser]

> On April 28, 2014, 11:06 a.m., Mark Gaiser wrote:
> > applets/systemtray/plugin/protocols/plasmoid/plasmoidtask.cpp, lines 61-63
> > 
> >
> > Ahh, this is tricky to follow. You are invoking the init function here 
> > _and_ earlier on using a singleShot timer if this translates to the 
> > Host::init method. If that's the case then i think you might be safe in 
> > removing this line and the singleshot and just call init() from the 
> > constructor.
> 
> Marco Martin wrote:
> no, it isn't init of Host.
> this is the init() of m_taskGraphicsObject, ie the init() of the 
> AppletInterface of the loaded plasmoid (is something that shouldn't even be 
> neded in theory, but preloading the applets seems to avoid some crash 
> scenarios)

Right, now it makes sense. Thank you for the clarification.


> On April 28, 2014, 11:06 a.m., Mark Gaiser wrote:
> > applets/systemtray/plugin/host.cpp, line 120
> > 
> >
> > just init(); ?
> > I don't see a reason why you need a single shot timer here.
> 
> Marco Martin wrote:
> this causes the tasks to be loaded after the applet is done, in turn, 
> making the whole workspace ui to load faster (and task icons appearing after 
> when the rest is done) is a little hack that does huge difference in 
> perceived startup speed

Sounds like a good (undocumented) reason :)


> On April 28, 2014, 11:06 a.m., Mark Gaiser wrote:
> > applets/systemtray/plugin/tasklistmodel.cpp, line 99
> > 
> >
> > Are you sure this is ok?
> > For example, if you add one row (and have none) this should translate 
> > to:
> > beginInsertRows(QModelIndex(), 0, 1);
> > 
> > I think you need something like:
> > int startCount = (rowCount() - 1 >= 0) ? rowCount() : 0;
> > beginInsertRows(QModelIndex(), startCount, startCount + 1);
> 
> David Edmundson wrote:
> beginInsertRows(QModelIndex(), 0, 1)
> 
> is saying you are inserting 2 rows, starting at 0 ending at 1. It's an 
> annoying QAbstractItemModel API.

Ok, makes sense now. I'm always confused by the start and end numbers.


> On April 28, 2014, 11:06 a.m., Mark Gaiser wrote:
> > applets/systemtray/plugin/tasklistmodel.cpp, line 103
> > 
> >
> > Remove, the model will notify QML. No need to do this yourself.
> 
> David Edmundson wrote:
> I have it as a property, so I need to tell Qt the property has changed.
> 
> I could do connect(this, SIGNAL("rowsInserted(...), this, 
> SIGNAL(rowCountChanged());
> I could do connect(this, SIGNAL("rowsRemoved(...), this, 
> SIGNAL(rowCountChanged());
> + rowsMoved + layoutChanged + modelReset.. 
> so I deemed this easier.

As explained in person. The view will know this so the line can be removed.


> On April 28, 2014, 11:06 a.m., Mark Gaiser wrote:
> > applets/systemtray/plugin/tasklistmodel.cpp, line 115
> > 
> >
> > Remove, the model will notify QML. No need to do this yourself.
> 
> Marco Martin wrote:
> the model does, but the exported count property needs to be notified by 
> hand

If you keep the property, yes. But there is no need for the property to exist 
at all because the view knows the model count.


- Mark


---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/117813/#review56724
---


On April 27, 2014, 10:51 p.m., David Edmundson wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/117813/
> ---
> 
> (Updated April 27, 2014, 10:51 p.m.)
> 
> 
> Review request for Plasma.
> 
> 
> Repository: plasma-workspace
> 
> 
> Description
> ---
> 
> Port QQmlListProperty to QAbstractListModel.
> QQmlListProperty only has a signal that the list has changed.This means when 
> used in a ListView every delegate has to be redone whenever a single item is 
> inserted or removed rather than just moved.
> 
> Given TaskDelegate is not the simplest of things this has a performance gain, 
> most noticeably on startup. Also rather than sorting all items after an 
> insert items are inserted in the right place using qLowerBound. Now we have 
> the correct signals we can remove the compression, they won't add anything. 
> 
> 
> Other commits:
> 
> Avoid constructing a QString for comparing, use QLatin1String for == 
> operators.
> 
> Remove useless include
> 
> Do not construct a map inside a lessThan funct

Review Request 117837: move argument parsing logic inside application from libs

[Bhushan Shah]

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/117837/
---

Review request for Plasma, Shantanu Tushar, Sinny Kumari, and Sujith Haridasan.


Repository: plasma-mediacenter


Description
---

Argument parsing should not be done inside library.


Diffs
-

  libs/mediacenter/playlistmodel.h 0de7c56 
  libs/mediacenter/playlistmodel.cpp b233f18 
  shells/newshell/mainwindow.cpp c1be61a 

Diff: https://git.reviewboard.kde.org/r/117837/diff/


Testing
---

Opened single file with pmc and multiple files with it.


Thanks,

Bhushan Shah

___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: Polishing sprint in May: dates?

[Luca Beltrame]

In data lunedì 28 aprile 2014 13:35:37, Sebastian Kügler ha scritto:

> focus for a while on immediate user feedback -- let's face it, many people
> just start testing once we declare stable.

With my forum hat on: yes, the vast majority at least, according to my 
(limited!) experience.

-- 
Luca Beltrame, Ph.D.
Translational Genomics Unit, Department of Oncology
IRCCS Istituto di Ricerche Farmacologiche "Mario Negri"


signature.asc
Description: This is a digitally signed message part.
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: RFC: oxygen and air plasma themes in oxygen.git?

[Marco Martin]

On Monday 28 April 2014, Martin Gräßlin wrote:
> > so, i would proceed moving them there (not even sure how much if any
> > history is needed especially becasue binary files)
> 
> does it make sense there? Neither the oxygen nor the air theme are in any
> way integrated with other Oxygen elements. Maybe moving to kdeartwork or
> something like that would be better?

boh, their difference was by design, so they are integrated in a way as "that 
was the vision"

as in kde-artwork, it would probably make sense there, but at the moment is 
strictly kde4 oriented and even worse still on subversion (with commits 
disabled since a while, as far i know)

so yeah, kde-arwork should be resurrected in a sensible way, but it's probably 
beyond scope of the first release.

-- 
Marco Martin
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: RFC: oxygen and air plasma themes in oxygen.git?

[Martin Gräßlin]

On Monday 28 April 2014 13:42:51 Marco Martin wrote:
> Hi all,
> right now the oxygen and air themes are still in pasma-framework (for lack
> of better place atm)
> since for transition purpose is good to still offer an old look, i think the
> oxygen repo would be a good place for it (especially in future releases
> when we'll have some kind of mega-theme support)
> 
> so, i would proceed moving them there (not even sure how much if any history
> is needed especially becasue binary files)

does it make sense there? Neither the oxygen nor the air theme are in any way 
integrated with other Oxygen elements. Maybe moving to kdeartwork or something 
like that would be better?

Cheers
Martin

signature.asc
Description: This is a digitally signed message part.
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

RFC: oxygen and air plasma themes in oxygen.git?

[Marco Martin]

Hi all,
right now the oxygen and air themes are still in pasma-framework (for lack of 
better place atm)
since for transition purpose is good to still offer an old look, i think the 
oxygen repo would be a good place for it (especially in future releases when 
we'll have some kind of mega-theme support)

so, i would proceed moving them there (not even sure how much if any history 
is needed especially becasue binary files)

comments?

-- 
Marco Martin
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: Polishing sprint in May: dates?

[Sebastian Kügler]

On Monday, April 28, 2014 13:33:14 Marco Martin wrote:
> On Monday 28 April 2014, Sebastian Kügler wrote:
> > On Monday, April 28, 2014 12:59:15 Marco Martin wrote:
> > > On Monday 28 April 2014, Sebastian Kügler wrote:
> > > 
> > >
> > > we could try to do a planning one shortly after release..
> >
> > 
> >
> > But then, we have Akademy as well, where this would fit wonderfully...
> 
> true enough
> (was just thinking that is very late this year)

Maybe that's not that bad. Normally, after major releases, it's good to focus 
for a while on immediate user feedback -- let's face it, many people just 
start testing once we declare stable.
-- 
sebas

http://www.kde.org | http://vizZzion.org | GPG Key ID: 9119 0EF9
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: Polishing sprint in May: dates?

[Marco Martin]

On Monday 28 April 2014, Sebastian Kügler wrote:
> On Monday, April 28, 2014 12:59:15 Marco Martin wrote:
> > On Monday 28 April 2014, Sebastian Kügler wrote:
> > > We just discussed this during our hangout, and due to being rather late
> > > with all preps, we decided to cancel this sprint, and concentrate on
> > > bugs from our remote locations.
> > 
> > we could try to do a planning one shortly after release..
> 
> But then, we have Akademy as well, where this would fit wonderfully...

true enough ;)
(was just thinking that is very late this year)

-- 
Marco Martin
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: Polishing sprint in May: dates?

[Sebastian Kügler]

On Monday, April 28, 2014 12:59:15 Marco Martin wrote:
> On Monday 28 April 2014, Sebastian Kügler wrote:
> > We just discussed this during our hangout, and due to being rather late
> > with all preps, we decided to cancel this sprint, and concentrate on bugs
> > from our remote locations.
> 
> we could try to do a planning one shortly after release..

But then, we have Akademy as well, where this would fit wonderfully...
-- 
sebas

http://www.kde.org | http://vizZzion.org | GPG Key ID: 9119 0EF9
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: Review Request 117813: Make the system tray faster

[Marco Martin]

> On April 28, 2014, 11:06 a.m., Mark Gaiser wrote:
> > applets/systemtray/plugin/host.cpp, line 120
> > 
> >
> > just init(); ?
> > I don't see a reason why you need a single shot timer here.

this causes the tasks to be loaded after the applet is done, in turn, making 
the whole workspace ui to load faster (and task icons appearing after when the 
rest is done) is a little hack that does huge difference in perceived startup 
speed


> On April 28, 2014, 11:06 a.m., Mark Gaiser wrote:
> > applets/systemtray/plugin/protocols/plasmoid/plasmoidtask.cpp, lines 61-63
> > 
> >
> > Ahh, this is tricky to follow. You are invoking the init function here 
> > _and_ earlier on using a singleShot timer if this translates to the 
> > Host::init method. If that's the case then i think you might be safe in 
> > removing this line and the singleshot and just call init() from the 
> > constructor.

no, it isn't init of Host.
this is the init() of m_taskGraphicsObject, ie the init() of the 
AppletInterface of the loaded plasmoid (is something that shouldn't even be 
neded in theory, but preloading the applets seems to avoid some crash scenarios)


> On April 28, 2014, 11:06 a.m., Mark Gaiser wrote:
> > applets/systemtray/plugin/tasklistmodel.cpp, line 115
> > 
> >
> > Remove, the model will notify QML. No need to do this yourself.

the model does, but the exported count property needs to be notified by hand


- Marco


---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/117813/#review56724
---


On April 27, 2014, 10:51 p.m., David Edmundson wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/117813/
> ---
> 
> (Updated April 27, 2014, 10:51 p.m.)
> 
> 
> Review request for Plasma.
> 
> 
> Repository: plasma-workspace
> 
> 
> Description
> ---
> 
> Port QQmlListProperty to QAbstractListModel.
> QQmlListProperty only has a signal that the list has changed.This means when 
> used in a ListView every delegate has to be redone whenever a single item is 
> inserted or removed rather than just moved.
> 
> Given TaskDelegate is not the simplest of things this has a performance gain, 
> most noticeably on startup. Also rather than sorting all items after an 
> insert items are inserted in the right place using qLowerBound. Now we have 
> the correct signals we can remove the compression, they won't add anything. 
> 
> 
> Other commits:
> 
> Avoid constructing a QString for comparing, use QLatin1String for == 
> operators.
> 
> Remove useless include
> 
> Do not construct a map inside a lessThan function
> 
> lessThan functions have to be fast.
> Also Map -> Hash as we're not using order here.
> 
> 
> Diffs
> -
> 
>   applets/systemtray/package/contents/ui/ExpandedRepresentation.qml 2ef180b 
>   applets/systemtray/package/contents/ui/PlasmoidItem.qml 0eb1687 
>   applets/systemtray/package/contents/ui/StatusNotifierItem.qml fc889a8 
>   applets/systemtray/package/contents/ui/TaskDelegate.qml 913d8f1 
>   applets/systemtray/package/contents/ui/TaskListDelegate.qml 5501e02 
>   applets/systemtray/plugin/CMakeLists.txt f6e23b4 
>   applets/systemtray/plugin/host.h 02c5bbe 
>   applets/systemtray/plugin/host.cpp eafd0b6 
>   applets/systemtray/plugin/protocols/plasmoid/plasmoidtask.cpp 2b846f2 
>   applets/systemtray/plugin/tasklistmodel.h PRE-CREATION 
>   applets/systemtray/plugin/tasklistmodel.cpp PRE-CREATION 
> 
> Diff: https://git.reviewboard.kde.org/r/117813/diff/
> 
> 
> Testing
> ---
> 
> Seems to work :)
> 
> see branch davidedmundson/faster_systray to test
> 
> 
> Thanks,
> 
> David Edmundson
> 
>

___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: Minutes Monday Plasma hangout

[Ivan Čukić]

On Monday 28 Apr 2014 12:50:41 Sebastian Kügler wrote:
> Plasma Hangout 28-4-2014
> 
> Present: Antonis, David, Jonathan, Marco, Martin G., Martin K., Rohan,
> Sebastian, Aleix

Ivan:
 - Working on kamd database stuff


-- 
 A program that has not been tested does not work.
  -- Bjarne Stroustrup

___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: Review Request 117813: Make the system tray faster

[David Edmundson]

> On April 28, 2014, 11:06 a.m., Mark Gaiser wrote:
> > applets/systemtray/plugin/tasklistmodel.cpp, line 103
> > 
> >
> > Remove, the model will notify QML. No need to do this yourself.

I have it as a property, so I need to tell Qt the property has changed.

I could do connect(this, SIGNAL("rowsInserted(...), this, 
SIGNAL(rowCountChanged());
I could do connect(this, SIGNAL("rowsRemoved(...), this, 
SIGNAL(rowCountChanged());
+ rowsMoved + layoutChanged + modelReset.. 
so I deemed this easier.


> On April 28, 2014, 11:06 a.m., Mark Gaiser wrote:
> > applets/systemtray/plugin/tasklistmodel.cpp, line 99
> > 
> >
> > Are you sure this is ok?
> > For example, if you add one row (and have none) this should translate 
> > to:
> > beginInsertRows(QModelIndex(), 0, 1);
> > 
> > I think you need something like:
> > int startCount = (rowCount() - 1 >= 0) ? rowCount() : 0;
> > beginInsertRows(QModelIndex(), startCount, startCount + 1);

beginInsertRows(QModelIndex(), 0, 1)

is saying you are inserting 2 rows, starting at 0 ending at 1. It's an annoying 
QAbstractItemModel API.


- David


---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/117813/#review56724
---


On April 27, 2014, 10:51 p.m., David Edmundson wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/117813/
> ---
> 
> (Updated April 27, 2014, 10:51 p.m.)
> 
> 
> Review request for Plasma.
> 
> 
> Repository: plasma-workspace
> 
> 
> Description
> ---
> 
> Port QQmlListProperty to QAbstractListModel.
> QQmlListProperty only has a signal that the list has changed.This means when 
> used in a ListView every delegate has to be redone whenever a single item is 
> inserted or removed rather than just moved.
> 
> Given TaskDelegate is not the simplest of things this has a performance gain, 
> most noticeably on startup. Also rather than sorting all items after an 
> insert items are inserted in the right place using qLowerBound. Now we have 
> the correct signals we can remove the compression, they won't add anything. 
> 
> 
> Other commits:
> 
> Avoid constructing a QString for comparing, use QLatin1String for == 
> operators.
> 
> Remove useless include
> 
> Do not construct a map inside a lessThan function
> 
> lessThan functions have to be fast.
> Also Map -> Hash as we're not using order here.
> 
> 
> Diffs
> -
> 
>   applets/systemtray/package/contents/ui/ExpandedRepresentation.qml 2ef180b 
>   applets/systemtray/package/contents/ui/PlasmoidItem.qml 0eb1687 
>   applets/systemtray/package/contents/ui/StatusNotifierItem.qml fc889a8 
>   applets/systemtray/package/contents/ui/TaskDelegate.qml 913d8f1 
>   applets/systemtray/package/contents/ui/TaskListDelegate.qml 5501e02 
>   applets/systemtray/plugin/CMakeLists.txt f6e23b4 
>   applets/systemtray/plugin/host.h 02c5bbe 
>   applets/systemtray/plugin/host.cpp eafd0b6 
>   applets/systemtray/plugin/protocols/plasmoid/plasmoidtask.cpp 2b846f2 
>   applets/systemtray/plugin/tasklistmodel.h PRE-CREATION 
>   applets/systemtray/plugin/tasklistmodel.cpp PRE-CREATION 
> 
> Diff: https://git.reviewboard.kde.org/r/117813/diff/
> 
> 
> Testing
> ---
> 
> Seems to work :)
> 
> see branch davidedmundson/faster_systray to test
> 
> 
> Thanks,
> 
> David Edmundson
> 
>

___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: Review Request 117813: Make the system tray faster

[Mark Gaiser]

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/117813/#review56724
---



applets/systemtray/plugin/host.cpp


Looks like it can all be removed.



applets/systemtray/plugin/host.cpp


just init(); ?
I don't see a reason why you need a single shot timer here.



applets/systemtray/plugin/protocols/plasmoid/plasmoidtask.cpp


Ahh, this is tricky to follow. You are invoking the init function here 
_and_ earlier on using a singleShot timer if this translates to the Host::init 
method. If that's the case then i think you might be safe in removing this line 
and the singleshot and just call init() from the constructor.



applets/systemtray/plugin/tasklistmodel.h


Not needed, the model will notify QML when the row count changes.



applets/systemtray/plugin/tasklistmodel.cpp


Perhaps try do these test before calling this function via qLowerBound.

It would certainly be faster, but also more complicated so i don't know if 
it's worth it.



applets/systemtray/plugin/tasklistmodel.cpp


I don't think you plan on changing this at runtime - ever. You might want 
to change this to a const and use initializer lists (a C++11 feature). It works 
like this:

const QHash roleNames { {Qt::UserRole, "name"}, ..., ...};



applets/systemtray/plugin/tasklistmodel.cpp


Are you sure this is ok?
For example, if you add one row (and have none) this should translate to:
beginInsertRows(QModelIndex(), 0, 1);

I think you need something like:
int startCount = (rowCount() - 1 >= 0) ? rowCount() : 0;
beginInsertRows(QModelIndex(), startCount, startCount + 1); 



applets/systemtray/plugin/tasklistmodel.cpp


Remove, the model will notify QML. No need to do this yourself.



applets/systemtray/plugin/tasklistmodel.cpp


Remove, the model will notify QML. No need to do this yourself.


- Mark Gaiser


On April 27, 2014, 10:51 p.m., David Edmundson wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/117813/
> ---
> 
> (Updated April 27, 2014, 10:51 p.m.)
> 
> 
> Review request for Plasma.
> 
> 
> Repository: plasma-workspace
> 
> 
> Description
> ---
> 
> Port QQmlListProperty to QAbstractListModel.
> QQmlListProperty only has a signal that the list has changed.This means when 
> used in a ListView every delegate has to be redone whenever a single item is 
> inserted or removed rather than just moved.
> 
> Given TaskDelegate is not the simplest of things this has a performance gain, 
> most noticeably on startup. Also rather than sorting all items after an 
> insert items are inserted in the right place using qLowerBound. Now we have 
> the correct signals we can remove the compression, they won't add anything. 
> 
> 
> Other commits:
> 
> Avoid constructing a QString for comparing, use QLatin1String for == 
> operators.
> 
> Remove useless include
> 
> Do not construct a map inside a lessThan function
> 
> lessThan functions have to be fast.
> Also Map -> Hash as we're not using order here.
> 
> 
> Diffs
> -
> 
>   applets/systemtray/package/contents/ui/ExpandedRepresentation.qml 2ef180b 
>   applets/systemtray/package/contents/ui/PlasmoidItem.qml 0eb1687 
>   applets/systemtray/package/contents/ui/StatusNotifierItem.qml fc889a8 
>   applets/systemtray/package/contents/ui/TaskDelegate.qml 913d8f1 
>   applets/systemtray/package/contents/ui/TaskListDelegate.qml 5501e02 
>   applets/systemtray/plugin/CMakeLists.txt f6e23b4 
>   applets/systemtray/plugin/host.h 02c5bbe 
>   applets/systemtray/plugin/host.cpp eafd0b6 
>   applets/systemtray/plugin/protocols/plasmoid/plasmoidtask.cpp 2b846f2 
>   applets/systemtray/plugin/tasklistmodel.h PRE-CREATION 
>   applets/systemtray/plugin/tasklistmodel.cpp PRE-CREATION 
> 
> Diff: https://git.reviewboard.kde.org/r/117813/diff/
> 
> 
> Testing
> ---
> 
> Seems to work :)
> 
> see branch davidedmundson/faster_systray to test
> 
> 
> Thanks,
> 
> David Edmundson
> 
>

___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: Polishing sprint in May: dates?

[Marco Martin]

On Monday 28 April 2014, Sebastian Kügler wrote:
> > http://doodle.com/4qygezcx9zfnvahb
> > 
> > Cheers,
> 
> We just discussed this during our hangout, and due to being rather late
> with all preps, we decided to cancel this sprint, and concentrate on bugs
> from our remote locations.

we could try to do a planning one shortly after release..

-- 
Marco Martin
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: Polishing sprint in May: dates?

[Sebastian Kügler]

On Wednesday, March 19, 2014 12:18:11 Sebastian Kügler wrote:
> Hey,
> 
> A bunch of us want to get together in May to put finishing touches in Plasma
> Next. This sprint is important because it allows us to sit down and fix
> remaining bugs.
> 
> I am proposing dates here, though this is meant as a "soft proposal", May is
> a busy month, and we can't realistically get everyone into the same place.
> But even just with a few of us, it will be worth it. We will meet in the
> Blue Systems Barcelona office, if possible for a week of frantic bugfixing.
> 
> I'd like to ask the KDE e.V. for funding of some of the costs, and some
> people might need visa, so let's start early, and get the planning done
> asap. We might not be able to get the e.V. to fund some of the costs, since
> we already had a sprint in January, but let's try nevertheless.
> 
> Please specify the dates that you could make to Barcelona until Friday
> evening, latest, so we can proceed with the planning.
> 
> http://doodle.com/4qygezcx9zfnvahb
> 
> Cheers,

We just discussed this during our hangout, and due to being rather late with 
all preps, we decided to cancel this sprint, and concentrate on bugs from our 
remote locations.

Cheers,
-- 
sebas

http://www.kde.org | http://vizZzion.org | GPG Key ID: 9119 0EF9
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Minutes Monday Plasma hangout

[Sebastian Kügler]

Plasma Hangout 28-4-2014

Present: Antonis, David, Jonathan, Marco, Martin G., Martin K., Rohan, 
Sebastian, Aleix

Antonis:
  - refactored Plasmate into handlers (separately tested, has tests)
  
David:
  - worked on SDDM upstream
  - themeing copied from Plasma
  - working architectural problems (greeter running as root)
  - optimizations in the systemtray
  - worked on "useful stuff" during frameworks sprint

Jonathan:
  - new frameworks release coming up: please make sure everything is in beta 
shape next tuesday
  - Cleaning up Oxygen: what's really needed, co-installability

Aleix:
  - frameworks sprint
  - changed install locations from kde5/services and kde5/servicetypes is now 
in kservices5 and kservicetypes5: You will need clean builds!
  - install location of some components has changed dekstopthemes are now 
inside plasma/ directory

Marco:
  - pounding on dialog (sizing behavior and stability)
  - Would like to know about API issues, speak no or stfu forever (until 
Plasma 6)!
  - streamlined bugzilla components (in line now with other frameworks)
  - splashscreen branch in plasma-workspace davidedmundson/newsplash planned 
for merge
  - to create a breeze repository with mouse theme and other bits as they 
arrive
  - plans bugfixing
  - sprint in May: cancelled, due to being behind on preps too much, 
concentrate on remote bugfixing
  
Martin G:
  - bugfixing, bugfixing ... and bugfixing

Martin K:
  - framworks bug triaging
  - working on memleak in notifications' dialog handling
  - testing new breeze theme with Andrew and Jens

Rohan:
  - working with David to land SDDM on the ISO
  - shortcuts fixed on the ISO
  - kicker is also on the ISO now
  
Sebastian:
  - was on vacation
  - last week: churned through emails, poked a bit at jenkins, should be no 
more red now
  - caught up with builds (during frameworks sprint :/): starts now again
  - I should be fully useful again now, will work on bugs

-- 
sebas

http://www.kde.org | http://vizZzion.org | GPG Key ID: 9119 0EF9
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: Review Request 117757: Improve Kickoff's scrolling when changing categories

[Commit Hook]

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/117757/#review56727
---


This review has been submitted with commit 
e619f39d466b716c8fb48bd5f4663a72193d2f13 by Martin Klapetek to branch master.

- Commit Hook


On April 24, 2014, 8:14 p.m., Martin Klapetek wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/117757/
> ---
> 
> (Updated April 24, 2014, 8:14 p.m.)
> 
> 
> Review request for Plasma.
> 
> 
> Repository: plasma-desktop
> 
> 
> Description
> ---
> 
> Use case: I'm in All Applications, I scroll down, select Utilities, category 
> switches inside Utilities and the scrollbar is at the same bottom position 
> from my first scroll, which means that I see the apps in Utilities that are 
> at the bottom instead starting from the top.
> 
> These are two changes: always scroll the view to top when going into a 
> subcategory and then always center on the parent category when going level 
> up. The breadcrumbs however store QModelIndex (which makes sense) and 
> QModelIndex cannot be used from QML, so I added a helper method to the model 
> that returns a row of the given index and the view then centers on that row 
> when going category up. Might not be good at all but imho works&looks better.
> 
> 
> Diffs
> -
> 
>   applets/kickoff/core/applicationmodel.h a039773 
>   applets/kickoff/core/applicationmodel.cpp 3a5f0bb 
>   applets/kickoff/package/contents/ui/ApplicationsView.qml 09b7618 
> 
> Diff: https://git.reviewboard.kde.org/r/117757/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Martin Klapetek
> 
>

___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: Review Request 117757: Improve Kickoff's scrolling when changing categories

[Martin Klapetek]

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/117757/
---

(Updated April 28, 2014, 10:16 a.m.)


Status
--

This change has been marked as submitted.


Review request for Plasma.


Repository: plasma-desktop


Description
---

Use case: I'm in All Applications, I scroll down, select Utilities, category 
switches inside Utilities and the scrollbar is at the same bottom position from 
my first scroll, which means that I see the apps in Utilities that are at the 
bottom instead starting from the top.

These are two changes: always scroll the view to top when going into a 
subcategory and then always center on the parent category when going level up. 
The breadcrumbs however store QModelIndex (which makes sense) and QModelIndex 
cannot be used from QML, so I added a helper method to the model that returns a 
row of the given index and the view then centers on that row when going 
category up. Might not be good at all but imho works&looks better.


Diffs
-

  applets/kickoff/core/applicationmodel.h a039773 
  applets/kickoff/core/applicationmodel.cpp 3a5f0bb 
  applets/kickoff/package/contents/ui/ApplicationsView.qml 09b7618 

Diff: https://git.reviewboard.kde.org/r/117757/diff/


Testing
---


Thanks,

Martin Klapetek

___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: Review Request 117813: Make the system tray faster

[Sebastian Kügler]

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/117813/#review56719
---


I like muchos. Haven't tested it yet, as I'm still resurrecting my builds, so 
I'm ok with shipping it, but don't feel comfortable "ship it"ing it myself. :)

Good work!


applets/systemtray/package/contents/ui/StatusNotifierItem.qml


The answer is easy: It also wastes memory! ;-)

More seriously, I think it's a leftover from the old implementation and can 
be removed throughout.



applets/systemtray/package/contents/ui/TaskDelegate.qml


It is used to determine whether we're a delegate in the hidden section (so 
with text), or in the panel section (square).



applets/systemtray/plugin/host.cpp


no spaces around Task



applets/systemtray/plugin/tasklistmodel.h


"a model representing items for the system tray"?



applets/systemtray/plugin/tasklistmodel.cpp


see above


- Sebastian Kügler


On April 27, 2014, 10:51 p.m., David Edmundson wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/117813/
> ---
> 
> (Updated April 27, 2014, 10:51 p.m.)
> 
> 
> Review request for Plasma.
> 
> 
> Repository: plasma-workspace
> 
> 
> Description
> ---
> 
> Port QQmlListProperty to QAbstractListModel.
> QQmlListProperty only has a signal that the list has changed.This means when 
> used in a ListView every delegate has to be redone whenever a single item is 
> inserted or removed rather than just moved.
> 
> Given TaskDelegate is not the simplest of things this has a performance gain, 
> most noticeably on startup. Also rather than sorting all items after an 
> insert items are inserted in the right place using qLowerBound. Now we have 
> the correct signals we can remove the compression, they won't add anything. 
> 
> 
> Other commits:
> 
> Avoid constructing a QString for comparing, use QLatin1String for == 
> operators.
> 
> Remove useless include
> 
> Do not construct a map inside a lessThan function
> 
> lessThan functions have to be fast.
> Also Map -> Hash as we're not using order here.
> 
> 
> Diffs
> -
> 
>   applets/systemtray/package/contents/ui/ExpandedRepresentation.qml 2ef180b 
>   applets/systemtray/package/contents/ui/PlasmoidItem.qml 0eb1687 
>   applets/systemtray/package/contents/ui/StatusNotifierItem.qml fc889a8 
>   applets/systemtray/package/contents/ui/TaskDelegate.qml 913d8f1 
>   applets/systemtray/package/contents/ui/TaskListDelegate.qml 5501e02 
>   applets/systemtray/plugin/CMakeLists.txt f6e23b4 
>   applets/systemtray/plugin/host.h 02c5bbe 
>   applets/systemtray/plugin/host.cpp eafd0b6 
>   applets/systemtray/plugin/protocols/plasmoid/plasmoidtask.cpp 2b846f2 
>   applets/systemtray/plugin/tasklistmodel.h PRE-CREATION 
>   applets/systemtray/plugin/tasklistmodel.cpp PRE-CREATION 
> 
> Diff: https://git.reviewboard.kde.org/r/117813/diff/
> 
> 
> Testing
> ---
> 
> Seems to work :)
> 
> see branch davidedmundson/faster_systray to test
> 
> 
> Thanks,
> 
> David Edmundson
> 
>

___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: Review Request 117813: Make the system tray faster

[Marco Martin]

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/117813/#review56721
---


very good direction, just a couple of minor issues


applets/systemtray/plugin/host.cpp


did you try disabling and then reenabling a category? (didn't try the 
patch, just not sure looking at it how correctly reenables stuff that doesn't 
keep track of?)



applets/systemtray/plugin/tasklistmodel.h


as a convention, models exported in qml export their size as "count" (just 
because the primitive ListModel calls it that way)


- Marco Martin


On April 27, 2014, 10:51 p.m., David Edmundson wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/117813/
> ---
> 
> (Updated April 27, 2014, 10:51 p.m.)
> 
> 
> Review request for Plasma.
> 
> 
> Repository: plasma-workspace
> 
> 
> Description
> ---
> 
> Port QQmlListProperty to QAbstractListModel.
> QQmlListProperty only has a signal that the list has changed.This means when 
> used in a ListView every delegate has to be redone whenever a single item is 
> inserted or removed rather than just moved.
> 
> Given TaskDelegate is not the simplest of things this has a performance gain, 
> most noticeably on startup. Also rather than sorting all items after an 
> insert items are inserted in the right place using qLowerBound. Now we have 
> the correct signals we can remove the compression, they won't add anything. 
> 
> 
> Other commits:
> 
> Avoid constructing a QString for comparing, use QLatin1String for == 
> operators.
> 
> Remove useless include
> 
> Do not construct a map inside a lessThan function
> 
> lessThan functions have to be fast.
> Also Map -> Hash as we're not using order here.
> 
> 
> Diffs
> -
> 
>   applets/systemtray/package/contents/ui/ExpandedRepresentation.qml 2ef180b 
>   applets/systemtray/package/contents/ui/PlasmoidItem.qml 0eb1687 
>   applets/systemtray/package/contents/ui/StatusNotifierItem.qml fc889a8 
>   applets/systemtray/package/contents/ui/TaskDelegate.qml 913d8f1 
>   applets/systemtray/package/contents/ui/TaskListDelegate.qml 5501e02 
>   applets/systemtray/plugin/CMakeLists.txt f6e23b4 
>   applets/systemtray/plugin/host.h 02c5bbe 
>   applets/systemtray/plugin/host.cpp eafd0b6 
>   applets/systemtray/plugin/protocols/plasmoid/plasmoidtask.cpp 2b846f2 
>   applets/systemtray/plugin/tasklistmodel.h PRE-CREATION 
>   applets/systemtray/plugin/tasklistmodel.cpp PRE-CREATION 
> 
> Diff: https://git.reviewboard.kde.org/r/117813/diff/
> 
> 
> Testing
> ---
> 
> Seems to work :)
> 
> see branch davidedmundson/faster_systray to test
> 
> 
> Thanks,
> 
> David Edmundson
> 
>

___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: Review Request 117450: Add Milou to the default panel layout

[Martin Klapetek]

On Mon, Apr 28, 2014 at 11:16 AM, Sebastian Kügler  wrote:

> On Friday, April 25, 2014 17:21:55 Thomas Pfeiffer wrote:
> > On Thursday 24 April 2014 11:53:49 Sebastian Kügler wrote:
> > > Type to search is used in other applications as well, without input
> bar:
> > >
> > > - Firefox: search in webpages
> >
> > Which version are you referring to? In Firefox 28, when I start typing
> > without  having anything focused, nothing happens. No search starts.
>
> That seems to have been yanked (could've been long time ago, I haven't used
> Firefox a lot lately).
>

FYI, this was always optional and default was always (and probably still
is) off.

Cheers
-- 
Martin Klapetek | KDE Developer
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: Review Request 117813: Make the system tray faster

[Martin Klapetek]

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/117813/#review56714
---


Overall +1 from me


applets/systemtray/package/contents/ui/TaskDelegate.qml


No.



applets/systemtray/plugin/host.cpp


This reminds me - didn't we decide at the Jan sprint that the popup (the 
"hidden" here?) should contain /all/ the systray plasmoids?

Also sorry for hijacking the review, we could just fix it at once?



applets/systemtray/plugin/protocols/plasmoid/plasmoidtask.cpp


Whitespace leftover it seems



applets/systemtray/plugin/tasklistmodel.h


remove ;)



applets/systemtray/plugin/tasklistmodel.cpp


const QModelIndex& index -> const QModelIndex &index



applets/systemtray/plugin/tasklistmodel.cpp


const QModelIndex& parent -> const QModelIndex &parent

also Q_UNUSED(parent) to spare compiler warning



applets/systemtray/plugin/tasklistmodel.cpp


Task* task -> Task *task



applets/systemtray/plugin/tasklistmodel.cpp


Task* task -> Task *task


- Martin Klapetek


On April 28, 2014, 12:51 a.m., David Edmundson wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/117813/
> ---
> 
> (Updated April 28, 2014, 12:51 a.m.)
> 
> 
> Review request for Plasma.
> 
> 
> Repository: plasma-workspace
> 
> 
> Description
> ---
> 
> Port QQmlListProperty to QAbstractListModel.
> QQmlListProperty only has a signal that the list has changed.This means when 
> used in a ListView every delegate has to be redone whenever a single item is 
> inserted or removed rather than just moved.
> 
> Given TaskDelegate is not the simplest of things this has a performance gain, 
> most noticeably on startup. Also rather than sorting all items after an 
> insert items are inserted in the right place using qLowerBound. Now we have 
> the correct signals we can remove the compression, they won't add anything. 
> 
> 
> Other commits:
> 
> Avoid constructing a QString for comparing, use QLatin1String for == 
> operators.
> 
> Remove useless include
> 
> Do not construct a map inside a lessThan function
> 
> lessThan functions have to be fast.
> Also Map -> Hash as we're not using order here.
> 
> 
> Diffs
> -
> 
>   applets/systemtray/package/contents/ui/ExpandedRepresentation.qml 2ef180b 
>   applets/systemtray/package/contents/ui/PlasmoidItem.qml 0eb1687 
>   applets/systemtray/package/contents/ui/StatusNotifierItem.qml fc889a8 
>   applets/systemtray/package/contents/ui/TaskDelegate.qml 913d8f1 
>   applets/systemtray/package/contents/ui/TaskListDelegate.qml 5501e02 
>   applets/systemtray/plugin/CMakeLists.txt f6e23b4 
>   applets/systemtray/plugin/host.h 02c5bbe 
>   applets/systemtray/plugin/host.cpp eafd0b6 
>   applets/systemtray/plugin/protocols/plasmoid/plasmoidtask.cpp 2b846f2 
>   applets/systemtray/plugin/tasklistmodel.h PRE-CREATION 
>   applets/systemtray/plugin/tasklistmodel.cpp PRE-CREATION 
> 
> Diff: https://git.reviewboard.kde.org/r/117813/diff/
> 
> 
> Testing
> ---
> 
> Seems to work :)
> 
> see branch davidedmundson/faster_systray to test
> 
> 
> Thanks,
> 
> David Edmundson
> 
>

___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: Review Request 117813: Make the system tray faster

[Aleix Pol Gonzalez]

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/117813/#review56718
---


All in all, looks like a good improvement to me.


applets/systemtray/package/contents/ui/TaskDelegate.qml


!



applets/systemtray/package/contents/ui/TaskDelegate.qml


!


- Aleix Pol Gonzalez


On April 27, 2014, 10:51 p.m., David Edmundson wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/117813/
> ---
> 
> (Updated April 27, 2014, 10:51 p.m.)
> 
> 
> Review request for Plasma.
> 
> 
> Repository: plasma-workspace
> 
> 
> Description
> ---
> 
> Port QQmlListProperty to QAbstractListModel.
> QQmlListProperty only has a signal that the list has changed.This means when 
> used in a ListView every delegate has to be redone whenever a single item is 
> inserted or removed rather than just moved.
> 
> Given TaskDelegate is not the simplest of things this has a performance gain, 
> most noticeably on startup. Also rather than sorting all items after an 
> insert items are inserted in the right place using qLowerBound. Now we have 
> the correct signals we can remove the compression, they won't add anything. 
> 
> 
> Other commits:
> 
> Avoid constructing a QString for comparing, use QLatin1String for == 
> operators.
> 
> Remove useless include
> 
> Do not construct a map inside a lessThan function
> 
> lessThan functions have to be fast.
> Also Map -> Hash as we're not using order here.
> 
> 
> Diffs
> -
> 
>   applets/systemtray/package/contents/ui/ExpandedRepresentation.qml 2ef180b 
>   applets/systemtray/package/contents/ui/PlasmoidItem.qml 0eb1687 
>   applets/systemtray/package/contents/ui/StatusNotifierItem.qml fc889a8 
>   applets/systemtray/package/contents/ui/TaskDelegate.qml 913d8f1 
>   applets/systemtray/package/contents/ui/TaskListDelegate.qml 5501e02 
>   applets/systemtray/plugin/CMakeLists.txt f6e23b4 
>   applets/systemtray/plugin/host.h 02c5bbe 
>   applets/systemtray/plugin/host.cpp eafd0b6 
>   applets/systemtray/plugin/protocols/plasmoid/plasmoidtask.cpp 2b846f2 
>   applets/systemtray/plugin/tasklistmodel.h PRE-CREATION 
>   applets/systemtray/plugin/tasklistmodel.cpp PRE-CREATION 
> 
> Diff: https://git.reviewboard.kde.org/r/117813/diff/
> 
> 
> Testing
> ---
> 
> Seems to work :)
> 
> see branch davidedmundson/faster_systray to test
> 
> 
> Thanks,
> 
> David Edmundson
> 
>

___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: Review Request 117450: Add Milou to the default panel layout

[Marco Martin]

On Monday 28 April 2014, Sebastian Kügler wrote:

> Can you show a screenshot? It should show the short hostname and the distro
> name. Perhaps a bug?
> 
> > Plus, I'm sure we can find better places for branding than the place
> > where the  filter bar used to be.
> 
> Well, it's still less elements in the UI, that's a win in my books.

wouldn't be separing the menu and milou more elements in the ui then?

anyways, if the consensus is that they are two things functionally different 
enough, i have then no objections in adding milou in the panel by default and 
should be done soon

> > > - kmail's move to folder
> > 
> > Sorry, I don't know what you mean by "move to folder". Could you please
> > elaborate?
> 
> Hit m on an email in kmail, it will pop up a folder selection dialog. Then
> just type a substring of one folder name.

wow, very useful, but in almost 10 years of use, never noticed this thing 
exists :p

Cheers, 
Marco Martin
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: Fwd: KDE Frameworks Release Cycle

[Sebastian Kügler]

On Sunday, April 27, 2014 15:55:07 Albert Astals Cid wrote:
> El Diumenge, 27 d'abril de 2014, a les 15:15:32, David Faure va escriure:
> > FYI.
> 
> Interesting fact here that original the mail was just sent to k-f-d and
> k-c-d.
> 
> I am seeing similar patterns in the plasma land, where they went their own
> way  with the releasing discussion and only sent to this list after the
> discussion happened (or that's my impression) (Note i'm not complaining of
> being left aside since actually i was there in person for the KF5
> dicussion).
> 
> I'm just raising the question if we want to:
>  a) Try to make the KF5 and plasma people work more in the release-team
> list  when discussing about releases
>  b) Rename the release-team list to kde-applications-release-team or
> something  like that to make it clear it is about "KDE Applications" side
> of the previous three "Applications, Plaform and Workspaces" sides of a
> release
>  c) Disband the relase-team altogether.
> 
> I'd like a) to happen but i can see if being hard so i'm open to anything 
> people want

a), but instead of doing the full discussion on release-team, I think it makes 
sense to first find out what the team doing the software itself wants, and in 
that phase, the discussion is best held on, e.g. plasma-devel, and when 
there's a consensus is reached, taken to release-team for further discussion.

That is in fact how I see the past discussions. If it felt like a finished plan 
is presented to the release-team, that was not, at least my, intention, and we 
can make sure in the future it's clearer.
-- 
sebas

http://www.kde.org | http://vizZzion.org | GPG Key ID: 9119 0EF9
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: Review Request 117450: Add Milou to the default panel layout

[Sebastian Kügler]

On Friday, April 25, 2014 17:21:55 Thomas Pfeiffer wrote:
> On Thursday 24 April 2014 11:53:49 Sebastian Kügler wrote:
> > One has to take history into account here. While cleaning up the Kickoff
> > UI, I've removed the "branding" button. This small text is a useful
> > replacement>
> > for that:
> >
> > - Distros want to have the distro name somewhere in the UI
> > - Users need the name for support inquiries (otherwise, finding out the
> > name of the OS is hard and different for distros, barring command line)
> From what I've seen, it doesn't show the distro name, but the full
> hostname,  which is a technical thing which ordinary users probably won't
> understand anyway.

Can you show a screenshot? It should show the short hostname and the distro 
name. Perhaps a bug?

> Plus, I'm sure we can find better places for branding than the place where
> the  filter bar used to be.

Well, it's still less elements in the UI, that's a win in my books.

> > Just yanking it out creates new problems.
> >
> > Type to search is used in other applications as well, without input bar:
> >
> > - Firefox: search in webpages
> 
> Which version are you referring to? In Firefox 28, when I start typing
> without  having anything focused, nothing happens. No search starts.

That seems to have been yanked (could've been long time ago, I haven't used 
Firefox a lot lately).

> > - kmail's move to folder
> 
> Sorry, I don't know what you mean by "move to folder". Could you please 
> elaborate?

Hit m on an email in kmail, it will pop up a folder selection dialog. Then 
just type a substring of one folder name.

> > The underlying issue might be that it's understood as search, while it
> > really is _filtering_, and then it happens to be not so uncommon anymore
> > to
> > do it like this. The fact that we're using runners for it is entirely not
> > interesting to the user, it also means that we should not just enable more
> > runners here. Kickoff is simply not a search-optimized UI. I agree with
> > Vishesh that as Kickoff is now, search has no place in there.
> 
> "Filter as you type" is common in GNOME/GTK software, but not in KDE
> software.  If we want to establish this as a new pattern, it should be
> implemented in more than just one place (plus some place in kmail which I
> apparently wasn't aware of yet).

I think it's a useful pattern, and we should patternize it, which probably 
means documenting. I'd suggest to try with the current implementation in 
Kickoff, and see what user feedback we get (i.e. bugreports).

Cheers,
-- 
sebas

http://www.kde.org | http://vizZzion.org | GPG Key ID: 9119 0EF9
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: Qt Quick Controls style

[Marco Martin]

On Thursday 24 April 2014, Andrew Lake wrote:
> Hello all,
> 
> We completed the visual design for the UI controls recently in the VDG
> forums. After a little finalization work, it's available in my scratch repo
> (git.kde.org:scratch/alake/BreezeStyle). I know a general Breeze project
> repo was recently suggested, which would work just fine I think. We can
> move this into that repo once it's set up.

btw, i see that you started both a breezestyle and breezewindec scratch repos.

maybe one of the two may be promoted as just "breeze" repo?
(sysadmin ticket to move i it i guess)

or i can start an empty one to then migrate stuff inside as we go (since 
master would have to contain releasable stuff)

-- 
Marco Martin
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: Re: KDE VDG forums and "Ask a Dev" thread

[Martin Gräßlin]

On Monday 28 April 2014 10:54:24 Jens Reuterberg wrote:
> Well either that or we did it a week event - I mean it's not like it's
> gonna be ten thousand questions and most of them will be
> considering design, layout and usability - so say if a dev promised to
> pop in once a day and answer the questions that's cropped up?

wouldn't it make more sense to have a different thread then per dev who 
answers. E.g. having a long living thread for all general KWin things people 
want to know?

signature.asc
Description: This is a digitally signed message part.
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: Review Request 117779: fix crash when textureNode->texture() is null

[Martin Gräßlin]

> On April 26, 2014, 8:22 a.m., Martin Gräßlin wrote:
> > I'd rather not want to see us hide this crash. There is an underlying 
> > problem which needs a more proper fix. I recently hit this problem myself 
> > on one on my systems and a crash one can reproduce is as good as fixed ;-)

fix for underlying problem: https://git.reviewboard.kde.org/r/117824/


- Martin


---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/117779/#review56587
---


On April 26, 2014, 12:51 a.m., Alexander Richardson wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/117779/
> ---
> 
> (Updated April 26, 2014, 12:51 a.m.)
> 
> 
> Review request for kde-workspace and Plasma.
> 
> 
> Repository: plasma-framework
> 
> 
> Description
> ---
> 
> fix crash when textureNode->texture() is null
> 
> I get this crash very frequently on my system. This is probably only fixing
> the symptom and not the underlying issue, however at least plasma no longer
> crashes every few minutes
> 
> 
> Diffs
> -
> 
>   src/declarativeimports/core/windowthumbnail.cpp 
> 59255f75994adb96b30fb503c759b2e9110ab708 
> 
> Diff: https://git.reviewboard.kde.org/r/117779/diff/
> 
> 
> Testing
> ---
> 
> No longer crashes
> 
> 
> Thanks,
> 
> Alexander Richardson
> 
>

___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Review Request 117824: Discard window thumbnail pixmap after texture got destroyed by SceneGraph

[Martin Gräßlin]

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/117824/
---

Review request for Plasma and Alexander Richardson.


Bugs: 333482
http://bugs.kde.org/show_bug.cgi?id=333482


Repository: plasma-framework


Description
---

Discard window thumbnail pixmap after texture got destroyed by SceneGraph

If the window holding the WindowThumbnail item goes away the texture hold
by the node is destroyed but our bound window pixmap is not yet freed.
This results in incorrect state the next time the WindowThumbnail is
shown.

To get back into a clean state discardPixmap() is called if there is no
texture but a bound low level pixmap.

BUG: 333482


Diffs
-

  src/declarativeimports/core/windowthumbnail.cpp 
d1a7fef1fc5fd119592710d80274d2abe0c8b3b1 

Diff: https://git.reviewboard.kde.org/r/117824/diff/


Testing
---


Thanks,

Martin Gräßlin

___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: KDE VDG forums and "Ask a Dev" thread

[Jens Reuterberg]

Well either that or we did it a week event - I mean it's not like it's 
gonna be ten thousand questions and most of them will be 
considering design, layout and usability - so say if a dev promised to 
pop in once a day and answer the questions that's cropped up?
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: Review Request 117779: fix crash when textureNode->texture() is null

[Martin Gräßlin]

> On April 28, 2014, 10:24 a.m., Martin Gräßlin wrote:
> > did you notice any pattern which triggers the crash? That could help me to 
> > reproduce and find the root cause.

I found the condition myself:

1. show thumbnail
2. let it hide again
3. show thumbnail for same task

when switching to another task crash does not happen.


- Martin


---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/117779/#review56707
---


On April 26, 2014, 12:51 a.m., Alexander Richardson wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/117779/
> ---
> 
> (Updated April 26, 2014, 12:51 a.m.)
> 
> 
> Review request for kde-workspace and Plasma.
> 
> 
> Repository: plasma-framework
> 
> 
> Description
> ---
> 
> fix crash when textureNode->texture() is null
> 
> I get this crash very frequently on my system. This is probably only fixing
> the symptom and not the underlying issue, however at least plasma no longer
> crashes every few minutes
> 
> 
> Diffs
> -
> 
>   src/declarativeimports/core/windowthumbnail.cpp 
> 59255f75994adb96b30fb503c759b2e9110ab708 
> 
> Diff: https://git.reviewboard.kde.org/r/117779/diff/
> 
> 
> Testing
> ---
> 
> No longer crashes
> 
> 
> Thanks,
> 
> Alexander Richardson
> 
>

___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: KDE VDG forums and "Ask a Dev" thread

[Marco Martin]

On Monday 28 April 2014, Jens Reuterberg wrote:

> 
> So essentially:
> If there is anyone who knows that they don't feel up for it. Tell me,
> it's cool. You guys have a ton of work.
> If its anyone who likes to go first that would be awesome too - like a
> short introduction of you and what you do - and then open for questions.


Sounds good.
how would it proceed? would be like a previous agreement with a developer in 
particular that would be like (completely random example)
the developer(s) of Dolphin agreed to answer questions on that thread for that 
given day?

-- 
Marco Martin
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: plasma-framework in kdereview

[Luigi Toscano]

On Friday 25 of April 2014 15:43:39 Marco Martin wrote:
> On Friday 25 April 2014 15:24:50 Luigi Toscano wrote:
> > On Friday 25 of April 2014 15:14:46 Àlex Fiestas wrote:
> > > Moving plasma-framework to frameworks means that we will loose
> > > flexibility
> > > since we won't be able to break api/abi.
> > > 
> > > So, do we really have to move it there? Imho would be prudent to keep it
> > > somewhere else where api/abi stability is not mandatory.
> > > 
> > > Also, right now there is only one user of this framework
> > > (plasma-desktop),
> > > I would wait until at least we have 2 more shells based on it to commit
> > > to any stability.
> > 
> > Wasn't/isn't libplasma supposed to be used also by other applications
> > (amarok was the main user I guess)?
> 
> It always was.
> wether they want to use it or not it's their problem
> 
> and this attitude of "pest dependency" deeply bothers me and makes me not
> much really motivated to keep working on it.
> 
> yes, it has a lot of dependencies and is not optimal.
> f i would have it primarly as ligthtweight libraries i would split it at
> lest in 3-4 parts, but i think it's better preventing the fragmentation of
> little libraries in this case


No, no, I was not talking against 3rd-party usage - it's just another point 
for API and ABI stability :)

Ciao
-- 
Luigi
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Re: Review Request 117779: fix crash when textureNode->texture() is null

[Martin Gräßlin]

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/117779/#review56707
---


did you notice any pattern which triggers the crash? That could help me to 
reproduce and find the root cause.

- Martin Gräßlin


On April 26, 2014, 12:51 a.m., Alexander Richardson wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/117779/
> ---
> 
> (Updated April 26, 2014, 12:51 a.m.)
> 
> 
> Review request for kde-workspace and Plasma.
> 
> 
> Repository: plasma-framework
> 
> 
> Description
> ---
> 
> fix crash when textureNode->texture() is null
> 
> I get this crash very frequently on my system. This is probably only fixing
> the symptom and not the underlying issue, however at least plasma no longer
> crashes every few minutes
> 
> 
> Diffs
> -
> 
>   src/declarativeimports/core/windowthumbnail.cpp 
> 59255f75994adb96b30fb503c759b2e9110ab708 
> 
> Diff: https://git.reviewboard.kde.org/r/117779/diff/
> 
> 
> Testing
> ---
> 
> No longer crashes
> 
> 
> Thanks,
> 
> Alexander Richardson
> 
>

___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

Review Request 117822: Add safety checks to XCB functions in WindowThumbnail

[Martin Gräßlin]

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/117822/
---

Review request for Plasma.


Repository: plasma-framework


Description
---

Add safety checks to XCB functions in WindowThumbnail

Prevents XCB warnings about BadWindow when a tooltip is shown for the
first time.


Diffs
-

  src/declarativeimports/core/windowthumbnail.cpp 
d1a7fef1fc5fd119592710d80274d2abe0c8b3b1 

Diff: https://git.reviewboard.kde.org/r/117822/diff/


Testing
---


Thanks,

Martin Gräßlin

___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel

KDE VDG forums and "Ask a Dev" thread

[Jens Reuterberg]

Ok I've been thinking of getting a "Ask a dev" thread going. As many of 
you know I hope - the work is puttering on in the open forums and tbh 
plans of moving all the secret work into public is well under way.
The ideal being to have a community with the VDG replaced with 
Community Organizers who can help getting the right idea to the right 
dev and the right dev to the forum. This was essentially planned by me 
to be happening this autumn but things are going faster than expected.


Anyway I wanted to set up a "Ask a dev" thread - essentially an AMA 
where people post direct questions to a thing or an aspect and I try to 
hound one of you to answer (the questions will of course revolve around 
Plasma)


So essentially:
If there is anyone who knows that they don't feel up for it. Tell me, 
it's cool. You guys have a ton of work.
If its anyone who likes to go first that would be awesome too - like a 
short introduction of you and what you do - and then open for questions.


/Jens 
___
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel