Re: Review Request 127236: Leverage QDir to know what's in a KIconThemeDir

[Aleix Pol Gonzalez]

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/127236/
---

(Updated April 30, 2016, 3:44 a.m.)


Status
--

This change has been discarded.


Review request for KDE Frameworks and Christoph Feck.


Repository: kiconthemes


Description
---

At the moment we're playing Battleship to see if an icon is present in a 
subdirectory. This means that we are checking on every directory if there's an 
icon that matches the size with a said name on every request.

This can be seen easily with strace:
```
$ strace kwrite |& grep ENOENT | wc -l
6212
```
After the patch: 
```
$ strace kwrite |& grep ENOENT | wc -l
1993
```
We reduce these accesses to let QDir keep the list of files inside the 
directory (that was already being generated at some point, it just was being 
discarded).


Diffs
-

  src/kicontheme.cpp 0996054 

Diff: https://git.reviewboard.kde.org/r/127236/diff/


Testing
---

Builds, tests still pass, applications start noticeably faster.


Thanks,

Aleix Pol Gonzalez

___
Kde-frameworks-devel mailing list
Kde-frameworks-devel@kde.org
https://mail.kde.org/mailman/listinfo/kde-frameworks-devel

Re: Review Request 127788: Fix leak in KNS::Engine

[Aleix Pol Gonzalez]

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/127788/
---

(Updated April 30, 2016, 1:56 a.m.)


Status
--

This change has been marked as submitted.


Review request for KDE Frameworks and Jeremy Whiting.


Changes
---

Submitted with commit a061fd2abdbe2f03a6b61b2ae871f673b5cade51 by Aleix Pol to 
branch master.


Repository: knewstuff


Description
---

A local variable was shadowing the actual variable declaration.


Diffs
-

  src/core/engine.cpp c8d0579 

Diff: https://git.reviewboard.kde.org/r/127788/diff/


Testing
---


Thanks,

Aleix Pol Gonzalez

___
Kde-frameworks-devel mailing list
Kde-frameworks-devel@kde.org
https://mail.kde.org/mailman/listinfo/kde-frameworks-devel

Re: Review Request 127655: Fix KAboutData::applicationData() to sync to current Q*Application metadata

[Albert Astals Cid]

> On April 29, 2016, 3:15 a.m., Michael Pyne wrote:
> > I think I disagree with the idea of overwriting KAboutData properties if 
> > they are already set by the user. Alex, any thoughts?
> > 
> > In the event the KAboutData doesn't already exist I think automatically 
> > setting it up makes sense, and QCoreApplication is a good source. But I 
> > would rather flag property conflicts than to break ties when developer 
> > selects two different values for same property, as that's change in 
> > behavior might break other parts of code that rely on KAboutData not 
> > changing values.
> > 
> > Would this partial solution be OK for the problem you're running into?

> I think I disagree with the idea of overwriting KAboutData properties if they 
> are already set by the user. Alex, any thoughts?

I agree with Michael, it seems strage it overwriting what you may have set in 
setAboutData.


- Albert


---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/127655/#review95002
---


On April 28, 2016, 1:04 a.m., Friedrich W. H. Kossebau wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/127655/
> ---
> 
> (Updated April 28, 2016, 1:04 a.m.)
> 
> 
> Review request for KDE Frameworks, Alex Richardson and Michael Pyne.
> 
> 
> Repository: kcoreaddons
> 
> 
> Description
> ---
> 
> KAboutData is passed as values on getting and setting the "applicationData",
> and it only makes sense to have its properties be a transparent access
> to the actual mirrored Q*Application metadata.
> 
> Even more as there is code in KF5 (e.g. KXMLGUI) which relies on 
> KAboutData::applicationData(),
> without requiring the user to use KAboutData::setApplicationData().
> 
> 
> Diffs
> -
> 
>   autotests/kaboutdatatest.cpp f31e7f3 
>   src/lib/kaboutdata.h 97c0f2b 
>   src/lib/kaboutdata.cpp ceb0c06 
> 
> Diff: https://git.reviewboard.kde.org/r/127655/diff/
> 
> 
> Testing
> ---
> 
> Added autotests pass.
> 
> 
> Thanks,
> 
> Friedrich W. H. Kossebau
> 
>

___
Kde-frameworks-devel mailing list
Kde-frameworks-devel@kde.org
https://mail.kde.org/mailman/listinfo/kde-frameworks-devel

Re: Review Request 127788: Fix leak in KNS::Engine

[Albert Astals Cid]

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/127788/#review95033
---


Fix it, then Ship it!





src/core/engine.cpp (line 155)


I'd add a
delete m_atticaProviderManager;
prior to this line just in case.


- Albert Astals Cid


On April 29, 2016, 12:45 p.m., Aleix Pol Gonzalez wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/127788/
> ---
> 
> (Updated April 29, 2016, 12:45 p.m.)
> 
> 
> Review request for KDE Frameworks and Jeremy Whiting.
> 
> 
> Repository: knewstuff
> 
> 
> Description
> ---
> 
> A local variable was shadowing the actual variable declaration.
> 
> 
> Diffs
> -
> 
>   src/core/engine.cpp c8d0579 
> 
> Diff: https://git.reviewboard.kde.org/r/127788/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Aleix Pol Gonzalez
> 
>

___
Kde-frameworks-devel mailing list
Kde-frameworks-devel@kde.org
https://mail.kde.org/mailman/listinfo/kde-frameworks-devel

Re: Review Request 127770: Increase maximum string length in KSycoca database

[Jos van den Oever]

> On apr 28, 2016, 9:12 a.m., Milian Wolff wrote:
> > src/sycoca/ksycocautils.cpp, line 29
> > 
> >
> > as a follow-up cleanup I suggest to introduce an enum to hold this 
> > magic value

I can do that if the code stays. I made another rr to remove it: 
https://git.reviewboard.kde.org/r/127786/


- Jos


---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/127770/#review94946
---


On apr 28, 2016, 7:01 a.m., Jos van den Oever wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/127770/
> ---
> 
> (Updated apr 28, 2016, 7:01 a.m.)
> 
> 
> Review request for KDE Frameworks and David Faure.
> 
> 
> Repository: kservice
> 
> 
> Description
> ---
> 
> The code for reading and writing of strings in KSycoca is not symmetrical. 
> Strings of any length can be written, but only strings of less than 8192 
> bytes may be read. This limit is set in KSycocaUtilsPrivate::read. The limit 
> is probably there to avoid out-of-memory situations.
> 
> On my system I have a lot of XDG data dirs. The length of the environment 
> variable is currently 4092 bytes. KSycocaBuild saves that as UTF-16 which 
> needs 8184 bytes. KBuildSycoca save that without complaint but complains when 
> reading it.
> 
> 
> The simplest solution here is to simply increase the magic number 8192 to 
> e.g. 65528. This is just a temporary buffer.
> 
> Or we just check the size of the whole cache file (e.g. < 100M) and remove 
> all other limits. That would simplify
> 
>KSycocaUtilsPrivate::read(*str, header.prefixes);
> 
> to
> 
>*str >> header.prefixes;
> 
> This patch chooses the first option.
> 
> 
> Diffs
> -
> 
>   src/sycoca/ksycocautils.cpp 1ba75e8 
> 
> Diff: https://git.reviewboard.kde.org/r/127770/diff/
> 
> 
> Testing
> ---
> 
> Ran unit tests on KService. All but one of the previously failing tests on 
> NixOS is now fixed.
> 
> 
> Thanks,
> 
> Jos van den Oever
> 
>

___
Kde-frameworks-devel mailing list
Kde-frameworks-devel@kde.org
https://mail.kde.org/mailman/listinfo/kde-frameworks-devel

Re: Review Request 127788: Fix leak in KNS::Engine

[David Edmundson]

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/127788/#review95019
---


Ship it!




Ship It!

- David Edmundson


On April 29, 2016, 12:45 p.m., Aleix Pol Gonzalez wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/127788/
> ---
> 
> (Updated April 29, 2016, 12:45 p.m.)
> 
> 
> Review request for KDE Frameworks and Jeremy Whiting.
> 
> 
> Repository: knewstuff
> 
> 
> Description
> ---
> 
> A local variable was shadowing the actual variable declaration.
> 
> 
> Diffs
> -
> 
>   src/core/engine.cpp c8d0579 
> 
> Diff: https://git.reviewboard.kde.org/r/127788/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Aleix Pol Gonzalez
> 
>

___
Kde-frameworks-devel mailing list
Kde-frameworks-devel@kde.org
https://mail.kde.org/mailman/listinfo/kde-frameworks-devel

Review Request 127788: Fix leak in KNS::Engine

[Aleix Pol Gonzalez]

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/127788/
---

Review request for KDE Frameworks and Jeremy Whiting.


Repository: knewstuff


Description
---

A local variable was shadowing the actual variable declaration.


Diffs
-

  src/core/engine.cpp c8d0579 

Diff: https://git.reviewboard.kde.org/r/127788/diff/


Testing
---


Thanks,

Aleix Pol Gonzalez

___
Kde-frameworks-devel mailing list
Kde-frameworks-devel@kde.org
https://mail.kde.org/mailman/listinfo/kde-frameworks-devel

Re: Review Request 127786: Remove custom read functions for QString and QStringList

[Jos van den Oever]

> On apr 29, 2016, 10:56 a.m., David Faure wrote:
> > This is not about trust and attacks, this is about not allocating 4 GB of 
> > RAM when reading a corrupted binary file.
> 
> Jos van den Oever wrote:
> That will only happen if the file or stream is 4 GB. `QDataStream 
> >>(QDataStream , QString )` allocates while reading in 1 MiB 
> chunks.

http://code.qt.io/cgit/qt/qtbase.git/tree/src/corelib/tools/qstring.cpp#n8637


- Jos


---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/127786/#review95012
---


On apr 29, 2016, 10:22 a.m., Jos van den Oever wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/127786/
> ---
> 
> (Updated apr 29, 2016, 10:22 a.m.)
> 
> 
> Review request for KDE Frameworks, David Faure and Milian Wolff.
> 
> 
> Repository: kservice
> 
> 
> Description
> ---
> 
> Writing KBuildSycoca is done with <<. Up till now there were special 'safe' 
> functions for reading QString and QStringList. They only limited the size of 
> QString and the number of allowed entries in QStringList. The cache file is 
> created by the trusted system. If file size is an attack vector, these safe 
> functions are useful and we should keep them.
> 
> This patch is three commits:
> 
> 1)  Use the standard read function for reading QStringList
> 
> 
> 2)  Use the standard read function for reading QString
> 
> 
> 3)  Remove redundant #include
> 
> ksycocaentry.h is included via kservice.h
> 
> 
> Diffs
> -
> 
>   src/CMakeLists.txt f4d09d5 
>   src/services/kservicegroup.h c046314 
>   src/services/kservicetypefactory.cpp 2edc57c 
>   src/sycoca/kctimefactory.cpp a8c7846 
>   src/sycoca/ksycoca.cpp 5d43ef4 
>   src/sycoca/ksycoca_p.h 119c3ee 
>   src/sycoca/ksycocaentry.cpp 5fbd158 
>   src/sycoca/ksycocautils.cpp 84998b7 
>   src/sycoca/ksycocautils_p.h aad9d50 
> 
> Diff: https://git.reviewboard.kde.org/r/127786/diff/
> 
> 
> Testing
> ---
> 
> All tests still pass.
> 
> 
> Thanks,
> 
> Jos van den Oever
> 
>

___
Kde-frameworks-devel mailing list
Kde-frameworks-devel@kde.org
https://mail.kde.org/mailman/listinfo/kde-frameworks-devel

Re: Review Request 127786: Remove custom read functions for QString and QStringList

[Jos van den Oever]

> On apr 29, 2016, 10:56 a.m., David Faure wrote:
> > This is not about trust and attacks, this is about not allocating 4 GB of 
> > RAM when reading a corrupted binary file.

That will only happen if the file or stream is 4 GB. `QDataStream 
>>(QDataStream , QString )` allocates while reading in 1 MiB 
chunks.


- Jos


---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/127786/#review95012
---


On apr 29, 2016, 10:22 a.m., Jos van den Oever wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/127786/
> ---
> 
> (Updated apr 29, 2016, 10:22 a.m.)
> 
> 
> Review request for KDE Frameworks, David Faure and Milian Wolff.
> 
> 
> Repository: kservice
> 
> 
> Description
> ---
> 
> Writing KBuildSycoca is done with <<. Up till now there were special 'safe' 
> functions for reading QString and QStringList. They only limited the size of 
> QString and the number of allowed entries in QStringList. The cache file is 
> created by the trusted system. If file size is an attack vector, these safe 
> functions are useful and we should keep them.
> 
> This patch is three commits:
> 
> 1)  Use the standard read function for reading QStringList
> 
> 
> 2)  Use the standard read function for reading QString
> 
> 
> 3)  Remove redundant #include
> 
> ksycocaentry.h is included via kservice.h
> 
> 
> Diffs
> -
> 
>   src/CMakeLists.txt f4d09d5 
>   src/services/kservicegroup.h c046314 
>   src/services/kservicetypefactory.cpp 2edc57c 
>   src/sycoca/kctimefactory.cpp a8c7846 
>   src/sycoca/ksycoca.cpp 5d43ef4 
>   src/sycoca/ksycoca_p.h 119c3ee 
>   src/sycoca/ksycocaentry.cpp 5fbd158 
>   src/sycoca/ksycocautils.cpp 84998b7 
>   src/sycoca/ksycocautils_p.h aad9d50 
> 
> Diff: https://git.reviewboard.kde.org/r/127786/diff/
> 
> 
> Testing
> ---
> 
> All tests still pass.
> 
> 
> Thanks,
> 
> Jos van den Oever
> 
>

___
Kde-frameworks-devel mailing list
Kde-frameworks-devel@kde.org
https://mail.kde.org/mailman/listinfo/kde-frameworks-devel

Re: Review Request 127786: Remove custom read functions for QString and QStringList

[David Faure]

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/127786/#review95012
---



This is not about trust and attacks, this is about not allocating 4 GB of RAM 
when reading a corrupted binary file.

- David Faure


On April 29, 2016, 10:22 a.m., Jos van den Oever wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/127786/
> ---
> 
> (Updated April 29, 2016, 10:22 a.m.)
> 
> 
> Review request for KDE Frameworks, David Faure and Milian Wolff.
> 
> 
> Repository: kservice
> 
> 
> Description
> ---
> 
> Writing KBuildSycoca is done with <<. Up till now there were special 'safe' 
> functions for reading QString and QStringList. They only limited the size of 
> QString and the number of allowed entries in QStringList. The cache file is 
> created by the trusted system. If file size is an attack vector, these safe 
> functions are useful and we should keep them.
> 
> This patch is three commits:
> 
> 1)  Use the standard read function for reading QStringList
> 
> 
> 2)  Use the standard read function for reading QString
> 
> 
> 3)  Remove redundant #include
> 
> ksycocaentry.h is included via kservice.h
> 
> 
> Diffs
> -
> 
>   src/CMakeLists.txt f4d09d5 
>   src/services/kservicegroup.h c046314 
>   src/services/kservicetypefactory.cpp 2edc57c 
>   src/sycoca/kctimefactory.cpp a8c7846 
>   src/sycoca/ksycoca.cpp 5d43ef4 
>   src/sycoca/ksycoca_p.h 119c3ee 
>   src/sycoca/ksycocaentry.cpp 5fbd158 
>   src/sycoca/ksycocautils.cpp 84998b7 
>   src/sycoca/ksycocautils_p.h aad9d50 
> 
> Diff: https://git.reviewboard.kde.org/r/127786/diff/
> 
> 
> Testing
> ---
> 
> All tests still pass.
> 
> 
> Thanks,
> 
> Jos van den Oever
> 
>

___
Kde-frameworks-devel mailing list
Kde-frameworks-devel@kde.org
https://mail.kde.org/mailman/listinfo/kde-frameworks-devel

Review Request 127786: Remove custom read functions for QString and QStringList

[Jos van den Oever]

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/127786/
---

Review request for KDE Frameworks, David Faure and Milian Wolff.


Repository: kservice


Description
---

Writing KBuildSycoca is done with <<. Up till now there were special 'safe' 
functions for reading QString and QStringList. They only limited the size of 
QString and the number of allowed entries in QStringList. The cache file is 
created by the trusted system. If file size is an attack vector, these safe 
functions are useful and we should keep them.

This patch is three commits:

1)  Use the standard read function for reading QStringList


2)  Use the standard read function for reading QString


3)  Remove redundant #include

ksycocaentry.h is included via kservice.h


Diffs
-

  src/CMakeLists.txt f4d09d5 
  src/services/kservicegroup.h c046314 
  src/services/kservicetypefactory.cpp 2edc57c 
  src/sycoca/kctimefactory.cpp a8c7846 
  src/sycoca/ksycoca.cpp 5d43ef4 
  src/sycoca/ksycoca_p.h 119c3ee 
  src/sycoca/ksycocaentry.cpp 5fbd158 
  src/sycoca/ksycocautils.cpp 84998b7 
  src/sycoca/ksycocautils_p.h aad9d50 

Diff: https://git.reviewboard.kde.org/r/127786/diff/


Testing
---

All tests still pass.


Thanks,

Jos van den Oever

___
Kde-frameworks-devel mailing list
Kde-frameworks-devel@kde.org
https://mail.kde.org/mailman/listinfo/kde-frameworks-devel

Jenkins-kde-ci: krunner master kf5-qt5 » Linux,gcc - Build # 37 - Unstable!

[no-reply]

GENERAL INFO

BUILD UNSTABLE
Build URL: 
https://build.kde.org/job/krunner%20master%20kf5-qt5/PLATFORM=Linux,compiler=gcc/37/
Project: PLATFORM=Linux,compiler=gcc
Date of build: Fri, 29 Apr 2016 09:46:24 +
Build duration: 9 min 10 sec

CHANGE SET
Revision af4ef9d77373084da236e34b97ab0965696300bd by scripty: (SVN_SILENT made 
messages (.desktop file) - always resolve ours)
  change: edit templates/runner/%{APPNAMELC}.desktop


JUNIT RESULTS

Name: (root) Failed: 1 test(s), Passed: 0 test(s), Skipped: 0 test(s), Total: 1 
test(s)Failed: TestSuite.krunner-runnercontexttest

COBERTURA RESULTS

Cobertura Coverage Report
  

By packages
  ___
Kde-frameworks-devel mailing list
Kde-frameworks-devel@kde.org
https://mail.kde.org/mailman/listinfo/kde-frameworks-devel

Re: Review Request 127779: use system colors for monochrome icons

[Marco Martin]

> On April 28, 2016, 5:36 p.m., Aleix Pol Gonzalez wrote:
> > src/kiconloader.cpp, line 861
> > 
> >
> > We will need that and expose it properly, as we'll have to request the 
> > icon cache to clean if it's one of such icon themes, then force 
> > re-rendering.

latest version exposes it via KIconTheme::followsColorScheme() (maybe would be 
better followsSystemColorScheme?)


- Marco


---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/127779/#review94984
---


On April 29, 2016, 9:25 a.m., Marco Martin wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/127779/
> ---
> 
> (Updated April 29, 2016, 9:25 a.m.)
> 
> 
> Review request for KDE Frameworks and Plasma.
> 
> 
> Repository: kiconthemes
> 
> 
> Description
> ---
> 
> Breeze uses stylesheet information to color its icons with some "named" 
> colors that change depending from the current system color scheme, this is 
> already used since some time in the icons used by the Plasma shell.
> But QWidget applications have the same problem, when the user changes the 
> color scheme from breeze to breeze dark (or any color scheme), most of the 
> icons will be black on black.
> This patch clones (a bit simplified and taking only the most "important" 
> colors) the logic used by Plasma::Svg to color icons with the stylesheet.
> 
> even tough it's doing more things at icon generation, an application that 
> uses a lot of icons like Dolphin doesn't seem to have noticeable startup time 
> difference, even when the image cache is not present yet, so i hope is not an 
> unacceptable performance tradeoff (successive loads are unchanged as are from 
> the image cache).
> 
> still some questions and things that can be optimized, like
> 
> * an optional key in the theme metadata file to explicitly enable this, to 
> not have it running in themes that don't care?
> 
> * can i use karchive in this framework?, so it would work with svgz icons as 
> well
> 
> * right now to refresh icons at runtime it depends from a patch in the colors 
> kcm to emit iconchanges as well, alternatively kiconloader could watch for 
> kcolorscheme changes as well, but i don't think is strictly necessary
> 
> 
> Diffs
> -
> 
>   CMakeLists.txt 2e838e8 
>   autotests/coloredsvgicon.svg PRE-CREATION 
>   autotests/kiconloader_unittest.cpp 0e47cc8 
>   autotests/resources.qrc a19c963 
>   src/CMakeLists.txt 0e30a35 
>   src/kiconloader.cpp 75ab482 
>   src/kicontheme.h 3190665 
>   src/kicontheme.cpp 0996054 
> 
> Diff: https://git.reviewboard.kde.org/r/127779/diff/
> 
> 
> Testing
> ---
> 
> 
> File Attachments
> 
> 
> dadel1.png
>   
> https://git.reviewboard.kde.org/media/uploaded/files/2016/04/28/0fc42425-947c-479e-9759-09c7a703a456__dadel1.png
> 
> 
> Thanks,
> 
> Marco Martin
> 
>

___
Kde-frameworks-devel mailing list
Kde-frameworks-devel@kde.org
https://mail.kde.org/mailman/listinfo/kde-frameworks-devel

Re: Review Request 127779: use system colors for monochrome icons

[Marco Martin]

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/127779/
---

(Updated April 29, 2016, 9:25 a.m.)


Review request for KDE Frameworks and Plasma.


Repository: kiconthemes


Description
---

Breeze uses stylesheet information to color its icons with some "named" colors 
that change depending from the current system color scheme, this is already 
used since some time in the icons used by the Plasma shell.
But QWidget applications have the same problem, when the user changes the color 
scheme from breeze to breeze dark (or any color scheme), most of the icons will 
be black on black.
This patch clones (a bit simplified and taking only the most "important" 
colors) the logic used by Plasma::Svg to color icons with the stylesheet.

even tough it's doing more things at icon generation, an application that uses 
a lot of icons like Dolphin doesn't seem to have noticeable startup time 
difference, even when the image cache is not present yet, so i hope is not an 
unacceptable performance tradeoff (successive loads are unchanged as are from 
the image cache).

still some questions and things that can be optimized, like

* an optional key in the theme metadata file to explicitly enable this, to not 
have it running in themes that don't care?

* can i use karchive in this framework?, so it would work with svgz icons as 
well

* right now to refresh icons at runtime it depends from a patch in the colors 
kcm to emit iconchanges as well, alternatively kiconloader could watch for 
kcolorscheme changes as well, but i don't think is strictly necessary


Diffs (updated)
-

  CMakeLists.txt 2e838e8 
  autotests/coloredsvgicon.svg PRE-CREATION 
  autotests/kiconloader_unittest.cpp 0e47cc8 
  autotests/resources.qrc a19c963 
  src/CMakeLists.txt 0e30a35 
  src/kiconloader.cpp 75ab482 
  src/kicontheme.h 3190665 
  src/kicontheme.cpp 0996054 

Diff: https://git.reviewboard.kde.org/r/127779/diff/


Testing
---


File Attachments


dadel1.png
  
https://git.reviewboard.kde.org/media/uploaded/files/2016/04/28/0fc42425-947c-479e-9759-09c7a703a456__dadel1.png


Thanks,

Marco Martin

___
Kde-frameworks-devel mailing list
Kde-frameworks-devel@kde.org
https://mail.kde.org/mailman/listinfo/kde-frameworks-devel