Re: Web Shortcuts KCM

2014-08-05 Thread Eike Hein



On 08/04/2014 10:09 AM, David Faure wrote:

So yep, that's not going away any time soon ;)


Alright, so that leaves the licensing problem, right? Do I
need to rewrite the KCM? Can I even? Do we contact Yves
Arrouye for relicensing?


Cheers,
Eike

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


Re: Review Request 119588: Remove Qt5 debugfull hack from ECMConfig.cmake

2014-08-05 Thread Aleix Pol Gonzalez

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/119588/#review63888
---

Ship it!


Looks reasonable to me.

- Aleix Pol Gonzalez


On Aug. 5, 2014, 9:09 p.m., Alex Merry wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/119588/
> ---
> 
> (Updated Aug. 5, 2014, 9:09 p.m.)
> 
> 
> Review request for Extra Cmake Modules and KDE Frameworks.
> 
> 
> Repository: extra-cmake-modules
> 
> 
> Description
> ---
> 
> The original purpose of this was to set the QT_NO_DEBUG macro if the old
> DebugFull configuration was used. We got rid of DebugFull (instead just
> using Debug), so it is no longer relevant, and this hack never belonged
> in ECMConfig.cmake anyway (it should have been in KDECompilerSettings).
> 
> CHANGELOG: ECM now works when the required CMake version is set to
> 3.0.0 or higher, and does not require Qt5Core to be available.
> 
> BUG: 331849
> 
> 
> Diffs
> -
> 
>   ECMConfig.cmake.in ea47abf0d8b025dd9e424b0aea26986d1f288279 
> 
> Diff: https://git.reviewboard.kde.org/r/119588/diff/
> 
> 
> Testing
> ---
> 
> kimageformats configures when the required CMake version is set to 3.0.0 in 
> (before, the configure step would fail).
> 
> oxygen-fonts configures when /usr/lib/cmake/Qt5Core is removed (before, the 
> configure step would fail).
> 
> 
> Thanks,
> 
> Alex Merry
> 
>

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


Re: KDE Frameworks 5.1.0 release

2014-08-05 Thread Luigi Toscano
Martin Klapetek ha scritto:
> On Sat, Aug 2, 2014 at 3:48 PM, David Faure  > wrote:
> 
> I just tagged and packed 5.1.0.
> 
> 
> Bugzilla versions also need to be updated. Is there some easy batch way to add
> new versions to bugzilla products? If not, I'll spend the time tomorrow
> evening adding them to all frameworks-* products.

Write a script with https://fedorahosted.org/python-bugzilla/ and kindly offer
it to our sysadmins :)

(maybe it already exists)

Ciao
-- 
Luigi

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


Re: KDE Frameworks 5.1.0 release

2014-08-05 Thread Martin Klapetek
On Sat, Aug 2, 2014 at 3:48 PM, David Faure  wrote:

> I just tagged and packed 5.1.0.
>

Bugzilla versions also need to be updated. Is there some easy batch way to
add new versions to bugzilla products? If not, I'll spend the time tomorrow
evening adding them to all frameworks-* products.

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


Re: Review Request 119505: Instance our onw QFileDialog instead of using getExistingDirectoryUrl

2014-08-05 Thread Àlex Fiestas

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/119505/
---

(Updated ago. 5, 2014, 10:01 p.m.)


Review request for KDE Frameworks.


Repository: kio


Description
---

getExistingDirectoryUrl only works for local files so we have to implement it 
on our own for now.

As for Qt, QFileDialog::getExistingDirectoryUrl calls QUrl.toLocalFile, if we 
remove that, further down the code path _qt_get_directory is called which 
checks that the file exists but since Qt is not able to talk all our kios (for 
example smb) it will return false.
So at the moment we are forced to implement it ourselves.


Diffs (updated)
-

  src/widgets/kurlrequester.cpp cf0b0c7 

Diff: https://git.reviewboard.kde.org/r/119505/diff/


Testing
---


Thanks,

Àlex Fiestas

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


Re: Making KFileMetaData a framework

2014-08-05 Thread Kevin Ottens
Hello,

On Tuesday 05 August 2014 18:36:24 Vishesh Handa wrote:
> I would appreciate it if everyone could review the code once before it gets
> into frameworks.

metainfo.yaml should have "release: false" until it's part of a KF release. 
Also it should be integration and not functional (relies on plugins).

The "test" folder should be named "tests" (see directory structure policy).

Public headers should use <> style include. Also it seems you're not following 
the k convention but the namespace convention for your classes, then the 
includes in public headers should be namespaced as well (e.g. 
).

I'm not sure why ExtractorPluginManager is exported. A function in a namespace 
would be enough, there's no point in instantiating a manager by hand from the 
client code perspective, at best looks like leaking an implementation detail.

Similarly the ExtractorPlugin naming is odd in the public API as it states it 
is necessarily a plugin (implementation detail). I'd rename it 
ExtractorInterface, I'd drop the suffix altogether or I'd keep ExtractorPlugin 
for plugin implementors while they'd be wrapped in Extractor instances on the 
client code side (I think that's actually my favorite solution).

AFAICT there's no reason for ExtractorPlugin to inherit from QObject at that 
point. Same thing for ExtractorPluginManager.

Creating by hand the ExtractionResult, then passing it by pointer to the 
extract method looks odd. I'd expect calling extract() with a bunch of 
parameters and getting a result back (another reason for wrapping plugins on 
the client side).

The whole API is synchronous which we probably don't want. It'd be better to 
have an async API (much better to build up sync on top of async than the 
contrary).

What about the thread safety? At a glance I would say ExtractorPluginManager 
is not

That's about it after a quick glance while tired. Keep us posted for a second 
round.

Regards.
-- 
Kévin Ottens, http://ervin.ipsquad.net

KDAB - proud supporter of KDE, http://www.kdab.com



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


Re: Review Request 119588: Remove Qt5 debugfull hack from ECMConfig.cmake

2014-08-05 Thread Alex Merry

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/119588/
---

(Updated Aug. 5, 2014, 9:09 p.m.)


Review request for Extra Cmake Modules and KDE Frameworks.


Changes
---

Adding frameworks on the suspicion that nobody reads kde-build-system.


Repository: extra-cmake-modules


Description
---

The original purpose of this was to set the QT_NO_DEBUG macro if the old
DebugFull configuration was used. We got rid of DebugFull (instead just
using Debug), so it is no longer relevant, and this hack never belonged
in ECMConfig.cmake anyway (it should have been in KDECompilerSettings).

CHANGELOG: ECM now works when the required CMake version is set to
3.0.0 or higher, and does not require Qt5Core to be available.

BUG: 331849


Diffs
-

  ECMConfig.cmake.in ea47abf0d8b025dd9e424b0aea26986d1f288279 

Diff: https://git.reviewboard.kde.org/r/119588/diff/


Testing
---

kimageformats configures when the required CMake version is set to 3.0.0 in 
(before, the configure step would fail).

oxygen-fonts configures when /usr/lib/cmake/Qt5Core is removed (before, the 
configure step would fail).


Thanks,

Alex Merry

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


Re: Review Request 119267: Adding KWindowSystem::setOnActivities(WId win, const QStringList &activities) method

2014-08-05 Thread Ivan Čukić

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/119267/
---

(Updated Aug. 5, 2014, 8:21 p.m.)


Review request for KDE Frameworks, kwin and Martin Gräßlin.


Changes
---

Introduced KDE_ALL_ACTIVITIES_UUID


Repository: kwindowsystem


Description
---

Currently, the library only has the method for retrieving a list of activities 
a window belongs to.

This is adding a method which provides changing the list of activities for a 
window.


Diffs (updated)
-

  autotests/kwindowinfox11test.cpp 50ce806 
  src/kwindowinfo_x11.cpp 041dfd3 
  src/kwindowsystem.h 0b58e71 
  src/kwindowsystem.cpp fb59603 
  src/kwindowsystem_p.h 8861844 
  src/kwindowsystem_p_x11.h 9baa6ae 
  src/kwindowsystem_x11.cpp 2016820 
  src/netwm.h 2d812a7 
  src/netwm.cpp 1daad1e 
  src/netwm_p.h a201cb6 

Diff: https://git.reviewboard.kde.org/r/119267/diff/


Testing
---

Yes, works with the new activity switcher.


Thanks,

Ivan Čukić

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


Re: Review Request 119152: Do not define QT_DISABLE_DEPRECATED_BEFORE

2014-08-05 Thread Kevin Funk

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/119152/
---

(Updated Aug. 5, 2014, 8:06 p.m.)


Status
--

This change has been discarded.


Review request for KDE Frameworks and Stephen Kelly.


Repository: kdelibs4support


Description
---

Do not define QT_DISABLE_DEPRECATED_BEFORE

When defining it in your own project, you'll get:
:0:0: warning:
"QT_DISABLE_DEPRECATED_BEFORE" redefined [enabled by default]

Is there any reason for it to be there?


Diffs
-

  cmake/modules/ECMQt4To5Porting.cmake 4204fa541790aa38c74b9d6f0b2111af2157b2bc 

Diff: https://git.reviewboard.kde.org/r/119152/diff/


Testing
---


Thanks,

Kevin Funk

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


Re: Making KFileMetaData a framework

2014-08-05 Thread David Edmundson
In general that's some of the tidied code I've seen in a long time.

Comments below. One major, most not.


TypeInfo/PropertyInfo/SimpleExtractionResult need a working &operator=
otherwise we shallow copy d.

I can cause a crash in dump.cpp


Extracting info from a file can take a long time, right now you spawn
a separate execuatble for Baloo; but for other uses (i.e dolphin
getting info on a non-indexed file) you might want to add a helper
API; either a new thread or a service like baloo-file-extractor and a
wrapper round calling it.



ExtractorPluginManager::Private::allExtractors
what's going on with the variable "plugins"? You're always checking if
an empty list contains things. Looks leftover from something.



In properties.h I would space out the enum so that in the future when
you insert extra ones you can put it alongside the relevant category
without breaking API

i.e
//Audio
BitRate = 100,
Channels,
Duration,

// Documents
Author =200,
Title,
...

Otherwise it'll be an ungrouped mess in future releases if you ever
have to add another audio property.

-
ExtractionPluginManager possibly shouldn't be a QObject?

It /might/ be a good idea to make ExtractionPluginManager static so
you don't load the plugins every time (which can be a bit expensive)




dump.cpp should check there's at least one arg.
(I know it's a test, but you explicitly check for >=2 but not <1)


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


Making KFileMetaData a framework

2014-08-05 Thread Vishesh Handa
Hello people

I would like to make KFileMetaData a framework. It's a simple library for 
extracting file metadata and text. It depends on the i18n and KArchive (can be 
made optional), along with a number of external dependencies which are 
optional.

The code is of decent quality, and we have a certain number of unit tests. 
It's currently only used by Baloo.

I would appreciate it if everyone could review the code once before it gets 
into frameworks.
 
-- 
Vishesh Handa
___
Kde-frameworks-devel mailing list
Kde-frameworks-devel@kde.org
https://mail.kde.org/mailman/listinfo/kde-frameworks-devel


Re: Review Request 119382: Port kio-mtp to KF5

2014-08-05 Thread Jan Grulich

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/119382/
---

(Updated Srp. 5, 2014, 12:36 odp.)


Status
--

This change has been marked as submitted.


Review request for KDE Frameworks and Philipp Schmidt.


Repository: kio-mtp


Description
---

I ported kio-mtp to KF5 including:
kDebug -> qCDebug
KUrl -> QUrl

And it doesn't need Kdelibs4Support.

Also one thing which should be probably considered is including kio-mtp in 
kio-extras.


Diffs
-

  kio_mtp.cpp 6e587fa 
  kio_mtp_helpers.h b66adbd 
  CMakeLists.txt cb27440 
  devicecache.h 1fa3932 
  devicecache.cpp 54ca56f 
  filecache.h 590e799 
  filecache.cpp dfd4fe6 
  kio_mtp.h b05f09b 
  kio_mtp_helpers.cpp 0f9ddbc 

Diff: https://git.reviewboard.kde.org/r/119382/diff/


Testing
---

My phone is visible in Dolphin and I was able to finally copy my photos from 
phone to my laptop.


Thanks,

Jan Grulich

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


Re: KAction Warning

2014-08-05 Thread Christoph Cullmann
> On Saturday 19 July 2014 23:57:15 Christoph Cullmann wrote:
> > Hi,
> > 
> > perhaps that is a stupid question, but last time I read up in the docs,
> > they
> > didn't help me:
> > 
> > KatePart spits out a lot of:
> > 
> > kate(15138)/(default) KXMLGUIFactoryPrivate::saveDefaultActionProperties:
> > Shortcut for KAction  "smart_newline" "Insert Smart Newline" set with
> > QShortcut::setShortcut()! See KAction documentation. kate(15138)/(default)
> > KXMLGUIFactoryPrivate::saveDefaultActionProperties: Shortcut for KAction
> > "tools_indent" "&Indent" set with QShortcut::setShortcut()! See KAction
> > documentation. ...
> > 
> > How to fix that?
> 
> Warning improved in
> http://commits.kde.org/kxmlgui/11cb423c0de0090ea398efa5d1bccd2a39e97987
> 
> -> use KAction::setDefaultShortcut(s)
Thanks, will fix the wrong usages ;)

Greetings
Christoph 

-- 
- Dr.-Ing. Christoph Cullmann -
AbsInt Angewandte Informatik GmbH  Email: cullm...@absint.com
Science Park 1 Tel:   +49-681-38360-22
66123 Saarbrücken  Fax:   +49-681-38360-20
GERMANYWWW:   http://www.AbsInt.com

Geschäftsführung: Dr.-Ing. Christian Ferdinand
Eingetragen im Handelsregister des Amtsgerichts Saarbrücken, HRB 11234
___
Kde-frameworks-devel mailing list
Kde-frameworks-devel@kde.org
https://mail.kde.org/mailman/listinfo/kde-frameworks-devel


Re: KAction Warning

2014-08-05 Thread laurent Montel
Le mardi 5 août 2014 08:19:35 David Faure a écrit :
> On Saturday 19 July 2014 23:57:15 Christoph Cullmann wrote:
> > Hi,
> > 
> > perhaps that is a stupid question, but last time I read up in the docs,
> > they didn't help me:
> > 
> > KatePart spits out a lot of:
> > 
> > kate(15138)/(default) KXMLGUIFactoryPrivate::saveDefaultActionProperties:
> > Shortcut for KAction  "smart_newline" "Insert Smart Newline" set with
> > QShortcut::setShortcut()! See KAction documentation. kate(15138)/(default)
> > KXMLGUIFactoryPrivate::saveDefaultActionProperties: Shortcut for KAction
> > "tools_indent" "&Indent" set with QShortcut::setShortcut()! See KAction
> > documentation. ...
> > 
> > How to fix that?
> 
> Warning improved in
> http://commits.kde.org/kxmlgui/11cb423c0de0090ea398efa5d1bccd2a39e97987
> 
> -> use KAction::setDefaultShortcut(s)

s/KAction::setDefaultShortcut/KActionCollection::setDefaultShortcut/ :)



-- 
Laurent Montel | laurent.mon...@kdab.com | KDE/Qt Senior Software Engineer
KDAB (France) S.A.S., a KDAB Group company
Tel. France +33 (0)4 90 84 08 53,  http://www.kdab.fr


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


Re: Review Request 118514: Add KColumnResizer to KWidgetsAddons

2014-08-05 Thread Aurélien Gâteau

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/118514/
---

(Updated Aug. 5, 2014, 7:50 a.m.)


Status
--

This change has been marked as submitted.


Review request for KDE Frameworks and Christoph Feck.


Repository: kwidgetsaddons


Description
---

I have been asked to add ColumnResizer to KWidgetAddons. ColumnResizer is a 
helper class I created long ago to help maintain uniform widths across layouts. 
You can learn more about it from the blog post I wrote back then: 
http://agateau.com/2011/clean-up-your-layouts-with-columnresizer/ . I 
simplified the API a bit, added some documentation, and here it is.


Diffs
-

  autotests/CMakeLists.txt 854e916 
  autotests/kcolumnresizertest-forms.ui PRE-CREATION 
  autotests/kcolumnresizertest-grid-and-form.ui PRE-CREATION 
  autotests/kcolumnresizertest-grids.ui PRE-CREATION 
  autotests/kcolumnresizertest.h PRE-CREATION 
  autotests/kcolumnresizertest.cpp PRE-CREATION 
  docs/pics/kcolumnresizer.png PRE-CREATION 
  src/CMakeLists.txt 27b9084 
  src/kcolumnresizer.h PRE-CREATION 
  src/kcolumnresizer.cpp PRE-CREATION 
  tests/CMakeLists.txt eccf887 
  tests/kcolumnresizertest.h PRE-CREATION 
  tests/kcolumnresizertest.cpp PRE-CREATION 
  tests/kcolumnresizertest.ui PRE-CREATION 

Diff: https://git.reviewboard.kde.org/r/118514/diff/


Testing
---

Manual test program behaves as expected.


Thanks,

Aurélien Gâteau

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


Re: Review Request 119564: Add define to re-enable Qt functionality we depend on.

2014-08-05 Thread Axel Rasmussen

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/119564/
---

(Updated Aug. 5, 2014, 7:07 a.m.)


Status
--

This change has been discarded.


Review request for Build System, Extra Cmake Modules and KDE Frameworks.


Bugs: 337472
http://bugs.kde.org/show_bug.cgi?id=337472


Repository: extra-cmake-modules


Description
---

Upstream Qt commit e112c2e altered the way QExplicitlySharedDataPointer behaves 
by default, such that it no longer uses a `static_cast` to cast compatible 
pointers. A nontrivial amount of KDE code (I encountered the build error in 
KService, although there are other users) depends on this functionality, so 
this commit adds a define to the build system which re-enables the code in Qt.


Diffs
-

  kde-modules/KDECompilerSettings.cmake fdc930e 

Diff: https://git.reviewboard.kde.org/r/119564/diff/


Testing
---

I applied this patch, attempted to build KService, and the build succeeded, 
rather than exhibiting the compile errors found in the log provided on the bug 
report.


Thanks,

Axel Rasmussen

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


Re: Review Request 119564: Add define to re-enable Qt functionality we depend on.

2014-08-05 Thread Kevin Funk


> On Aug. 2, 2014, 9:04 a.m., Alex Merry wrote:
> > I would rather change the code. The Qt behaviour was changed for a reason, 
> > to prevent accidental use of dangerous behaviour, and I'm not too keen on 
> > undoing that move for all software that uses KDECompilerSettings.
> 
> Sune Vuorela wrote:
> agreed.
> 
> Kevin Funk wrote:
> Yep. I'm also *strongly* in favor of adjusting the code instead of 
> enabling the define.
> 
> In fact, I thought I've fixed all of KF5. (It isn't?). 
> There are some compile errors in code *using* KF5, which I'm trying to 
> port ASAP.
> 
> Axel Rasmussen wrote:
> Would we prefer just using `static_cast` like 
> QExplicitlySharedDataPointer used to do, or shall we use `dynamic_cast` and 
> check for `NULL` results to be extra safe? I am willing to write up a patch 
> that fixes this issue in KService over the next day or two.
> 
> Alex Merry wrote:
> I'd expect most of the cases to be suitable for `static_cast` (if you can 
> guarantee what it will be from the context). If in doubt, state your 
> uncertainty in the review request description so that reviewers know to take 
> extra care.

Note: KService has already been fixed two weeks ago -- see 
https://git.reviewboard.kde.org/r/119241/


- Kevin


---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/119564/#review63665
---


On Aug. 1, 2014, 4:06 p.m., Axel Rasmussen wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/119564/
> ---
> 
> (Updated Aug. 1, 2014, 4:06 p.m.)
> 
> 
> Review request for Build System, Extra Cmake Modules and KDE Frameworks.
> 
> 
> Bugs: 337472
> http://bugs.kde.org/show_bug.cgi?id=337472
> 
> 
> Repository: extra-cmake-modules
> 
> 
> Description
> ---
> 
> Upstream Qt commit e112c2e altered the way QExplicitlySharedDataPointer 
> behaves by default, such that it no longer uses a `static_cast` to cast 
> compatible pointers. A nontrivial amount of KDE code (I encountered the build 
> error in KService, although there are other users) depends on this 
> functionality, so this commit adds a define to the build system which 
> re-enables the code in Qt.
> 
> 
> Diffs
> -
> 
>   kde-modules/KDECompilerSettings.cmake fdc930e 
> 
> Diff: https://git.reviewboard.kde.org/r/119564/diff/
> 
> 
> Testing
> ---
> 
> I applied this patch, attempted to build KService, and the build succeeded, 
> rather than exhibiting the compile errors found in the log provided on the 
> bug report.
> 
> 
> Thanks,
> 
> Axel Rasmussen
> 
>

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