KTabWidget vs QTabWidget

[Jeremy Whiting]

Hey all,

In looking into fixing the remaining issues in Okular's frameworks
branch I realized that in part of the effort to port it away from
KDELibs4Support it got some functionality removed. It was ported from
KTabWidget to QTabWidget but QTabWidget doesn't seem to support drag
and drop the way KTabWidget did. In looking at the KTabWidget
documentation on api.kde.org it still says that KTabWidget is
preferred over QTabWidget [1]. If that's the case why did it end up in
KDELibs4Support?

In reading Qt documentation about drag and drop [2] it seems that you
need to subclass widgets in order to specify any additional mime types
that should be handled by a drop event (which okular made use of so
you could drop documents on it's tab bar to open them). Without
KTabWidget we lose that feature completely unless we subclass
QTabWidget (which we have in KTabWidget... so why not just use it...).
Am I missing something? If not I suggest we reconsider and maybe
move/copy? KTabWidget into KF5::WidgetsAddons as it still provides
functionality we want/need in some cases. I'm not sure what would be
BC or SC in this case tbh (or maybe users of KTabWidget should just
keep using KDELibs4Support?)

BR,
Jeremy


1. 
http://api.kde.org/frameworks-api/frameworks5-apidocs/kdelibs4support/html/classKTabWidget.html
2. http://doc.qt.io/qt-5/dnd.html
___
Kde-frameworks-devel mailing list
Kde-frameworks-devel@kde.org
https://mail.kde.org/mailman/listinfo/kde-frameworks-devel

Re: Review Request 125279: KSycoca: change DB filename to include language and sha1 of the dirs it's built from.

[Albert Astals Cid]

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/125279/#review85643
---

Ship it!


Ship It!

- Albert Astals Cid


On set. 16, 2015, 11:30 p.m., David Faure wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/125279/
> ---
> 
> (Updated set. 16, 2015, 11:30 p.m.)
> 
> 
> Review request for KDE Frameworks and Albert Astals Cid.
> 
> 
> Repository: kservice
> 
> 
> Description
> ---
> 
> This will prevent sycoca-rebuild ping-pong if two apps with different settings
> would share the same file (and keep finding that it's wrong for them),
> and it fixes Albert's bug that "LANG=de kcmshell5 --list" doesn't show German
> translations for the strings coming from desktop files.
> 
> 
> Diffs
> -
> 
>   autotests/ksycocatest.cpp 7c2d91e056726540b8a3a5c679d9b2a93f023c50 
>   docs/kbuildsycoca5/man-kbuildsycoca5.8.docbook 
> 3419e42cb29c6699bc17e6dd46bbc523139c59eb 
>   src/sycoca/kbuildsycoca.cpp 3685211e9da68f14516ec2b3d9a7e6b4f559b6f3 
>   src/sycoca/ksycoca.h c561dfd1550fd28e73144af6d3b2fa9008b17f59 
>   src/sycoca/ksycoca.cpp c5465a828da615e87220304e3f8b160d471edbc7 
> 
> Diff: https://git.reviewboard.kde.org/r/125279/diff/
> 
> 
> Testing
> ---
> 
> unittests, after adjusting ksycocatest which was checking for the old 
> behavior (same file).
> 
> Albert: there we are, finally ;) Your bug should be fixed (I don't have 
> translations installed to test it, though). Thanks for all the reviews!
> 
> 
> Thanks,
> 
> David Faure
> 
>

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

Re: Building KF in version 5.14 on OSX fails for framework kdesignerplugin

[Marko Käning]

Dear Kurt,

On 16 Sep 2015, at 15:20 , Kurt Hindenburg  wrote:
> Git commit 6cbf534c3cd2c5bdda819e092c9778c27e814161 by Kurt Hindenburg.

thanks for your commit!

Looking forward to release 5.15 then, David.
:)

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

Re: Review Request 125300: Fix lock in KProtocolManager

[David Faure]

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/125300/#review85628
---


Recursive mutexes are more costly and a symptom of bad programming. I'd much 
rather that we fix the recursive locking.

Any chance for a unittest? ;)

- David Faure


On Sept. 18, 2015, 4:38 p.m., Aleix Pol Gonzalez wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/125300/
> ---
> 
> (Updated Sept. 18, 2015, 4:38 p.m.)
> 
> 
> Review request for KDE Frameworks.
> 
> 
> Bugs: 350890
> https://bugs.kde.org/show_bug.cgi?id=350890
> 
> 
> Repository: kio
> 
> 
> Description
> ---
> 
> We had a non-recursive mutex and we kept requesting it. Eventually it locked 
> because we were locking the mutex when already locked.
> 
> An alternative fix could be to rearrange the code and release it sooner, but 
> it doesn't seem worth it. (although I had to remove 2 asserts and I feel 
> uneasy about it).
> 
> 
> Diffs
> -
> 
>   src/core/kprotocolmanager.cpp 294ebdf 
> 
> Diff: https://git.reviewboard.kde.org/r/125300/diff/
> 
> 
> Testing
> ---
> 
> Tests pass, cannot reproduce the attached bug anymore.
> 
> 
> 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 125300: Fix lock in KProtocolManager

[Aleix Pol Gonzalez]

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/125300/
---

Review request for KDE Frameworks.


Bugs: 350890
https://bugs.kde.org/show_bug.cgi?id=350890


Repository: kio


Description
---

We had a non-recursive mutex and we kept requesting it. Eventually it locked 
because we were locking the mutex when already locked.

An alternative fix could be to rearrange the code and release it sooner, but it 
doesn't seem worth it. (although I had to remove 2 asserts and I feel uneasy 
about it).


Diffs
-

  src/core/kprotocolmanager.cpp 294ebdf 

Diff: https://git.reviewboard.kde.org/r/125300/diff/


Testing
---

Tests pass, cannot reproduce the attached bug anymore.


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 125259: Support multiple X servers in the NETWM classes

[Martin Gräßlin]

> On Sept. 18, 2015, 2:21 p.m., Thomas Lübking wrote:
> > src/platforms/xcb/netwm_p.h, line 31
> > 
> >
> > errrhemmm ... the idea was actually to keep things more in sync and not 
> > have to match variable, string assignment and netAtomCount by hand, so I 
> > would have actually gone for including an atom header twice on different 
> > macro definition.
> > 
> > Also the presented system drops scope (KDE_NET_WM, NET_WM, WM and ... 
> > UTF8_STRING ;-)
> > 
> > Sorry :-(

sorry, now you have lost me with what you actually want.


- Martin


---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/125259/#review85619
---


On Sept. 18, 2015, 10:19 a.m., Martin Gräßlin wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/125259/
> ---
> 
> (Updated Sept. 18, 2015, 10:19 a.m.)
> 
> 
> Review request for KDE Frameworks.
> 
> 
> Repository: kwindowsystem
> 
> 
> Description
> ---
> 
> So far on first creation of a NETRootInfo or NETWinInfo a static
> initialization of atoms was performed. This meant that the NET classes
> are only able to interact with the X server the first NET class is
> connected to.
> 
> Normally applications don't need to interact with multiple X servers.
> An exception is kwin_wayland which needs a connection to its Xwayland
> server and on the x11 backend to the X server it renders to. So far
> KWin could not use the NET classes for it, causing the rendering window
> to e.g. not have a window title.
> 
> This change removes the implicit constraint on one X server by
> creating a hash of connection and atoms. For each created NET class
> we check whether we have already resolved the atoms, if yes we reuse
> otherwise we create them.
> 
> For the normal use case of one X11 connection the change should be
> rather minimal. Instead of performing the check whether the static
> atoms have been created, it now is a check whether the atoms for the
> connection have been created. The atoms are kept in a
> QSharedDataPointer ensuring that we don't needless copy the atoms into
> each class.
> 
> CHANGELOG: Allow interacting with multiple X servers in the NETWM classes.
> 
> [autotests] NETWM tests get a new X server per test
> 
> Making use of new feature of allowing multiple X connections:
> start X server in init() instead of initTestCase().
> 
> 
> Diffs
> -
> 
>   autotests/netwininfotestwm.cpp a98e12fd690b0250337c7588e1525af1d0dda38c 
>   autotests/netrootinfotestwm.cpp e918873a5281f6ecbcc1769de3dadcf8f6222f6a 
>   autotests/netwininfotestclient.cpp a5b10376b943ea914a9521b5c07f9ef13a65d2f1 
>   src/platforms/xcb/netwm.cpp d99a925ad2b99d5e60ab1dafcb01400b4a6a4c93 
>   src/platforms/xcb/netwm_p.h 917a86ed5b6c83f36e73bbc346360b943d457243 
> 
> Diff: https://git.reviewboard.kde.org/r/125259/diff/
> 
> 
> Testing
> ---
> 
> see adjusted unit tests. Also tried it in kwin_wayland
> 
> 
> Thanks,
> 
> Martin Gräßlin
> 
>

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

Re: Review Request 125259: Support multiple X servers in the NETWM classes

[Thomas Lübking]

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/125259/#review85619
---



src/platforms/xcb/netwm_p.h (line 31)


errrhemmm ... the idea was actually to keep things more in sync and not 
have to match variable, string assignment and netAtomCount by hand, so I would 
have actually gone for including an atom header twice on different macro 
definition.

Also the presented system drops scope (KDE_NET_WM, NET_WM, WM and ... 
UTF8_STRING ;-)

Sorry :-(


- Thomas Lübking


On Sept. 18, 2015, 8:19 vorm., Martin Gräßlin wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/125259/
> ---
> 
> (Updated Sept. 18, 2015, 8:19 vorm.)
> 
> 
> Review request for KDE Frameworks.
> 
> 
> Repository: kwindowsystem
> 
> 
> Description
> ---
> 
> So far on first creation of a NETRootInfo or NETWinInfo a static
> initialization of atoms was performed. This meant that the NET classes
> are only able to interact with the X server the first NET class is
> connected to.
> 
> Normally applications don't need to interact with multiple X servers.
> An exception is kwin_wayland which needs a connection to its Xwayland
> server and on the x11 backend to the X server it renders to. So far
> KWin could not use the NET classes for it, causing the rendering window
> to e.g. not have a window title.
> 
> This change removes the implicit constraint on one X server by
> creating a hash of connection and atoms. For each created NET class
> we check whether we have already resolved the atoms, if yes we reuse
> otherwise we create them.
> 
> For the normal use case of one X11 connection the change should be
> rather minimal. Instead of performing the check whether the static
> atoms have been created, it now is a check whether the atoms for the
> connection have been created. The atoms are kept in a
> QSharedDataPointer ensuring that we don't needless copy the atoms into
> each class.
> 
> CHANGELOG: Allow interacting with multiple X servers in the NETWM classes.
> 
> [autotests] NETWM tests get a new X server per test
> 
> Making use of new feature of allowing multiple X connections:
> start X server in init() instead of initTestCase().
> 
> 
> Diffs
> -
> 
>   autotests/netwininfotestwm.cpp a98e12fd690b0250337c7588e1525af1d0dda38c 
>   autotests/netrootinfotestwm.cpp e918873a5281f6ecbcc1769de3dadcf8f6222f6a 
>   autotests/netwininfotestclient.cpp a5b10376b943ea914a9521b5c07f9ef13a65d2f1 
>   src/platforms/xcb/netwm.cpp d99a925ad2b99d5e60ab1dafcb01400b4a6a4c93 
>   src/platforms/xcb/netwm_p.h 917a86ed5b6c83f36e73bbc346360b943d457243 
> 
> Diff: https://git.reviewboard.kde.org/r/125259/diff/
> 
> 
> Testing
> ---
> 
> see adjusted unit tests. Also tried it in kwin_wayland
> 
> 
> Thanks,
> 
> Martin Gräßlin
> 
>

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

Re: Review Request 125278: KSycoca: add a q pointer to remove more singleton usage

[David Faure]

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/125278/
---

(Updated Sept. 18, 2015, 12:21 p.m.)


Status
--

This change has been marked as submitted.


Review request for KDE Frameworks and Albert Astals Cid.


Changes
---

Submitted with commit 72a14b5a9aadb4dfb6a1978c7d02385175aca815 by David Faure 
to branch master.


Repository: kservice


Description
---

This change is needed for an upcoming unittest which needs to create
another KSycoca instance.


Diffs
-

  src/sycoca/ksycoca.cpp c5465a828da615e87220304e3f8b160d471edbc7 
  src/sycoca/ksycoca_p.h a86148a20fccf038e2d46f5d2d6d460be94910da 

Diff: https://git.reviewboard.kde.org/r/125278/diff/


Testing
---

ctest.

Albert: don't despair, this is the one-before-last patch ;)


Thanks,

David Faure

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

Re: Review Request 125261: Add mimeTypes() to KPluginMetaData

[Sebastian Kügler]

> On Sept. 16, 2015, 9:44 p.m., Sebastian Kügler wrote:
> > autotests/kpluginmetadatatest.cpp, line 110
> > 
> >
> > Should be a list, so enclosed with [ ].
> > 
> > In the header docs, you also say "string", so maybe I'm overlooking 
> > something?
> 
> Alex Richardson wrote:
> You're right, it should be a list. `KPluginMetaData::readStringList()` 
> treats a single string as a list with one element so this does not cause an 
> error and I didn't notice.
> 
> Possibly it should output a warning in that case?

Yes, but as long as it even silently does the right thing, it's not so much of 
an issue. I guess a warning would not hurt, anyway.


- Sebastian


---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/125261/#review85519
---


On Sept. 16, 2015, 4:53 p.m., Alex Richardson wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/125261/
> ---
> 
> (Updated Sept. 16, 2015, 4:53 p.m.)
> 
> 
> Review request for KDE Frameworks, David Faure and Sebastian Kügler.
> 
> 
> Repository: kcoreaddons
> 
> 
> Description
> ---
> 
> When loading a .desktop file this will parse the XDG MimeType= key
> 
> Contrary to KService we don't merge these with ServiceTypes but rather
> have them as a separate property.
> 
> 
> Diffs
> -
> 
>   autotests/data/fakeplugin.desktop 2e186869bf0cb39277ca3f9d2e8603d64d0223c6 
>   autotests/kpluginmetadatatest.cpp 3af5e1b842b0bc231a5ac001112e141f751d2ff5 
>   src/lib/plugin/desktopfileparser.h 98d47ddf0f877c4a25928026b3d5fe169cfc9e75 
>   src/lib/plugin/desktopfileparser.cpp 
> 0b03eb154deb58840c91c12658780c0d492b593c 
>   src/lib/plugin/kpluginmetadata.h 183b0d0583259f7ed74e97858a68c5c388fd0a9a 
>   src/lib/plugin/kpluginmetadata.cpp b13d6dd52827cc03d9473600aa4d2bab8a95a1d4 
> 
> Diff: https://git.reviewboard.kde.org/r/125261/diff/
> 
> 
> Testing
> ---
> 
> Used this for a WIP port of Okular to new Plugin loading. Could also be used 
> by KDevelop instead of the custom X-KDevelop-SupportedMimeTypes property
> 
> Requires https://git.reviewboard.kde.org/r/125263/ to ensure that there are 
> no regressions
> 
> 
> Thanks,
> 
> Alex Richardson
> 
>

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

Re: Review Request 125275: KSycoca: remove all self() methods for factories, store them in KSycoca instead.

[David Faure]

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/125275/
---

(Updated Sept. 18, 2015, 8:54 a.m.)


Status
--

This change has been marked as submitted.


Review request for KDE Frameworks and Albert Astals Cid.


Changes
---

Submitted with commit d5b83aac4475d9d96468dc9ae1b4a9c35b0f72a1 by David Faure 
to branch master.


Repository: kservice


Description
---

This removes the whole template magic and QThreadStorage for each factory,
and goes to a more object oriented design where the factories belong to
a ksycoca instance (= one per thread, plus one for rebuilding).

This even makes it possible to create another ksycoca instance with its
own factories, for special unittests (coming up).


Diffs
-

  autotests/kmimeassociationstest.cpp 5550938abe4d9c9c1ee698ac1b225312690e4aa6 
  src/services/kmimetypefactory.cpp ba07aa0bc9b5d528454cf426b1feadb049402123 
  src/services/kmimetypetrader.cpp 702ef2913d3e218f53bb8b935e3d5ea19d66f7cc 
  src/services/kplugininfo.cpp 93dcb8fc35930d89e4e60efea755b018d329794f 
  src/services/kservice.cpp 045a2dad2775decf6bf427873b9fd1e292a1fe87 
  src/services/kservicefactory.cpp 9b0e0c199818fea774c08a4f8fab5aca417927c8 
  src/services/kservicegroup.cpp 50eed4a5c4d1917ce9a9d4a3ed617a5749b122d3 
  src/services/kservicegroupfactory.cpp 
08e0bdcd765ab08afdb71cabc640fd21a73f4218 
  src/services/kservicetype.cpp 0ad453937d892d55725401dba024f95b95e91405 
  src/services/kservicetypefactory.cpp 62fd230b4f7a6558c707d257796c5967f39c3607 
  src/services/kservicetypeprofile.cpp e4953a6333d044b77f983705036409a25c4fcb1c 
  src/services/kservicetypetrader.cpp ca964e920bd7a9e0e6ab901d99b713b7ea7d3de3 
  src/sycoca/kbuildsycoca.cpp 3685211e9da68f14516ec2b3d9a7e6b4f559b6f3 
  src/sycoca/kbuildsycoca_p.h c24413398694de29a788dd9498435572f5f605a3 
  src/sycoca/ksycoca.h c561dfd1550fd28e73144af6d3b2fa9008b17f59 
  src/sycoca/ksycoca.cpp c5465a828da615e87220304e3f8b160d471edbc7 
  src/sycoca/ksycoca_p.h a86148a20fccf038e2d46f5d2d6d460be94910da 
  src/sycoca/ksycocafactory_p.h c54bfc7b1ce917f11b6451f30863b374f2d94eb3 
  src/sycoca/vfolder_menu.cpp b17ef2d371a2804ebe6029a0b7abfbebe4298d50 
  src/sycoca/vfolder_menu_p.h b90f5efad903b73cbea7b40c142cc9e55e8c5bed 

Diff: https://git.reviewboard.kde.org/r/125275/diff/


Testing
---

ctest (3 times)


Thanks,

David Faure

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

Re: Review Request 125259: Support multiple X servers in the NETWM classes

[Martin Gräßlin]

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/125259/
---

(Updated Sept. 18, 2015, 10:19 a.m.)


Review request for KDE Frameworks.


Changes
---

Changed to an enum approach. I tried to get some logical grouping into the enum 
- if someone has suggestions to further split them, please let me know.

The atoms are still all fetched on start to reduce roundtrips.


Repository: kwindowsystem


Description
---

So far on first creation of a NETRootInfo or NETWinInfo a static
initialization of atoms was performed. This meant that the NET classes
are only able to interact with the X server the first NET class is
connected to.

Normally applications don't need to interact with multiple X servers.
An exception is kwin_wayland which needs a connection to its Xwayland
server and on the x11 backend to the X server it renders to. So far
KWin could not use the NET classes for it, causing the rendering window
to e.g. not have a window title.

This change removes the implicit constraint on one X server by
creating a hash of connection and atoms. For each created NET class
we check whether we have already resolved the atoms, if yes we reuse
otherwise we create them.

For the normal use case of one X11 connection the change should be
rather minimal. Instead of performing the check whether the static
atoms have been created, it now is a check whether the atoms for the
connection have been created. The atoms are kept in a
QSharedDataPointer ensuring that we don't needless copy the atoms into
each class.

CHANGELOG: Allow interacting with multiple X servers in the NETWM classes.

[autotests] NETWM tests get a new X server per test

Making use of new feature of allowing multiple X connections:
start X server in init() instead of initTestCase().


Diffs (updated)
-

  autotests/netwininfotestwm.cpp a98e12fd690b0250337c7588e1525af1d0dda38c 
  autotests/netrootinfotestwm.cpp e918873a5281f6ecbcc1769de3dadcf8f6222f6a 
  autotests/netwininfotestclient.cpp a5b10376b943ea914a9521b5c07f9ef13a65d2f1 
  src/platforms/xcb/netwm.cpp d99a925ad2b99d5e60ab1dafcb01400b4a6a4c93 
  src/platforms/xcb/netwm_p.h 917a86ed5b6c83f36e73bbc346360b943d457243 

Diff: https://git.reviewboard.kde.org/r/125259/diff/


Testing
---

see adjusted unit tests. Also tried it in kwin_wayland


Thanks,

Martin Gräßlin

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

Re: Review Request 125274: KBuildSycoca: remove writing of the ksycoca5stamp file.

[David Faure]

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/125274/
---

(Updated Sept. 18, 2015, 8:08 a.m.)


Status
--

This change has been marked as submitted.


Review request for KDE Frameworks and Albert Astals Cid.


Changes
---

Submitted with commit b83f4d76e2ae5d53a075d6c28da33b8a050605b7 by David Faure 
to branch master.


Repository: kservice


Description
---

The code for reading it was already removed as part of b0c8fd8e64f,
the removal of the --checkstamps option.

The feature still exists anyway, a timestamp of "at least everything until
that time is in the DB" is in the sycoca header, and it used by the timestamp
check in ksycoca.cpp. This was an unnecessary separate file.

REVIEW: 125274


Diffs
-

  src/sycoca/kbuildsycoca.cpp a009d62f24da093988066e8850f9ace5d6f0766d 
  docs/kbuildsycoca5/man-kbuildsycoca5.8.docbook 
3419e42cb29c6699bc17e6dd46bbc523139c59eb 

Diff: https://git.reviewboard.kde.org/r/125274/diff/


Testing
---


Thanks,

David Faure

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

Re: Review Request 125272: KSycoca: rebuild sycoca in-process in most unittests as well.

[David Faure]

---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/125272/
---

(Updated Sept. 18, 2015, 7:17 a.m.)


Status
--

This change has been marked as submitted.


Review request for KDE Frameworks and Albert Astals Cid.


Changes
---

Submitted with commit 46b6bee1e745cecac2007c797ce2f3056497fe03 by David Faure 
to branch master.


Repository: kservice


Description
---

This tests what happens now in applications, in order to catch any possible
issues with the in-process rebuild.
On the other hand, I left the QProcess usage in ksycocatest, so that this
one ensures the kbuildsycoca5 binary still works.

This change caught a nasty code path where the sycoca-header-check
during rebuild would actually trigger a rebuild, which would deadlock
on the lockfile. This bug was already in 5.14.0 but there the symptom
was harder to spot due to the process separation (constant rebuild, bug 352787).


Diffs
-

  autotests/kmimeassociationstest.cpp 5550938abe4d9c9c1ee698ac1b225312690e4aa6 
  autotests/kservicetest.cpp ffd963dfc99fb52ce9c8546defc7148c1a982858 
  autotests/ksycocadicttest.cpp 52ce426fe2a0020f1952389420493f3ff6f32092 
  autotests/ksycocathreadtest.cpp 29a096fea2b1d315a4ca13e241a383d410fd157a 
  src/sycoca/ksycoca.cpp c5465a828da615e87220304e3f8b160d471edbc7 

Diff: https://git.reviewboard.kde.org/r/125272/diff/


Testing
---

unittests


Thanks,

David Faure

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

Re: Review Request 125279: KSycoca: change DB filename to include language and sha1 of the dirs it's built from.

[David Faure]

> On Sept. 17, 2015, 10:23 p.m., Albert Astals Cid wrote:
> > src/sycoca/ksycoca.cpp, line 688
> > 
> >
> > Do we want to qStableSort the list just in case?

We definitely don't. Order does matter. You don't see the same desktop files if 
e.g. /opt is before or after /usr.


- David


---
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/125279/#review85597
---


On Sept. 16, 2015, 11:30 p.m., David Faure wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/125279/
> ---
> 
> (Updated Sept. 16, 2015, 11:30 p.m.)
> 
> 
> Review request for KDE Frameworks and Albert Astals Cid.
> 
> 
> Repository: kservice
> 
> 
> Description
> ---
> 
> This will prevent sycoca-rebuild ping-pong if two apps with different settings
> would share the same file (and keep finding that it's wrong for them),
> and it fixes Albert's bug that "LANG=de kcmshell5 --list" doesn't show German
> translations for the strings coming from desktop files.
> 
> 
> Diffs
> -
> 
>   autotests/ksycocatest.cpp 7c2d91e056726540b8a3a5c679d9b2a93f023c50 
>   docs/kbuildsycoca5/man-kbuildsycoca5.8.docbook 
> 3419e42cb29c6699bc17e6dd46bbc523139c59eb 
>   src/sycoca/kbuildsycoca.cpp 3685211e9da68f14516ec2b3d9a7e6b4f559b6f3 
>   src/sycoca/ksycoca.h c561dfd1550fd28e73144af6d3b2fa9008b17f59 
>   src/sycoca/ksycoca.cpp c5465a828da615e87220304e3f8b160d471edbc7 
> 
> Diff: https://git.reviewboard.kde.org/r/125279/diff/
> 
> 
> Testing
> ---
> 
> unittests, after adjusting ksycocatest which was checking for the old 
> behavior (same file).
> 
> Albert: there we are, finally ;) Your bug should be fixed (I don't have 
> translations installed to test it, though). Thanks for all the reviews!
> 
> 
> Thanks,
> 
> David Faure
> 
>

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