Re: [Interest] Deploying QtWebEngine on Linux with Qt 5.6.0

[Thiago Macieira]

On sexta-feira, 24 de junho de 2016 15:40:17 PDT Mike Jackson wrote:
> I am running Ubuntu 14.04 and Qt 5.6 downloaded from downloads.qt.io. I
> have built my application
> and the application runs but when I try to invoke the QWebEngine classes I
> just get an empty widget and the following error in the terminal:
> 
> /opt/DREAM3D_SDK/Qt5.6.0/5.6/gcc_64/libexec/QtWebEngineProcess: error while
> loading shared libraries: libQt5WebEngineCore.so.5: cannot open shared
> object file: No such file or directory
> 
> The same thing happens when I package my application. I have tried all
> sorts of ways of configuring the qt.conf directory, moving libs into other
> folders, renaming executables but nothing seems to work. Has anyone else
> run into a situation like this? Can anyone make recommendations on how to
> package QWebEngine?

You need to set the LD_LIBRARY_PATH environment variable before your 
application is started. So replace your application with a shell script that 
sets the environment and then starts the application.

-- 
Thiago Macieira - thiago.macieira (AT) intel.com
  Software Architect - Intel Open Source Technology Center

___
Interest mailing list
Interest@qt-project.org
http://lists.qt-project.org/mailman/listinfo/interest

Re: [Interest] 5.8 Features?

[Kevin Mcintyre]

I had this issue after I installed QT a couple of times (the first time I
didn't install with version with Android)

So looking back on my notes I sent to "Interest" - I resolved

"via export LD_LIBRARY_PATH=/home/ubuntu/Qt/5.6/gcc_64/lib in my .bashrc
file."

On Fri, Jun 24, 2016 at 12:59 PM, Jason H  wrote:

> I've solved these, to various exents, or Android and iOS.
> I would share my code as longas everyone promises to not laughs at my
> code. I do not know what is "best" or everyone. I just wrote what I had to
> fill the requirement I had.
>
>
> *Sent:* Friday, June 24, 2016 at 12:20 PM
> *From:* ekke 
> *To:* interest@qt-project.org
>
> *Subject:* Re: [Interest] 5.8 Features?
> Am 24.06.16 um 18:02 schrieb Jason H:
>
> 6 months of latency would be great.
> But the things I talk about are pretty basic on mobile:
> - Foreground/background lifecycle events,
> - Screen wake locks,
> - Notifications (local / remote)
>
> These have been aound since before Qt targeted mobile and are sorely STILL
> missing from Qt.
>
> Things upcoming that I wouldn't complain about having to implement myself:
> - Fingerprint scanning
> As this is relatively new for Android and iOS platforms. Though the Atrix
> (2010) had a fingerprint scanner, but only Android 6 had a platform API.
> iPhone had it as of the 5S.
>
> It's like Qt is on mobile only if you want to put things on the screen and
> do AJAX. But if you really want to do anything really "mobile" you're on
> your own. We still can't control the video recording parameters on iOS
> (Thanks to my company, it will land in 5.6.2 -- was supposed to land in
> 5.6.1). Qt can only really be accurately described to be a Cross-platform
> UI on mobile. Outside of that, you're writing Java and Obj-C. So call it
> cross-platform for mobile is a stretch. I urging Qt to focus on eliminating
> the asterisks, so it's proper Mobile (capital M) platform.
>
> With that said though, Qt's abstraction of various platform services is a
> godsend. The fact that ReactNative gives you access to AVFoundation doesn't
> do a whole lot when you have to write ReactNative that targets AVFoundation
> and more code to target android.media SDK and handle the intricacies of
> both in your own code base. So I think the Qt approach is right. I just
> want more of it. :-)
>
>
> +1
>
> another point: I think that there are many developers out there already
> implementing common missing features in Java and ObjectiveC
>
> would be great to collect and exchange this to help each other - don't
> know where's the best place and it should be promoted by Qt
>
>
> *Sent:* Friday, June 24, 2016 at 9:26 AM
> *From:* "Xavier Bigand" 
> 
> *To:* "Robert Iakobashvili" 
> 
> *Cc:* interest  
> *Subject:* Re: [Interest] 5.8 Features?
> Like you said I think that the iOS and Android progress too fast and on an
> other cadence than Qt.
> We should not forget that Qt has to create a unified cross platform API,
> that is necessary harder than creating a new one for one platform.
>
> I think that a latency of 6 months to a year is still reasonable for Qt
> depending on how it fall with releases.
>
> In my opinion if you need something faster, you may have to consider to
> implement features your self. We started our application with 4.8 and
> necessitas and Qt was much slower than now to integrate new features
> provided by mobile devices. Some features like DPI retrieving wasn't
> correctly implemented so because it was a blocker for us, we fixed it by
> calling the native API on Android.
>
>
>
>
>
> 2016-06-24 15:00 GMT+02:00 Robert Iakobashvili :
>>
>> On Tue, Jun 21, 2016 at 10:55 PM, Jason H  wrote:
>> > I feel like the last few releases have been run by the trolls, and not
>> the users of Qt. I was hoping open governance would enable the community to
>> direct Qt development, but I seem to have misinterpreted what it means. I'm
>> looking for what's going into 5.8.. not much listed on the releases page.
>> >
>> > I'd like to suggest that mobile get some much needed love.
>> > - Application state transitions; Foreground, background
>> > - Background processing API
>> > - Screen wake lock API
>> > - In-app Notifications: local, remote
>> >
>> > While I have those characterized as "mobile" there are things like
>> notifications occurring on desktop platforms.
>> >
>> > Any thoughts?
>> >
>>
>> Agree with Jason that mobile support needs more love
>> and adding "Native, native, native ..."
>>
>> However, it could be that progress made at iOS and Android side is too
>> fast and
>> our expectations from Qt are too high?
>>
>> As any cross-platform framework Qt has its limitations.
>> Still, it has good integration points to allow additions of native code.
>>
>> jm4c to add.
>>
>> Kind regards,
>> Robert
>> ___
>> Interest mailing list
>> Interest@qt-project.org
>> http://lists.qt-project.org/ma

[Interest] Fwd: Re: QUndoStack

[Igor Mironchik]

 Forwarded Message 
Subject:Re: [Interest] QUndoStack
Date:   Fri, 24 Jun 2016 19:38:22 +0300
From:   Igor Mironchik 
To: Bob Hood 



Your approach is possible. But guess that you move rect in scene by 
mouse, then auto redo() is ugly. :)


24.06.2016 18:53 пользователь "Bob Hood" > написал:


   On 6/24/2016 9:42 AM, Alan Ezust wrote:

   There is another thread on this very subject from last week:

   http://lists.qt-project.org/pipermail/interest/2016-June/023114.html
   http://lists.qt-project.org/pipermail/interest/2016-June/023116.html

   I agree with you, the QUndoStack design is really odd and the
   need to execute commands while pushing (in my view) makes it
   less flexible than it could be. I wonder if we will hear from
   anyone who likes this design?


   It has been my experience that this is the typical design of an Undo
   system.  You invoke redo() as you inject the undo action into the
   stack, effectively killing two birds with one stone.  You don't push
   an action onto the stack unless it has actually been applied.  If
   the Undo stack did not apply the action, there's no contract in
   place to guarantee that the action has been, or will be, applied in
   another context.  A subsequent undo() on an action that was never
   applied means you might (in the best case) corrupt the application's
   data.

   It's an effective design, requiring less code, and guaranteeing a
   valid stack state.

   Just my $0.02.
   ___
   Interest mailing list
   Interest@qt-project.org 
   http://lists.qt-project.org/mailman/listinfo/interest



___
Interest mailing list
Interest@qt-project.org
http://lists.qt-project.org/mailman/listinfo/interest

[Interest] Fwd: Re: QUndoStack

[Igor Mironchik]

 Forwarded Message 
Subject:Re: [Interest] QUndoStack
Date:   Fri, 24 Jun 2016 19:35:44 +0300
From:   Igor Mironchik 
To: Alan Ezust 



How I missed it?! Thank you for the links. I solved it by undone flag in 
commands, so redo() does nothing untill undo() done.


24.06.2016 18:42 пользователь "Alan Ezust" > написал:


   There is another thread on this very subject from last week:

   http://lists.qt-project.org/pipermail/interest/2016-June/023114.html
   http://lists.qt-project.org/pipermail/interest/2016-June/023116.html

   I agree with you, the QUndoStack design is really odd and the need
   to execute commands while pushing (in my view) makes it less
   flexible than it could be. I wonder if we will hear from anyone who
   likes this design?



   On Fri, Jun 24, 2016 at 5:22 AM, Igor Mironchik
   mailto:igor.mironc...@gmail.com>> wrote:

   Hello.

   I'm adding undo stack to my app. But in the documentation I found:

   voidQUndoStack::push(QUndoCommand *cmd)

   Pushes cmd on the stack or merges it with the most recently
   executed command. In either case, executes cmd by calling its
   redo () function.

   And in sources:

   void QUndoStack::push(QUndoCommand *cmd)
   {
Q_D(QUndoStack);
cmd->redo();

   So my question is why just pushed to the stack undo command
   executes? I.e. why cmd->redo() calls in push()?

   If I understand all correctly then cmd->redo() should be called
   only in QUndoStack::redo() method. And in push I just want to
   push new undo command to the stack...


   ___
   Interest mailing list
   Interest@qt-project.org 
   http://lists.qt-project.org/mailman/listinfo/interest




___
Interest mailing list
Interest@qt-project.org
http://lists.qt-project.org/mailman/listinfo/interest

Re: [Interest] 5.8 Features?

[Jason H]

I've solved these, to various exents, or Android and iOS.

I would share my code as longas everyone promises to not laughs at my code. I do not know what is "best" or everyone. I just wrote what I had to fill the requirement I had. 

 

 



Sent: Friday, June 24, 2016 at 12:20 PM
From: ekke 
To: interest@qt-project.org
Subject: Re: [Interest] 5.8 Features?



Am 24.06.16 um 18:02 schrieb Jason H:




6 months of latency would be great.

But the things I talk about are pretty basic on mobile:

- Foreground/background lifecycle events,

- Screen wake locks,

- Notifications (local / remote)

 

These have been aound since before Qt targeted mobile and are sorely STILL missing from Qt.

 

Things upcoming that I wouldn't complain about having to implement myself:

- Fingerprint scanning

As this is relatively new for Android and iOS platforms. Though the Atrix (2010) had a fingerprint scanner, but only Android 6 had a platform API. iPhone had it as of the 5S. 

 

It's like Qt is on mobile only if you want to put things on the screen and do AJAX. But if you really want to do anything really "mobile" you're on your own. We still can't control the video recording parameters on iOS (Thanks to my company, it will land in 5.6.2 -- was supposed to land in 5.6.1). Qt can only really be accurately described to be a Cross-platform UI on mobile. Outside of that, you're writing Java and Obj-C. So call it cross-platform for mobile is a stretch. I urging Qt to focus on eliminating the asterisks, so it's proper Mobile (capital M) platform. 

 

With that said though, Qt's abstraction of various platform services is a godsend. The fact that ReactNative gives you access to AVFoundation doesn't do a whole lot when you have to write ReactNative that targets AVFoundation and more code to target android.media SDK and handle the intricacies of both in your own code base. So I think the Qt approach is right. I just want more of it. :-)

 



+1

another point: I think that there are many developers out there already implementing common missing features in Java and ObjectiveC

would be great to collect and exchange this to help each other - don't know where's the best place and it should be promoted by Qt



 

Sent: Friday, June 24, 2016 at 9:26 AM
From: "Xavier Bigand" 
To: "Robert Iakobashvili" 
Cc: interest 
Subject: Re: [Interest] 5.8 Features?




Like you said I think that the iOS and Android progress too fast and on an other cadence than Qt.
We should not forget that Qt has to create a unified cross platform API, that is necessary harder than creating a new one for one platform.

 

I think that a latency of 6 months to a year is still reasonable for Qt depending on how it fall with releases.

 

In my opinion if you need something faster, you may have to consider to implement features your self. We started our application with 4.8 and necessitas and Qt was much slower than now to integrate new features provided by mobile devices. Some features like DPI retrieving wasn't correctly implemented so because it was a blocker for us, we fixed it by calling the native API on Android.

 

 

 

 


 
2016-06-24 15:00 GMT+02:00 Robert Iakobashvili :

On Tue, Jun 21, 2016 at 10:55 PM, Jason H  wrote:
> I feel like the last few releases have been run by the trolls, and not the users of Qt. I was hoping open governance would enable the community to direct Qt development, but I seem to have misinterpreted what it means. I'm looking for what's going into 5.8.. not much listed on the releases page.
>
> I'd like to suggest that mobile get some much needed love.
> - Application state transitions; Foreground, background
> - Background processing API
> - Screen wake lock API
> - In-app Notifications: local, remote
>
> While I have those characterized as "mobile" there are things like notifications occurring on desktop platforms.
>
> Any thoughts?
>

Agree with Jason that mobile support needs more love
and adding "Native, native, native ..."

However, it could be that progress made at iOS and Android side is too fast and
our expectations from Qt are too high?

As any cross-platform framework Qt has its limitations.
Still, it has good integration points to allow additions of native code.

jm4c to add.

Kind regards,
Robert
___
Interest mailing list
Interest@qt-project.org
http://lists.qt-project.org/mailman/listinfo/interest

 

 
--


Xavier


___ Interest mailing list Interest@qt-project.org http://lists.qt-project.org/mailman/listinfo/interest




 

 
 

___
Interest mailing list
Interest@qt-project.org
http://lists.qt-project.org/mailman/listinfo/interest



 

___ Interest mailing list Interest@qt-project.org http://lists.qt-project.org/mailman/listinfo/interest




___
Interest mailin

[Interest] Deploying QtWebEngine on Linux with Qt 5.6.0

[Mike Jackson]

I am running Ubuntu 14.04 and Qt 5.6 downloaded from downloads.qt.io. I
have built my application
and the application runs but when I try to invoke the QWebEngine classes I
just get an empty widget and the following error in the terminal:

/opt/DREAM3D_SDK/Qt5.6.0/5.6/gcc_64/libexec/QtWebEngineProcess: error while
loading shared libraries: libQt5WebEngineCore.so.5: cannot open shared
object file: No such file or directory

The same thing happens when I package my application. I have tried all
sorts of ways of configuring the qt.conf directory, moving libs into other
folders, renaming executables but nothing seems to work. Has anyone else
run into a situation like this? Can anyone make recommendations on how to
package QWebEngine?

I have copied over all the resources and translations so that is not the
issue.

Thanks.
-- 
Mike Jackson
___
Interest mailing list
Interest@qt-project.org
http://lists.qt-project.org/mailman/listinfo/interest

Re: [Interest] Native event filter in QtService

[Konstantin Shegunov]

On Fri, Jun 24, 2016 at 5:56 PM, Julius Bullinger <
julius.bullin...@asctec.de> wrote:

> It still doesn’t work like in a QCodeApplication because I only receive 
> windows_dispatcher_MSG
> and no windows_generic_MSG, but this may be Window’s fault.
>

One thing I forgot to mention, you could try running it with the --fake
switch to see if you get those messages you're after (--fake will simply
run it as a regular console application).
Comparing the results from that with QCoreApplication should hint whether
the problem is with my code, or if Windows just doesn't send these for
services.

Kind regards.
___
Interest mailing list
Interest@qt-project.org
http://lists.qt-project.org/mailman/listinfo/interest

Re: [Interest] Native event filter in QtService

[Konstantin Shegunov]

On Fri, Jun 24, 2016 at 5:56 PM, Julius Bullinger <
julius.bullin...@asctec.de> wrote:

> Replacing QtService with your QtDaemon was relatively straight forward,
> and as a result I can receive native events now!
>

Excellent! I'm glad it works for you and thanks for the contribution, I'll
be approving the pull request shortly.

(It still doesn’t work like in a QCodeApplication because I only
receive windows_dispatcher_MSG
> and no windows_generic_MSG, but this may be Window’s fault.)
>
>
Well, I don't know, to be honest. I don't mess with the event loop at all.
The only "strange" stuff that I do is to start additional thread so not to
block main() when the service is started. This should not interfere with
the native events you receive though ...

Kind regards.
___
Interest mailing list
Interest@qt-project.org
http://lists.qt-project.org/mailman/listinfo/interest

Re: [Interest] QUndoStack

[william.croc...@analog.com]

On 06/24/2016 11:52 AM, Bob Hood wrote:

On 6/24/2016 9:42 AM, Alan Ezust wrote:

There is another thread on this very subject from last week:

http://lists.qt-project.org/pipermail/interest/2016-June/023114.html
http://lists.qt-project.org/pipermail/interest/2016-June/023116.html

I agree with you, the QUndoStack design is really odd and the need to execute
commands while pushing (in my view) makes it less flexible than it could be. I
wonder if we will hear from anyone who likes this design?


It has been my experience that this is the typical design of an Undo system. You
invoke redo() as you inject the undo action into the stack, effectively killing
two birds with one stone. You don't push an action onto the stack unless it has
actually been applied. If the Undo stack did not apply the action, there's no
contract in place to guarantee that the action has been, or will be, applied in
another context. A subsequent undo() on an action that was never applied means
you might (in the best case) corrupt the application's data.

It's an effective design, requiring less code, and guaranteeing a valid stack
state.



I have been using it for a while and find it acceptable as well.
Create a command to do something, push it onto the stack and you are done.
I had an issue with "macros" until I just determined that I needed to use them 
all of the time.



Just my $0.02.
___
Interest mailing list
Interest@qt-project.org
http://lists.qt-project.org/mailman/listinfo/interest



___
Interest mailing list
Interest@qt-project.org
http://lists.qt-project.org/mailman/listinfo/interest

Re: [Interest] 5.8 Features?

[ekke]

Am 24.06.16 um 18:02 schrieb Jason H:
> 6 months of latency would be great.
> But the things I talk about are pretty basic on mobile:
> - Foreground/background lifecycle events,
> - Screen wake locks,
> - Notifications (local / remote)
>  
> These have been aound since before Qt targeted mobile and are sorely
> STILL missing from Qt.
>  
> Things upcoming that I wouldn't complain about having to implement myself:
> - Fingerprint scanning
> As this is relatively new for Android and iOS platforms. Though the
> Atrix (2010) had a fingerprint scanner, but only Android 6 had a
> platform API. iPhone had it as of the 5S. 
>  
> It's like Qt is on mobile only if you want to put things on the screen
> and do AJAX. But if you really want to do anything really "mobile"
> you're on your own. We still can't control the video recording
> parameters on iOS (Thanks to my company, it will land in 5.6.2 -- was
> supposed to land in 5.6.1). Qt can only really be accurately described
> to be a Cross-platform UI on mobile. Outside of that, you're writing
> Java and Obj-C. So call it cross-platform for mobile is a stretch. I
> urging Qt to focus on eliminating the asterisks, so it's proper Mobile
> (capital M) platform. 
>  
> With that said though, Qt's abstraction of various platform services
> is a godsend. The fact that ReactNative gives you access to
> AVFoundation doesn't do a whole lot when you have to write ReactNative
> that targets AVFoundation and more code to target android.media
> SDK and handle the intricacies of both in your own code base. So I
> think the Qt approach is right. I just want more of it. :-)
>  
+1

another point: I think that there are many developers out there already
implementing common missing features in Java and ObjectiveC

would be great to collect and exchange this to help each other - don't
know where's the best place and it should be promoted by Qt
>  
> *Sent:* Friday, June 24, 2016 at 9:26 AM
> *From:* "Xavier Bigand" 
> *To:* "Robert Iakobashvili" 
> *Cc:* interest 
> *Subject:* Re: [Interest] 5.8 Features?
> Like you said I think that the iOS and Android progress too fast and
> on an other cadence than Qt.
> We should not forget that Qt has to create a unified cross platform
> API, that is necessary harder than creating a new one for one platform.
>  
> I think that a latency of 6 months to a year is still reasonable for
> Qt depending on how it fall with releases.
>  
> In my opinion if you need something faster, you may have to consider
> to implement features your self. We started our application with 4.8
> and necessitas and Qt was much slower than now to integrate new
> features provided by mobile devices. Some features like DPI retrieving
> wasn't correctly implemented so because it was a blocker for us, we
> fixed it by calling the native API on Android.
>  
>  
>  
>  
>  
> 2016-06-24 15:00 GMT+02:00 Robert Iakobashvili :
>
> On Tue, Jun 21, 2016 at 10:55 PM, Jason H  wrote:
> > I feel like the last few releases have been run by the trolls,
> and not the users of Qt. I was hoping open governance would enable
> the community to direct Qt development, but I seem to have
> misinterpreted what it means. I'm looking for what's going into
> 5.8.. not much listed on the releases page.
> >
> > I'd like to suggest that mobile get some much needed love.
> > - Application state transitions; Foreground, background
> > - Background processing API
> > - Screen wake lock API
> > - In-app Notifications: local, remote
> >
> > While I have those characterized as "mobile" there are things
> like notifications occurring on desktop platforms.
> >
> > Any thoughts?
> >
>
> Agree with Jason that mobile support needs more love
> and adding "Native, native, native ..."
>
> However, it could be that progress made at iOS and Android side is
> too fast and
> our expectations from Qt are too high?
>
> As any cross-platform framework Qt has its limitations.
> Still, it has good integration points to allow additions of native
> code.
>
> jm4c to add.
>
> Kind regards,
> Robert
> ___
> Interest mailing list
> Interest@qt-project.org
> http://lists.qt-project.org/mailman/listinfo/interest
>
>  
>  
> -- 
> Xavier
> ___ Interest mailing list
> Interest@qt-project.org
> http://lists.qt-project.org/mailman/listinfo/interest
>
>
> ___
> Interest mailing list
> Interest@qt-project.org
> http://lists.qt-project.org/mailman/listinfo/interest



___
Interest mailing list
Interest@qt-project.org
http://lists.qt-project.org/mailman/listinfo/interest

Re: [Interest] QMAKE_DEFAULT_INCDIRS = "/opt/local/include /usr/include" and building QtWebengine

[Thiago Macieira]

On sexta-feira, 24 de junho de 2016 12:59:13 PDT René J. V. Bertin wrote:
> -echo `echo "$1" | sed 's,^[^ ]* .*$,"&",'`
> +echo `echo $@ | sed 's,^[^ ]* .*$,&,'`

This has no effect on your issue and in fact makes it worse.

Previously, shellQuoteLines * printed *. Now it prints the list of files.

-- 
Thiago Macieira - thiago.macieira (AT) intel.com
  Software Architect - Intel Open Source Technology Center

___
Interest mailing list
Interest@qt-project.org
http://lists.qt-project.org/mailman/listinfo/interest

Re: [Interest] 5.8 Features?

[Jason H]

6 months of latency would be great.

But the things I talk about are pretty basic on mobile:

- Foreground/background lifecycle events,

- Screen wake locks,

- Notifications (local / remote)

 

These have been aound since before Qt targeted mobile and are sorely STILL missing from Qt.

 

Things upcoming that I wouldn't complain about having to implement myself:

- Fingerprint scanning

As this is relatively new for Android and iOS platforms. Though the Atrix (2010) had a fingerprint scanner, but only Android 6 had a platform API. iPhone had it as of the 5S. 

 

It's like Qt is on mobile only if you want to put things on the screen and do AJAX. But if you really want to do anything really "mobile" you're on your own. We still can't control the video recording parameters on iOS (Thanks to my company, it will land in 5.6.2 -- was supposed to land in 5.6.1). Qt can only really be accurately described to be a Cross-platform UI on mobile. Outside of that, you're writing Java and Obj-C. So call it cross-platform for mobile is a stretch. I urging Qt to focus on eliminating the asterisks, so it's proper Mobile (capital M) platform. 

 

With that said though, Qt's abstraction of various platform services is a godsend. The fact that ReactNative gives you access to AVFoundation doesn't do a whole lot when you have to write ReactNative that targets AVFoundation and more code to target android.media SDK and handle the intricacies of both in your own code base. So I think the Qt approach is right. I just want more of it. :-)

 

 

Sent: Friday, June 24, 2016 at 9:26 AM
From: "Xavier Bigand" 
To: "Robert Iakobashvili" 
Cc: interest 
Subject: Re: [Interest] 5.8 Features?




Like you said I think that the iOS and Android progress too fast and on an other cadence than Qt.
We should not forget that Qt has to create a unified cross platform API, that is necessary harder than creating a new one for one platform.

 

I think that a latency of 6 months to a year is still reasonable for Qt depending on how it fall with releases.

 

In my opinion if you need something faster, you may have to consider to implement features your self. We started our application with 4.8 and necessitas and Qt was much slower than now to integrate new features provided by mobile devices. Some features like DPI retrieving wasn't correctly implemented so because it was a blocker for us, we fixed it by calling the native API on Android.

 

 

 

 


 
2016-06-24 15:00 GMT+02:00 Robert Iakobashvili :

On Tue, Jun 21, 2016 at 10:55 PM, Jason H  wrote:
> I feel like the last few releases have been run by the trolls, and not the users of Qt. I was hoping open governance would enable the community to direct Qt development, but I seem to have misinterpreted what it means. I'm looking for what's going into 5.8.. not much listed on the releases page.
>
> I'd like to suggest that mobile get some much needed love.
> - Application state transitions; Foreground, background
> - Background processing API
> - Screen wake lock API
> - In-app Notifications: local, remote
>
> While I have those characterized as "mobile" there are things like notifications occurring on desktop platforms.
>
> Any thoughts?
>

Agree with Jason that mobile support needs more love
and adding "Native, native, native ..."

However, it could be that progress made at iOS and Android side is too fast and
our expectations from Qt are too high?

As any cross-platform framework Qt has its limitations.
Still, it has good integration points to allow additions of native code.

jm4c to add.

Kind regards,
Robert
___
Interest mailing list
Interest@qt-project.org
http://lists.qt-project.org/mailman/listinfo/interest

 

 
--


Xavier


___ Interest mailing list Interest@qt-project.org http://lists.qt-project.org/mailman/listinfo/interest



___
Interest mailing list
Interest@qt-project.org
http://lists.qt-project.org/mailman/listinfo/interest

Re: [Interest] QUndoStack

[Bob Hood]

On 6/24/2016 9:42 AM, Alan Ezust wrote:

There is another thread on this very subject from last week:

http://lists.qt-project.org/pipermail/interest/2016-June/023114.html
http://lists.qt-project.org/pipermail/interest/2016-June/023116.html

I agree with you, the QUndoStack design is really odd and the need to 
execute commands while pushing (in my view) makes it less flexible than it 
could be. I wonder if we will hear from anyone who likes this design?


It has been my experience that this is the typical design of an Undo system.  
You invoke redo() as you inject the undo action into the stack, effectively 
killing two birds with one stone.  You don't push an action onto the stack 
unless it has actually been applied.  If the Undo stack did not apply the 
action, there's no contract in place to guarantee that the action has been, or 
will be, applied in another context.  A subsequent undo() on an action that 
was never applied means you might (in the best case) corrupt the application's 
data.


It's an effective design, requiring less code, and guaranteeing a valid stack 
state.


Just my $0.02.
___
Interest mailing list
Interest@qt-project.org
http://lists.qt-project.org/mailman/listinfo/interest

Re: [Interest] QUndoStack

[Alan Ezust]

There is another thread on this very subject from last week:

http://lists.qt-project.org/pipermail/interest/2016-June/023114.html
http://lists.qt-project.org/pipermail/interest/2016-June/023116.html

I agree with you, the QUndoStack design is really odd and the need to
execute commands while pushing (in my view) makes it less flexible than it
could be. I wonder if we will hear from anyone who likes this design?



On Fri, Jun 24, 2016 at 5:22 AM, Igor Mironchik 
wrote:

> Hello.
>
> I'm adding undo stack to my app. But in the documentation I found:
>
> void QUndoStack::push(QUndoCommand  *cmd)
>
> Pushes cmd on the stack or merges it with the most recently executed
> command. In either case, executes cmd by calling its redo
> () function.
> And in sources:
>
> void QUndoStack::push(QUndoCommand *cmd)
> {
> Q_D(QUndoStack);
> cmd->redo();
>
> So my question is why just pushed to the stack undo command executes? I.e.
> why cmd->redo() calls in push()?
>
> If I understand all correctly then cmd->redo() should be called only in
> QUndoStack::redo() method. And in push I just want to push new undo command
> to the stack...
>
>
> ___
> Interest mailing list
> Interest@qt-project.org
> http://lists.qt-project.org/mailman/listinfo/interest
>
>
___
Interest mailing list
Interest@qt-project.org
http://lists.qt-project.org/mailman/listinfo/interest

Re: [Interest] Native event filter in QtService

[Julius Bullinger]

Hello Konstantin,

this looks really good! Replacing QtService with your QtDaemon was relatively 
straight forward, and as a result I can receive native events now! (It still 
doesn’t work like in a QCodeApplication because I only receive 
windows_dispatcher_MSG and no windows_generic_MSG, but this may be Window’s 
fault.)

Thanks for this suggestion!

Best regards,
Julius

Von: Konstantin Shegunov [mailto:kshegu...@gmail.com]
Gesendet: Freitag, 24. Juni 2016 11:36 Uhr
An: Julius Bullinger 
Cc: Interests Qt 
Betreff: Re: [Interest] Native event filter in QtService

Hi,
If I may be so bold, I've wrapped up a (Qt 5) replacement for QtService 
recently. You might be interested to try it out (I've chosen the MIT license).
https://bitbucket.org/nye/qtdaemon

Kind regards.
___
Interest mailing list
Interest@qt-project.org
http://lists.qt-project.org/mailman/listinfo/interest

Re: [Interest] Problems with QDockWidgets in Qt5.6.1 and Qt5.7

[Chris Gripeos]

Just a heads up,


I just downloaded latest Qt from github and verified that all the docking 
issues we've seen in Qt5.6.1 and Qt5.7.0 are fixed.


It must have been the CL I mentioned below.


- Chris




From: Interest 
 on behalf of 
Chris Gripeos 
Sent: Thursday, June 23, 2016 8:33:23 PM
To: interest@qt-project.org
Subject: Re: [Interest] Problems with QDockWidgets in Qt5.6.1 and Qt5.7


I wonder if this is related to my problem:


https://github.com/qt/qtbase/commit/f54bd20f6afbbd7bcbc55b3f3f59ed15f91f8ecd

[https://avatars3.githubusercontent.com/u/959326?v=3&s=200]

Revert "End the drag if the dockwidget is being hidden" · 
qt/qtbase@f54bd20
github.com
This reverts commit 75b705fec8e9517047d7dfa98203edff69f2bf8a and 
798e0064e9be78f8320ff25a9af50d1b5e5badb1. This broke drag and drop of 
QDockWidget if there is no animations (QTBUG-54185) and this...





From: Interest 
 on behalf of 
Chris Gripeos 
Sent: Thursday, June 23, 2016 7:25:12 PM
To: interest@qt-project.org
Subject: [Interest] Problems with QDockWidgets in Qt5.6.1 and Qt5.7


Hello,


I'm seeing some new and strange behaviour after upgrading our app from Qt5.6.0 
to Qt5.6.1 and even Qt5.7.0 (5.6.1 and 5.7 have the same exact behaviour).


I have a QDockWidget with:


setAllowedAreas(Qt::AllDockWidgetAreas)


And I have a QMainWindow with:


setDockOptions(QMainWindow::AllowTabbedDocks | QMainWindow::AllowNestedDocks);
setTabPosition(Qt::AllDockWidgetAreas, QTabWidget::North);


The QMainWindow is populated with a Central Widget that has a minimum width, 
but still has plenty of room to add the QDockWidget to the left or right areas.


The QDockWidget has a minimum width and height, but still has plenty of room to 
be docked in the QMainWindow.



Now, when I drag the QDockWidget in the QMainWindow, Qt highlights the areas on 
the left, top, bottom, and right as I drag the QDockWidget around. This tells 
me that if the dock areas are highlighted I should be allowed to drop the 
QDockWidget on them.


The problem is that as soon as i drop the QDockWidget in those areas, the 
Widget only gets docked on the Right Area.


No matter which area I try to drop it on, it always ends up on the Right.


In Qt5.6.0 and earlier (we were on Qt4.8 before that):

- Qt would *only* highlight areas where a Widget was allowed to be docked. 
Qt5.6.1 seems to highlight all areas, even when not allowed (no-fit).

- Docking was possible on the highlighted areas without the QDockWidget always 
getting docked on the Right.


Has anybody else experience such a problem with Qt5.6.1 or Qt5.7.0?


Does anybody know what it might be that we're doing wrong?



Thanks in advance for any help with this problem!


- Chris

___
Interest mailing list
Interest@qt-project.org
http://lists.qt-project.org/mailman/listinfo/interest

Re: [Interest] Custom QSGGeometryNode Scene blending issue

[Roberto Garrido]

Hi Gunnar,
we were creating our texture using "quickWindow->createTextureFromId",
passing our FBO texture id as parameter, but forgot to add the "
CreateTextureOption::TextureHasAlphaChannel". I think we have the desired
effect now.

Thank you so much for helping us to narrow this down.
Cheers,
Robert.

On 24 June 2016 at 10:40, Gunnar Sletta  wrote:

> Since the window's background is being cleared when you draw the
> framebuffer, it sounds as the window's default framebuffer is bound at the
> time of clearing. That or that the FBO is indeed cleared to fully
> transparent and rendered without blending, thus overwriting the pixels
> window.
>
> To further debug this, I would have first checked the contents of the FBO,
> using QOpenGLFramebufferObject::toImage() or using glReadPixels if this is
> your own FBO, and then look at the blending that happens in the scene graph
> if that one is ok. What does the QSGTexture instance look like that you use
> to wrap the FBO. Does it report that there is alpha?
>
> QSG_VISUALIZE=overdraw will tell you whether the FBO is drawn in the scene
> graph using blending or not.
>
> Or simply use QQuickFramebufferObject, of course :)
>
> cheers,
> Gunnar
>
> > On 23 Jun 2016, at 17:26, Roberto Garrido <
> robertogarridomar...@gmail.com> wrote:
> >
> > Gunnar,
> > thanks for your detailed response.
> > We don't discard going for the QQuickFramebufferObject soon, but
> currently what we have is the Item/QSGGeometryNode setup.
> > I've been doing some tests, and I cannot manage to get the desired
> effect. The most silly example I can think of is to clear our buffer with a
> fully transparent color (as you suggested), then render a fullscreen quad
> with also a fully transparent color (using premultiplied alpha -> 1, 1 -
> SRC_ALPHA). That should give us fully transparent item, and the qml scene
> behind should be shown. But what we see is the desktop itself, not the QML
> items behind. Does it give you any clue on what could I be doing wrong?
> >
> > Our node (derived from QSGGeometryNode) takes a QSGTextureMaterial (with
> the default shader for this type of material), and we have also activated
> the QSGMaterial::Blending flag.
> >
> > Thanks,
> > Robert.
> >
> > On 7 June 2016 at 09:41, Gunnar Sletta  wrote:
> > Hi Roberto,
> >
> > The opacity of any geometry node is decided by the opacity that comes
> out of the fragment shader of its material. Depending on how the alpha
> logic exists in your custom GL renderer, there are two ways to do this.
> >
> > Option 1: If most of the FBO is partially transparent, or if there are
> bits and pieces that are overlaid on top of the rest of the UI, like
> controls on a HUD, then you want to render into an FBO with an alpha
> buffer, clear it with fully transparent, draw your custom GL stuff and then
> alpha blend the entire FBO on top of the QML scene. You do alpha blending
> of the fbo content by setting the QSGMaterial::Blending flag and returning
> a premultiplied alpha color from the fragment shader. The
> QSGTextureMaterial has a fragment shader that does the right thing.
> >
> > Option 2: If you have an opaque FBO with certain sections, say a given
> rectangle, fully transparent, you can draw the FBO as fully opaque, but
> create a geometry that excludes the fully transparent region. This has the
> benefit you will be drawing more opaque stuff which is generally faster for
> the GPU to handle, compared to blended stuff.
> >
> > It is of course possible to do a mixture of 1 and 2, by creating two
> separate materials, both referring to the FBO. For the fully opaque
> sections of the FBO, you use an opaque material to draw it which is fast.
> For the semi-transparent sections of the FBO you use a different geometry
> node with a material that enables blending. The fully transparent regions
> are fully excluded from either geometry, so those cost nothing. (something
> like the MaskedImage I have here:
> https://github.com/qtproject/playground-scenegraph/tree/master/shapes)
> >
> > Btw, if you used QQuickFramebufferObject instead of rolling your own
> custom Item/node/material, you would get option 1 by default :)
> >
> > cheers,
> > Gunnar
> >
> > > On 06 Jun 2016, at 18:05, Roberto Garrido <
> robertogarridomar...@gmail.com> wrote:
> > >
> > > Hi all!
> > > We have a custom QQuickItem which in turn creates a custom
> QSGGeometryNode. We reimplement the QSGGeometryNode's preprocess() method
> in order to draw our scene to a FrameBufferObject with a texture attached.
> This OpenGL texture ID is shared with the QSGGeometryNode's material. This
> way, we can render our custom OpenGL scene into a QML item. So far, so good.
> > >
> > > However, we want to use alpha blending on some parts of our scene, in
> order to let the user see the QML items behind our custom item. Something
> similar to what Item::opacity value is doing, but only for a specific area
> of our Item. It's like having a hole inside our Item that allows the user
> to see w

Re: [Interest] 5.8 Features?

[Xavier Bigand]

Like you said I think that the iOS and Android progress too fast and on an
other cadence than Qt.
We should not forget that Qt has to create a unified cross platform API,
that is necessary harder than creating a new one for one platform.

I think that a latency of 6 months to a year is still reasonable for Qt
depending on how it fall with releases.

In my opinion if you need something faster, you may have to consider to
implement features your self. We started our application with 4.8 and
necessitas and Qt was much slower than now to integrate new features
provided by mobile devices. Some features like DPI retrieving wasn't
correctly implemented so because it was a blocker for us, we fixed it by
calling the native API on Android.





2016-06-24 15:00 GMT+02:00 Robert Iakobashvili :

> On Tue, Jun 21, 2016 at 10:55 PM, Jason H  wrote:
> > I feel like the last few releases have been run by the trolls, and not
> the users of Qt. I was hoping open governance would enable the community to
> direct Qt development, but I seem to have misinterpreted what it means. I'm
> looking for what's going into 5.8.. not much listed on the releases page.
> >
> > I'd like to suggest that mobile get some much needed love.
> > - Application state transitions; Foreground, background
> > - Background processing API
> > - Screen wake lock API
> > - In-app Notifications: local, remote
> >
> > While I have those characterized as "mobile" there are things like
> notifications occurring on desktop platforms.
> >
> > Any thoughts?
> >
>
> Agree with Jason that mobile support needs more love
> and adding "Native, native, native ..."
>
> However, it could be that progress made at iOS and Android side is too
> fast and
> our expectations from Qt are too high?
>
> As any cross-platform framework Qt has its limitations.
> Still, it has good integration points to allow additions of native code.
>
> jm4c to add.
>
> Kind regards,
> Robert
> ___
> Interest mailing list
> Interest@qt-project.org
> http://lists.qt-project.org/mailman/listinfo/interest
>



-- 
Xavier
___
Interest mailing list
Interest@qt-project.org
http://lists.qt-project.org/mailman/listinfo/interest

Re: [Interest] Native event filter in QtService

[Tony Rietwyk]

Doh!  I looked but missed that paragraph.  Sorry for the noise! 

 

Tony

 

 

From: Interest
[mailto:interest-bounces+tony=rightsoft.com...@qt-project.org] On Behalf Of
Julius Bullinger
Sent: Friday, 24 June 2016 7:03 PM
To: interest@qt-project.org
Subject: Re: [Interest] Native event filter in QtService

 

Hi Tony,

 

thanks for your reply. Reading the documentation of
QAbstractEventDispatcher::installNativeEventFilter(), it says in the fourth
paragraph

 

If multiple event filters are installed, the filter that was
installed last is activated first.

 

Thus, I was under the impression that I should be able to add my own filter,
without overwriting the one installed by QtService. Is this incorrect?

 

Regards,

Julius

 

Von: Interest
[mailto:interest-bounces+julius.bullinger=asctec...@qt-project.org] Im
Auftrag von Tony Rietwyk
Gesendet: Freitag, 24. Juni 2016 10:57 Uhr
An: interest@qt-project.org
Betreff: Re: [Interest] Native event filter in QtService

 

Hi Julius, 

 

qtservice_win.cpp around line 830 at your reference [1] installs its own
nativeEventFilter - probably displacing yours.  

 

I suspect you'll need to merge the Qt filter into yours, and get rid of that
install.  

 

Regards, 

 

Tony

 

 

From: Interest
[mailto:interest-bounces+tony=rightsoft.com...@qt-project.org] On Behalf Of
Julius Bullinger
Sent: Friday, 24 June 2016 5:25 PM
To: interest@qt-project.org Interest
Subject: [Interest] Native event filter in QtService

 

I'm trying to write a small (windows only) service based on QtService [1]
listening for some USB device events. For this, I created a
NativeDeviceEventFilter class based on QAbstractNativeEventFilter. It works
perfectly when used in a QCoreApplication.

 

Now, when installing this event filter in my UpdaterService class, it isn't
being called.

 

I tried each of:

 

  application()->installNativeEventFilter(deviceEvent_);
 
  QCoreApplication::instance()->installNativeEventFilter(deviceEvent_);
 
 
QAbstractEventDispatcher::instance()->installNativeEventFilter(deviceEvent_)
;

 

both in the service's constructor and start() method, as well as using a
local instance of NativeDeviceEventFilter,

but none of these worked. The event just isn't registered at all.

 

Has anyone ever done something like this? Or is there another way to receive
native messages (MSG structs) in a QtService?

 

Best regards,

Julius

 

[1]: https://github.com/qtproject/qt-solutions/tree/master/qtservice

___
Interest mailing list
Interest@qt-project.org
http://lists.qt-project.org/mailman/listinfo/interest

Re: [Interest] 5.8 Features?

[Robert Iakobashvili]

On Tue, Jun 21, 2016 at 10:55 PM, Jason H  wrote:
> I feel like the last few releases have been run by the trolls, and not the 
> users of Qt. I was hoping open governance would enable the community to 
> direct Qt development, but I seem to have misinterpreted what it means. I'm 
> looking for what's going into 5.8.. not much listed on the releases page.
>
> I'd like to suggest that mobile get some much needed love.
> - Application state transitions; Foreground, background
> - Background processing API
> - Screen wake lock API
> - In-app Notifications: local, remote
>
> While I have those characterized as "mobile" there are things like 
> notifications occurring on desktop platforms.
>
> Any thoughts?
>

Agree with Jason that mobile support needs more love
and adding "Native, native, native ..."

However, it could be that progress made at iOS and Android side is too fast and
our expectations from Qt are too high?

As any cross-platform framework Qt has its limitations.
Still, it has good integration points to allow additions of native code.

jm4c to add.

Kind regards,
Robert
___
Interest mailing list
Interest@qt-project.org
http://lists.qt-project.org/mailman/listinfo/interest

[Interest] QUndoStack

[Igor Mironchik]

Hello.

I'm adding undo stack to my app. But in the documentation I found:

voidQUndoStack::push(QUndoCommand *cmd)

Pushes cmd on the stack or merges it with the most recently executed 
command. In either case, executes cmd by calling its redo 
() function.


And in sources:

void QUndoStack::push(QUndoCommand *cmd)
{
Q_D(QUndoStack);
cmd->redo();

So my question is why just pushed to the stack undo command executes? 
I.e. why cmd->redo() calls in push()?


If I understand all correctly then cmd->redo() should be called only in 
QUndoStack::redo() method. And in push I just want to push new undo 
command to the stack...


___
Interest mailing list
Interest@qt-project.org
http://lists.qt-project.org/mailman/listinfo/interest

Re: [Interest] QMAKE_DEFAULT_INCDIRS = "/opt/local/include /usr/include" and building QtWebengine

[René J . V . Bertin]

René J. V. Bertin wrote:


> I checked, the DEFAULT_INCDIRS and DEFAULT_LIBDIRS variables in
> qtbase/configure do not have quotes.

Some more on this: I did patch qtbase/configure to prune a number of 
undesirable 
paths from those 2 variables. Without that patch I do not get the quotes. I 
haven't been able to understand why the behaviour is different in those 2 
cases, 
but it may be related to the fact that my patch already converts the newlines 
to 
a series of spaces.

I found a fix though:

--- a/qtbase/configure
+++ b/qtbase/configure
@@ -122,7 +122,7 @@ shellQuoteLines()
 {
 # The call of the outer echo makes the shell word-split the output of
 # the nested pipe, thus effectively converting newlines to spaces.
-echo `echo "$1" | sed 's,^[^ ]* .*$,"&",'`
+echo `echo $@ | sed 's,^[^ ]* .*$,&,'`
 }

This makes it possible to call shellQuoteLines without quoting the argument 
(e.g. `shellQuoteLines $DEFAULT_INCDIRS`) and doesn't introduce any quotes in 
the procedure itself. In fact, I'm surprised that I wasn't getting quotes in 
the 
stock configure; probably because the original input doesn't have runs of 2 or 
more spaces.

Hope this helps :)

R.

___
Interest mailing list
Interest@qt-project.org
http://lists.qt-project.org/mailman/listinfo/interest

Re: [Interest] universal builds on OS X

[René J . V . Bertin]

On Friday June 24 2016 12:34:20 Konstantin Tokarev wrote:

>There is https://www.darlinghq.org, at some point it may become usable for 
>testing

Wow, I hadn't heard about that one for quite a while ... :)

R
___
Interest mailing list
Interest@qt-project.org
http://lists.qt-project.org/mailman/listinfo/interest

Re: [Interest] Native event filter in QtService

[Konstantin Shegunov]

Hi,
If I may be so bold, I've wrapped up a (Qt 5) replacement for QtService
recently. You might be interested to try it out (I've chosen the MIT
license).
https://bitbucket.org/nye/qtdaemon

Kind regards.
___
Interest mailing list
Interest@qt-project.org
http://lists.qt-project.org/mailman/listinfo/interest

Re: [Interest] universal builds on OS X

[Konstantin Tokarev]

24.06.2016, 12:20, "René J. V. Bertin" :
> Jean-Michaël Celerier wrote:
>
>>  Some people are working on a toolchain for cross-compiling OS X software;
>>  maybe
>>  it could be extended to iOS? It "only" requires the SDK that you can borrow
>>  from a
>>  friend's mac (or even by downloading xcode yourself).
>>
>>  Here's a link :
>>  https://github.com/tpoechtrager/osxcross
>
> I can see why one would want to do this for targeting iOS, but OS X? How are 
> you
> going to test your code? If you're going to run a VM for that you can just as
> well build in it ...

There is https://www.darlinghq.org, at some point it may become usable for 
testing

>
>>  But I guess the hardest part would be all the signing that Xcode does to
>>  allow
>>  you to run some code you wrote on your 800$ iphone...
>
> Guess why I never looked at developing for my phone ... and I only buy 
> 2nd-hand
> iPhones? ;)
>
> R.
>
> ___
> Interest mailing list
> Interest@qt-project.org
> http://lists.qt-project.org/mailman/listinfo/interest

-- 
Regards,
Konstantin
___
Interest mailing list
Interest@qt-project.org
http://lists.qt-project.org/mailman/listinfo/interest

Re: [Interest] 5.8 Features?

[Sze Howe Koh]

On 22 June 2016 at 21:39, Jason H  wrote:
>
> > Subject: Re: [Interest] 5.8 Features?
> >
> > On 22 June 2016 at 09:34, Thiago Macieira  wrote:
> > >
> > > On terça-feira, 21 de junho de 2016 22:55:07 PDT Jason H wrote:
> > > > I feel like the last few releases have been run by the trolls, and not 
> > > > the
> > > > users of Qt. I was hoping open governance would enable the community to
> > > > direct Qt development, but I seem to have misinterpreted what it means.
> >
> > See https://wiki.qt.io/The_Qt_Governance_Model
> >
> > "Open governance" means that all members of the community can raise
> > proposals and vote on them. It is also a meritocratic system (as
> > opposed to democratic), which means more weight is given to members of
> > higher rank. Put simply, rank is gained through contributions and
> > commitment to the Qt Project.
>
> Where and when does this voting occur?

Where: Often at the Development mailing list:
http://lists.qt-project.org/mailman/listinfo/development
When: Whenever someone posts a proposal.

Examples:
* Proposal for choosing the list of platforms to support for Qt 5.8:
http://thread.gmane.org/gmane.comp.lib.qt.devel/24938/focus=24965
* Proposal for polishing and releasing QtSingleApplication:
http://thread.gmane.org/gmane.comp.lib.qt.devel/26279/focus=26358

My apologies, "voting" was the wrong word to use, as Thiago explained
in his email. Rather than counting votes, the process is a
discussion/debate where the proposals with the highest merit get
accepted.


> I've been vocal about improving Qt on mobile since mobile was an option. I've 
> been hardcore on the shortcomings for about 20 months. During that time, I 
> saw 5.6 get released, 5.7 get released at the same time as 5.6.1, and a lot 
> of meaningless (to me) modules added. And now 5.8 is only 1.5 months from 
> 5.7? What_the_hell? I'd like to see Qt complete it's mobile initiative.
>
> As for contributing, the past 20 months I've been under _multiple_ commercial 
> licenses, one that was paying for a Desktop just to use Qt Charts. I think 
> that's pretty material.

First of all, thank you for supporting Qt's development by purchasing
commercial licenses.

I'd imagine that funds from commercial licenses are normally allocated
to the development and maintenance of the entire Qt framework. Start a
conversation with your sales rep, and see if you can request greater
focus on specific areas. (Disclaimer: I'm not affiliated with the Qt
Company and I don't speak on their behalf, so please correct me if
this advice is inappropriate.)

Please remember: The Qt Company's commercial decision making process
and the Qt Project's Open Governance are _not_ the same thing. In the
context of Open Governance, "contribution" refers mainly to code
contributions. I don't have insight into the commercial decision
making process.

Anyway, as Tuukka pointed out in his email, Qt is making steady
progress in the mobile world (although "mobile" is still very broad,
and it sounds like the progress to date isn't in the areas that you're
interested in). There are multiple ways to get the mobile features
that you want, sooner:

* The open source community way: Get involved personally. Test,
review, and provide feedback for relevant patches, e.g.
https://codereview.qt-project.org/#/c/149602/ . If a patch gets
neglected, ping the list for more attention. Even better, write and
submit your own patches.
* The commercial customer way: Talk to your Qt Company sales rep
directly to try to influence their decisions.
* Use a ready-made 3rd party toolkit to complement official Qt
features, e.g. V-Play: http://v-play.net/
* Get 3rd party professionals on board to implement the specific
features that you want, e.g.
https://woboq.com/blog/qdockwidget-changes-in-56.html -- "You can hire
Woboq to fix your own Qt bugs or implement new features inside Qt"


Regards,
Sze-Howe
___
Interest mailing list
Interest@qt-project.org
http://lists.qt-project.org/mailman/listinfo/interest

Re: [Interest] universal builds on OS X

[René J . V . Bertin]

Jean-Michaël Celerier wrote:

> Some people are working on a toolchain for cross-compiling OS X software;
> maybe
> it could be extended to iOS? It "only" requires the SDK that you can borrow
> from a
> friend's mac (or even by downloading xcode yourself).
> 
> Here's a link :
> https://github.com/tpoechtrager/osxcross

I can see why one would want to do this for targeting iOS, but OS X? How are 
you 
going to test your code? If you're going to run a VM for that you can just as 
well build in it ...

> But I guess the hardest part would be all the signing that Xcode does to
> allow
> you to run some code you wrote on your 800$ iphone...

Guess why I never looked at developing for my phone ... and I only buy 2nd-hand 
iPhones? ;)

R.

___
Interest mailing list
Interest@qt-project.org
http://lists.qt-project.org/mailman/listinfo/interest

Re: [Interest] Native event filter in QtService

[Bo Thorsen]

Hi all,

I'm copying this to devel, because it fits in a discussion a week or two 
ago.


Den 24-06-2016 kl. 10:56 skrev Tony Rietwyk:

qtservice_win.cpp around line 830 at your reference [1] installs its own
nativeEventFilter - probably displacing yours.

I suspect you'll need to merge the Qt filter into yours, and get rid of
that install.


That is impossible for commercial projects because the solutions are 
LGPL only. You can't copy LGPL code into your own code base unless you 
are using GPL or LGPL yourself.


And that's why I copied this message to devel, because there was a 
discussion on whether to add solutions to Qt Core or not. And one of the 
arguments against is that people can just use the solutions as they are. 
This is one case, where apparently that's not (at least currently) possible.


I can come up with a couple of workarounds for this, sure. But these are 
all brittle and may not work if the solutions code is changed.



*From:*Interest
[mailto:interest-bounces+tony=rightsoft.com...@qt-project.org] *On
Behalf Of *Julius Bullinger
*Sent:* Friday, 24 June 2016 5:25 PM
*To:* interest@qt-project.org Interest
*Subject:* [Interest] Native event filter in QtService

I’m trying to write a small (windows only) service based on QtService
[1] listening for some USB device events. For this, I created a
NativeDeviceEventFilterclass based on QAbstractNativeEventFilter. It
works perfectly when used in a QCoreApplication.

Now, when installing this event filter in my UpdaterServiceclass, it
isn’t being called.

I tried each of:

  application()->installNativeEventFilter(deviceEvent_);

  QCoreApplication::instance()->installNativeEventFilter(deviceEvent_);

QAbstractEventDispatcher::instance()->installNativeEventFilter(deviceEvent_);

both in the service’s constructor and start()method, as well as using a
local instance of NativeDeviceEventFilter,

but none of these worked. The event just isn’t registered at all.

Has anyone ever done something like this? Or is there another way to
receive native messages (MSG structs) in a QtService?


Bo Thorsen,
Director, Viking Software.

--
Viking Software
Qt and C++ developers for hire
http://www.vikingsoft.eu
___
Interest mailing list
Interest@qt-project.org
http://lists.qt-project.org/mailman/listinfo/interest

Re: [Interest] QMAKE_DEFAULT_INCDIRS = "/opt/local/include /usr/include" and building QtWebengine

[René J . V . Bertin]

Thiago Macieira wrote:

Hi,

> First of all, since this functionality changed very recently, what released
> version or Git commit are you at?

Sorry, this is with Qt 5.6.1 but I never tried to build QtWebEngine on Linux 
before.

I checked, the DEFAULT_INCDIRS and DEFAULT_LIBDIRS variables in 
qtbase/configure 
do not have quotes.

R.

___
Interest mailing list
Interest@qt-project.org
http://lists.qt-project.org/mailman/listinfo/interest

Re: [Interest] Native event filter in QtService

[Julius Bullinger]

Hi Tony,

thanks for your reply. Reading the documentation of 
QAbstractEventDispatcher::installNativeEventFilter(), it says in the fourth 
paragraph

If multiple event filters are installed, the filter that was 
installed last is activated first.

Thus, I was under the impression that I should be able to add my own filter, 
without overwriting the one installed by QtService. Is this incorrect?

Regards,
Julius

Von: Interest 
[mailto:interest-bounces+julius.bullinger=asctec...@qt-project.org] Im Auftrag 
von Tony Rietwyk
Gesendet: Freitag, 24. Juni 2016 10:57 Uhr
An: interest@qt-project.org
Betreff: Re: [Interest] Native event filter in QtService

Hi Julius,

qtservice_win.cpp around line 830 at your reference [1] installs its own 
nativeEventFilter - probably displacing yours.

I suspect you'll need to merge the Qt filter into yours, and get rid of that 
install.

Regards,

Tony


From: Interest [mailto:interest-bounces+tony=rightsoft.com...@qt-project.org] 
On Behalf Of Julius Bullinger
Sent: Friday, 24 June 2016 5:25 PM
To: interest@qt-project.org Interest
Subject: [Interest] Native event filter in QtService

I'm trying to write a small (windows only) service based on QtService [1] 
listening for some USB device events. For this, I created a 
NativeDeviceEventFilter class based on QAbstractNativeEventFilter. It works 
perfectly when used in a QCoreApplication.

Now, when installing this event filter in my UpdaterService class, it isn't 
being called.

I tried each of:


  application()->installNativeEventFilter(deviceEvent_);



  QCoreApplication::instance()->installNativeEventFilter(deviceEvent_);



  QAbstractEventDispatcher::instance()->installNativeEventFilter(deviceEvent_);

both in the service's constructor and start() method, as well as using a local 
instance of NativeDeviceEventFilter,
but none of these worked. The event just isn't registered at all.

Has anyone ever done something like this? Or is there another way to receive 
native messages (MSG structs) in a QtService?

Best regards,
Julius

[1]: https://github.com/qtproject/qt-solutions/tree/master/qtservice
___
Interest mailing list
Interest@qt-project.org
http://lists.qt-project.org/mailman/listinfo/interest

Re: [Interest] Native event filter in QtService

[Tony Rietwyk]

Hi Julius, 

 

qtservice_win.cpp around line 830 at your reference [1] installs its own
nativeEventFilter - probably displacing yours.  

 

I suspect you'll need to merge the Qt filter into yours, and get rid of that
install.  

 

Regards, 

 

Tony

 

 

From: Interest
[mailto:interest-bounces+tony=rightsoft.com...@qt-project.org] On Behalf Of
Julius Bullinger
Sent: Friday, 24 June 2016 5:25 PM
To: interest@qt-project.org Interest
Subject: [Interest] Native event filter in QtService

 

I'm trying to write a small (windows only) service based on QtService [1]
listening for some USB device events. For this, I created a
NativeDeviceEventFilter class based on QAbstractNativeEventFilter. It works
perfectly when used in a QCoreApplication.

 

Now, when installing this event filter in my UpdaterService class, it isn't
being called.

 

I tried each of:

 

  application()->installNativeEventFilter(deviceEvent_);
 
  QCoreApplication::instance()->installNativeEventFilter(deviceEvent_);
 
 
QAbstractEventDispatcher::instance()->installNativeEventFilter(deviceEvent_)
;

 

both in the service's constructor and start() method, as well as using a
local instance of NativeDeviceEventFilter,

but none of these worked. The event just isn't registered at all.

 

Has anyone ever done something like this? Or is there another way to receive
native messages (MSG structs) in a QtService?

 

Best regards,

Julius

 

[1]: https://github.com/qtproject/qt-solutions/tree/master/qtservice

___
Interest mailing list
Interest@qt-project.org
http://lists.qt-project.org/mailman/listinfo/interest

Re: [Interest] Custom QSGGeometryNode Scene blending issue

[Gunnar Sletta]

Since the window's background is being cleared when you draw the framebuffer, 
it sounds as the window's default framebuffer is bound at the time of clearing. 
That or that the FBO is indeed cleared to fully transparent and rendered 
without blending, thus overwriting the pixels window.

To further debug this, I would have first checked the contents of the FBO, 
using QOpenGLFramebufferObject::toImage() or using glReadPixels if this is your 
own FBO, and then look at the blending that happens in the scene graph if that 
one is ok. What does the QSGTexture instance look like that you use to wrap the 
FBO. Does it report that there is alpha?

QSG_VISUALIZE=overdraw will tell you whether the FBO is drawn in the scene 
graph using blending or not.

Or simply use QQuickFramebufferObject, of course :)

cheers,
Gunnar

> On 23 Jun 2016, at 17:26, Roberto Garrido  
> wrote:
> 
> Gunnar,
> thanks for your detailed response. 
> We don't discard going for the QQuickFramebufferObject soon, but currently 
> what we have is the Item/QSGGeometryNode setup.
> I've been doing some tests, and I cannot manage to get the desired effect. 
> The most silly example I can think of is to clear our buffer with a fully 
> transparent color (as you suggested), then render a fullscreen quad with also 
> a fully transparent color (using premultiplied alpha -> 1, 1 - SRC_ALPHA). 
> That should give us fully transparent item, and the qml scene behind should 
> be shown. But what we see is the desktop itself, not the QML items behind. 
> Does it give you any clue on what could I be doing wrong?
> 
> Our node (derived from QSGGeometryNode) takes a QSGTextureMaterial (with the 
> default shader for this type of material), and we have also activated the 
> QSGMaterial::Blending flag.
> 
> Thanks,
> Robert.
> 
> On 7 June 2016 at 09:41, Gunnar Sletta  wrote:
> Hi Roberto,
> 
> The opacity of any geometry node is decided by the opacity that comes out of 
> the fragment shader of its material. Depending on how the alpha logic exists 
> in your custom GL renderer, there are two ways to do this.
> 
> Option 1: If most of the FBO is partially transparent, or if there are bits 
> and pieces that are overlaid on top of the rest of the UI, like controls on a 
> HUD, then you want to render into an FBO with an alpha buffer, clear it with 
> fully transparent, draw your custom GL stuff and then alpha blend the entire 
> FBO on top of the QML scene. You do alpha blending of the fbo content by 
> setting the QSGMaterial::Blending flag and returning a premultiplied alpha 
> color from the fragment shader. The QSGTextureMaterial has a fragment shader 
> that does the right thing.
> 
> Option 2: If you have an opaque FBO with certain sections, say a given 
> rectangle, fully transparent, you can draw the FBO as fully opaque, but 
> create a geometry that excludes the fully transparent region. This has the 
> benefit you will be drawing more opaque stuff which is generally faster for 
> the GPU to handle, compared to blended stuff.
> 
> It is of course possible to do a mixture of 1 and 2, by creating two separate 
> materials, both referring to the FBO. For the fully opaque sections of the 
> FBO, you use an opaque material to draw it which is fast. For the 
> semi-transparent sections of the FBO you use a different geometry node with a 
> material that enables blending. The fully transparent regions are fully 
> excluded from either geometry, so those cost nothing. (something like the 
> MaskedImage I have here: 
> https://github.com/qtproject/playground-scenegraph/tree/master/shapes)
> 
> Btw, if you used QQuickFramebufferObject instead of rolling your own custom 
> Item/node/material, you would get option 1 by default :)
> 
> cheers,
> Gunnar
> 
> > On 06 Jun 2016, at 18:05, Roberto Garrido  
> > wrote:
> >
> > Hi all!
> > We have a custom QQuickItem which in turn creates a custom QSGGeometryNode. 
> > We reimplement the QSGGeometryNode's preprocess() method in order to draw 
> > our scene to a FrameBufferObject with a texture attached. This OpenGL 
> > texture ID is shared with the QSGGeometryNode's material. This way, we can 
> > render our custom OpenGL scene into a QML item. So far, so good.
> >
> > However, we want to use alpha blending on some parts of our scene, in order 
> > to let the user see the QML items behind our custom item. Something similar 
> > to what Item::opacity value is doing, but only for a specific area of our 
> > Item. It's like having a hole inside our Item that allows the user to see 
> > what's behind. The information about the specific region that should be 
> > transparent is handled inside our custom OpenGL renderer.
> >
> > I've been trying to find out what Qt is doing in order to change the 
> > opacity of a QML item, with no luck.
> >
> > Any idea on how can we achieve this effect, with our Item/Node 
> > configuration?
> > Thanks in advance,
> > Robert.
> >
> > --
> > Website: http://robertogarrido.com
> > Twitter: htt

[Interest] Native event filter in QtService

[Julius Bullinger]

I'm trying to write a small (windows only) service based on QtService [1] 
listening for some USB device events. For this, I created a 
NativeDeviceEventFilter class based on QAbstractNativeEventFilter. It works 
perfectly when used in a QCoreApplication.

Now, when installing this event filter in my UpdaterService class, it isn't 
being called.

I tried each of:


  application()->installNativeEventFilter(deviceEvent_);



  QCoreApplication::instance()->installNativeEventFilter(deviceEvent_);



  QAbstractEventDispatcher::instance()->installNativeEventFilter(deviceEvent_);

both in the service's constructor and start() method, as well as using a local 
instance of NativeDeviceEventFilter,
but none of these worked. The event just isn't registered at all.

Has anyone ever done something like this? Or is there another way to receive 
native messages (MSG structs) in a QtService?

Best regards,
Julius

[1]: https://github.com/qtproject/qt-solutions/tree/master/qtservice
___
Interest mailing list
Interest@qt-project.org
http://lists.qt-project.org/mailman/listinfo/interest