From backup.rlacko at gmail.com  Wed Apr  2 14:00:49 2014
From: backup.rlacko at gmail.com (Roman Lacko)
Date: Wed, 2 Apr 2014 14:00:49 +0200
Subject: [PySide] PySide 1.2.1 Windows builds for Python 3.4
Message-ID: <CAHC6v6h9vN-jBHJeipacBPsMVzKPaTPpO5k6vqbakz=6j4cUsg@mail.gmail.com>

Hi,

I have uploaded PySide 1.2.1 Windows builds for Python 3.4 [1]


Regards
-Roman

[1] http://qt-project.org/wiki/PySide_Binaries_Windows
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.qt-project.org/pipermail/pyside/attachments/20140402/51cab230/attachment.html>

From backup.rlacko at gmail.com  Wed Apr  2 14:19:25 2014
From: backup.rlacko at gmail.com (Roman Lacko)
Date: Wed, 2 Apr 2014 14:19:25 +0200
Subject: [PySide] PySide 1.2.1 Windows builds for Python 3.4
In-Reply-To: <CAHC6v6h9vN-jBHJeipacBPsMVzKPaTPpO5k6vqbakz=6j4cUsg@mail.gmail.com>
References: <CAHC6v6h9vN-jBHJeipacBPsMVzKPaTPpO5k6vqbakz=6j4cUsg@mail.gmail.com>
Message-ID: <CAHC6v6jyczC+B4oz3JQLxwpGUggoFiZGq4z9NRXq3LCFox4xww@mail.gmail.com>

if you have problems installing PySide on Python 3.4 via pip, try following
command:

$ pip install pyside --allow-all-external -f
http://download.qt-project.org/official_releases/pyside/

Regards
-Roman

2014-04-02 14:00 GMT+02:00 Roman Lacko <backup.rlacko at gmail.com>:

> Hi,
>
> I have uploaded PySide 1.2.1 Windows builds for Python 3.4 [1]
>
>
> Regards
> -Roman
>
> [1] http://qt-project.org/wiki/PySide_Binaries_Windows
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.qt-project.org/pipermail/pyside/attachments/20140402/766fc3d1/attachment.html>

From techtonik at gmail.com  Wed Apr  2 16:20:09 2014
From: techtonik at gmail.com (anatoly techtonik)
Date: Wed, 2 Apr 2014 17:20:09 +0300
Subject: [PySide] PySide 1.2.1 Windows builds for Python 3.4
In-Reply-To: <CAHC6v6h9vN-jBHJeipacBPsMVzKPaTPpO5k6vqbakz=6j4cUsg@mail.gmail.com>
References: <CAHC6v6h9vN-jBHJeipacBPsMVzKPaTPpO5k6vqbakz=6j4cUsg@mail.gmail.com>
Message-ID: <CAPkN8x+jEaktGy0zRCUs1YkqOuV-gzMQvEuyXr2T+4S+FUEhBw@mail.gmail.com>

Cool. I am glad to see this.  =)

On Wed, Apr 2, 2014 at 3:00 PM, Roman Lacko <backup.rlacko at gmail.com> wrote:
> Hi,
>
> I have uploaded PySide 1.2.1 Windows builds for Python 3.4 [1]
>
>
> Regards
> -Roman
>
> [1] http://qt-project.org/wiki/PySide_Binaries_Windows
>
>
> _______________________________________________
> PySide mailing list
> PySide at qt-project.org
> http://lists.qt-project.org/mailman/listinfo/pyside
>



-- 
anatoly t.


From sean at seanfisk.com  Wed Apr  2 18:21:46 2014
From: sean at seanfisk.com (Sean Fisk)
Date: Wed, 2 Apr 2014 12:21:46 -0400
Subject: [PySide] PySide 1.2.1 Windows builds for Python 3.4
In-Reply-To: <CAPkN8x+jEaktGy0zRCUs1YkqOuV-gzMQvEuyXr2T+4S+FUEhBw@mail.gmail.com>
References: <CAHC6v6h9vN-jBHJeipacBPsMVzKPaTPpO5k6vqbakz=6j4cUsg@mail.gmail.com>
	<CAPkN8x+jEaktGy0zRCUs1YkqOuV-gzMQvEuyXr2T+4S+FUEhBw@mail.gmail.com>
Message-ID: <CAF-FwG_MaVT70RA64rGOq_aQvwMuyfj5RCpOhDGxNfTfbSmRgA@mail.gmail.com>

This is great! Thanks so much.


--
Sean Fisk


On Wed, Apr 2, 2014 at 10:20 AM, anatoly techtonik <techtonik at gmail.com>wrote:

> Cool. I am glad to see this.  =)
>
> On Wed, Apr 2, 2014 at 3:00 PM, Roman Lacko <backup.rlacko at gmail.com>
> wrote:
> > Hi,
> >
> > I have uploaded PySide 1.2.1 Windows builds for Python 3.4 [1]
> >
> >
> > Regards
> > -Roman
> >
> > [1] http://qt-project.org/wiki/PySide_Binaries_Windows
> >
> >
> > _______________________________________________
> > PySide mailing list
> > PySide at qt-project.org
> > http://lists.qt-project.org/mailman/listinfo/pyside
> >
>
>
>
> --
> anatoly t.
> _______________________________________________
> PySide mailing list
> PySide at qt-project.org
> http://lists.qt-project.org/mailman/listinfo/pyside
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.qt-project.org/pipermail/pyside/attachments/20140402/d2817617/attachment.html>

From jonasthiem at googlemail.com  Thu Apr  3 00:34:50 2014
From: jonasthiem at googlemail.com (Jonas Thiem)
Date: Thu, 03 Apr 2014 00:34:50 +0200
Subject: [PySide] Please add proper copyright notice and LGPL 2.1 link to
 licensing page
Message-ID: <533C908A.3070307@googlemail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi guys,

Please be so kind and fix the licensing page by adding a proper PySide
(C) ... note and an actual link to the LGPL 2.1 here:
http://qt-project.org/wiki/About-PySide

People may want to attribute you correctly with the full copyright
notice and the license text as the LGPL actually requires for any
program that links or depends on PySide (no matter if PySide is
actually shipped with it or not), so it would be nice if you could
give them the necessary information on that page to do that.

Basically a sub page this: http://qt-project.org/doc/qt-4.8/lgpl.html
(Proper full copyright line + the full license) should be linked from
your Licensing info paragraph here:
http://qt-project.org/wiki/About-PySide

If PySide also contains the shiboken poject (that's not entirely clear
on that page either), please add a note about that too, its licensing
and its copyright notice assuming it is LGPL too.

Regards,
Jonas Thiem
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1
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=Xq+z
-----END PGP SIGNATURE-----


From techtonik at gmail.com  Thu Apr  3 07:49:14 2014
From: techtonik at gmail.com (anatoly techtonik)
Date: Thu, 3 Apr 2014 08:49:14 +0300
Subject: [PySide] Please add proper copyright notice and LGPL 2.1 link
 to licensing page
In-Reply-To: <533C908A.3070307@googlemail.com>
References: <533C908A.3070307@googlemail.com>
Message-ID: <CAPkN8xK_5eUn+k7jUVsP1yN7GY-LPhuj35tr92A-JsBRDeTzbw@mail.gmail.com>

On Thu, Apr 3, 2014 at 1:34 AM, Jonas Thiem <jonasthiem at googlemail.com> wrote:
>
> Please be so kind and fix the licensing page by adding a proper PySide
> (C) ... note and an actual link to the LGPL 2.1 here:
> http://qt-project.org/wiki/About-PySide

It is not a licensing page as far as I see it, but if you can find it
(probably in source code repository) feel free to add a link - it
is a wiki after all and you are a contributor already. =)

> People may want to attribute you correctly with the full copyright
> notice and the license text as the LGPL actually requires for any
> program that links or depends on PySide (no matter if PySide is
> actually shipped with it or not), so it would be nice if you could
> give them the necessary information on that page to do that.

LGPL doesn't require you to include copyright notice or attribute
anybody, or this page should be fixed:
https://tldrlegal.com/license/gnu-lesser-general-public-license-v2.1-(lgpl-2.1)

> Basically a sub page this: http://qt-project.org/doc/qt-4.8/lgpl.html
> (Proper full copyright line + the full license) should be linked from
> your Licensing info paragraph here:
> http://qt-project.org/wiki/About-PySide

PySide is not completely Qt. It uses Qt and link to it - right. In any
case - if you dig the proper link and insert it- nobody will not object.

This one:
https://qt.gitorious.org/pyside/pyside/source/d2a47ab8f27af7e74d34797464da85c128c17c37:COPYING


From jonasthiem at googlemail.com  Thu Apr  3 08:14:43 2014
From: jonasthiem at googlemail.com (Jonas Thiem)
Date: Thu, 03 Apr 2014 08:14:43 +0200
Subject: [PySide] Please add proper copyright notice and LGPL 2.1 link
 to licensing page
In-Reply-To: <CAPkN8xK_5eUn+k7jUVsP1yN7GY-LPhuj35tr92A-JsBRDeTzbw@mail.gmail.com>
References: <533C908A.3070307@googlemail.com>
	<CAPkN8xK_5eUn+k7jUVsP1yN7GY-LPhuj35tr92A-JsBRDeTzbw@mail.gmail.com>
Message-ID: <533CFC53.1050701@googlemail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 04/03/2014 07:49 AM, anatoly techtonik wrote:
> LGPL doesn't require you to include copyright notice or attribute 
> anybody, or this page should be fixed: 
> https://tldrlegal.com/license/gnu-lesser-general-public-license-v2.1-(lgpl-2.1)

>
>
> 
That's
probably incorrect (however I'm not a lawyer). Anyway:

The only reason that a Python PySide program is not linked to PySide is
that the program happens to be Python instead of C++, not that the
program would operate any differently by itself. And such a program
still can't run without PySide, similar to an actual linked program in
the classical C++ linking sense.

Concluding, I guess a python program purely using PySide might be
considered similar to a C++ program linking PySide.

But to any program linking an LGPL lib, I'm pretty sure this statement
of Section 6 of the LGPL 2.1 applies:

"You must give prominent notice with each copy of the work that the
Library is used in it and that the Library and its use are covered by
this License. You must supply a copy of this License. If the work
during execution displays copyright notices, you must include the
copyright notice for the Library among them, as well as a reference
directing the user to the copy of this License."

Feel free to correct me if you have a deeper understanding of the LGPL.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1
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=Xr0Q
-----END PGP SIGNATURE-----


From maxime.perrotin at esa.int  Thu Apr  3 11:43:14 2014
From: maxime.perrotin at esa.int (Maxime Perrotin)
Date: Thu, 03 Apr 2014 11:43:14 +0200
Subject: [PySide] ESA Summer of Code - an opportunity for PySide?
Message-ID: <29728_1396518573_533D2EAD_29728_2995_1_533D2D32.5060507@esa.int>

Dear all,

It may be of interest of the PySide community to submit a project to the 
ESA summer of code (SOCIS):
http://sophia.estec.esa.int/socis/

The 2014 edition was just launched. In a similar way to the GSoC, SOCIS 
allows students to be paid to participate actively during 3 months to an 
open-source project.

Past editions showed that we get top-level students. If you think that 
it is a good idea to put someone on e.g. Qt5 bindings for PySide, then 
please feel free to submit a proposal.
Application form for new projects is open until 15th April.

There is a selection process, no guarantee to be selected of course, but 
it is always worth trying.... we can fund up to 23 projects a year - 
last year we received 26 proposals.

Best regards,
Maxime






This message and any attachments are intended for the use of the addressee or addressees only.
The unauthorised disclosure, use, dissemination or copying (either in whole or in part) of its
content is not permitted.
If you received this message in error, please notify the sender and delete it from your system.
Emails can be altered and their integrity cannot be guaranteed by the sender.

Please consider the environment before printing this email.



From ydkj14 at gmail.com  Fri Apr  4 22:50:37 2014
From: ydkj14 at gmail.com (Martin B.)
Date: Fri, 4 Apr 2014 22:50:37 +0200
Subject: [PySide] How can I add a QComboBox to a QTreeView?
Message-ID: <CAEMTNLNU8gwJKuWS3_GgLFEifE9CeaQppAvgFszJr_Mh7zkpCw@mail.gmail.com>

I've created a QTreeView and a QStandardItemModel.
Then I add several rows and replace the item in the second column with a
combobox.

item = QStandardItem(text2)
model.appendRow([ QStandardItem(text1), item ])

cbx = QComboBox()
cbx.addItem("A")
cbx.addItem("B")

tvw.setIndexWidget(model.indexFromItem(item), cbx)

The problem is that the combobox doesn't show up. Instead text2 is shown.

What is wrong with this code?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.qt-project.org/pipermail/pyside/attachments/20140404/715d312a/attachment.html>

From arjunaravikumar at gmail.com  Fri Apr  4 22:58:33 2014
From: arjunaravikumar at gmail.com (Arjuna Ravikumar)
Date: Fri, 4 Apr 2014 13:58:33 -0700
Subject: [PySide] PySide Slot name bug?
Message-ID: <CAE0HNPumzM1V21kPqWuTyD4qPxvPwbS-ptkb-Lj+PGii_aEhYg@mail.gmail.com>

Hi,

I'm relatively new to PyQt and PySide but am tasked with moving our tools
over from PyQt to PySide at our studio.

I've noticed a problem with PySide slot decorators as I've described in
this StackOverflow post here:
http://stackoverflow.com/questions/22850064/pyqt-pyside-using-named-slots

Can someone confirm if this is indeed a bug or if there is a workaround I'm
unaware of.

If it is indeed a bug, is it realistic to wait for a fix?

Thanks!
Arjuna
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.qt-project.org/pipermail/pyside/attachments/20140404/4207cd23/attachment.html>

From ltanure at gmail.com  Mon Apr  7 17:28:23 2014
From: ltanure at gmail.com (Lucas Tanure)
Date: Mon, 7 Apr 2014 12:28:23 -0300
Subject: [PySide] PySide Slot name bug?
In-Reply-To: <CAE0HNPumzM1V21kPqWuTyD4qPxvPwbS-ptkb-Lj+PGii_aEhYg@mail.gmail.com>
References: <CAE0HNPumzM1V21kPqWuTyD4qPxvPwbS-ptkb-Lj+PGii_aEhYg@mail.gmail.com>
Message-ID: <CAJyon0t7SehDyoVmsLo1QsxB4L41OY0Un0D3e_xvNV10-rSZzg@mail.gmail.com>

Do like this :

@QtCore.Slot(str)
def on_pushButton_clicked(self, name = "on_pushButton_clicked"):
    print "pressed button"



--
Lucas A. Tanure Alves
+55 (19) 988176559


On Fri, Apr 4, 2014 at 5:58 PM, Arjuna Ravikumar
<arjunaravikumar at gmail.com>wrote:

> Hi,
>
> I'm relatively new to PyQt and PySide but am tasked with moving our tools
> over from PyQt to PySide at our studio.
>
> I've noticed a problem with PySide slot decorators as I've described in
> this StackOverflow post here:
> http://stackoverflow.com/questions/22850064/pyqt-pyside-using-named-slots
>
> Can someone confirm if this is indeed a bug or if there is a workaround
> I'm unaware of.
>
> If it is indeed a bug, is it realistic to wait for a fix?
>
> Thanks!
> Arjuna
>
> _______________________________________________
> PySide mailing list
> PySide at qt-project.org
> http://lists.qt-project.org/mailman/listinfo/pyside
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.qt-project.org/pipermail/pyside/attachments/20140407/4acdabfd/attachment.html>

From tismer at stackless.com  Sat Apr 12 21:50:12 2014
From: tismer at stackless.com (Christian Tismer)
Date: Sat, 12 Apr 2014 21:50:12 +0200
Subject: [PySide] Install via pip fails on OS X Mavericks
In-Reply-To: <CAHC6v6ihZP-Hgf6e+5gLK+1ohOrPo=T9fPKWPU_0O_WrkjbKFw@mail.gmail.com>
References: <CAGj3X_CFjewgkN+oMoF+h75pmV4pGmZa77aN8E76CskSVVnvrg@mail.gmail.com>	<CAF-FwG8b+dAhvtZB6wumknXp2xQxe=e=yZ95q-iaaPMz-8G+Jg@mail.gmail.com>	<CA+c+8J37E-_PNp9gaWi11qE3vSSUy8bM9-z+MEuJqnKjwh_J1Q@mail.gmail.com>	<CAF-FwG_H_xJx8EhVwYphcQrzuDn18V+EVOba-ETCz68tX4fJ1w@mail.gmail.com>	<CA+c+8J2+9F0BnM-fHPGb9581yVkG6gQFicm6g02gPtF1iGw8cQ@mail.gmail.com>
	<CAHC6v6ihZP-Hgf6e+5gLK+1ohOrPo=T9fPKWPU_0O_WrkjbKFw@mail.gmail.com>
Message-ID: <534998F4.6000401@stackless.com>

Hi Roman,

for use in my projects, I cloned shiboken and pyside-setup,
just in order to build 1.2.1 on Mavericks.

Problem is: The fix for Mavericks is in shiboken, but only reachable
when building the dev version  1.3.0dev. But a dev version is not for PyPi.

My quick hack was to tag the latest version of shiboken with "1.2.1fix",
then I changed the 1.2.1 version in setup.py like so:

    '1.2.1': [
        ["shiboken", "1.2.1fix"],
        ["pyside", "1.2.1"],
        ["pyside-tools", "0.2.15"],
        ["pyside-examples", "master"],
    ],

and then the build works.

$ python setup.py bdist_wheel --version=1.2.1

I agree this is not totally correct, should officially better be 1.2.2
but for us this works.

I would really like to produce wheels for OS X, finally. But it is kind of
a showstopper that fixes get in, but pip does not see this.
Actually, this is not really a new pyside version, but just a fix to
shiboken.

What do you think, and how can I help/make it happen?
Should I provide a patch to create 1.2.2 ?

cheers - Chris


On 04.03.14 01:05, Roman Lacko wrote:
> Hi,
>
> You can also install PySide with pip directli from git repository.
>
> This should work, but have not tested with latest version:
> pip install git+ssh://github.com/PySide/pyside-setup.git
> <https://github.com/username/repo.git>
>
> -Roman
>
> 2014-03-03 23:23 GMT+01:00 Erica Cane <ericacane at splitu.com
> <mailto:ericacane at splitu.com>>:
>
>     Thanks Sean. I am indeed interested in compiling from source in a
>     virtualenv. That said, the ideal solution would be for the PyPi
>     package to be updated with the fix for Mavericks. :)
>
>     Thanks again!
>     Erica
>
>
>     On Sun, Mar 2, 2014 at 10:44 PM, Sean Fisk <sean at seanfisk.com
>     <mailto:sean at seanfisk.com>> wrote:
>
>         When compiling from source, you can install into a virtualenv.
>         When using Homebrew, it’s not possible to install into a
>         virtualenv.
>
>         I don’t use Homebrew’s Python, and don’t install any other
>         Python packages other than PySide. Then, inside of the
>         virtualenv, I use a PTH file
>         <http://docs.python.org/2/library/site.html> to point to
>         Homebrew’s |site-packages/| directory. Unfortunately, this
>         also includes any other Python packages installed by Homebrew
>         if you decide to install them. It’s not the best option, but
>         it works well. I can send you my PTH-file-writing-script if
>         you are interested.
>
>         If you are interested in compiling from source into a
>         virtualenv, my team is working on improving the PySide
>         BuildScripts <https://github.com/PySide/BuildScripts> repo and
>         I can send you what we have as well.
>
>         If you are interested in either of these, expect a reply from
>         me tomorrow as it is pretty late in my neck of the woods.
>
>         Cheers,
>
>
>
>         --
>         Sean Fisk
>
>
>         On Mon, Mar 3, 2014 at 1:29 AM, Erica Cane
>         <ericacane at splitu.com <mailto:ericacane at splitu.com>> wrote:
>
>             Thanks for the responses Ryan and Sean.
>
>             Is it possible to use homebrew or source to install into a
>             virtualenv?
>
>
>             On Sun, Mar 2, 2014 at 10:06 PM, Sean Fisk
>             <sean at seanfisk.com <mailto:sean at seanfisk.com>> wrote:
>
>                 Hi Erica,
>
>                 This won’t fix the PyPi issue, but I usually install
>                 on Mac OS X using Homebrew <http://brew.sh/>. Though
>                 that has had some recent issues as well.
>
>                 Cheers,
>
>
>
>                 --
>                 Sean Fisk
>
>
>                 On Sun, Mar 2, 2014 at 6:47 PM, Erica B.
>                 <ericacane at splitu.com <mailto:ericacane at splitu.com>>
>                 wrote:
>
>                     Hi all,
>
>                     I'm attempting to install pyside via pip on my
>                     Macbook Pro running OS X 10.9.2 (Mavericks) and am
>                     running into some issues. The install fails with
>                     the following error:
>
>                         /build/PySide/sources/shiboken/ext/sparsehash/google/dense_hash_map:97:10:
>                         fatal error: 'tr1/functional' file not found
>                         #include HASH_FUN_H                  //
>                         defined in config.h
>                                  ^
>                         /build/PySide/sources/shiboken/ext/sparsehash/google/sparsehash/sparseconfig.h:18:24:
>                         note: expanded from macro 'HASH_FUN_H'
>                             #define HASH_FUN_H <tr1/functional>
>                                                ^
>                         1 error generated.
>                         make[2]: ***
>                         [libshiboken/CMakeFiles/libshiboken.dir/sbkconverter.cpp.o]
>                         Error 1
>                         make[1]: ***
>                         [libshiboken/CMakeFiles/libshiboken.dir/all]
>                         Error 2
>                         make: *** [all] Error 2
>                         error: Error compiling shiboken
>
>
>                     After some searching, it seems this is a Mavericks
>                     error and may have been fixed in November
>                     (https://codereview.qt-project.org/#change,69324),
>                     but that change hasn't made its way to PyPi
>                     (https://pypi.python.org/pypi/PySide). If this is
>                     the case, perhaps the PyPi package could use a
>                     refresh. If not, any ideas what can be done to fix
>                     the error?
>
>                     Thanks!
>                     Erica
>
>
>                     _______________________________________________
>
>                     PySide mailing list
>                     PySide at qt-project.org <mailto:PySide at qt-project.org>
>                     http://lists.qt-project.org/mailman/listinfo/pyside
>
>
>
>
>
>
>     _______________________________________________
>     PySide mailing list
>     PySide at qt-project.org <mailto:PySide at qt-project.org>
>     http://lists.qt-project.org/mailman/listinfo/pyside
>
>
>
>
> _______________________________________________
> PySide mailing list
> PySide at qt-project.org
> http://lists.qt-project.org/mailman/listinfo/pyside


-- 
Christian Tismer             :^)   <mailto:tismer at stackless.com>
Software Consulting          :     Have a break! Take a ride on Python's
Karl-Liebknecht-Str. 121     :    *Starship* http://starship.python.net/
14482 Potsdam                :     PGP key -> http://pgp.uni-mainz.de
phone +49 173 24 18 776  fax +49 (30) 700143-0023
PGP 0x57F3BF04       9064 F4E1 D754 C2FF 1619  305B C09C 5A3B 57F3 BF04
      whom do you want to sponsor today?   http://www.stackless.com/

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.qt-project.org/pipermail/pyside/attachments/20140412/b6705c4b/attachment.html>

From backup.rlacko at gmail.com  Sun Apr 13 17:29:17 2014
From: backup.rlacko at gmail.com (Roman Lacko)
Date: Sun, 13 Apr 2014 17:29:17 +0200
Subject: [PySide] Install via pip fails on OS X Mavericks
In-Reply-To: <534998F4.6000401@stackless.com>
References: <CAGj3X_CFjewgkN+oMoF+h75pmV4pGmZa77aN8E76CskSVVnvrg@mail.gmail.com>
	<CAF-FwG8b+dAhvtZB6wumknXp2xQxe=e=yZ95q-iaaPMz-8G+Jg@mail.gmail.com>
	<CA+c+8J37E-_PNp9gaWi11qE3vSSUy8bM9-z+MEuJqnKjwh_J1Q@mail.gmail.com>
	<CAF-FwG_H_xJx8EhVwYphcQrzuDn18V+EVOba-ETCz68tX4fJ1w@mail.gmail.com>
	<CA+c+8J2+9F0BnM-fHPGb9581yVkG6gQFicm6g02gPtF1iGw8cQ@mail.gmail.com>
	<CAHC6v6ihZP-Hgf6e+5gLK+1ohOrPo=T9fPKWPU_0O_WrkjbKFw@mail.gmail.com>
	<534998F4.6000401@stackless.com>
Message-ID: <CAHC6v6h5c4Ja8WH6jN-StaiEyKU342odTx3ESchZ+hpnmxVcZg@mail.gmail.com>

Hi Christian,

If your osx fixes are allready merged in official shiboken repository at
gitorious and John as main maintainer and release manager agreed, we can
start the release process.

John, is there any time schedule for new release ?

Thanks
Roman


2014-04-12 21:50 GMT+02:00 Christian Tismer <tismer at stackless.com>:

>  Hi Roman,
>
> for use in my projects, I cloned shiboken and pyside-setup,
> just in order to build 1.2.1 on Mavericks.
>
> Problem is: The fix for Mavericks is in shiboken, but only reachable
> when building the dev version  1.3.0dev. But a dev version is not for PyPi.
>
> My quick hack was to tag the latest version of shiboken with "1.2.1fix",
> then I changed the 1.2.1 version in setup.py like so:
>
>     '1.2.1': [
>         ["shiboken", "1.2.1fix"],
>         ["pyside", "1.2.1"],
>         ["pyside-tools", "0.2.15"],
>         ["pyside-examples", "master"],
>     ],
>
> and then the build works.
>
> $ python setup.py bdist_wheel --version=1.2.1
>
> I agree this is not totally correct, should officially better be 1.2.2
> but for us this works.
>
> I would really like to produce wheels for OS X, finally. But it is kind of
> a showstopper that fixes get in, but pip does not see this.
> Actually, this is not really a new pyside version, but just a fix to
> shiboken.
>
> What do you think, and how can I help/make it happen?
> Should I provide a patch to create 1.2.2 ?
>
> cheers - Chris
>
>
> On 04.03.14 01:05, Roman Lacko wrote:
>
>  Hi,
>
> You can also install PySide with pip directli from git repository.
>
> This should work, but have not tested with latest version:
>  pip install git+ssh://github.com/PySide/pyside-setup.git<https://github.com/username/repo.git>
>
> -Roman
>
> 2014-03-03 23:23 GMT+01:00 Erica Cane <ericacane at splitu.com>:
>
>> Thanks Sean. I am indeed interested in compiling from source in a
>> virtualenv. That said, the ideal solution would be for the PyPi package to
>> be updated with the fix for Mavericks. :)
>>
>>  Thanks again!
>> Erica
>>
>>
>> On Sun, Mar 2, 2014 at 10:44 PM, Sean Fisk <sean at seanfisk.com> wrote:
>>
>>>  When compiling from source, you can install into a virtualenv. When
>>> using Homebrew, it’s not possible to install into a virtualenv.
>>>
>>> I don’t use Homebrew’s Python, and don’t install any other Python
>>> packages other than PySide. Then, inside of the virtualenv, I use a PTH
>>> file <http://docs.python.org/2/library/site.html> to point to
>>> Homebrew’s site-packages/ directory. Unfortunately, this also includes
>>> any other Python packages installed by Homebrew if you decide to install
>>> them. It’s not the best option, but it works well. I can send you my
>>> PTH-file-writing-script if you are interested.
>>>
>>> If you are interested in compiling from source into a virtualenv, my
>>> team is working on improving the PySide BuildScripts<https://github.com/PySide/BuildScripts>repo and I can send you what we have as well.
>>>
>>> If you are interested in either of these, expect a reply from me
>>> tomorrow as it is pretty late in my neck of the woods.
>>>
>>> Cheers,
>>>
>>>
>>>  --
>>> Sean Fisk
>>>
>>>
>>> On Mon, Mar 3, 2014 at 1:29 AM, Erica Cane <ericacane at splitu.com> wrote:
>>>
>>>>  Thanks for the responses Ryan and Sean.
>>>>
>>>>  Is it possible to use homebrew or source to install into a virtualenv?
>>>>
>>>>
>>>> On Sun, Mar 2, 2014 at 10:06 PM, Sean Fisk <sean at seanfisk.com> wrote:
>>>>
>>>>>  Hi Erica,
>>>>>
>>>>> This won’t fix the PyPi issue, but I usually install on Mac OS X using
>>>>> Homebrew <http://brew.sh/>. Though that has had some recent issues as
>>>>> well.
>>>>>
>>>>> Cheers,
>>>>>
>>>>>
>>>>>  --
>>>>> Sean Fisk
>>>>>
>>>>>
>>>>>  On Sun, Mar 2, 2014 at 6:47 PM, Erica B. <ericacane at splitu.com>wrote:
>>>>>
>>>>>>   Hi all,
>>>>>>
>>>>>>  I'm attempting to install pyside via pip on my Macbook Pro running
>>>>>> OS X 10.9.2 (Mavericks) and am running into some issues. The install fails
>>>>>> with the following error:
>>>>>>
>>>>>>  /build/PySide/sources/shiboken/ext/sparsehash/google/dense_hash_
>>>>>>> map:97:10: fatal error: 'tr1/functional' file not found
>>>>>>> #include HASH_FUN_H                  // defined in config.h
>>>>>>>          ^
>>>>>>> /build/PySide/sources/shiboken/ext/sparsehash/google/sparsehash/sparseconfig.h:18:24:
>>>>>>> note: expanded from macro 'HASH_FUN_H'
>>>>>>>     #define HASH_FUN_H <tr1/functional>
>>>>>>>                        ^
>>>>>>> 1 error generated.
>>>>>>> make[2]: ***
>>>>>>> [libshiboken/CMakeFiles/libshiboken.dir/sbkconverter.cpp.o] Error 1
>>>>>>> make[1]: *** [libshiboken/CMakeFiles/libshiboken.dir/all] Error 2
>>>>>>> make: *** [all] Error 2
>>>>>>> error: Error compiling shiboken
>>>>>>
>>>>>>
>>>>>>  After some searching, it seems this is a Mavericks error and may
>>>>>> have been fixed in November (
>>>>>> https://codereview.qt-project.org/#change,69324), but that change
>>>>>> hasn't made its way to PyPi (https://pypi.python.org/pypi/PySide).
>>>>>> If this is the case, perhaps the PyPi package could use a refresh. If not,
>>>>>> any ideas what can be done to fix the error?
>>>>>>
>>>>>>  Thanks!
>>>>>>  Erica
>>>>>>
>>>>>>
>>>>>>  _______________________________________________
>>>>>>
>>>>>> PySide mailing list
>>>>>> PySide at qt-project.org
>>>>>> http://lists.qt-project.org/mailman/listinfo/pyside
>>>>>>
>>>>>>
>>>>>
>>>>
>>>
>>
>> _______________________________________________
>> PySide mailing list
>> PySide at qt-project.org
>> http://lists.qt-project.org/mailman/listinfo/pyside
>>
>>
>
>
> _______________________________________________
> PySide mailing listPySide at qt-project.orghttp://lists.qt-project.org/mailman/listinfo/pyside
>
>
>
> --
> Christian Tismer             :^)   <mailto:tismer at stackless.com> <tismer at stackless.com>
> Software Consulting          :     Have a break! Take a ride on Python's
> Karl-Liebknecht-Str. 121     :    *Starship* http://starship.python.net/
> 14482 Potsdam                :     PGP key -> http://pgp.uni-mainz.de
> phone +49 173 24 18 776  fax +49 (30) 700143-0023
> PGP 0x57F3BF04       9064 F4E1 D754 C2FF 1619  305B C09C 5A3B 57F3 BF04
>       whom do you want to sponsor today?   http://www.stackless.com/
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.qt-project.org/pipermail/pyside/attachments/20140413/89200259/attachment.html>

From jpe at wingware.com  Sun Apr 13 19:49:44 2014
From: jpe at wingware.com (John Ehresman)
Date: Sun, 13 Apr 2014 13:49:44 -0400
Subject: [PySide] Install via pip fails on OS X Mavericks
In-Reply-To: <CAHC6v6h5c4Ja8WH6jN-StaiEyKU342odTx3ESchZ+hpnmxVcZg@mail.gmail.com>
References: <CAGj3X_CFjewgkN+oMoF+h75pmV4pGmZa77aN8E76CskSVVnvrg@mail.gmail.com>	<CAF-FwG8b+dAhvtZB6wumknXp2xQxe=e=yZ95q-iaaPMz-8G+Jg@mail.gmail.com>	<CA+c+8J37E-_PNp9gaWi11qE3vSSUy8bM9-z+MEuJqnKjwh_J1Q@mail.gmail.com>	<CAF-FwG_H_xJx8EhVwYphcQrzuDn18V+EVOba-ETCz68tX4fJ1w@mail.gmail.com>	<CA+c+8J2+9F0BnM-fHPGb9581yVkG6gQFicm6g02gPtF1iGw8cQ@mail.gmail.com>	<CAHC6v6ihZP-Hgf6e+5gLK+1ohOrPo=T9fPKWPU_0O_WrkjbKFw@mail.gmail.com>	<534998F4.6000401@stackless.com>
	<CAHC6v6h5c4Ja8WH6jN-StaiEyKU342odTx3ESchZ+hpnmxVcZg@mail.gmail.com>
Message-ID: <534ACE38.7040701@wingware.com>

On 4/13/14, 11:29 AM, Roman Lacko wrote:
> If your osx fixes are allready merged in official shiboken repository at
> gitorious and John as main maintainer and release manager agreed, we can
> start the release process.
>
> John, is there any time schedule for new release ?

I think we should do a release soon (maybe in the next couple of days), 
for this and for Python 3.4 support.

Christian mentioned wheels -- AFAIK wheels are not yet suitable as a 
binary distribution format for extension modules on OS X or for Linux. 
I also would not assume that a binary compiled on Mavericks will work on 
earlier OS X releases.  I generally compile on the earliest OS X or 
Linux version the binaries need to run on.

Cheers,

John


From matthew.brett at gmail.com  Sun Apr 13 19:54:24 2014
From: matthew.brett at gmail.com (Matthew Brett)
Date: Sun, 13 Apr 2014 10:54:24 -0700
Subject: [PySide] Install via pip fails on OS X Mavericks
In-Reply-To: <534ACE38.7040701@wingware.com>
References: <CAGj3X_CFjewgkN+oMoF+h75pmV4pGmZa77aN8E76CskSVVnvrg@mail.gmail.com>
	<CAF-FwG8b+dAhvtZB6wumknXp2xQxe=e=yZ95q-iaaPMz-8G+Jg@mail.gmail.com>
	<CA+c+8J37E-_PNp9gaWi11qE3vSSUy8bM9-z+MEuJqnKjwh_J1Q@mail.gmail.com>
	<CAF-FwG_H_xJx8EhVwYphcQrzuDn18V+EVOba-ETCz68tX4fJ1w@mail.gmail.com>
	<CA+c+8J2+9F0BnM-fHPGb9581yVkG6gQFicm6g02gPtF1iGw8cQ@mail.gmail.com>
	<CAHC6v6ihZP-Hgf6e+5gLK+1ohOrPo=T9fPKWPU_0O_WrkjbKFw@mail.gmail.com>
	<534998F4.6000401@stackless.com>
	<CAHC6v6h5c4Ja8WH6jN-StaiEyKU342odTx3ESchZ+hpnmxVcZg@mail.gmail.com>
	<534ACE38.7040701@wingware.com>
Message-ID: <CAH6Pt5rHUJs+MSJw_BnhoncxUEXkdMQpezD4dK0SFpFNoCpyKA@mail.gmail.com>

Hi,

On Sun, Apr 13, 2014 at 10:49 AM, John Ehresman <jpe at wingware.com> wrote:
> On 4/13/14, 11:29 AM, Roman Lacko wrote:
>> If your osx fixes are allready merged in official shiboken repository at
>> gitorious and John as main maintainer and release manager agreed, we can
>> start the release process.
>>
>> John, is there any time schedule for new release ?
>
> I think we should do a release soon (maybe in the next couple of days),
> for this and for Python 3.4 support.
>
> Christian mentioned wheels -- AFAIK wheels are not yet suitable as a
> binary distribution format for extension modules on OS X or for Linux.
> I also would not assume that a binary compiled on Mavericks will work on
> earlier OS X releases.  I generally compile on the earliest OS X or
> Linux version the binaries need to run on.

OSX binary wheels should work fine, and they are getting picked up pretty fast:

https://github.com/MacPython/wiki/wiki/Spinning-wheels

I am happy to build a wheel for you if you lack a 10.6 machine, but
I've built wheels for numpy, scipy and matplotlib on Mavericks and had
no problem on 10.6.

I agree that Linux wheels are likely to be dangerous, but I haven't
tried them myself.

Cheers,

Matthew


From jpe at wingware.com  Sun Apr 13 20:08:42 2014
From: jpe at wingware.com (John Ehresman)
Date: Sun, 13 Apr 2014 14:08:42 -0400
Subject: [PySide] Install via pip fails on OS X Mavericks
In-Reply-To: <CAH6Pt5rHUJs+MSJw_BnhoncxUEXkdMQpezD4dK0SFpFNoCpyKA@mail.gmail.com>
References: <CAGj3X_CFjewgkN+oMoF+h75pmV4pGmZa77aN8E76CskSVVnvrg@mail.gmail.com>	<CAF-FwG8b+dAhvtZB6wumknXp2xQxe=e=yZ95q-iaaPMz-8G+Jg@mail.gmail.com>	<CA+c+8J37E-_PNp9gaWi11qE3vSSUy8bM9-z+MEuJqnKjwh_J1Q@mail.gmail.com>	<CAF-FwG_H_xJx8EhVwYphcQrzuDn18V+EVOba-ETCz68tX4fJ1w@mail.gmail.com>	<CA+c+8J2+9F0BnM-fHPGb9581yVkG6gQFicm6g02gPtF1iGw8cQ@mail.gmail.com>	<CAHC6v6ihZP-Hgf6e+5gLK+1ohOrPo=T9fPKWPU_0O_WrkjbKFw@mail.gmail.com>	<534998F4.6000401@stackless.com>	<CAHC6v6h5c4Ja8WH6jN-StaiEyKU342odTx3ESchZ+hpnmxVcZg@mail.gmail.com>	<534ACE38.7040701@wingware.com>
	<CAH6Pt5rHUJs+MSJw_BnhoncxUEXkdMQpezD4dK0SFpFNoCpyKA@mail.gmail.com>
Message-ID: <534AD2AA.5070808@wingware.com>

On 4/13/14, 1:54 PM, Matthew Brett wrote:
> OSX binary wheels should work fine, and they are getting picked up pretty fast:
>
> https://github.com/MacPython/wiki/wiki/Spinning-wheels
>
> I am happy to build a wheel for you if you lack a 10.6 machine, but
> I've built wheels for numpy, scipy and matplotlib on Mavericks and had
> no problem on 10.6.

I thought I saw discussion on the distutils list of the potentially 
different abi's on OS X, but maybe this has been resolved.

The C++ standard library version may be an issue when supporting 
different OS X versions.  I don't know that there will be problems, but 
I've found it's better to think that there might be version related 
problems than to assume things will work.

Cheers,

John


From tismer at stackless.com  Sun Apr 13 21:07:49 2014
From: tismer at stackless.com (Christian Tismer)
Date: Sun, 13 Apr 2014 21:07:49 +0200
Subject: [PySide] Install via pip fails on OS X Mavericks
In-Reply-To: <CAH6Pt5rHUJs+MSJw_BnhoncxUEXkdMQpezD4dK0SFpFNoCpyKA@mail.gmail.com>
References: <CAGj3X_CFjewgkN+oMoF+h75pmV4pGmZa77aN8E76CskSVVnvrg@mail.gmail.com>
	<CAF-FwG8b+dAhvtZB6wumknXp2xQxe=e=yZ95q-iaaPMz-8G+Jg@mail.gmail.com>
	<CA+c+8J37E-_PNp9gaWi11qE3vSSUy8bM9-z+MEuJqnKjwh_J1Q@mail.gmail.com>
	<CAF-FwG_H_xJx8EhVwYphcQrzuDn18V+EVOba-ETCz68tX4fJ1w@mail.gmail.com>
	<CA+c+8J2+9F0BnM-fHPGb9581yVkG6gQFicm6g02gPtF1iGw8cQ@mail.gmail.com>
	<CAHC6v6ihZP-Hgf6e+5gLK+1ohOrPo=T9fPKWPU_0O_WrkjbKFw@mail.gmail.com>
	<534998F4.6000401@stackless.com>
	<CAHC6v6h5c4Ja8WH6jN-StaiEyKU342odTx3ESchZ+hpnmxVcZg@mail.gmail.com>
	<534ACE38.7040701@wingware.com>
	<CAH6Pt5rHUJs+MSJw_BnhoncxUEXkdMQpezD4dK0SFpFNoCpyKA@mail.gmail.com>
Message-ID: <534AE085.1080503@stackless.com>

Hi guys,

I have built both eggs and wheels for PySide on Mountain Lion and
Mavericks.

Using the latest XCode, I could not manage to change
MACOSX_DEPLOYMENT_TARGET.
When changing that to something like 10.6, I always get the complaint that

distutils.errors.DistutilsPlatformError: $MACOSX_DEPLOYMENT_TARGET
mismatch: now "10.6" but "10.9" during configure

So I ended up with doing all builds without setting that, once on 10.9,
and once on a virtual 10.8. But all eggs and wheels are working fine.

It is probably a bit redundant, but for sure very correct to use such
exacly built wheels.

cheers - Chris


On 13.04.14 19:54, Matthew Brett wrote:
> Hi,
>
> On Sun, Apr 13, 2014 at 10:49 AM, John Ehresman <jpe at wingware.com> wrote:
>> On 4/13/14, 11:29 AM, Roman Lacko wrote:
>>> If your osx fixes are allready merged in official shiboken repository at
>>> gitorious and John as main maintainer and release manager agreed, we can
>>> start the release process.
>>>
>>> John, is there any time schedule for new release ?
>> I think we should do a release soon (maybe in the next couple of days),
>> for this and for Python 3.4 support.
>>
>> Christian mentioned wheels -- AFAIK wheels are not yet suitable as a
>> binary distribution format for extension modules on OS X or for Linux.
>> I also would not assume that a binary compiled on Mavericks will work on
>> earlier OS X releases.  I generally compile on the earliest OS X or
>> Linux version the binaries need to run on.
> OSX binary wheels should work fine, and they are getting picked up pretty fast:
>
> https://github.com/MacPython/wiki/wiki/Spinning-wheels
>
> I am happy to build a wheel for you if you lack a 10.6 machine, but
> I've built wheels for numpy, scipy and matplotlib on Mavericks and had
> no problem on 10.6.
>
> I agree that Linux wheels are likely to be dangerous, but I haven't
> tried them myself.
>
> Cheers,
>
> Matthew


-- 
Christian Tismer             :^)   <mailto:tismer at stackless.com>
Software Consulting          :     Have a break! Take a ride on Python's
Karl-Liebknecht-Str. 121     :    *Starship* http://starship.python.net/
14482 Potsdam                :     PGP key -> http://pgp.uni-mainz.de
phone +49 173 24 18 776  fax +49 (30) 700143-0023
PGP 0x57F3BF04       9064 F4E1 D754 C2FF 1619  305B C09C 5A3B 57F3 BF04
      whom do you want to sponsor today?   http://www.stackless.com/



From tismer at stackless.com  Sun Apr 13 22:45:35 2014
From: tismer at stackless.com (Christian Tismer)
Date: Sun, 13 Apr 2014 22:45:35 +0200
Subject: [PySide] Install via pip fails on OS X Mavericks
In-Reply-To: <CA+c+8J2+9F0BnM-fHPGb9581yVkG6gQFicm6g02gPtF1iGw8cQ@mail.gmail.com>
References: <CAGj3X_CFjewgkN+oMoF+h75pmV4pGmZa77aN8E76CskSVVnvrg@mail.gmail.com>	<CAF-FwG8b+dAhvtZB6wumknXp2xQxe=e=yZ95q-iaaPMz-8G+Jg@mail.gmail.com>	<CA+c+8J37E-_PNp9gaWi11qE3vSSUy8bM9-z+MEuJqnKjwh_J1Q@mail.gmail.com>	<CAF-FwG_H_xJx8EhVwYphcQrzuDn18V+EVOba-ETCz68tX4fJ1w@mail.gmail.com>
	<CA+c+8J2+9F0BnM-fHPGb9581yVkG6gQFicm6g02gPtF1iGw8cQ@mail.gmail.com>
Message-ID: <534AF76F.2040204@stackless.com>

Hi Erica,

yesterday, I have built PySide 1.2.1 wheels and eggs for
python 3.3, 3.4, 2.7 on OS X 10.8 and 10.9.

You can use my downloads from here:

https://bitbucket.org/pydica/pyside-setup/downloads

Installation is not as nice as with pip, but it is fast and works well:

Example after downloading and running from a virtual env:

$ pip install wheel
$ wheel install
~/src/pyside-setup/dist/PySide-1.2.1-cp34-cp34m-macosx_10_9_x86_64.whl
$ python py3/bin/pyside_postinstall.py -install

Let me know how it works for you.

ciao - Chris


On 03.03.14 23:23, Erica Cane wrote:
> Thanks Sean. I am indeed interested in compiling from source in a
> virtualenv. That said, the ideal solution would be for the PyPi
> package to be updated with the fix for Mavericks. :)
>
> Thanks again!
> Erica
>
>
> On Sun, Mar 2, 2014 at 10:44 PM, Sean Fisk <sean at seanfisk.com
> <mailto:sean at seanfisk.com>> wrote:
>
>     When compiling from source, you can install into a virtualenv.
>     When using Homebrew, it’s not possible to install into a virtualenv.
>
>     I don’t use Homebrew’s Python, and don’t install any other Python
>     packages other than PySide. Then, inside of the virtualenv, I use
>     a PTH file <http://docs.python.org/2/library/site.html> to point
>     to Homebrew’s |site-packages/| directory. Unfortunately, this also
>     includes any other Python packages installed by Homebrew if you
>     decide to install them. It’s not the best option, but it works
>     well. I can send you my PTH-file-writing-script if you are interested.
>
>     If you are interested in compiling from source into a virtualenv,
>     my team is working on improving the PySide BuildScripts
>     <https://github.com/PySide/BuildScripts> repo and I can send you
>     what we have as well.
>
>     If you are interested in either of these, expect a reply from me
>     tomorrow as it is pretty late in my neck of the woods.
>
>     Cheers,
>
>
>
>     --
>     Sean Fisk
>
>
>     On Mon, Mar 3, 2014 at 1:29 AM, Erica Cane <ericacane at splitu.com
>     <mailto:ericacane at splitu.com>> wrote:
>
>         Thanks for the responses Ryan and Sean.
>
>         Is it possible to use homebrew or source to install into a
>         virtualenv?
>
>
>         On Sun, Mar 2, 2014 at 10:06 PM, Sean Fisk <sean at seanfisk.com
>         <mailto:sean at seanfisk.com>> wrote:
>
>             Hi Erica,
>
>             This won’t fix the PyPi issue, but I usually install on
>             Mac OS X using Homebrew <http://brew.sh/>. Though that has
>             had some recent issues as well.
>
>             Cheers,
>
>
>
>             --
>             Sean Fisk
>
>
>             On Sun, Mar 2, 2014 at 6:47 PM, Erica B.
>             <ericacane at splitu.com <mailto:ericacane at splitu.com>> wrote:
>
>                 Hi all,
>
>                 I'm attempting to install pyside via pip on my Macbook
>                 Pro running OS X 10.9.2 (Mavericks) and am running
>                 into some issues. The install fails with the following
>                 error:
>
>                     /build/PySide/sources/shiboken/ext/sparsehash/google/dense_hash_map:97:10:
>                     fatal error: 'tr1/functional' file not found
>                     #include HASH_FUN_H                  // defined in
>                     config.h
>                              ^
>                     /build/PySide/sources/shiboken/ext/sparsehash/google/sparsehash/sparseconfig.h:18:24:
>                     note: expanded from macro 'HASH_FUN_H'
>                         #define HASH_FUN_H <tr1/functional>
>                                            ^
>                     1 error generated.
>                     make[2]: ***
>                     [libshiboken/CMakeFiles/libshiboken.dir/sbkconverter.cpp.o]
>                     Error 1
>                     make[1]: ***
>                     [libshiboken/CMakeFiles/libshiboken.dir/all] Error 2
>                     make: *** [all] Error 2
>                     error: Error compiling shiboken
>
>
>                 After some searching, it seems this is a Mavericks
>                 error and may have been fixed in November
>                 (https://codereview.qt-project.org/#change,69324), but
>                 that change hasn't made its way to PyPi
>                 (https://pypi.python.org/pypi/PySide). If this is the
>                 case, perhaps the PyPi package could use a refresh. If
>                 not, any ideas what can be done to fix the error?
>
>                 Thanks!
>                 Erica
>
>
>                 _______________________________________________
>
>                 PySide mailing list
>                 PySide at qt-project.org <mailto:PySide at qt-project.org>
>                 http://lists.qt-project.org/mailman/listinfo/pyside
>
>
>
>
>
>
>
> _______________________________________________
> PySide mailing list
> PySide at qt-project.org
> http://lists.qt-project.org/mailman/listinfo/pyside


-- 
Christian Tismer             :^)   <mailto:tismer at stackless.com>
Software Consulting          :     Have a break! Take a ride on Python's
Karl-Liebknecht-Str. 121     :    *Starship* http://starship.python.net/
14482 Potsdam                :     PGP key -> http://pgp.uni-mainz.de
phone +49 173 24 18 776  fax +49 (30) 700143-0023
PGP 0x57F3BF04       9064 F4E1 D754 C2FF 1619  305B C09C 5A3B 57F3 BF04
      whom do you want to sponsor today?   http://www.stackless.com/

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.qt-project.org/pipermail/pyside/attachments/20140413/f35531a5/attachment.html>

From matthew.brett at gmail.com  Sun Apr 13 23:27:51 2014
From: matthew.brett at gmail.com (Matthew Brett)
Date: Sun, 13 Apr 2014 14:27:51 -0700
Subject: [PySide] Install via pip fails on OS X Mavericks
In-Reply-To: <534AE085.1080503@stackless.com>
References: <CAGj3X_CFjewgkN+oMoF+h75pmV4pGmZa77aN8E76CskSVVnvrg@mail.gmail.com>
	<CAF-FwG8b+dAhvtZB6wumknXp2xQxe=e=yZ95q-iaaPMz-8G+Jg@mail.gmail.com>
	<CA+c+8J37E-_PNp9gaWi11qE3vSSUy8bM9-z+MEuJqnKjwh_J1Q@mail.gmail.com>
	<CAF-FwG_H_xJx8EhVwYphcQrzuDn18V+EVOba-ETCz68tX4fJ1w@mail.gmail.com>
	<CA+c+8J2+9F0BnM-fHPGb9581yVkG6gQFicm6g02gPtF1iGw8cQ@mail.gmail.com>
	<CAHC6v6ihZP-Hgf6e+5gLK+1ohOrPo=T9fPKWPU_0O_WrkjbKFw@mail.gmail.com>
	<534998F4.6000401@stackless.com>
	<CAHC6v6h5c4Ja8WH6jN-StaiEyKU342odTx3ESchZ+hpnmxVcZg@mail.gmail.com>
	<534ACE38.7040701@wingware.com>
	<CAH6Pt5rHUJs+MSJw_BnhoncxUEXkdMQpezD4dK0SFpFNoCpyKA@mail.gmail.com>
	<534AE085.1080503@stackless.com>
Message-ID: <CAH6Pt5oYExC_wf6w95_TL+5G7MLeWuXaxxp==t9_8P4j_1qRnw@mail.gmail.com>

Hi,

On Sun, Apr 13, 2014 at 12:07 PM, Christian Tismer <tismer at stackless.com> wrote:
> Hi guys,
>
> I have built both eggs and wheels for PySide on Mountain Lion and
> Mavericks.
>
> Using the latest XCode, I could not manage to change
> MACOSX_DEPLOYMENT_TARGET.
> When changing that to something like 10.6, I always get the complaint that
>
> distutils.errors.DistutilsPlatformError: $MACOSX_DEPLOYMENT_TARGET
> mismatch: now "10.6" but "10.9" during configure
>
> So I ended up with doing all builds without setting that, once on 10.9,
> and once on a virtual 10.8. But all eggs and wheels are working fine.
>
> It is probably a bit redundant, but for sure very correct to use such
> exacly built wheels.

The trick is to build the wheels with the Python.org python, which
declares its SDK as 10.6.

You can then rename the wheels if you want them to install on - say -
system python, etc - see the page and the PEP for details.

Here are some wheels suitable renamed so they will install into the
most common combinations of Python source / OSX version:

http://kerbin.bic.berkeley.edu/wheelhouse/

I don't know of any ABI compatibility issues - there are links in the
page I pointed to, with discussion about that, but I don't think these
worries were well founded.

Of course, the best thing is to test...

Cheers,

Matthew


From tismer at stackless.com  Mon Apr 14 03:11:38 2014
From: tismer at stackless.com (Christian Tismer)
Date: Mon, 14 Apr 2014 03:11:38 +0200
Subject: [PySide] Install via pip fails on OS X Mavericks
In-Reply-To: <CAH6Pt5oYExC_wf6w95_TL+5G7MLeWuXaxxp==t9_8P4j_1qRnw@mail.gmail.com>
References: <CAGj3X_CFjewgkN+oMoF+h75pmV4pGmZa77aN8E76CskSVVnvrg@mail.gmail.com>
	<CAF-FwG8b+dAhvtZB6wumknXp2xQxe=e=yZ95q-iaaPMz-8G+Jg@mail.gmail.com>
	<CA+c+8J37E-_PNp9gaWi11qE3vSSUy8bM9-z+MEuJqnKjwh_J1Q@mail.gmail.com>
	<CAF-FwG_H_xJx8EhVwYphcQrzuDn18V+EVOba-ETCz68tX4fJ1w@mail.gmail.com>
	<CA+c+8J2+9F0BnM-fHPGb9581yVkG6gQFicm6g02gPtF1iGw8cQ@mail.gmail.com>
	<CAHC6v6ihZP-Hgf6e+5gLK+1ohOrPo=T9fPKWPU_0O_WrkjbKFw@mail.gmail.com>
	<534998F4.6000401@stackless.com>
	<CAHC6v6h5c4Ja8WH6jN-StaiEyKU342odTx3ESchZ+hpnmxVcZg@mail.gmail.com>
	<534ACE38.7040701@wingware.com>
	<CAH6Pt5rHUJs+MSJw_BnhoncxUEXkdMQpezD4dK0SFpFNoCpyKA@mail.gmail.com>
	<534AE085.1080503@stackless.com>
	<CAH6Pt5oYExC_wf6w95_TL+5G7MLeWuXaxxp==t9_8P4j_1qRnw@mail.gmail.com>
Message-ID: <534B35CA.6010203@stackless.com>

Hi, oh-oh...

On 13.04.14 23:27, Matthew Brett wrote:
> Hi,
>
> On Sun, Apr 13, 2014 at 12:07 PM, Christian Tismer <tismer at stackless.com> wrote:
>> Hi guys,
>>
>> I have built both eggs and wheels for PySide on Mountain Lion and
>> Mavericks.
>>
>> Using the latest XCode, I could not manage to change
>> MACOSX_DEPLOYMENT_TARGET.
>> When changing that to something like 10.6, I always get the complaint that
>>
>> distutils.errors.DistutilsPlatformError: $MACOSX_DEPLOYMENT_TARGET
>> mismatch: now "10.6" but "10.9" during configure
>>
>> So I ended up with doing all builds without setting that, once on 10.9,
>> and once on a virtual 10.8. But all eggs and wheels are working fine.
>>
>> It is probably a bit redundant, but for sure very correct to use such
>> exacly built wheels.
> The trick is to build the wheels with the Python.org python, which
> declares its SDK as 10.6.

Yes, I saw that now:
I used homebrew python without thinking.
And sure, whenever I build Python myself, I do use the 10.6 setting.

Now I understood that this setting defines the minimum for extension
modules, and PySide still is one. Forgot that completely, because
it compiles so long.

After looking into the Cellar, everything was clear: :-)

/usr/local/Cellar/python3/3.4.0_1/Frameworks/Python.framework/Versions/3.4/lib/python3.4/_sysconfigdata.py:
'MACOSX_DEPLOYMENT_TARGET': '10.9',

This setting is inserted unconditionally from the OS version in the
python formula. Yay!

>
> You can then rename the wheels if you want them to install on - say -
> system python, etc - see the page and the PEP for details.
>
> Here are some wheels suitable renamed so they will install into the
> most common combinations of Python source / OSX version:
>
> http://kerbin.bic.berkeley.edu/wheelhouse/
>
> I don't know of any ABI compatibility issues - there are links in the
> page I pointed to, with discussion about that, but I don't think these
> worries were well founded.
>
> Of course, the best thing is to test...
>

So, the versions that I built are fine, just a bit over-done.

Well, these versions also minimize possible incompatibilities ;-)

Thank you for solving the knot in my head.

Btw., I was not aware that the wheels may be just renamed. I assumed
that some more knowledge is sitting inside to verify the right version.
Sadly, things are way more primitive that I hoped for.

cheers & thanks -- Chris

-- 
Christian Tismer             :^)   <mailto:tismer at stackless.com>
Software Consulting          :     Have a break! Take a ride on Python's
Karl-Liebknecht-Str. 121     :    *Starship* http://starship.python.net/
14482 Potsdam                :     PGP key -> http://pgp.uni-mainz.de
phone +49 173 24 18 776  fax +49 (30) 700143-0023
PGP 0x57F3BF04       9064 F4E1 D754 C2FF 1619  305B C09C 5A3B 57F3 BF04
      whom do you want to sponsor today?   http://www.stackless.com/



From jmohler at gamry.com  Mon Apr 14 22:03:07 2014
From: jmohler at gamry.com (Joel B. Mohler)
Date: Mon, 14 Apr 2014 16:03:07 -0400
Subject: [PySide] Install via pip fails on OS X Mavericks
In-Reply-To: <534ACE38.7040701@wingware.com>
References: <CAGj3X_CFjewgkN+oMoF+h75pmV4pGmZa77aN8E76CskSVVnvrg@mail.gmail.com>	<CAF-FwG8b+dAhvtZB6wumknXp2xQxe=e=yZ95q-iaaPMz-8G+Jg@mail.gmail.com>	<CA+c+8J37E-_PNp9gaWi11qE3vSSUy8bM9-z+MEuJqnKjwh_J1Q@mail.gmail.com>	<CAF-FwG_H_xJx8EhVwYphcQrzuDn18V+EVOba-ETCz68tX4fJ1w@mail.gmail.com>	<CA+c+8J2+9F0BnM-fHPGb9581yVkG6gQFicm6g02gPtF1iGw8cQ@mail.gmail.com>	<CAHC6v6ihZP-Hgf6e+5gLK+1ohOrPo=T9fPKWPU_0O_WrkjbKFw@mail.gmail.com>	<534998F4.6000401@stackless.com>	<CAHC6v6h5c4Ja8WH6jN-StaiEyKU342odTx3ESchZ+hpnmxVcZg@mail.gmail.com>
	<534ACE38.7040701@wingware.com>
Message-ID: <534C3EFB.8050802@gamry.com>

On 4/13/2014 1:49 PM, John Ehresman wrote:
> On 4/13/14, 11:29 AM, Roman Lacko wrote:
>> >If your osx fixes are allready merged in official shiboken repository at
>> >gitorious and John as main maintainer and release manager agreed, we can
>> >start the release process.
>> >
>> >John, is there any time schedule for new release ?
> I think we should do a release soon (maybe in the next couple of days),
> for this and for Python 3.4 support.

Will the release include patches from 
https://codereview.qt-project.org/#q,status:open+project:pyside/pyside,n,z 
?  In particular, I'm mostly concerned about my own patch there.

Joel






From jonasthiem at googlemail.com  Wed Apr 16 23:08:47 2014
From: jonasthiem at googlemail.com (Jonas Thiem)
Date: Wed, 16 Apr 2014 23:08:47 +0200
Subject: [PySide] Trouble understanding QStandardItemModel
Message-ID: <CAD7Z2ho148E0W5a6SBxki5Az60JqaKhnmnS36h+3pXqnXLGKcw@mail.gmail.com>

Hi,

Trying to understand QStandardItemModel I came up with this:

from PySide.QtCore import *
from PySide.QtGui import *

model = QStandardItemModel()
model.invisibleRootItem().appendRow(QStandardItem("test"))
assert(model.rowCount() == 1)
assert(model.columnCount() == 1)
mIndex = model.createIndex(0, 0)
print(model.data(model.index(0, 0)))
assert(model.itemFromIndex(mIndex) != None)

The last assert fails with PySide 1.2.1 and Python 3.3.5 on Windows 7.
Is that supposed to happen? What would be the proper approach to
simply get the first most topleft item in a QStandardItemModel?

Regards,
Jonas Thiem


From pyside at mlists.thewrittenword.com  Thu Apr 17 00:21:17 2014
From: pyside at mlists.thewrittenword.com (Albert Chin)
Date: Wed, 16 Apr 2014 17:21:17 -0500
Subject: [PySide] Problems building pyside 1.2.1 on AIX (shiboken problem)
Message-ID: <20140416222116.GA5604@thewrittenword.com>

I've tried building pyside-1.2.1 on AIX. I've been able to
successfully build it on Solaris, HP-UX, and Linux. The build on AIX
errors with:
  cd /opt/build/pyside-qt4.8+1.2.1/PySide/QtGui && /opt/TWWfsw/gcc44/bin/g++   -DQT_CORE_LIB -DQT_GUI_LIB -DQT_NO_DEBUG -DQtGui_EXPORTS -O2 -fPIC -I/opt/TWWfsw/libqt48/include/gcc44 -I/opt/TWWfsw/libqt48/include/gcc44/QtGui -I/opt/TWWfsw/libqt48/include/gcc44/QtCore -I/opt/build/pyside-qt4.8+1.2.1/PySide/QtGui/QtGui -I/opt/build/pyside-qt4.8+1.2.1/PySide/QtGui -I/opt/build/pyside-qt4.8+1.2.1/PySide/QtGui/PySide/QtGui -I/opt/build/pyside-qt4.8+1.2.1/PySide -I/opt/TWWfsw/libqt48/lib/python27/include/shiboken -I/opt/build/pyside-qt4.8+1.2.1/libpyside -I/opt/TWWfsw/python27/include/python2.7 -I/opt/build/pyside-qt4.8+1.2.1/PySide/QtCore/PySide/QtCore    -o CMakeFiles/QtGui.dir/PySide/QtGui/qkeyevent_wrapper.cpp.o -c /opt/build/pyside-qt4.8+1.2.1/PySide/QtGui/PySide/QtGui/qkeyevent_wrapper.cpp
  /opt/build/pyside-qt4.8+1.2.1/PySide/QtGui/PySide/QtGui/qkeyevent_wrapper.cpp: In function 'PyObject* Sbk_QKeyEvent_get_autor(PyObject*, void*)':
  /opt/build/pyside-qt4.8+1.2.1/PySide/QtGui/PySide/QtGui/qkeyevent_wrapper.cpp:686: error: attempt to take address of bit-field structure member 'QKeyEvent::autor'
  /opt/build/pyside-qt4.8+1.2.1/PySide/QtGui/PySide/QtGui/qkeyevent_wrapper.cpp: In function 'int Sbk_QKeyEvent_set_autor(PyObject*, PyObject*, void*)':
  /opt/build/pyside-qt4.8+1.2.1/PySide/QtGui/PySide/QtGui/qkeyevent_wrapper.cpp:706: error: attempt to take address of bit-field structure member 'QKeyEvent::autor'
  gmake[2]: *** [PySide/QtGui/CMakeFiles/QtGui.dir/PySide/QtGui/qkeyevent_wrapper.cpp.o] Error 1
  gmake[2]: Leaving directory `/opt/build/pyside-qt4.8+1.2.1'

Looking at PySide/QtGui/PySide/QtGui/qkeyevent_wrapper.cpp:
  static PyObject* Sbk_QKeyEvent_get_autor(PyObject* self, void*)
  {
      ::QKeyEvent* cppSelf = 0;
      SBK_UNUSED(cppSelf)
      if (!Shiboken::Object::isValid(self))
          return 0;
      cppSelf = ((::QKeyEvent*)Shiboken::Conversions::cppPointer(SbkPySide_QtGuiTypes[SBK_QKEYEVENT_IDX], (SbkObject*)self));
      PyObject* pyOut = Shiboken::Conversions::copyToPython(Shiboken::Conversions::PrimitiveTypeConverter<uint>(), &cppSelf->autor);
      return pyOut;
  }

On Linux, the same function appears as:
  static PyObject* Sbk_QKeyEvent_get_autor(PyObject* self, void*)
  {
      ::QKeyEvent* cppSelf = 0;
      SBK_UNUSED(cppSelf)
      if (!Shiboken::Object::isValid(self))
          return 0;
      cppSelf = ((::QKeyEvent*)Shiboken::Conversions::cppPointer(SbkPySide_QtGuiTypes[SBK_QKEYEVENT_IDX], (SbkObject*)self));
      uint cppOut_local = cppSelf->autor;
      PyObject* pyOut = Shiboken::Conversions::copyToPython(Shiboken::Conversions::PrimitiveTypeConverter<uint>(), &cppOut_local);
      return pyOut;
  }

I need to get shiboken to provide the same behavior on Linux as on
AIX. While debugging this, it seems that the alias for "uint" doesn't
map to "unsigned int" on AIX. In ApiExtractor/typesystem.cpp from
shiboken:
  bool TypeEntry::isCppPrimitive() const
  {
      if (!isPrimitive())
          return false;

      PrimitiveTypeEntry* aliasedType = ((PrimitiveTypeEntry*)this)->basicAliasedTypeEntry();
      QByteArray typeName = (aliasedType ? aliasedType->name() : m_name).toAscii();

On AIX, aliasedType is NULL while on Solaris, HP-UX, and Linux, it is
not. I presume this means there is on alias from "uint" to "unsigned
int". Where in the code should I be looking further?

-- 
albert chin (china at thewrittenword.com)


From schampailler at skynet.be  Thu Apr 17 08:20:55 2014
From: schampailler at skynet.be (S. Champailler)
Date: Thu, 17 Apr 2014 08:20:55 +0200 (CEST)
Subject: [PySide] Trouble understanding QStandardItemModel
In-Reply-To: <CAD7Z2ho148E0W5a6SBxki5Az60JqaKhnmnS36h+3pXqnXLGKcw@mail.gmail.com>
References: <CAD7Z2ho148E0W5a6SBxki5Az60JqaKhnmnS36h+3pXqnXLGKcw@mail.gmail.com>
Message-ID: <1644806592.93345.1397715655832.open-xchange@webmail.nmp.skynet.be>

Tried your code, same error.

I'm just wondering if model.data(...) default role is not preventing something
to work in QStandardItemModel....

Stefan

> Le 16 avril 2014 à 23:08, Jonas Thiem <jonasthiem at googlemail.com> a écrit :
>
>
> Hi,
>
> Trying to understand QStandardItemModel I came up with this:
>
> from PySide.QtCore import *
> from PySide.QtGui import *
>
> model = QStandardItemModel()
> model.invisibleRootItem().appendRow(QStandardItem("test"))
> assert(model.rowCount() == 1)
> assert(model.columnCount() == 1)
> mIndex = model.createIndex(0, 0)
> print(model.data(model.index(0, 0)))
> assert(model.itemFromIndex(mIndex) != None)
>
> The last assert fails with PySide 1.2.1 and Python 3.3.5 on Windows 7.
> Is that supposed to happen? What would be the proper approach to
> simply get the first most topleft item in a QStandardItemModel?
>
> Regards,
> Jonas Thiem
> _______________________________________________
> PySide mailing list
> PySide at qt-project.org
> http://lists.qt-project.org/mailman/listinfo/pyside


From robinds at student.matnat.uio.no  Thu Apr 17 11:18:38 2014
From: robinds at student.matnat.uio.no (Robin Skahjem-Eriksen)
Date: Thu, 17 Apr 2014 11:18:38 +0200
Subject: [PySide] QTextLayout::glyphRuns() is missing in Pyside 1.2
Message-ID: <20140417091838.GB26357@uio.no>

I'm starting to learn Qt, by using PySide and it is really exciting and fun. I'm so glad this project exists, as I know I would probably not have been using them right now if I had to learn C++.

Unfortunately, I've encountered a missing method in PySide: The binding to QTextLayout::glyphRuns(). I have seen that this is available in PyQt, but I really would prefer to have it available through PySide.

Can this please be fixed by someone who knows C++ and the source code for the PySide project?
If this is fixed, can I please receive a notice, so I can rebuild the and install the package with the patch included?

I hate to bother anybody, but this feature actually seems completely crucial to my project.
If there is anything I can do to speed up the process or aid it in any way, please let me know.

Sincerely,
Robin Skahjem-Eriksen



From greatrgb at gmail.com  Thu Apr 17 15:16:14 2014
From: greatrgb at gmail.com (Tony Barbieri)
Date: Thu, 17 Apr 2014 09:16:14 -0400
Subject: [PySide] Trouble understanding QStandardItemModel
In-Reply-To: <1644806592.93345.1397715655832.open-xchange@webmail.nmp.skynet.be>
References: <CAD7Z2ho148E0W5a6SBxki5Az60JqaKhnmnS36h+3pXqnXLGKcw@mail.gmail.com>
	<1644806592.93345.1397715655832.open-xchange@webmail.nmp.skynet.be>
Message-ID: <CAJhmvsQr=CWhHPe4ZP8g=4MJmmXVy7_fz7w5hEw0OWyBHekcrA@mail.gmail.com>

You could try model.itemFromIndex(model.index(0, 0)), that should return
the item.  You also don't need to use the invisibleRootItem to append a
row, you should be able to just call model.appendRow(QStandardItem("test")).


On Thu, Apr 17, 2014 at 2:20 AM, S. Champailler <schampailler at skynet.be>wrote:

> Tried your code, same error.
>
> I'm just wondering if model.data(...) default role is not preventing
> something
> to work in QStandardItemModel....
>
> Stefan
>
> > Le 16 avril 2014 à 23:08, Jonas Thiem <jonasthiem at googlemail.com> a
> écrit :
> >
> >
> > Hi,
> >
> > Trying to understand QStandardItemModel I came up with this:
> >
> > from PySide.QtCore import *
> > from PySide.QtGui import *
> >
> > model = QStandardItemModel()
> > model.invisibleRootItem().appendRow(QStandardItem("test"))
> > assert(model.rowCount() == 1)
> > assert(model.columnCount() == 1)
> > mIndex = model.createIndex(0, 0)
> > print(model.data(model.index(0, 0)))
> > assert(model.itemFromIndex(mIndex) != None)
> >
> > The last assert fails with PySide 1.2.1 and Python 3.3.5 on Windows 7.
> > Is that supposed to happen? What would be the proper approach to
> > simply get the first most topleft item in a QStandardItemModel?
> >
> > Regards,
> > Jonas Thiem
> > _______________________________________________
> > PySide mailing list
> > PySide at qt-project.org
> > http://lists.qt-project.org/mailman/listinfo/pyside
> _______________________________________________
> PySide mailing list
> PySide at qt-project.org
> http://lists.qt-project.org/mailman/listinfo/pyside
>



-- 
-tony
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.qt-project.org/pipermail/pyside/attachments/20140417/7586ff6d/attachment.html>

From schampailler at skynet.be  Thu Apr 17 15:34:06 2014
From: schampailler at skynet.be (S. Champailler)
Date: Thu, 17 Apr 2014 15:34:06 +0200 (CEST)
Subject: [PySide] Trouble understanding QStandardItemModel
In-Reply-To: <CAJhmvsQr=CWhHPe4ZP8g=4MJmmXVy7_fz7w5hEw0OWyBHekcrA@mail.gmail.com>
References: <CAD7Z2ho148E0W5a6SBxki5Az60JqaKhnmnS36h+3pXqnXLGKcw@mail.gmail.com>
	<1644806592.93345.1397715655832.open-xchange@webmail.nmp.skynet.be>
	<CAJhmvsQr=CWhHPe4ZP8g=4MJmmXVy7_fz7w5hEw0OWyBHekcrA@mail.gmail.com>
Message-ID: <1348659476.104119.1397741646294.open-xchange@webmail.nmp.skynet.be>

Indeed, according to the documentation, createIndex seems to be there for
buidling one's own models. The regular operation is model.index(r,c)

stF

> Le 17 avril 2014 à 15:16, Tony Barbieri <greatrgb at gmail.com> a écrit :
>
>
> You could try model.itemFromIndex(model.index(0, 0)), that should return
> the item.  You also don't need to use the invisibleRootItem to append a
> row, you should be able to just call model.appendRow(QStandardItem("test")).
>
>
> On Thu, Apr 17, 2014 at 2:20 AM, S. Champailler <schampailler at skynet.be>wrote:
>
> > Tried your code, same error.
> >
> > I'm just wondering if model.data(...) default role is not preventing
> > something
> > to work in QStandardItemModel....
> >
> > Stefan
> >
> > > Le 16 avril 2014 à 23:08, Jonas Thiem <jonasthiem at googlemail.com> a
> > écrit :
> > >
> > >
> > > Hi,
> > >
> > > Trying to understand QStandardItemModel I came up with this:
> > >
> > > from PySide.QtCore import *
> > > from PySide.QtGui import *
> > >
> > > model = QStandardItemModel()
> > > model.invisibleRootItem().appendRow(QStandardItem("test"))
> > > assert(model.rowCount() == 1)
> > > assert(model.columnCount() == 1)
> > > mIndex = model.createIndex(0, 0)
> > > print(model.data(model.index(0, 0)))
> > > assert(model.itemFromIndex(mIndex) != None)
> > >
> > > The last assert fails with PySide 1.2.1 and Python 3.3.5 on Windows 7.
> > > Is that supposed to happen? What would be the proper approach to
> > > simply get the first most topleft item in a QStandardItemModel?
> > >
> > > Regards,
> > > Jonas Thiem
> > > _______________________________________________
> > > PySide mailing list
> > > PySide at qt-project.org
> > > http://lists.qt-project.org/mailman/listinfo/pyside
> > _______________________________________________
> > PySide mailing list
> > PySide at qt-project.org
> > http://lists.qt-project.org/mailman/listinfo/pyside
> >
>
>
>
> --
> -tony


From jpe at wingware.com  Thu Apr 17 17:22:37 2014
From: jpe at wingware.com (John Ehresman)
Date: Thu, 17 Apr 2014 11:22:37 -0400
Subject: [PySide] Problems building pyside 1.2.1 on AIX (shiboken
	problem)
In-Reply-To: <20140416222116.GA5604@thewrittenword.com>
References: <20140416222116.GA5604@thewrittenword.com>
Message-ID: <534FF1BD.8040609@wingware.com>

On 4/16/14, 6:21 PM, Albert Chin wrote:
> I need to get shiboken to provide the same behavior on Linux as on
> AIX. While debugging this, it seems that the alias for "uint" doesn't
> map to "unsigned int" on AIX. In ApiExtractor/typesystem.cpp from
> shiboken:
>    bool TypeEntry::isCppPrimitive() const
>    {
>        if (!isPrimitive())
>            return false;
>
>        PrimitiveTypeEntry* aliasedType = ((PrimitiveTypeEntry*)this)->basicAliasedTypeEntry();
>        QByteArray typeName = (aliasedType ? aliasedType->name() : m_name).toAscii();
>
> On AIX, aliasedType is NULL while on Solaris, HP-UX, and Linux, it is
> not. I presume this means there is on alias from "uint" to "unsigned
> int". Where in the code should I be looking further?

I suspect this means the "typedef unsigned int uint" isn't being seen 
when shiboken scans the header files during generation on AIX while it 
is seen on the other platforms.  I don't think the uint type is special 
cased inside of shiboken.

Cheers,

John



From pyside at mlists.thewrittenword.com  Thu Apr 17 20:58:27 2014
From: pyside at mlists.thewrittenword.com (Albert Chin)
Date: Thu, 17 Apr 2014 13:58:27 -0500
Subject: [PySide] Problems building pyside 1.2.1 on AIX (shiboken
	problem)
In-Reply-To: <534FF1BD.8040609@wingware.com>
References: <20140416222116.GA5604@thewrittenword.com>
	<534FF1BD.8040609@wingware.com>
Message-ID: <20140417185827.GA16984@thewrittenword.com>

On Thu, Apr 17, 2014 at 11:22:37AM -0400, John Ehresman wrote:
> On 4/16/14, 6:21 PM, Albert Chin wrote:
> > I need to get shiboken to provide the same behavior on Linux as on
> > AIX. While debugging this, it seems that the alias for "uint" doesn't
> > map to "unsigned int" on AIX. In ApiExtractor/typesystem.cpp from
> > shiboken:
> >    bool TypeEntry::isCppPrimitive() const
> >    {
> >        if (!isPrimitive())
> >            return false;
> >
> >        PrimitiveTypeEntry* aliasedType = ((PrimitiveTypeEntry*)this)->basicAliasedTypeEntry();
> >        QByteArray typeName = (aliasedType ? aliasedType->name() : m_name).toAscii();
> >
> > On AIX, aliasedType is NULL while on Solaris, HP-UX, and Linux, it
> > is not. I presume this means there is on alias from "uint" to
> > "unsigned int". Where in the code should I be looking further?
> 
> I suspect this means the "typedef unsigned int uint" isn't being
> seen when shiboken scans the header files during generation on AIX
> while it is seen on the other platforms.  I don't think the uint
> type is special cased inside of shiboken.

Correct. The problem is that on AIX we have:
  typedef unsigned int uint_t;
  typedef uint_t uint;

How do we handle this?

-- 
albert chin (china at thewrittenword.com)


From jpe at wingware.com  Thu Apr 17 21:07:00 2014
From: jpe at wingware.com (John Ehresman)
Date: Thu, 17 Apr 2014 15:07:00 -0400
Subject: [PySide] Problems building pyside 1.2.1 on AIX (shiboken
	problem)
In-Reply-To: <20140417185827.GA16984@thewrittenword.com>
References: <20140416222116.GA5604@thewrittenword.com>	<534FF1BD.8040609@wingware.com>
	<20140417185827.GA16984@thewrittenword.com>
Message-ID: <53502654.1090202@wingware.com>

On 4/17/14, 2:58 PM, Albert Chin wrote:
>> I suspect this means the "typedef unsigned int uint" isn't being
>> seen when shiboken scans the header files during generation on AIX
>> while it is seen on the other platforms.  I don't think the uint
>> type is special cased inside of shiboken.
>
> Correct. The problem is that on AIX we have:
>    typedef unsigned int uint_t;
>    typedef uint_t uint;

The question is why you don't get an alias from uint to uint_t and then 
another one from uint_t to unsigned int.  I don't know the answer.  You 
might try putting together a minimal test case and then seeing what the 
difference is in it.

Cheers,

John


From pyside at mlists.thewrittenword.com  Thu Apr 17 21:33:15 2014
From: pyside at mlists.thewrittenword.com (Albert Chin)
Date: Thu, 17 Apr 2014 14:33:15 -0500
Subject: [PySide] Problems building pyside 1.2.1 on AIX (shiboken
	problem)
In-Reply-To: <53502654.1090202@wingware.com>
References: <20140416222116.GA5604@thewrittenword.com>
	<534FF1BD.8040609@wingware.com>
	<20140417185827.GA16984@thewrittenword.com>
	<53502654.1090202@wingware.com>
Message-ID: <20140417193315.GB16984@thewrittenword.com>

On Thu, Apr 17, 2014 at 03:07:00PM -0400, John Ehresman wrote:
> On 4/17/14, 2:58 PM, Albert Chin wrote:
> >> I suspect this means the "typedef unsigned int uint" isn't being
> >> seen when shiboken scans the header files during generation on AIX
> >> while it is seen on the other platforms.  I don't think the uint
> >> type is special cased inside of shiboken.
> >
> > Correct. The problem is that on AIX we have:
> >    typedef unsigned int uint_t;
> >    typedef uint_t uint;
> 
> The question is why you don't get an alias from uint to uint_t and then 
> another one from uint_t to unsigned int.  I don't know the answer.  You 
> might try putting together a minimal test case and then seeing what the 
> difference is in it.

We do get an alias from uint to uint_t. But, oddly, not from uint_t to
unsigned int. At least, that's what I've gathered from my debugging so
far.

-- 
albert chin (china at thewrittenword.com)


From pyside at mlists.thewrittenword.com  Fri Apr 18 05:29:45 2014
From: pyside at mlists.thewrittenword.com (Albert Chin)
Date: Thu, 17 Apr 2014 22:29:45 -0500
Subject: [PySide] Problems building pyside 1.2.1 on AIX (shiboken
	problem)
In-Reply-To: <20140417193315.GB16984@thewrittenword.com>
References: <20140416222116.GA5604@thewrittenword.com>
	<534FF1BD.8040609@wingware.com>
	<20140417185827.GA16984@thewrittenword.com>
	<53502654.1090202@wingware.com>
	<20140417193315.GB16984@thewrittenword.com>
Message-ID: <20140418032945.GA14428@thewrittenword.com>

On Thu, Apr 17, 2014 at 02:33:15PM -0500, Albert Chin wrote:
> On Thu, Apr 17, 2014 at 03:07:00PM -0400, John Ehresman wrote:
> > On 4/17/14, 2:58 PM, Albert Chin wrote:
> > >> I suspect this means the "typedef unsigned int uint" isn't being
> > >> seen when shiboken scans the header files during generation on AIX
> > >> while it is seen on the other platforms.  I don't think the uint
> > >> type is special cased inside of shiboken.
> > >
> > > Correct. The problem is that on AIX we have:
> > >    typedef unsigned int uint_t;
> > >    typedef uint_t uint;
> > 
> > The question is why you don't get an alias from uint to uint_t and then 
> > another one from uint_t to unsigned int.  I don't know the answer.  You 
> > might try putting together a minimal test case and then seeing what the 
> > difference is in it.
> 
> We do get an alias from uint to uint_t. But, oddly, not from uint_t to
> unsigned int. At least, that's what I've gathered from my debugging so
> far.

Testing the attached patch. Seems shiboken was ok all along.

-- 
albert chin (china at thewrittenword.com)
-------------- next part --------------
A non-text attachment was scrubbed...
Name: pyside.patch
Type: text/x-diff
Size: 557 bytes
Desc: not available
URL: <http://lists.qt-project.org/pipermail/pyside/attachments/20140417/d9af7838/attachment.patch>

From pyside at mlists.thewrittenword.com  Fri Apr 18 13:51:19 2014
From: pyside at mlists.thewrittenword.com (Albert Chin)
Date: Fri, 18 Apr 2014 06:51:19 -0500
Subject: [PySide] Problems building pyside 1.2.1 on AIX (shiboken
	problem)
In-Reply-To: <20140418032945.GA14428@thewrittenword.com>
References: <20140416222116.GA5604@thewrittenword.com>
	<534FF1BD.8040609@wingware.com>
	<20140417185827.GA16984@thewrittenword.com>
	<53502654.1090202@wingware.com>
	<20140417193315.GB16984@thewrittenword.com>
	<20140418032945.GA14428@thewrittenword.com>
Message-ID: <20140418115119.GC14428@thewrittenword.com>

On Thu, Apr 17, 2014 at 10:29:45PM -0500, Albert Chin wrote:
> On Thu, Apr 17, 2014 at 02:33:15PM -0500, Albert Chin wrote:
> > On Thu, Apr 17, 2014 at 03:07:00PM -0400, John Ehresman wrote:
> > > On 4/17/14, 2:58 PM, Albert Chin wrote:
> > > >> I suspect this means the "typedef unsigned int uint" isn't being
> > > >> seen when shiboken scans the header files during generation on AIX
> > > >> while it is seen on the other platforms.  I don't think the uint
> > > >> type is special cased inside of shiboken.
> > > >
> > > > Correct. The problem is that on AIX we have:
> > > >    typedef unsigned int uint_t;
> > > >    typedef uint_t uint;
> > > 
> > > The question is why you don't get an alias from uint to uint_t and then 
> > > another one from uint_t to unsigned int.  I don't know the answer.  You 
> > > might try putting together a minimal test case and then seeing what the 
> > > difference is in it.
> > 
> > We do get an alias from uint to uint_t. But, oddly, not from uint_t to
> > unsigned int. At least, that's what I've gathered from my debugging so
> > far.
> 
> Testing the attached patch. Seems shiboken was ok all along.

This patch fixed the issue for me. Can someone commit?

-- 
albert chin (china at thewrittenword.com)


From 2281570025 at qq.com  Fri Apr 18 17:50:05 2014
From: 2281570025 at qq.com (=?ISO-8859-1?B?aU1hdGg=?=)
Date: Fri, 18 Apr 2014 23:50:05 +0800
Subject: [PySide] make the cursor on the screen invisible after it has
	stopped
Message-ID: <tencent_6147CB21139232CC2EEF4774@qq.com>

first, sorry for my poor english !


I want to make the cursor on the screen become invisible after it has stopped moving for 10s. I think this problem can be solved easily if there is a signal like positionChanged(QPoint lastPos , QPoint currentPos) existed,lastPos means the last position of the cursor (hot spot) of the primary screen in global screen coordinates, currentPos means the current  position, this signal should be emitted once the cursor stopped moving .Note the Mouse move events here should occur even when a mouse button is not pressed down, i.e. there is mouse tracking with cursor.In Qt, mouse tracking could be enabled with QWidget::setMouseTracking(), however my problem is not restricted to Qt, it is system wide, I want to do this on Windows now ,anyone knows how to enable mouse tracking here ?
any other solution is also welcome !
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.qt-project.org/pipermail/pyside/attachments/20140418/ad59b107/attachment.html>

From mw_triad at users.sourceforge.net  Fri Apr 18 18:16:31 2014
From: mw_triad at users.sourceforge.net (Matthew Woehlke)
Date: Fri, 18 Apr 2014 12:16:31 -0400
Subject: [PySide] make the cursor on the screen invisible after it has
	stopped
In-Reply-To: <tencent_6147CB21139232CC2EEF4774__13335.3569465273$1397836259$gmane$org@qq.com>
References: <tencent_6147CB21139232CC2EEF4774__13335.3569465273$1397836259$gmane$org@qq.com>
Message-ID: <lirj50$aa2$1@ger.gmane.org>

On 2014-04-18 11:50, iMath wrote:
> I want to make the cursor on the screen become invisible after it has
> stopped moving for 10s. [...] my problem is not restricted to Qt, it
> is system wide

Probably you can't do this portably, and even less likely in Python. 
However, I am suspicious of your *wanting* to do this.

If you want it to happen for *your application* (which would be much 
less Evil), you can possibly turn on mouse tracking for your top level 
window(s) and use QApplication::setOverrideCursor to hide the cursor.

I'd do something like, on every mouse move, set a timer at e.g. 1s 
interval if not set already and reset a counter... then in timer, 
increment the counter to see if your 10s interval is elapsed, and if so, 
kill the timer so that you aren't wasting CPU when the machine is idle.

-- 
Matthew



From backup.rlacko at gmail.com  Sun Apr 20 00:05:26 2014
From: backup.rlacko at gmail.com (Roman Lacko)
Date: Sun, 20 Apr 2014 00:05:26 +0200
Subject: [PySide] Fixing examples qt resource files under Python 3
Message-ID: <CAHC6v6gnhJLAZki7Jy2Qp08_pKES0BKpbh0t+NJePVLDdiwEUA@mail.gmail.com>

Hi,

Here is simple helper script to regenarate recursively all existing qt
resource files in examples folder distributed in official pyside packages
built with pyside-setup.

It will be included in next release but it could be usefull if someone has
problems running examples under Python 3, where by default the qt resource
files are generated for Python 2 only.

Just run the follwing script with Python where the PySide is installed:

$ python regenerate_examples_qt_resources.py

Regards
-Roman

====== regenerate_examples_qt_resources.py ======
import sys
import os
import subprocess
from PySide._utils import get_pyside_dir


PY_2 = sys.version_info[0] < 3


pyside_dir = get_pyside_dir()
pyside_rcc_path = os.path.join(pyside_dir, 'pyside-rcc')
pyside_rcc_options = '-py2' if PY_2 else '-py3'


def regenerate_qt_resources(src):
    names = os.listdir(src)
    for name in names:
        srcname = os.path.join(src, name)
        if os.path.isdir(srcname):
            regenerate_qt_resources(srcname)
        elif srcname.endswith('.qrc'):
            # Replace last occurence of '.qrc' in srcname
            srcname_split = srcname.rsplit('.qrc', 1)
            dstname = '_rc.py'.join(srcname_split)
            if os.path.exists(dstname):
                print('Regenerating %s from %s' % \
                    (dstname, os.path.basename(srcname)))
                subprocess.call([pyside_rcc_path,
                                 pyside_rcc_options,
                                 srcname, '-o', dstname])


if __name__ == '__main__':
    examples_dir = os.path.join(pyside_dir, 'examples')
    if os.path.exists(examples_dir):
        regenerate_qt_resources(examples_dir)
====== /regenerate_examples_qt_resources.py ======
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.qt-project.org/pipermail/pyside/attachments/20140420/8d42e190/attachment.html>

From 2281570025 at qq.com  Sun Apr 20 07:21:53 2014
From: 2281570025 at qq.com (=?ISO-8859-1?B?aU1hdGg=?=)
Date: Sun, 20 Apr 2014 13:21:53 +0800
Subject: [PySide] hide system cursor in system wide
Message-ID: <tencent_2D2637302317EE84040F4E3F@qq.com>

I want to hide system cursor for 10s for some reason ,but I found 


cursor.setShape(Qt.BlankCursor)


can only hide mouse cursor that is associated with QWidgets ,not in system wide ,i.e. when mouse cursor is hovering on QWidgets, it is invisible ,otherwise it is visible ,so is there any way to hide system cursor in system wide?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.qt-project.org/pipermail/pyside/attachments/20140420/8e99aa9d/attachment.html>

From thoromyr at mac.com  Sun Apr 20 13:11:06 2014
From: thoromyr at mac.com (Tim Doty)
Date: Sun, 20 Apr 2014 06:11:06 -0500
Subject: [PySide] hide system cursor in system wide
In-Reply-To: <tencent_2D2637302317EE84040F4E3F@qq.com>
References: <tencent_2D2637302317EE84040F4E3F@qq.com>
Message-ID: <98804565-83A9-4FF4-AC39-274485FED631@mac.com>

What you want cannot be done in pyside, it requires programming for the windowing environment you are using, or configuring that windowing environment. It isn’t something I have cared about for years, but this is a common feature of windowing systems (Qt is a toolkit, not a windowing system).

In short, there is no answer for this question that resides in pyside.

TIm Doty

On Apr 20, 2014, at 12:21 AM, iMath <2281570025 at qq.com> wrote:

> 
> 
> I want to hide system cursor for 10s for some reason ,but I found 
> 
> cursor.setShape(Qt.BlankCursor)
> 
> can only hide mouse cursor that is associated with QWidgets ,not in system wide ,i.e. when mouse cursor is hovering on QWidgets, it is invisible ,otherwise it is visible ,so is there any way to hide system cursor in system wide?
> _______________________________________________
> PySide mailing list
> PySide at qt-project.org
> http://lists.qt-project.org/mailman/listinfo/pyside



From jonasthiem at googlemail.com  Mon Apr 21 12:15:17 2014
From: jonasthiem at googlemail.com (Jonas Thiem)
Date: Mon, 21 Apr 2014 12:15:17 +0200
Subject: [PySide] Icons with Windows/PySide
Message-ID: <5354EFB5.5050302@googlemail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi,

It appears the windows install of PySide 1.2.1 for Python 3.3 has no
working icons, like QIcon.fromTheme("document-new").

Is there a way to fix that?

Regards,
Jonas Thiem
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1
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=oeVT
-----END PGP SIGNATURE-----


From igor.mironchik at gmail.com  Sun Apr 20 08:57:42 2014
From: igor.mironchik at gmail.com (igor.mironchik at gmail.com)
Date: Sun, 20 Apr 2014 08:57:42 +0200
Subject: [PySide] [Interest] hide system cursor in system wide
In-Reply-To: <tencent_2D2637302317EE84040F4E3F@qq.com>
References: <tencent_2D2637302317EE84040F4E3F@qq.com>
Message-ID: <C71514269D1D4AD0AA5A276C6B4EB3E3@GMI>

I’m not sure, but... Try to use QApplication::setOverrideCursor().

May be it will hide cursor in system wide...

From: iMath 
Sent: Sunday, April 20, 2014 7:21 AM
To: interest ; pyside ; pyqt 
Subject: [Interest] hide system cursor in system wide



I want to hide system cursor for 10s for some reason ,but I found 

cursor.setShape(Qt.BlankCursor)

can only hide mouse cursor that is associated with QWidgets ,not in system wide ,i.e. when mouse cursor is hovering on QWidgets, it is invisible ,otherwise it is visible ,so is there any way to hide system cursor in system wide?


--------------------------------------------------------------------------------
_______________________________________________
Interest mailing list
Interest at qt-project.org
http://lists.qt-project.org/mailman/listinfo/interest
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.qt-project.org/pipermail/pyside/attachments/20140420/b3cd8519/attachment.html>

From redteam316 at gmail.com  Sun Apr 20 09:15:24 2014
From: redteam316 at gmail.com (Jonathan Greig)
Date: Sun, 20 Apr 2014 02:15:24 -0500
Subject: [PySide] [Interest] hide system cursor in system wide
In-Reply-To: <C71514269D1D4AD0AA5A276C6B4EB3E3@GMI>
References: <tencent_2D2637302317EE84040F4E3F@qq.com>
	<C71514269D1D4AD0AA5A276C6B4EB3E3@GMI>
Message-ID: <CABZKpPBQzSO6yAyOx1TMjTOwpTrNOLUHKEG1Lt9KOyxaemTg3A@mail.gmail.com>

No, it won't.
On Apr 20, 2014 12:58 AM, <igor.mironchik at gmail.com> wrote:

>   I’m not sure, but... Try to use QApplication::setOverrideCursor().
>
> May be it will hide cursor in system wide...
>
>  *From:* iMath <2281570025 at qq.com>
> *Sent:* Sunday, April 20, 2014 7:21 AM
> *To:* interest <interest at qt-project.org> ; pyside <pyside at qt-project.org>;
> pyqt <pyqt at riverbankcomputing.com>
> *Subject:* [Interest] hide system cursor in system wide
>
>
>
> I want to hide system cursor for 10s for some reason ,but I found
>
> cursor.setShape(Qt.BlankCursor)
>
> can only hide mouse cursor that is associated with QWidgets ,not in system
> wide ,i.e. when mouse cursor is hovering on QWidgets, it is invisible
> ,otherwise it is visible ,so is there any way to hide system cursor in
> system wide?
>
> ------------------------------
> _______________________________________________
> Interest mailing list
> Interest at qt-project.org
> http://lists.qt-project.org/mailman/listinfo/interest
>
> _______________________________________________
> Interest mailing list
> Interest at qt-project.org
> http://lists.qt-project.org/mailman/listinfo/interest
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.qt-project.org/pipermail/pyside/attachments/20140420/6342b682/attachment.html>

From techtonik at gmail.com  Mon Apr 21 20:32:35 2014
From: techtonik at gmail.com (anatoly techtonik)
Date: Mon, 21 Apr 2014 21:32:35 +0300
Subject: [PySide] ESA Summer of Code - an opportunity for PySide?
In-Reply-To: <29728_1396518573_533D2EAD_29728_2995_1_533D2D32.5060507@esa.int>
References: <29728_1396518573_533D2EAD_29728_2995_1_533D2D32.5060507@esa.int>
Message-ID: <CAPkN8xL4EuUB1PH3bAGY-TJbapKzfUxBYF_Lepck0YCpFR=wGw@mail.gmail.com>

On Thu, Apr 3, 2014 at 12:43 PM, Maxime Perrotin
<maxime.perrotin at esa.int> wrote:
> Dear all,
>
> It may be of interest of the PySide community to submit a project to the
> ESA summer of code (SOCIS):
> http://sophia.estec.esa.int/socis/
>
> The 2014 edition was just launched. In a similar way to the GSoC, SOCIS
> allows students to be paid to participate actively during 3 months to an
> open-source project.
>
> Past editions showed that we get top-level students. If you think that
> it is a good idea to put someone on e.g. Qt5 bindings for PySide, then
> please feel free to submit a proposal.
> Application form for new projects is open until 15th April.
>
> There is a selection process, no guarantee to be selected of course, but
> it is always worth trying.... we can fund up to 23 projects a year -
> last year we received 26 proposals.

Thanks for the proposal. I feel it is a little bit late to jump in. If
you didn't get
the reply that's probably because there are not many people who are located
in Europe and could mentor the efforts.

I'd think about providing some stipends for mentors from former development
team. I think that's possible to arrange under EU Horizon 2020 program. If I
remember it properly, the end of April is a deadline for proposals.


From maxime1008 at gmail.com  Mon Apr 21 20:58:51 2014
From: maxime1008 at gmail.com (Maxime Perrotin)
Date: Mon, 21 Apr 2014 20:58:51 +0200
Subject: [PySide] ESA Summer of Code - an opportunity for PySide?
In-Reply-To: <CAPkN8xL4EuUB1PH3bAGY-TJbapKzfUxBYF_Lepck0YCpFR=wGw@mail.gmail.com>
References: <29728_1396518573_533D2EAD_29728_2995_1_533D2D32.5060507@esa.int>
	<CAPkN8xL4EuUB1PH3bAGY-TJbapKzfUxBYF_Lepck0YCpFR=wGw@mail.gmail.com>
Message-ID: <CAETTRT7Qb8RppzK8tBGOvWG3v8hLS0sf1L5TC51bafB9qVZQuA@mail.gmail.com>

Mentors don't need to be in Europe (they do not get paid). Only students do.
If that is of interest to you, please let me know. The selection process is
still ongoing.




2014-04-21 20:32 GMT+02:00 anatoly techtonik <techtonik at gmail.com>:

> On Thu, Apr 3, 2014 at 12:43 PM, Maxime Perrotin
> <maxime.perrotin at esa.int> wrote:
> > Dear all,
> >
> > It may be of interest of the PySide community to submit a project to the
> > ESA summer of code (SOCIS):
> > http://sophia.estec.esa.int/socis/
> >
> > The 2014 edition was just launched. In a similar way to the GSoC, SOCIS
> > allows students to be paid to participate actively during 3 months to an
> > open-source project.
> >
> > Past editions showed that we get top-level students. If you think that
> > it is a good idea to put someone on e.g. Qt5 bindings for PySide, then
> > please feel free to submit a proposal.
> > Application form for new projects is open until 15th April.
> >
> > There is a selection process, no guarantee to be selected of course, but
> > it is always worth trying.... we can fund up to 23 projects a year -
> > last year we received 26 proposals.
>
> Thanks for the proposal. I feel it is a little bit late to jump in. If
> you didn't get
> the reply that's probably because there are not many people who are located
> in Europe and could mentor the efforts.
>
> I'd think about providing some stipends for mentors from former development
> team. I think that's possible to arrange under EU Horizon 2020 program. If
> I
> remember it properly, the end of April is a deadline for proposals.
> _______________________________________________
> PySide mailing list
> PySide at qt-project.org
> http://lists.qt-project.org/mailman/listinfo/pyside
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.qt-project.org/pipermail/pyside/attachments/20140421/e4395393/attachment.html>

From pelujamo at gmail.com  Tue Apr 22 10:55:34 2014
From: pelujamo at gmail.com (Peter Pan)
Date: Tue, 22 Apr 2014 10:55:34 +0200
Subject: [PySide]  QMetaObject::invokeMethod doesn't find the method
Message-ID: <CABmHngmaKFcUxzj05BN9s72zy_0TJx8Fh=vyzba4sdgc=2QgKQ@mail.gmail.com>

Hi,

I'm trying to understand how QMetaObject::invokeMethod can call a method of
an object with parameters (this object will later run in a different thread
and that will therefore come in handy). I'm failing so far. I also posted a
stackexchange question (
http://stackoverflow.com/questions/23127844/qmetaobjectinvokemethod-doesnt-find-the-method)
but no answers solved the full problem.

What works (calling a method without a parameter):

from PySide import QtCore

class Example(QtCore.QObject):
    def __init__(self):
        super().__init__()


    @QtCore.Slot()
    def beep(self):
        print('beep')

if __name__ == '__main__':
    t = Example()
    QtCore.QMetaObject.invokeMethod(t, 'beep', QtCore.Qt.AutoConnection)


What doesn't work is calling with a parameter. I tried:

from PySide import QtCore

class Example(QtCore.QObject):
    def __init__(self):

        super().__init__()

    @QtCore.Slot(str)
    def beep(self, text):
        print(text)

@QtCore.Slot(str)
def beep(text):
    print(text)

if __name__ == '__main__':
    t = Example()

    QtCore.QMetaObject.invokeMethod(t, 'beep', QtCore.Qt.AutoConnection,
QtCore.QGenericArgument('text', 'hello'))
    # or that but both fails
    QtCore.QMetaObject.invokeMethod(t, 'beep', QtCore.Qt.AutoConnection,
QtCore.QGenericArgument('self', t), QtCore.QGenericArgument('text',
'hello'))

and get the output
QMetaObject::invokeMethod: No such method Example::beep(text)
QMetaObject::invokeMethod: No such method Example::beep(self,text)

So how can I use QMetaObject::invokeMethod to invoke a method of a QObject
derived class with parameters? I use PySide 1.2.1 on Python 3.3 64 Bit on
Windows 7.

Cheers
Jan
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.qt-project.org/pipermail/pyside/attachments/20140422/b200c103/attachment.html>

From jonasthiem at googlemail.com  Tue Apr 22 16:30:49 2014
From: jonasthiem at googlemail.com (Jonas Thiem)
Date: Tue, 22 Apr 2014 16:30:49 +0200
Subject: [PySide] Get icons working on windows
Message-ID: <CAD7Z2hq+xi5RUFfKAGg243Ps6Gdk52njFx==8wQichyhiCQ9iQ@mail.gmail.com>

Hi,

I tried to get icons working on Windows. To do that, I did the following steps:

1.) I created an "icons" subfolder (where the current working
directory of my application is when running)
2.) I built tango with ./configure && make on linux, and copied the
resulting folder into "icons\Tango\" (with the index.theme inside
etc.)

Then I used the following code in my application startup:

        if QIcon.themeName() == "":
            QIcon.setThemeSearchPaths(os.path.abspath("icons"))
            QIcon.setThemeName("Tango")
            assert(QIcon.themeName() == "Tango")
        assert(len(QIcon.themeName()) > 0)

No assert fails, no error happens or anything. But all the stock icons
from my PySide application are still missing. (They work perfectly
fine on Linux/Gnome 3)

Since there is no error or warning or anything, I don't know what to
do from here. Any suggestions?

Regards,
Jonas Thiem


From jonasthiem at googlemail.com  Tue Apr 22 18:17:31 2014
From: jonasthiem at googlemail.com (Jonas Thiem)
Date: Tue, 22 Apr 2014 18:17:31 +0200
Subject: [PySide] Get icons working on windows
In-Reply-To: <CAD7Z2hq+xi5RUFfKAGg243Ps6Gdk52njFx==8wQichyhiCQ9iQ@mail.gmail.com>
References: <CAD7Z2hq+xi5RUFfKAGg243Ps6Gdk52njFx==8wQichyhiCQ9iQ@mail.gmail.com>
Message-ID: <CAD7Z2hrDd1EC0OYoTGM5Y8R17jtn5Vai=NQ960LgJj4EWXksHQ@mail.gmail.com>

Yea my call to QIcon.setThemeSearchPaths() was just utterly wrong.
(Argument should be list and not a string, and also it may be a good
idea to retain the old paths I guess)

With that fixed it works :-)

Sorry for bothering!

Regards,
Jonas Thiem


On Tue, Apr 22, 2014 at 4:30 PM, Jonas Thiem <jonasthiem at googlemail.com> wrote:
> Hi,
>
> I tried to get icons working on Windows. To do that, I did the following steps:
>
> 1.) I created an "icons" subfolder (where the current working
> directory of my application is when running)
> 2.) I built tango with ./configure && make on linux, and copied the
> resulting folder into "icons\Tango\" (with the index.theme inside
> etc.)
>
> Then I used the following code in my application startup:
>
>         if QIcon.themeName() == "":
>             QIcon.setThemeSearchPaths(os.path.abspath("icons"))
>             QIcon.setThemeName("Tango")
>             assert(QIcon.themeName() == "Tango")
>         assert(len(QIcon.themeName()) > 0)
>
> No assert fails, no error happens or anything. But all the stock icons
> from my PySide application are still missing. (They work perfectly
> fine on Linux/Gnome 3)
>
> Since there is no error or warning or anything, I don't know what to
> do from here. Any suggestions?
>
> Regards,
> Jonas Thiem


From tismer at stackless.com  Fri Apr 25 03:21:20 2014
From: tismer at stackless.com (Christian Tismer)
Date: Fri, 25 Apr 2014 03:21:20 +0200
Subject: [PySide] Where is PySide 1.2.2 ???
Message-ID: <5359B890.7080706@stackless.com>

Hey guys,

this is an unfriendly email by intent. My patience is over.
This is not my regular style, be assured.


I am annoyed, and I am yelling out loud!
=============================

There is now, after way too many months, PySide 1.2.2 on Gerrit,
actually since 2014-04-22. Way, way too late.

Yes, there have been refinements to the demos, but that is not
the point: We have a working solution in the repos, and we did
not make sure to deliver it just in time!

This is a situation that I cannot understand and do not support:


What do we actually want to achieve?
==========================

I want people to use PySide, easily, without any worries, and without
having to wait eons to build PySide at home.
No idea what others want, but this bottleneck must definitely go away.
Now.

Mac OS/X Mavericks is waiting for its trivial patch since
October 22, 2013 since it was released.

We now have 2014-04-25, six months later, and still, "pip install pyside"
crashes on OS/X Mavericks!!

How comes? Who does not care? Why is PySide 1.2.2 not already here?
Who is waiting for what, exactly???


Procrastination does not help anybody!
===============================

I am sorry but whatever you say about it, I am annoyed and upset.
We have to deliver! If PySide should make any sense, we should make
sure that we always deliver in time after we have solved a problem.

If we don't deliver when we could, then we need to change our structure.

Yes, people are just volunteers. Me too. But that is no excuse.
Somebody has to step up and be the driving person for a release.


Stepping up
===========

And if a release does not happen until this Sunday (2014-04-27),
I will do a PySide 1.2.2 build for Mavericks and submit it under a
similar but
different name on PyPI. No matter what the consequences are, because
I am upset and annoyed, and you don't want me like this, for sure.


Please avoid this and deliver, right now!
==============================

Cheers - Chris

-- 
Christian Tismer             :^)   tismer at stackless.com
Software Consulting          :     http://www.stackless.com/
Karl-Liebknecht-Str. 121     :     http://www.pydica.net/
14482 Potsdam                :     GPG key -> 0xFB7BEE0E
phone +49 173 24 18 776  fax +49 (30) 700143-0023



From sdeibel at wingware.com  Fri Apr 25 03:43:09 2014
From: sdeibel at wingware.com (Stephan Deibel)
Date: Thu, 24 Apr 2014 21:43:09 -0400
Subject: [PySide] Where is PySide 1.2.2 ???
In-Reply-To: <5359B890.7080706@stackless.com>
References: <5359B890.7080706@stackless.com>
Message-ID: <5359BDAD.1030707@wingware.com>

Christian Tismer wrote:
> Yes, people are just volunteers. Me too. But that is no excuse.
> Somebody has to step up and be the driving person for a release.

I think in this context it's futile to get angry.  The only possibly 
effective approach is to take action yourself, in some form, to make 
releases happen more often and more smoothly.  The project certainly 
needs more volunteers that have the time and resources to put 
significant work into PySide.  I know finding time to work on PySide has 
been difficult for John, and I doubt this is any different for anyone 
else here.

At least there is now a 1.2.2 tag in the repository, so anyone with the 
time and resources could get into building binary distributions from that.

- Stephan



From tismer at stackless.com  Fri Apr 25 04:02:52 2014
From: tismer at stackless.com (Christian Tismer)
Date: Fri, 25 Apr 2014 04:02:52 +0200
Subject: [PySide] Where is PySide 1.2.2 ???
In-Reply-To: <5359BDAD.1030707@wingware.com>
References: <5359B890.7080706@stackless.com> <5359BDAD.1030707@wingware.com>
Message-ID: <C64FF36B-52E3-4E60-9555-155B9EE4B7CB@stackless.com>

Yes, probably you are right. 
I am of course volunteering to push this forward. 
It was just not clear why that patch needed six months. 

Of course, this was related to my chat with Stefan, today, and my
Starting over on PySide 5. 

Need to be crude, sometimes, not sorry. It helps me to do it. 

All the best - Chris


Von meinem iPad gesendet

> Am 25.04.2014 um 03:43 schrieb Stephan Deibel <sdeibel at wingware.com>:
> 
> Christian Tismer wrote:
>> Yes, people are just volunteers. Me too. But that is no excuse.
>> Somebody has to step up and be the driving person for a release.
> 
> I think in this context it's futile to get angry.  The only possibly effective approach is to take action yourself, in some form, to make releases happen more often and more smoothly.  The project certainly needs more volunteers that have the time and resources to put significant work into PySide.  I know finding time to work on PySide has been difficult for John, and I doubt this is any different for anyone else here.
> 
> At least there is now a 1.2.2 tag in the repository, so anyone with the time and resources could get into building binary distributions from that.
> 
> - Stephan
> 


From jpe at wingware.com  Fri Apr 25 16:21:38 2014
From: jpe at wingware.com (John Ehresman)
Date: Fri, 25 Apr 2014 10:21:38 -0400
Subject: [PySide] Where is PySide 1.2.2 ???
In-Reply-To: <5359B890.7080706@stackless.com>
References: <5359B890.7080706@stackless.com>
Message-ID: <535A6F72.2000404@wingware.com>

On 4/24/14, 9:21 PM, Christian Tismer wrote:
> There is now, after way too many months, PySide 1.2.2 on Gerrit,
> actually since 2014-04-22.

The release has been made and will be announced once all of the files 
have been uploaded.  Discussion of the release has taken place on the 
pyside-dev at googlegroups.com list.

Cheers,

John


From ochiu at teslamotors.com  Sat Apr 26 00:10:07 2014
From: ochiu at teslamotors.com (Otto Chiu)
Date: Fri, 25 Apr 2014 22:10:07 +0000
Subject: [PySide] Wrapping C anonymous enums
Message-ID: <459D4CF76F76B54697869793F5BEF3592D655D9B@SJC04-DAG01-N02.teslamotors.com>

Hi,
I am trying to create a binding and want to expose some anonymous C enums. I am having trouble with the enum-type. I tried to typedef the enum to a name and specify <enum-type name="typedefname" /> and I have also tried <enum-type identified-by-value="a_enum"/>. Both of them result in "enum 'abc' is specified in typesystem, but not declared", where abc is either the typedef name or the enum value. I have tried surrounding the #includes with extern "C" and without. What else am I missing?

Thanks,
Otto
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.qt-project.org/pipermail/pyside/attachments/20140425/fd63b7ae/attachment.html>

From tismer at stackless.com  Mon Apr 28 22:18:27 2014
From: tismer at stackless.com (Christian Tismer)
Date: Mon, 28 Apr 2014 22:18:27 +0200
Subject: [PySide] PySide 1.2.2 OS/X wheels
Message-ID: <535EB793.6040305@stackless.com>

Hi,

Wheels and eggs for python 2.6 2.7 3.3 3.4
compiled for OS/X 10.6 10.7 10.8 10.9 are built and uploaded.

Until the uploads appear on the official download site,
you can use this command to download and install pyside:

$ pip install -U PySide --use-wheel -f
https://bitbucket.org/pydica/pyside-setup/downloads

$ python (somepath)/bin/pyside_postinstall.py -install

There are two little problems in this build.

Buglet 1:
"pyside_postinstall.py" is in the bin folder, but does not work
because OSX 10.9 needs the shebang line which is missing.

Buglet 2:
There is one single combination missing:
- python 3.4 on OSX 10.7 crashes during its build.
 This needs to be fixed at python.org.

cheers - Chris

-- 
Christian Tismer             :^)   <mailto:tismer at stackless.com>
Software Consulting          :     Have a break! Take a ride on Python's
Karl-Liebknecht-Str. 121     :    *Starship* http://starship.python.net/
14482 Potsdam                :     PGP key -> http://pgp.uni-mainz.de
phone +49 173 24 18 776  fax +49 (30) 700143-0023
PGP 0x57F3BF04       9064 F4E1 D754 C2FF 1619  305B C09C 5A3B 57F3 BF04
      whom do you want to sponsor today?   http://www.stackless.com/


From matthew.brett at gmail.com  Mon Apr 28 22:22:00 2014
From: matthew.brett at gmail.com (Matthew Brett)
Date: Mon, 28 Apr 2014 13:22:00 -0700
Subject: [PySide] PySide 1.2.2 OS/X wheels
In-Reply-To: <535EB793.6040305@stackless.com>
References: <535EB793.6040305@stackless.com>
Message-ID: <CAH6Pt5rHQ1knk2c4naoNOPRUnncGbs=OLQF0fuPDWe-SxgeW7w@mail.gmail.com>

Hi,

On Mon, Apr 28, 2014 at 1:18 PM, Christian Tismer <tismer at stackless.com> wrote:
> Hi,
>
> Wheels and eggs for python 2.6 2.7 3.3 3.4
> compiled for OS/X 10.6 10.7 10.8 10.9 are built and uploaded.
>
> Until the uploads appear on the official download site,
> you can use this command to download and install pyside:
>
> $ pip install -U PySide --use-wheel -f
> https://bitbucket.org/pydica/pyside-setup/downloads
>
> $ python (somepath)/bin/pyside_postinstall.py -install
>
> There are two little problems in this build.
>
> Buglet 1:
> "pyside_postinstall.py" is in the bin folder, but does not work
> because OSX 10.9 needs the shebang line which is missing.
>
> Buglet 2:
> There is one single combination missing:
> - python 3.4 on OSX 10.7 crashes during its build.
>  This needs to be fixed at python.org.

Thank you so much for your work on this - it will make a huge
difference to using Qt on OSX.

Cheers,

Matthew


From tismer at stackless.com  Mon Apr 28 23:13:31 2014
From: tismer at stackless.com (Christian Tismer)
Date: Mon, 28 Apr 2014 23:13:31 +0200
Subject: [PySide] PySide 1.2.2 OS/X wheels
In-Reply-To: <CAH6Pt5rHQ1knk2c4naoNOPRUnncGbs=OLQF0fuPDWe-SxgeW7w@mail.gmail.com>
References: <535EB793.6040305@stackless.com>
	<CAH6Pt5rHQ1knk2c4naoNOPRUnncGbs=OLQF0fuPDWe-SxgeW7w@mail.gmail.com>
Message-ID: <535EC47B.3070903@stackless.com>

Hi Brett,

On 28/04/14 22:22, Matthew Brett wrote:
> Hi,
> 
> On Mon, Apr 28, 2014 at 1:18 PM, Christian Tismer <tismer at stackless.com> wrote:
>> Hi,
>>
>> Wheels and eggs for python 2.6 2.7 3.3 3.4
>> compiled for OS/X 10.6 10.7 10.8 10.9 are built and uploaded.
>>
>> Until the uploads appear on the official download site,
>> you can use this command to download and install pyside:
>>
>> $ pip install -U PySide --use-wheel -f
>> https://bitbucket.org/pydica/pyside-setup/downloads
>>
>> $ python (somepath)/bin/pyside_postinstall.py -install
>>
>> There are two little problems in this build.
>>
>> Buglet 1:
>> "pyside_postinstall.py" is in the bin folder, but does not work
>> because OSX 10.9 needs the shebang line which is missing.
>>
>> Buglet 2:
>> There is one single combination missing:
>> - python 3.4 on OSX 10.7 crashes during its build.
>>  This needs to be fixed at python.org.
> 
> Thank you so much for your work on this - it will make a huge
> difference to using Qt on OSX.

Thank you! Yes, that was what I always wanted to have for OSX.

Actually, I build things quite often in the last days, which was
a boring, manual task, and error-prone to get right.

So on Saturday I started hacking a script together that builds
everything from a single command line, builds all the 16 pythons
and then builds all the wheels and eggs in parallel batches.
This uses the original pyside_build as a sub-process.

Just the uploading to Bitbucket is still manual, until I find a way
to script that as well.
Maybe I should simply put that on a different site that I directly
control.

I will upload that script somewhere, soon.
It was interesting to make a bootstrap from nothing, that manages to
install everything alone for all python versions.

-- 
Christian Tismer             :^)   <mailto:tismer at stackless.com>
Software Consulting          :     Have a break! Take a ride on Python's
Karl-Liebknecht-Str. 121     :    *Starship* http://starship.python.net/
14482 Potsdam                :     PGP key -> http://pgp.uni-mainz.de
phone +49 173 24 18 776  fax +49 (30) 700143-0023
PGP 0x57F3BF04       9064 F4E1 D754 C2FF 1619  305B C09C 5A3B 57F3 BF04
      whom do you want to sponsor today?   http://www.stackless.com/


From matthew.brett at gmail.com  Mon Apr 28 23:23:52 2014
From: matthew.brett at gmail.com (Matthew Brett)
Date: Mon, 28 Apr 2014 14:23:52 -0700
Subject: [PySide] PySide 1.2.2 OS/X wheels
In-Reply-To: <535EC47B.3070903@stackless.com>
References: <535EB793.6040305@stackless.com>
	<CAH6Pt5rHQ1knk2c4naoNOPRUnncGbs=OLQF0fuPDWe-SxgeW7w@mail.gmail.com>
	<535EC47B.3070903@stackless.com>
Message-ID: <CAH6Pt5p49A_cOPOm08iCLjW2E=0i6Hj+8+-QW_ntKGnnsuTs9g@mail.gmail.com>

Hi,

On Mon, Apr 28, 2014 at 2:13 PM, Christian Tismer <tismer at stackless.com> wrote:
> Hi Brett,
>
> On 28/04/14 22:22, Matthew Brett wrote:
>> Hi,
>>
>> On Mon, Apr 28, 2014 at 1:18 PM, Christian Tismer <tismer at stackless.com> wrote:
>>> Hi,
>>>
>>> Wheels and eggs for python 2.6 2.7 3.3 3.4
>>> compiled for OS/X 10.6 10.7 10.8 10.9 are built and uploaded.
>>>
>>> Until the uploads appear on the official download site,
>>> you can use this command to download and install pyside:
>>>
>>> $ pip install -U PySide --use-wheel -f
>>> https://bitbucket.org/pydica/pyside-setup/downloads
>>>
>>> $ python (somepath)/bin/pyside_postinstall.py -install
>>>
>>> There are two little problems in this build.
>>>
>>> Buglet 1:
>>> "pyside_postinstall.py" is in the bin folder, but does not work
>>> because OSX 10.9 needs the shebang line which is missing.
>>>
>>> Buglet 2:
>>> There is one single combination missing:
>>> - python 3.4 on OSX 10.7 crashes during its build.
>>>  This needs to be fixed at python.org.
>>
>> Thank you so much for your work on this - it will make a huge
>> difference to using Qt on OSX.
>
> Thank you! Yes, that was what I always wanted to have for OSX.
>
> Actually, I build things quite often in the last days, which was
> a boring, manual task, and error-prone to get right.
>
> So on Saturday I started hacking a script together that builds
> everything from a single command line, builds all the 16 pythons
> and then builds all the wheels and eggs in parallel batches.
> This uses the original pyside_build as a sub-process.
>
> Just the uploading to Bitbucket is still manual, until I find a way
> to script that as well.
> Maybe I should simply put that on a different site that I directly
> control.
>
> I will upload that script somewhere, soon.
> It was interesting to make a bootstrap from nothing, that manages to
> install everything alone for all python versions.

The script sounds very useful it would be good to have it accessible somewhere?

Did you build the e.g. OSX 10.7 wheels specifically with the 10.7 SDK?
 Maybe it would have been simpler to build them all with the 10.6 SDK
and Python.org python, then rename them for the OSX / architecture
version?  That's what MinRK is doing for his OSX wheels here:
http://kerbin.bic.berkeley.edu/wheelhouse/

I guess you didn't build against the Python.org or System python -
because there are no '-intel' wheels?

Thanks again,

Matthew


From techtonik at gmail.com  Tue Apr 29 11:43:32 2014
From: techtonik at gmail.com (anatoly techtonik)
Date: Tue, 29 Apr 2014 12:43:32 +0300
Subject: [PySide] PySide 1.2.2 OS/X wheels
In-Reply-To: <535EC47B.3070903@stackless.com>
References: <535EB793.6040305@stackless.com>
	<CAH6Pt5rHQ1knk2c4naoNOPRUnncGbs=OLQF0fuPDWe-SxgeW7w@mail.gmail.com>
	<535EC47B.3070903@stackless.com>
Message-ID: <CAPkN8x+5DAqhm=EOrZnkHjzNaT_aSmLZibBvMhPU0EXes19fww@mail.gmail.com>

On Tue, Apr 29, 2014 at 12:13 AM, Christian Tismer <tismer at stackless.com> wrote:
>
> Thank you! Yes, that was what I always wanted to have for OSX.
>
> Actually, I build things quite often in the last days, which was
> a boring, manual task, and error-prone to get right.
>
> So on Saturday I started hacking a script together that builds
> everything from a single command line, builds all the 16 pythons
> and then builds all the wheels and eggs in parallel batches.
> This uses the original pyside_build as a sub-process.
>
> Just the uploading to Bitbucket is still manual, until I find a way
> to script that as well.
> Maybe I should simply put that on a different site that I directly
> control.
>
> I will upload that script somewhere, soon.
> It was interesting to make a bootstrap from nothing, that manages to
> install everything alone for all python versions.

There is:
https://github.com/PySide/pyside-setup
https://github.com/PySide/BuildScripts
Can anybody somehow link those two together.

I also have a lot of fun with reinventing and playing with bootstrapping
bicycles, which resulted in some conventions and public domain code
that I very much hope is reusable:
https://bitbucket.org/techtonik/locally

So, once I complete my SCons research, I may help to build a single
bootstrap.py script that does everything.
-- 
anatoly t.


From backup.rlacko at gmail.com  Tue Apr 29 12:11:20 2014
From: backup.rlacko at gmail.com (Roman Lacko)
Date: Tue, 29 Apr 2014 12:11:20 +0200
Subject: [PySide] PySide 1.2.2 OS/X wheels
In-Reply-To: <CAPkN8x+5DAqhm=EOrZnkHjzNaT_aSmLZibBvMhPU0EXes19fww@mail.gmail.com>
References: <535EB793.6040305@stackless.com>
	<CAH6Pt5rHQ1knk2c4naoNOPRUnncGbs=OLQF0fuPDWe-SxgeW7w@mail.gmail.com>
	<535EC47B.3070903@stackless.com>
	<CAPkN8x+5DAqhm=EOrZnkHjzNaT_aSmLZibBvMhPU0EXes19fww@mail.gmail.com>
Message-ID: <CAHC6v6iqBTR0D-AxT=WCJY083ciG-U6buLbf5d0XjuxDR-LPxg@mail.gmail.com>

2014-04-29 11:43 GMT+02:00 anatoly techtonik <techtonik at gmail.com>:

> On Tue, Apr 29, 2014 at 12:13 AM, Christian Tismer <tismer at stackless.com>
> wrote:
> >
> > Thank you! Yes, that was what I always wanted to have for OSX.
> >
> > Actually, I build things quite often in the last days, which was
> > a boring, manual task, and error-prone to get right.
> >
> > So on Saturday I started hacking a script together that builds
> > everything from a single command line, builds all the 16 pythons
> > and then builds all the wheels and eggs in parallel batches.
> > This uses the original pyside_build as a sub-process.
> >
> > Just the uploading to Bitbucket is still manual, until I find a way
> > to script that as well.
> > Maybe I should simply put that on a different site that I directly
> > control.
> >
> > I will upload that script somewhere, soon.
> > It was interesting to make a bootstrap from nothing, that manages to
> > install everything alone for all python versions.
>
> There is:
> https://github.com/PySide/pyside-setup
> https://github.com/PySide/BuildScripts
> Can anybody somehow link those two together.
>

https://github.com/PySide/pyside-setup is standard python setup project and
is used to build the python distribution packages, like wheels, eggs,
windows installers and source distributions. It support Linux, OSX and
Windows platforms.

There is also https://github.com/PySide/shiboken-setup, to support the
standalone Shiboken installation. It is based on pyside-setup.

AFAIK, the PySide/BuildScripts is used to build the PySide binaries on UNIX
platforms - but it does nothing else - no packaging and no Windows suppport.

I would recommend to use pyside-setup and shiboken-setup to build the
PySide and Shiboken binary distribution packages.


Regards
Roman


>
> I also have a lot of fun with reinventing and playing with bootstrapping
> bicycles, which resulted in some conventions and public domain code
> that I very much hope is reusable:
> https://bitbucket.org/techtonik/locally
>
> So, once I complete my SCons research, I may help to build a single
> bootstrap.py script that does everything.
> --
> anatoly t.
> _______________________________________________
> PySide mailing list
> PySide at qt-project.org
> http://lists.qt-project.org/mailman/listinfo/pyside
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.qt-project.org/pipermail/pyside/attachments/20140429/3f902765/attachment.html>

From jpe at wingware.com  Wed Apr 30 18:07:58 2014
From: jpe at wingware.com (John Ehresman)
Date: Wed, 30 Apr 2014 12:07:58 -0400
Subject: [PySide] PySide 1.2.2 released
Message-ID: <53611FDE.5000207@wingware.com>

I'm happy to announce the release of PySide 1.2.2, which is available at 
https://pypi.python.org/pypi/PySide/1.2.2  Shiboken 1.2.2 was also 
released and is at https://pypi.python.org/pypi/Shiboken/1.2.2

The most significant changes in the release are support for Python 3.4 
and for compiling on OS X 10.9.  Also new are binary distributions of 
PySide for OS X contributed by Christian Tismer and a binary 
distribution of Shiboken for Windows contributed by Roman Lacko.

Questions / feedback may be sent to the pyside mailing list at 
pyside at qt-project.org

Thanks goes to Roman Lacko, Christian Tismer, John Cummings, and Matthew 
Woehlke.

John



From a.richi at bluewin.ch  Wed Apr 30 18:09:08 2014
From: a.richi at bluewin.ch (Aaron Richiger)
Date: Wed, 30 Apr 2014 18:09:08 +0200
Subject: [PySide] PySide 1.2.2 released
In-Reply-To: <53611FDE.5000207@wingware.com>
References: <53611FDE.5000207@wingware.com>
Message-ID: <53612024.80406@bluewin.ch>

Great job and thanks a lot to all who worked on the release!!!
Aaron

Am 30.04.2014 18:07, schrieb John Ehresman:
> I'm happy to announce the release of PySide 1.2.2, which is available at
> https://pypi.python.org/pypi/PySide/1.2.2  Shiboken 1.2.2 was also
> released and is at https://pypi.python.org/pypi/Shiboken/1.2.2
>
> The most significant changes in the release are support for Python 3.4
> and for compiling on OS X 10.9.  Also new are binary distributions of
> PySide for OS X contributed by Christian Tismer and a binary
> distribution of Shiboken for Windows contributed by Roman Lacko.
>
> Questions / feedback may be sent to the pyside mailing list at
> pyside at qt-project.org
>
> Thanks goes to Roman Lacko, Christian Tismer, John Cummings, and Matthew
> Woehlke.
>
> John
>
> _______________________________________________
> PySide mailing list
> PySide at qt-project.org
> http://lists.qt-project.org/mailman/listinfo/pyside



From sean at seanfisk.com  Wed Apr 30 19:05:07 2014
From: sean at seanfisk.com (Sean Fisk)
Date: Wed, 30 Apr 2014 13:05:07 -0400
Subject: [PySide] PySide 1.2.2 released
In-Reply-To: <53611FDE.5000207@wingware.com>
References: <53611FDE.5000207@wingware.com>
Message-ID: <CAF-FwG-XuHMUph32_kTude0XswSLfw67a9ruhKv5L3LLsDMHfQ@mail.gmail.com>

This is awesome! Thanks to all who put in time to work on it.


--
Sean Fisk


On Wed, Apr 30, 2014 at 12:07 PM, John Ehresman <jpe at wingware.com> wrote:

> I'm happy to announce the release of PySide 1.2.2, which is available at
> https://pypi.python.org/pypi/PySide/1.2.2  Shiboken 1.2.2 was also
> released and is at https://pypi.python.org/pypi/Shiboken/1.2.2
>
> The most significant changes in the release are support for Python 3.4
> and for compiling on OS X 10.9.  Also new are binary distributions of
> PySide for OS X contributed by Christian Tismer and a binary
> distribution of Shiboken for Windows contributed by Roman Lacko.
>
> Questions / feedback may be sent to the pyside mailing list at
> pyside at qt-project.org
>
> Thanks goes to Roman Lacko, Christian Tismer, John Cummings, and Matthew
> Woehlke.
>
> John
>
> _______________________________________________
> PySide mailing list
> PySide at qt-project.org
> http://lists.qt-project.org/mailman/listinfo/pyside
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.qt-project.org/pipermail/pyside/attachments/20140430/ff95eb87/attachment.html>

From nmelchior at seegrid.com  Wed Apr 30 22:06:04 2014
From: nmelchior at seegrid.com (Nik A. Melchior)
Date: Wed, 30 Apr 2014 16:06:04 -0400
Subject: [PySide] PySide 1.2.2 released
In-Reply-To: <53611FDE.5000207@wingware.com>
References: <53611FDE.5000207@wingware.com>
Message-ID: <20140430200604.GA18918@fido.seegrid.com>

On Wed, Apr 30, 2014 at 12:07:58PM -0400, John Ehresman wrote:
> I'm happy to announce the release of PySide 1.2.2, which is available at
> https://pypi.python.org/pypi/PySide/1.2.2  Shiboken 1.2.2 was also
> released and is at https://pypi.python.org/pypi/Shiboken/1.2.2

It's good to see continuing development work on PySide.  Congratulations on
the new release.  Is there any chance of new development on pyside-tools?  Is
github the official repository for that code, or has it migrated to gerrit?

My pull request (fixing PYSIDE-130) is still languishing after a year.
There's another github pull request almost 2 years old.

What would be the best way to help get this code into the next release?

--
Nik Melchior
Sr. Roboticist       | 412-379-4500 x147
Seegrid Corporation  | www.seegrid.com
216 Park West Drive, Pittsburgh, PA 15275

Email Confidentiality Notice

The information contained in this transmission is confidential, proprietary or privileged and may be subject to protection under the law. This message is intended for the sole use of the individual or entity to whom it's addressed. If you are not the intended recipient, you are notified that any use, distribution or copying of the message is strictly prohibited and may subject you to criminal or civil penalties. If you received this transmission in error, please contact the sender immediately by replying to this email and delete the material from any computer.