From l.c.visser at science-applied.nl Tue May 1 11:37:01 2012 From: l.c.visser at science-applied.nl (Ludo Visser) Date: Tue, 1 May 2012 11:37:01 +0200 Subject: [PySide] QtUiTools.QUiLoader crash with custom widget In-Reply-To: <4F9EACE2.4030105@tylerlabs.com> References: <4F9E9A8C.1080802@damp.homeip.net> <4F9EACE2.4030105@tylerlabs.com> Message-ID: <025E0DF6-A834-4D3C-84AE-790314E81086@science-applied.nl> On Apr 30, 2012, at 5:16 pm, Tyler W. Wilson wrote: > Ditto on this issue for me. I just ran into this as well trying to add a > custom widget to a designer-based application. I had to move to PyQt4 > temporarily to make it work. > > Thanks, > Tyler > > On 4/30/2012 9:58 AM, Stefan wrote: >> Hi, >> >> I'm trying to create a QMainWindow in the Designer and load the ui file >> dynamically with QUiLoader in pyside 1.1.0. When I don't register any >> custom types it works, but obviously fails to create my derived class. >> When I register them it crashes. Here's a small example that crash for >> me on windows running python 2.7 and pyside 1.1.0: >> >> from PySide import QtCore, QtGui, QtUiTools >> import sys >> >> class MyWidget(QtGui.QWidget): >> def __init__(self, parent=None): >> super(MyWidget, self).__init__(parent) >> >> if __name__ == '__main__': >> application = QtGui.QApplication(sys.argv) >> >> loader = QtUiTools.QUiLoader() >> loader.registerCustomWidget(MyWidget) >> print loader.availableWidgets() >> print loader.createWidget('QLabel') >> print loader.createWidget('MyWidget') >> >> Running this prints the following: >> >> [...] >> >> So it the class is in there, it manages to create a QLabel, but then it >> crashes. >> >> Any ideas why? I don't have any pdb for python, so it's hard to get a >> callstack. >> >> /Stefan I'm not sure what is causing the crash, but one way to obtain the functionality you require is to subclass the QUiLoader class to handle your custom widgets (see http://www.mail-archive.com/pyside at lists.openbossa.org/msg00878.html). For example something like this (here, a top-level widget is used to wrap around the ui-loaded widget, and this top-level widget has an attribute specifying which custom widgets it wants to load; of course, different approaches are possible): def createWidget(self, className, parent = None, name = ""): if className in QtUiTools.QUiLoader.availableWidgets(self): widget = QtUiTools.QUiLoader.createWidget(self, className, parent, name) else: if hasattr(self.baseinstance, "customWidgets"): if className in self.baseinstance.customWidgets.keys(): widget = self.baseinstance.customWidgets[className](parent) else: raise KeyError("Unknown widget '%s'" % className) else: raise AttributeError("Trying to load custom widget '%s', but base instance '%s' does not specify custom widgets." % (className, repr(self.baseinstance))) if self.baseinstance is not None: setattr(self.baseinstance, name, widget) return widget -- Ludo C. Visser, M.Sc. Science Applied phone: +31 6 1066 3076 e-mail: l.c.visser at science-applied.nl From tyler at tylerlabs.com Tue May 1 17:42:31 2012 From: tyler at tylerlabs.com (Tyler W. Wilson) Date: Tue, 01 May 2012 11:42:31 -0400 Subject: [PySide] QtUiTools.QUiLoader crash with custom widget In-Reply-To: <025E0DF6-A834-4D3C-84AE-790314E81086@science-applied.nl> References: <4F9E9A8C.1080802@damp.homeip.net> <4F9EACE2.4030105@tylerlabs.com> <025E0DF6-A834-4D3C-84AE-790314E81086@science-applied.nl> Message-ID: <4FA00467.8020305@tylerlabs.com> On 5/1/2012 5:37 AM, Ludo Visser wrote: > On Apr 30, 2012, at 5:16 pm, Tyler W. Wilson wrote: > >> Ditto on this issue for me. I just ran into this as well trying to add a >> custom widget to a designer-based application. I had to move to PyQt4 >> temporarily to make it work. >> >> Thanks, >> Tyler >> >> On 4/30/2012 9:58 AM, Stefan wrote: >>> Hi, >>> >>> I'm trying to create a QMainWindow in the Designer and load the ui file >>> dynamically with QUiLoader in pyside 1.1.0. When I don't register any >>> custom types it works, but obviously fails to create my derived class. >>> When I register them it crashes. Here's a small example that crash for >>> me on windows running python 2.7 and pyside 1.1.0: >>> >>> from PySide import QtCore, QtGui, QtUiTools >>> import sys >>> >>> class MyWidget(QtGui.QWidget): >>> def __init__(self, parent=None): >>> super(MyWidget, self).__init__(parent) >>> >>> if __name__ == '__main__': >>> application = QtGui.QApplication(sys.argv) >>> >>> loader = QtUiTools.QUiLoader() >>> loader.registerCustomWidget(MyWidget) >>> print loader.availableWidgets() >>> print loader.createWidget('QLabel') >>> print loader.createWidget('MyWidget') >>> >>> Running this prints the following: >>> >>> [...] >>> >>> So it the class is in there, it manages to create a QLabel, but then it >>> crashes. >>> >>> Any ideas why? I don't have any pdb for python, so it's hard to get a >>> callstack. >>> >>> /Stefan > I'm not sure what is causing the crash, but one way to obtain the functionality you require is to subclass the QUiLoader class to handle your custom widgets (see http://www.mail-archive.com/pyside at lists.openbossa.org/msg00878.html). For example something like this (here, a top-level widget is used to wrap around the ui-loaded widget, and this top-level widget has an attribute specifying which custom widgets it wants to load; of course, different approaches are possible): > > def createWidget(self, className, parent = None, name = ""): > if className in QtUiTools.QUiLoader.availableWidgets(self): > widget = QtUiTools.QUiLoader.createWidget(self, className, parent, name) > else: > if hasattr(self.baseinstance, "customWidgets"): > if className in self.baseinstance.customWidgets.keys(): > widget = self.baseinstance.customWidgets[className](parent) > else: > raise KeyError("Unknown widget '%s'" % className) > else: > raise AttributeError("Trying to load custom widget '%s', but base instance '%s' does not specify custom widgets." % (className, repr(self.baseinstance))) > > if self.baseinstance is not None: > setattr(self.baseinstance, name, widget) > > return widget > > No doubt the issue might be worked around like shown above, but it seems like the base UILoader ought to handle this transparently, especially considering there is a section in .ui file that specifies the custom widgets. Plus, it does work in PyQt4. For example, my .ui includes this: QGradient QWidget
QGradient
1 So I would expect the base UILoader to look at this section and perform the proper imports to make these widgets available. But thank you for the tip on the work-around. - Tyler From jmohler at gamry.com Tue May 1 18:20:30 2012 From: jmohler at gamry.com (Joel B. Mohler) Date: Tue, 01 May 2012 12:20:30 -0400 Subject: [PySide] building PySide 1.1.1 against Qt 4.8.1 on windows Message-ID: <4FA00D4E.3070701@gamry.com> Hello, The old bugtracker has http://bugs.pyside.org/show_bug.cgi?id=1079 closed due to the move to qt-project. I'm now trying the build as according to http://qt-project.org/wiki/Building_PySide_on_Windows and running into this same bug. I'm going to report it on on the new tracker and make an effort to fix it myself, but I would appreciate if anybody else has some sort of progress on a fix to relay any pointers. Thanks, Joel From vasure at gmail.com Tue May 1 18:26:52 2012 From: vasure at gmail.com (Srini Kommoori) Date: Tue, 1 May 2012 09:26:52 -0700 Subject: [PySide] Web site transfer In-Reply-To: References: <4F9E7B49.9040404@nokia.com> <1335802700.31033.18.camel@erik-mac> <8789154.SZdH36y9H5@hugodesktop> Message-ID: Matti, Let me know if you need help in migration. I sent you my contacts in a dm. On Mon, Apr 30, 2012 at 12:34 PM, Alexander Jones wrote: > Are we not able to just migrate all of the old website content onto the > main Qt website in the longer term? Hosting a HTTP 301 'moved > permanently'-issuing website to keep old links alive indefinitely would be > very cheap. > > > On 30 April 2012 18:39, Hugo Parente Lima wrote: > >> On Monday, April 30, 2012 10:30:46 AM Srini Kommoori wrote: >> > I would like to propose S3/github for website hosting. It is pretty >> easy to >> > setup and we can automate to push the code/html pages very easily. >> >> I agree, github for website hosting. >> About the tarballs, we have tags on git for each release, so IMO tarballs >> are >> just another way of offer the same information and we can discard if it's >> causing troubles. >> >> > Here is my preferred service list: >> > S3/github for all static pages. >> > Blog - Blogger/Tumblr with multi user submits. >> > Wiki - [No change] Seems like already integrated with Qt main site. >> > Code - [No change] gitorious >> > >> > I can help and sponsor for the expenses(if any) related to the >> migration. >> > >> > thanks, >> > -Srini >> > >> > On Mon, Apr 30, 2012 at 9:18 AM, Erik Janssens >> > >> > wrote: >> > > Hello, >> > > >> > > We have a professional Weebly account, >> > > (http://www.weebly.com/) and can offer a site account for >> > > this purpose. >> > > >> > > A Weebly site can host web pages + images + forms. But it's >> > > not suited to offer source or binary packages for download. >> > > >> > > The downside of using Weebly is that the site should be rebuild >> > > from scratch, something I estimate to be a couple of hours work >> > > for someone experienced with CSS, html. >> > > >> > > The upside being that it's very easy to maintain. >> > > >> > > Just let me know if this would be interesting... >> > > >> > > Cheers, >> > > >> > > Erik >> > > >> > > On Mon, 2012-04-30 at 14:45 +0300, Matti Airas wrote: >> > > > Hi list, >> > > > >> > > > There's one final thing in the PySide migration to Qt add-on: the >> > > > website. I've been trying to have the migration arranged, but >> progress >> > > > has been very slow. Also, May 3 is my last work-day at Nokia, so >> that >> > > > complicates things further. >> > > > >> > > > PySide is currently hosted at a Virtual Machine provided by Nokia >> IT, >> > > > which has made it easy because there have not been any fees or >> bandwidth >> > > > quotas (and the connectivity has been quite excellent). >> Unfortunately, >> > > > that VM is leased only until end of June, so arrangements need to be >> > > > made pretty soon. >> > > > >> > > > Unfortunately, Qt Project is unwilling to provide a virtual machine >> for >> > > > PySide, and therefore a direct migration of the VM is not feasible. >> They >> > > > have proposed that they can host the static content (binaries, >> > > > documentation & doc packages) which account for vast majority of the >> > > > traffic anyway. Then, the current WordPress site contents (which >> isn't >> > > > much any longer, anyway) could be moved to the wiki, with a >> possibility >> > > > to have a static landing page as the home page. The news blog would >> then >> > > > need to be migrated to any free 3rd-party blogging platform such as >> > > > wordpress.com. This is probably doable and something that the >> project >> > > > could live with, but it does require additional effort and decrease >> the >> > > > level of service somewhat, both for the administrators (no access >> to WP >> > > > authoring tools) and for the end-users (URLs for binaries and docs >> > > > change, blog aggregation needs to be updated). In the long term, >> this >> > > > would be the approach requiring the least amount of maintenance. >> > > > >> > > > Another alternative which could be feasible is to have a VM >> elsewhere >> > > > and Qt Project would just point the DNS there. The VM requirements >> are >> > > > not huge: the current VM is a single-core one Xeon 2.5 GHz with 2 >> GB of >> > > > allocated memory and 40 GB of disk space. However, the site >> generates >> > > > quite some traffic - in April, 120 GB, and the number has been >> > > > increasing at a steady pace. Also, having a separate VM would >> require a >> > > > dedicated admin who would be willing to maintain the VM, take >> backups, >> > > >> > > etc. >> > > >> > > > I would like to hear your opinions on this one. In the first >> option, the >> > > > up front work and changes would be bigger but the long-term effort >> > > > lower, and no external costs would be incurred. In the latter >> option, >> > > > the migration would be less painful but having the VM would require >> more >> > > > maintenance in the future, and recurring payments would occur. >> > > > >> > > > Also, I'd like to call for volunteers to help with the migration >> and the >> > > > future maintenance. As said, my work obligation at Nokia is ending >> and >> > > > while I'm not vanishing altogether, it'd be helpful if someone else >> > > > could take the lead. Srini, you volunteered for the webmaster and >> > > > wikimaster duties in February - although this might not be what you >> had >> > > > in mind, would you still be interested? >> > > > >> > > > Also, if a separate VM is desired, a sponsor is needed for that. I >> > > > wonder if any companies would be interested in providing that, for a >> > > > modest logo or banner visibility on the web site? A basic 3rd-party >> VM >> > > > with that amount of traffic would cost some 40-50 EUR/USD a month. >> Not a >> > > > huge sum, but I wouldn't feel happy paying it out of my own pocket, >> > > > especially with the steadily increasing traffic. >> > > > >> > > > Cheers, >> > > > >> > > > Matti Airas >> > > > _______________________________________________ >> > > > 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 list >> PySide at qt-project.org >> http://lists.qt-project.org/mailman/listinfo/pyside >> >> > > > -- > Alexander Jones > Double Negative R&D > www.dneg.com > > > _______________________________________________ > PySide mailing list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From vasure at gmail.com Tue May 1 18:26:52 2012 From: vasure at gmail.com (Srini Kommoori) Date: Tue, 1 May 2012 09:26:52 -0700 Subject: [PySide] Web site transfer In-Reply-To: References: <4F9E7B49.9040404@nokia.com> <1335802700.31033.18.camel@erik-mac> <8789154.SZdH36y9H5@hugodesktop> Message-ID: Matti, Let me know if you need help in migration. I sent you my contacts in a dm. On Mon, Apr 30, 2012 at 12:34 PM, Alexander Jones wrote: > Are we not able to just migrate all of the old website content onto the > main Qt website in the longer term? Hosting a HTTP 301 'moved > permanently'-issuing website to keep old links alive indefinitely would be > very cheap. > > > On 30 April 2012 18:39, Hugo Parente Lima wrote: > >> On Monday, April 30, 2012 10:30:46 AM Srini Kommoori wrote: >> > I would like to propose S3/github for website hosting. It is pretty >> easy to >> > setup and we can automate to push the code/html pages very easily. >> >> I agree, github for website hosting. >> About the tarballs, we have tags on git for each release, so IMO tarballs >> are >> just another way of offer the same information and we can discard if it's >> causing troubles. >> >> > Here is my preferred service list: >> > S3/github for all static pages. >> > Blog - Blogger/Tumblr with multi user submits. >> > Wiki - [No change] Seems like already integrated with Qt main site. >> > Code - [No change] gitorious >> > >> > I can help and sponsor for the expenses(if any) related to the >> migration. >> > >> > thanks, >> > -Srini >> > >> > On Mon, Apr 30, 2012 at 9:18 AM, Erik Janssens >> > >> > wrote: >> > > Hello, >> > > >> > > We have a professional Weebly account, >> > > (http://www.weebly.com/) and can offer a site account for >> > > this purpose. >> > > >> > > A Weebly site can host web pages + images + forms. But it's >> > > not suited to offer source or binary packages for download. >> > > >> > > The downside of using Weebly is that the site should be rebuild >> > > from scratch, something I estimate to be a couple of hours work >> > > for someone experienced with CSS, html. >> > > >> > > The upside being that it's very easy to maintain. >> > > >> > > Just let me know if this would be interesting... >> > > >> > > Cheers, >> > > >> > > Erik >> > > >> > > On Mon, 2012-04-30 at 14:45 +0300, Matti Airas wrote: >> > > > Hi list, >> > > > >> > > > There's one final thing in the PySide migration to Qt add-on: the >> > > > website. I've been trying to have the migration arranged, but >> progress >> > > > has been very slow. Also, May 3 is my last work-day at Nokia, so >> that >> > > > complicates things further. >> > > > >> > > > PySide is currently hosted at a Virtual Machine provided by Nokia >> IT, >> > > > which has made it easy because there have not been any fees or >> bandwidth >> > > > quotas (and the connectivity has been quite excellent). >> Unfortunately, >> > > > that VM is leased only until end of June, so arrangements need to be >> > > > made pretty soon. >> > > > >> > > > Unfortunately, Qt Project is unwilling to provide a virtual machine >> for >> > > > PySide, and therefore a direct migration of the VM is not feasible. >> They >> > > > have proposed that they can host the static content (binaries, >> > > > documentation & doc packages) which account for vast majority of the >> > > > traffic anyway. Then, the current WordPress site contents (which >> isn't >> > > > much any longer, anyway) could be moved to the wiki, with a >> possibility >> > > > to have a static landing page as the home page. The news blog would >> then >> > > > need to be migrated to any free 3rd-party blogging platform such as >> > > > wordpress.com. This is probably doable and something that the >> project >> > > > could live with, but it does require additional effort and decrease >> the >> > > > level of service somewhat, both for the administrators (no access >> to WP >> > > > authoring tools) and for the end-users (URLs for binaries and docs >> > > > change, blog aggregation needs to be updated). In the long term, >> this >> > > > would be the approach requiring the least amount of maintenance. >> > > > >> > > > Another alternative which could be feasible is to have a VM >> elsewhere >> > > > and Qt Project would just point the DNS there. The VM requirements >> are >> > > > not huge: the current VM is a single-core one Xeon 2.5 GHz with 2 >> GB of >> > > > allocated memory and 40 GB of disk space. However, the site >> generates >> > > > quite some traffic - in April, 120 GB, and the number has been >> > > > increasing at a steady pace. Also, having a separate VM would >> require a >> > > > dedicated admin who would be willing to maintain the VM, take >> backups, >> > > >> > > etc. >> > > >> > > > I would like to hear your opinions on this one. In the first >> option, the >> > > > up front work and changes would be bigger but the long-term effort >> > > > lower, and no external costs would be incurred. In the latter >> option, >> > > > the migration would be less painful but having the VM would require >> more >> > > > maintenance in the future, and recurring payments would occur. >> > > > >> > > > Also, I'd like to call for volunteers to help with the migration >> and the >> > > > future maintenance. As said, my work obligation at Nokia is ending >> and >> > > > while I'm not vanishing altogether, it'd be helpful if someone else >> > > > could take the lead. Srini, you volunteered for the webmaster and >> > > > wikimaster duties in February - although this might not be what you >> had >> > > > in mind, would you still be interested? >> > > > >> > > > Also, if a separate VM is desired, a sponsor is needed for that. I >> > > > wonder if any companies would be interested in providing that, for a >> > > > modest logo or banner visibility on the web site? A basic 3rd-party >> VM >> > > > with that amount of traffic would cost some 40-50 EUR/USD a month. >> Not a >> > > > huge sum, but I wouldn't feel happy paying it out of my own pocket, >> > > > especially with the steadily increasing traffic. >> > > > >> > > > Cheers, >> > > > >> > > > Matti Airas >> > > > _______________________________________________ >> > > > 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 list >> PySide at qt-project.org >> http://lists.qt-project.org/mailman/listinfo/pyside >> >> > > > -- > Alexander Jones > Double Negative R&D > www.dneg.com > > > _______________________________________________ > PySide mailing list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From bjelge at nefines.com Tue May 1 19:13:54 2012 From: bjelge at nefines.com (=?ISO-8859-1?Q?Bj=F8rn_Helge_Kj=F8snes?=) Date: Tue, 01 May 2012 19:13:54 +0200 Subject: [PySide] building PySide 1.1.1 against Qt 4.8.1 on windows In-Reply-To: <4FA00D4E.3070701@gamry.com> References: <4FA00D4E.3070701@gamry.com> Message-ID: <4FA019D2.70509@nefines.com> Hi, I have spent some time to figure out why this error is happening, but progress is slow due to total lack of experience with shiboken code and the type system and how to best debug this on Windows, so any help would be appreciated. So far I have found that it is the new open methods in QFile and in QFSFileEngine that not are added to the generated wrappers. It is only one open method added from the class and one for the parent class. The reason for these open methods not being added is that they have been marked by the apiextractor to have the Final flag set and therefor skipped in the CppGenerator::generateClass method. The code expecting to use these method to support overloaded methods in Python are generating code expecting these methods. The code where generating of code is skipped are in cppgenerator.cpp around line 250: else if ((!avoidProtectedHack() || !metaClass->hasPrivateDestructor()) && (func->isVirtual() || func->isAbstract())) writeVirtualMethodNative(s, func); The func->isVirtual() returns true because the final flag is set and not because it is a virtual method. I am now trying to figure out why the Final flag is set in these methods, so if anybody can point me to where this is done, please do so. Thanks, Bjorn Helge Kjosnes On 01.05.2012 18:20, Joel B. Mohler wrote: > Hello, > > The old bugtracker has http://bugs.pyside.org/show_bug.cgi?id=1079 > closed due to the move to qt-project. I'm now trying the build as > according to http://qt-project.org/wiki/Building_PySide_on_Windows and > running into this same bug. > > I'm going to report it on on the new tracker and make an effort to fix > it myself, but I would appreciate if anybody else has some sort of > progress on a fix to relay any pointers. > > Thanks, > Joel > > _______________________________________________ > PySide mailing list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside > From frank at ohufx.com Wed May 2 07:14:36 2012 From: frank at ohufx.com (Frank Rueter | OHUfx) Date: Wed, 02 May 2012 17:14:36 +1200 Subject: [PySide] creating custom setFrameStyle() Message-ID: <4FA0C2BC.1020609@ohufx.com> Hi everyone, I'm in the middle of creating a custom button by subclassing QWidget and reimplementing paintEvent(). I kinda have what I'm after except that I would like to give the whole button a "box/sunken" style. What is the best way to re-implement setFrameStyle() in my paintEvent? Or should I take another route? Thanks all! frank From frank at ohufx.com Wed May 2 07:44:09 2012 From: frank at ohufx.com (Frank Rueter | OHUfx) Date: Wed, 02 May 2012 17:44:09 +1200 Subject: [PySide] re-implementing setDown() for custom button Message-ID: <4FA0C9A9.5000909@ohufx.com> Hi, to follow up my previous mail, I also just noticed that I have to re-implement the QAbstractButton's setDown() function to make sure my custom button doesn't stay down after a drag&drop action. With a QPushButton I simply called self.setDown( False ) from inside the mouseMoveEvent, so now I'm trying to understand how to recreate this behaviour when using a custom paintEvent(). Any tips would be greatly appreciated. Cheers, frank From frank at ohufx.com Wed May 2 08:35:33 2012 From: frank at ohufx.com (Frank Rueter | OHUfx) Date: Wed, 02 May 2012 18:35:33 +1200 Subject: [PySide] re-implementing setDown() for custom button In-Reply-To: <4FA0C9A9.5000909@ohufx.com> References: <4FA0C9A9.5000909@ohufx.com> Message-ID: <4FA0D5B5.20604@ohufx.com> sorry for the monolog, but I just realised I will also have to make my custom button play along when it's parent's self.sender() function is called as well. The QSignalSender doesn't quite seem to be the right thing to use as it is implemented in the parent's constructor, however I would like to implement this in the button's class if possible to make it self contained. How would I go about that? Cheers, frank On 2/05/12 5:44 PM, Frank Rueter | OHUfx wrote: > Hi, > > to follow up my previous mail, I also just noticed that I have to > re-implement the QAbstractButton's setDown() function to make sure my > custom button doesn't stay down after a drag&drop action. > > With a QPushButton I simply called self.setDown( False ) from inside the > mouseMoveEvent, so now I'm trying to understand how to recreate this > behaviour when using a custom paintEvent(). > > Any tips would be greatly appreciated. > > Cheers, > frank > _______________________________________________ > PySide mailing list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside From bjelge at nefines.com Wed May 2 09:19:21 2012 From: bjelge at nefines.com (=?ISO-8859-1?Q?Bj=F8rn_Helge_Kj=F8snes?=) Date: Wed, 02 May 2012 09:19:21 +0200 Subject: [PySide] building PySide 1.1.1 against Qt 4.8.1 on windows a kind of solution In-Reply-To: <4FA019D2.70509@nefines.com> References: <4FA00D4E.3070701@gamry.com> <4FA019D2.70509@nefines.com> Message-ID: <4FA0DFF9.8090106@nefines.com> Hi, after some more digging I am able to compile and run PySide 1.1.1 against Qt 4.8.1. It is a ugly solution that can not be committed to the repos, but for if there are any in need for a version, you can replace a few lines of code in shiboken-1.1.1\generator\shiboken\cppgenerator.cpp file. Locate line 287 and replace: && (func->name() != "qt_metacall")) overloads.append(func); with: && (func->name() != "qt_metacall")){ if (func->toString().contains("open")) { if (func->isVirtual()) overloads.append(func); } else overloads.append(func); } and before any say anything about the method, I know it is ugly and should be avoided. I do it like this to avoid to many side effects until I understand how it should be done. May me anybody can help out here. The problem in Windows are that the code generated for the overload wrappers are not getting only the methods that there are generated headers and source code for. In QFile you have at least 6 open methods with 1 to 3 parameters that are overloaded. In PySide only 2 of these survives into the wrappers. The code writing method handling the overload do not see that some of the methods might get filtered out by the avoidProtectedHack. see line 258 in the cppgenerator.cpp file: else if ((!avoidProtectedHack() || !metaClass->hasPrivateDestructor()) && (func->isVirtual() || func->isAbstract())) writeVirtualMethodNative(s, func); The makes shiboken not render code for 3 out of 4 overloads of open in QFile. Can anybody tell me if this is what the avoid protected hack is all about, skipping implementation of methods in Windows that not meet certain requirements? If so, why is that done and is there anything I can do to help fix that? Is there anywhere I can find information of how things are implemented and why. I feel I start to understand how the code generating is done and I will probably start using this instead of my hand-crafted wrappers written in pycxx when I can find the time. Kind regards, Bjorn Helge Kjosnes On 01.05.2012 19:13, Bjørn Helge Kjøsnes wrote: > Hi, > > I have spent some time to figure out why this error is happening, but > progress is slow due to total lack of experience with shiboken code and > the type system and how to best debug this on Windows, so any help would > be appreciated. > > So far I have found that it is the new open methods in QFile and in > QFSFileEngine that not are added to the generated wrappers. It is only > one open method added from the class and one for the parent class. The > reason for these open methods not being added is that they have been > marked by the apiextractor to have the Final flag set and therefor > skipped in the CppGenerator::generateClass method. The code expecting to > use these method to support overloaded methods in Python are generating > code expecting these methods. > > The code where generating of code is skipped are in cppgenerator.cpp > around line 250: > else if ((!avoidProtectedHack() || > !metaClass->hasPrivateDestructor()) > && (func->isVirtual() || func->isAbstract())) > writeVirtualMethodNative(s, func); > > The func->isVirtual() returns true because the final flag is set and not > because it is a virtual method. > > I am now trying to figure out why the Final flag is set in these > methods, so if anybody can point me to where this is done, please do so. > > Thanks, > Bjorn Helge Kjosnes > > On 01.05.2012 18:20, Joel B. Mohler wrote: >> Hello, >> >> The old bugtracker has http://bugs.pyside.org/show_bug.cgi?id=1079 >> closed due to the move to qt-project. I'm now trying the build as >> according to http://qt-project.org/wiki/Building_PySide_on_Windows and >> running into this same bug. >> >> I'm going to report it on on the new tracker and make an effort to fix >> it myself, but I would appreciate if anybody else has some sort of >> progress on a fix to relay any pointers. >> >> Thanks, >> Joel >> >> _______________________________________________ >> 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 > From matti.p.airas at nokia.com Wed May 2 09:26:30 2012 From: matti.p.airas at nokia.com (Matti Airas) Date: Wed, 02 May 2012 10:26:30 +0300 Subject: [PySide] Web site transfer In-Reply-To: References: <4F9E7B49.9040404@nokia.com> <1335802700.31033.18.camel@erik-mac> Message-ID: <4FA0E1A6.6070902@nokia.com> On 30.04.2012 20:30, ext Srini Kommoori wrote: > I would like to propose S3/github for website hosting. It is pretty > easy to setup and we can automate to push the code/html pages very > easily. > > Here is my preferred service list: > S3/github for all static pages. > Blog - Blogger/Tumblr with multi user submits. > Wiki - [No change] Seems like already integrated with Qt main site. > Code - [No change] gitorious > > I can help and sponsor for the expenses(if any) related to the migration. Since there were no objections for getting rid of the current VM setup, I'd rather than using S3/github just use the offered hosting by Qt Project. We'll get a managed environment and don't need to worry about any costs. I don't have strong opinions about the blogging platform. I'm sure e.g. Blogger would work just fine. Cheers, ma. -------------- next part -------------- An HTML attachment was scrubbed... URL: From jmohler at gamry.com Wed May 2 15:23:35 2012 From: jmohler at gamry.com (Joel B. Mohler) Date: Wed, 02 May 2012 09:23:35 -0400 Subject: [PySide] building PySide 1.1.1 against Qt 4.8.1 on windows a kind of solution In-Reply-To: <4FA0DFF9.8090106@nefines.com> References: <4FA00D4E.3070701@gamry.com> <4FA019D2.70509@nefines.com> <4FA0DFF9.8090106@nefines.com> Message-ID: <4FA13557.9020106@gamry.com> On 5/2/2012 3:19 AM, Bjørn Helge Kjøsnes wrote: > Hi, > > after some more digging I am able to compile and run PySide 1.1.1 > against Qt 4.8.1. > > It is a ugly solution that can not be committed to the repos, but for if > there are any in need for a version, you can replace a few lines of code > in shiboken-1.1.1\generator\shiboken\cppgenerator.cpp file. > > Locate line 287 and replace: > && (func->name() != "qt_metacall")) > overloads.append(func); > > with: > && (func->name() != "qt_metacall")){ > if (func->toString().contains("open")) { > if (func->isVirtual()) > overloads.append(func); > } > else > overloads.append(func); > } Bjørn, Thanks, I confirm that this replacement works for me as well (not that that makes it any more legitimate as you point out). Unfortunately, all I can do is confirm what you've done. I have no more inspiration about what is going on behind the scenes here. Note that I've reported this bug in the in new tracker at https://bugreports.qt-project.org/browse/PYSIDE-63 Joel From frank at ohufx.com Thu May 3 08:23:52 2012 From: frank at ohufx.com (Frank Rueter | OHUfx) Date: Thu, 03 May 2012 18:23:52 +1200 Subject: [PySide] re-implementing setDown() for custom button In-Reply-To: <4FA0D5B5.20604@ohufx.com> References: <4FA0C9A9.5000909@ohufx.com> <4FA0D5B5.20604@ohufx.com> Message-ID: <4FA22478.9030007@ohufx.com> Any takers? I'm kinda stuck with this one unfortunately. On 2/05/12 6:35 PM, Frank Rueter | OHUfx wrote: > sorry for the monolog, but I just realised I will also have to make my > custom button play along when it's parent's self.sender() function is > called as well. > The QSignalSender doesn't quite seem to be the right thing to use as it > is implemented in the parent's constructor, however I would like to > implement this in the button's class if possible to make it self contained. > > How would I go about that? > > Cheers, > frank > > On 2/05/12 5:44 PM, Frank Rueter | OHUfx wrote: >> Hi, >> >> to follow up my previous mail, I also just noticed that I have to >> re-implement the QAbstractButton's setDown() function to make sure my >> custom button doesn't stay down after a drag&drop action. >> >> With a QPushButton I simply called self.setDown( False ) from inside the >> mouseMoveEvent, so now I'm trying to understand how to recreate this >> behaviour when using a custom paintEvent(). >> >> Any tips would be greatly appreciated. >> >> Cheers, >> frank >> _______________________________________________ >> 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 From frank at ohufx.com Thu May 3 09:11:45 2012 From: frank at ohufx.com (Frank Rueter | OHUfx) Date: Thu, 03 May 2012 19:11:45 +1200 Subject: [PySide] re-implementing setDown() for custom button In-Reply-To: References: <4FA0C9A9.5000909@ohufx.com> <4FA0D5B5.20604@ohufx.com> <4FA22478.9030007@ohufx.com> Message-ID: <4FA22FB1.6050306@ohufx.com> Indeed, my apologies. Will strip out the relevant parts and post an example soon. On 3/05/12 7:06 PM, Srini Kommoori wrote: > Code snippet would help in understanding what you are asking. > > For most of my cases this works great: > pushButton.clicked.connect(findFile) > > If you are trying to implement your own style button, QPushButton also > takes icon. > > > On Wed, May 2, 2012 at 11:23 PM, Frank Rueter | OHUfx > wrote: > > Any takers? > I'm kinda stuck with this one unfortunately. > > > On 2/05/12 6:35 PM, Frank Rueter | OHUfx wrote: > > sorry for the monolog, but I just realised I will also have to > make my > > custom button play along when it's parent's self.sender() > function is > > called as well. > > The QSignalSender doesn't quite seem to be the right thing to > use as it > > is implemented in the parent's constructor, however I would like to > > implement this in the button's class if possible to make it self > contained. > > > > How would I go about that? > > > > Cheers, > > frank > > > > On 2/05/12 5:44 PM, Frank Rueter | OHUfx wrote: > >> Hi, > >> > >> to follow up my previous mail, I also just noticed that I have to > >> re-implement the QAbstractButton's setDown() function to make > sure my > >> custom button doesn't stay down after a drag&drop action. > >> > >> With a QPushButton I simply called self.setDown( False ) from > inside the > >> mouseMoveEvent, so now I'm trying to understand how to recreate > this > >> behaviour when using a custom paintEvent(). > >> > >> Any tips would be greatly appreciated. > >> > >> Cheers, > >> frank > >> _______________________________________________ > >> 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 list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From frank at ohufx.com Thu May 3 10:37:45 2012 From: frank at ohufx.com (Frank Rueter | OHUfx) Date: Thu, 03 May 2012 20:37:45 +1200 Subject: [PySide] re-implementing setDown() for custom button In-Reply-To: <4FA22FB1.6050306@ohufx.com> References: <4FA0C9A9.5000909@ohufx.com> <4FA0D5B5.20604@ohufx.com> <4FA22478.9030007@ohufx.com> <4FA22FB1.6050306@ohufx.com> Message-ID: <4FA243D9.9000601@ohufx.com> Here is some example code that I ripped out of my main code: http://pastebin.com/jXZXXebW It compares a sub classed QPushButton with one I painted from scratch. I was actually able to solve some of the problems in the process (and for some reason the QObject.sender() works as well now) , but the one thing that still doesn't work with the new button is when I try to set a drag action like this: drag.start( Qt.CopyAction ) The drag action for the "new button" is in line 192 which doesn't seem to have any effect when dragging the button, while the one for the "old button" in line 27 works as expected and turns the mouse cursor into a little plus sign when dragging. What am I missing? Cheers, frank On 3/05/12 7:11 PM, Frank Rueter | OHUfx wrote: > Indeed, my apologies. > Will strip out the relevant parts and post an example soon. > > On 3/05/12 7:06 PM, Srini Kommoori wrote: >> Code snippet would help in understanding what you are asking. >> >> For most of my cases this works great: >> pushButton.clicked.connect(findFile) >> >> If you are trying to implement your own style button, QPushButton >> also takes icon. >> >> >> On Wed, May 2, 2012 at 11:23 PM, Frank Rueter | OHUfx >> > wrote: >> >> Any takers? >> I'm kinda stuck with this one unfortunately. >> >> >> On 2/05/12 6:35 PM, Frank Rueter | OHUfx wrote: >> > sorry for the monolog, but I just realised I will also have to >> make my >> > custom button play along when it's parent's self.sender() >> function is >> > called as well. >> > The QSignalSender doesn't quite seem to be the right thing to >> use as it >> > is implemented in the parent's constructor, however I would like to >> > implement this in the button's class if possible to make it >> self contained. >> > >> > How would I go about that? >> > >> > Cheers, >> > frank >> > >> > On 2/05/12 5:44 PM, Frank Rueter | OHUfx wrote: >> >> Hi, >> >> >> >> to follow up my previous mail, I also just noticed that I have to >> >> re-implement the QAbstractButton's setDown() function to make >> sure my >> >> custom button doesn't stay down after a drag&drop action. >> >> >> >> With a QPushButton I simply called self.setDown( False ) from >> inside the >> >> mouseMoveEvent, so now I'm trying to understand how to >> recreate this >> >> behaviour when using a custom paintEvent(). >> >> >> >> Any tips would be greatly appreciated. >> >> >> >> Cheers, >> >> frank >> >> _______________________________________________ >> >> 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 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 -------------- next part -------------- An HTML attachment was scrubbed... URL: From frank at ohufx.com Thu May 3 10:51:35 2012 From: frank at ohufx.com (Frank Rueter | OHUfx) Date: Thu, 03 May 2012 20:51:35 +1200 Subject: [PySide] re-implementing setDown() for custom button In-Reply-To: <4FA243D9.9000601@ohufx.com> References: <4FA0C9A9.5000909@ohufx.com> <4FA0D5B5.20604@ohufx.com> <4FA22478.9030007@ohufx.com> <4FA22FB1.6050306@ohufx.com> <4FA243D9.9000601@ohufx.com> Message-ID: <4FA24717.4060307@ohufx.com> DOH. Minutes after posting this I realised that the little plus on the cursor only shows up on a "droppable" area, which my custom button is not (and the default one is I guess). In other words I just solved all the problem myself by trying to strip it all down for a test to post. Sorry, will try and be a bit more accountable next time. On 3/05/12 8:37 PM, Frank Rueter | OHUfx wrote: > Here is some example code that I ripped out of my main code: > http://pastebin.com/jXZXXebW > > It compares a sub classed QPushButton with one I painted from scratch. > > I was actually able to solve some of the problems in the process (and > for some reason the QObject.sender() works as well now) , but the one > thing that still doesn't work with the new button is when I try to set > a drag action like this: > drag.start( Qt.CopyAction ) > > The drag action for the "new button" is in line 192 which doesn't seem > to have any effect when dragging the button, while the one for the > "old button" in line 27 works as expected and turns the mouse cursor > into a little plus sign when dragging. > > What am I missing? > > > Cheers, > frank > > > On 3/05/12 7:11 PM, Frank Rueter | OHUfx wrote: >> Indeed, my apologies. >> Will strip out the relevant parts and post an example soon. >> >> On 3/05/12 7:06 PM, Srini Kommoori wrote: >>> Code snippet would help in understanding what you are asking. >>> >>> For most of my cases this works great: >>> pushButton.clicked.connect(findFile) >>> >>> If you are trying to implement your own style button, QPushButton >>> also takes icon. >>> >>> >>> On Wed, May 2, 2012 at 11:23 PM, Frank Rueter | OHUfx >>> > wrote: >>> >>> Any takers? >>> I'm kinda stuck with this one unfortunately. >>> >>> >>> On 2/05/12 6:35 PM, Frank Rueter | OHUfx wrote: >>> > sorry for the monolog, but I just realised I will also have to >>> make my >>> > custom button play along when it's parent's self.sender() >>> function is >>> > called as well. >>> > The QSignalSender doesn't quite seem to be the right thing to >>> use as it >>> > is implemented in the parent's constructor, however I would >>> like to >>> > implement this in the button's class if possible to make it >>> self contained. >>> > >>> > How would I go about that? >>> > >>> > Cheers, >>> > frank >>> > >>> > On 2/05/12 5:44 PM, Frank Rueter | OHUfx wrote: >>> >> Hi, >>> >> >>> >> to follow up my previous mail, I also just noticed that I have to >>> >> re-implement the QAbstractButton's setDown() function to make >>> sure my >>> >> custom button doesn't stay down after a drag&drop action. >>> >> >>> >> With a QPushButton I simply called self.setDown( False ) from >>> inside the >>> >> mouseMoveEvent, so now I'm trying to understand how to >>> recreate this >>> >> behaviour when using a custom paintEvent(). >>> >> >>> >> Any tips would be greatly appreciated. >>> >> >>> >> Cheers, >>> >> frank >>> >> _______________________________________________ >>> >> 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 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 list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside -------------- next part -------------- An HTML attachment was scrubbed... URL: From vasure at gmail.com Thu May 3 11:49:37 2012 From: vasure at gmail.com (Srini Kommoori) Date: Thu, 3 May 2012 02:49:37 -0700 Subject: [PySide] re-implementing setDown() for custom button In-Reply-To: <4FA24717.4060307@ohufx.com> References: <4FA0C9A9.5000909@ohufx.com> <4FA0D5B5.20604@ohufx.com> <4FA22478.9030007@ohufx.com> <4FA22FB1.6050306@ohufx.com> <4FA243D9.9000601@ohufx.com> <4FA24717.4060307@ohufx.com> Message-ID: Thats great. One thing I observed is: for your custom button, if you want a nice/native copy icon to appear - set following in your dragEnterEvent(I would prefer to do have dragMoveEvent) event.setDropAction(QtCore.Qt.CopyAction) event.accept() -Srini On Thu, May 3, 2012 at 1:51 AM, Frank Rueter | OHUfx wrote: > DOH. Minutes after posting this I realised that the little plus on the > cursor only shows up on a "droppable" area, which my custom button is not > (and the default one is I guess). > In other words I just solved all the problem myself by trying to strip it > all down for a test to post. > Sorry, will try and be a bit more accountable next time. > > > > > On 3/05/12 8:37 PM, Frank Rueter | OHUfx wrote: > > Here is some example code that I ripped out of my main code: > http://pastebin.com/jXZXXebW > > It compares a sub classed QPushButton with one I painted from scratch. > > I was actually able to solve some of the problems in the process (and for > some reason the QObject.sender() works as well now) , but the one thing > that still doesn't work with the new button is when I try to set a drag > action like this: > drag.start( Qt.CopyAction ) > > The drag action for the "new button" is in line 192 which doesn't seem to > have any effect when dragging the button, while the one for the "old > button" in line 27 works as expected and turns the mouse cursor into a > little plus sign when dragging. > > What am I missing? > > > Cheers, > frank > > > On 3/05/12 7:11 PM, Frank Rueter | OHUfx wrote: > > Indeed, my apologies. > Will strip out the relevant parts and post an example soon. > > On 3/05/12 7:06 PM, Srini Kommoori wrote: > > Code snippet would help in understanding what you are asking. > > For most of my cases this works great: > pushButton.clicked.connect(findFile) > > If you are trying to implement your own style button, QPushButton also > takes icon. > > > On Wed, May 2, 2012 at 11:23 PM, Frank Rueter | OHUfx wrote: > >> Any takers? >> I'm kinda stuck with this one unfortunately. >> >> >> On 2/05/12 6:35 PM, Frank Rueter | OHUfx wrote: >> > sorry for the monolog, but I just realised I will also have to make my >> > custom button play along when it's parent's self.sender() function is >> > called as well. >> > The QSignalSender doesn't quite seem to be the right thing to use as it >> > is implemented in the parent's constructor, however I would like to >> > implement this in the button's class if possible to make it self >> contained. >> > >> > How would I go about that? >> > >> > Cheers, >> > frank >> > >> > On 2/05/12 5:44 PM, Frank Rueter | OHUfx wrote: >> >> Hi, >> >> >> >> to follow up my previous mail, I also just noticed that I have to >> >> re-implement the QAbstractButton's setDown() function to make sure my >> >> custom button doesn't stay down after a drag&drop action. >> >> >> >> With a QPushButton I simply called self.setDown( False ) from inside >> the >> >> mouseMoveEvent, so now I'm trying to understand how to recreate this >> >> behaviour when using a custom paintEvent(). >> >> >> >> Any tips would be greatly appreciated. >> >> >> >> Cheers, >> >> frank >> >> _______________________________________________ >> >> 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 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 > > > > _______________________________________________ > PySide mailing listPySide at qt-project.orghttp://lists.qt-project.org/mailman/listinfo/pyside > > > _______________________________________________ > PySide mailing list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From frank at ohufx.com Thu May 3 22:32:28 2012 From: frank at ohufx.com (Frank Rueter | OHUfx) Date: Fri, 04 May 2012 08:32:28 +1200 Subject: [PySide] re-implementing setDown() for custom button In-Reply-To: References: <4FA0C9A9.5000909@ohufx.com> <4FA0D5B5.20604@ohufx.com> <4FA22478.9030007@ohufx.com> <4FA22FB1.6050306@ohufx.com> <4FA243D9.9000601@ohufx.com> <4FA24717.4060307@ohufx.com> Message-ID: <4FA2EB5C.8080604@ohufx.com> Hi Srini, >>(I would prefer to do have dragMoveEvent) I'm not quite sure what you mean. Can you elaborate? I did what you suggested but can't see any change (maybe because I'm on osx?). Cheers, frank On 3/05/12 9:49 PM, Srini Kommoori wrote: > Thats great. > > One thing I observed is: for your custom button, if you want a > nice/native copy icon to appear - set following in your > dragEnterEvent(I would prefer to do have dragMoveEvent) > > event.setDropAction(QtCore.Qt.CopyAction) > event.accept() > > -Srini > > On Thu, May 3, 2012 at 1:51 AM, Frank Rueter | OHUfx > wrote: > > DOH. Minutes after posting this I realised that the little plus on > the cursor only shows up on a "droppable" area, which my custom > button is not (and the default one is I guess). > In other words I just solved all the problem myself by trying to > strip it all down for a test to post. > Sorry, will try and be a bit more accountable next time. > > > > > On 3/05/12 8:37 PM, Frank Rueter | OHUfx wrote: >> Here is some example code that I ripped out of my main code: >> http://pastebin.com/jXZXXebW >> >> It compares a sub classed QPushButton with one I painted from >> scratch. >> >> I was actually able to solve some of the problems in the process >> (and for some reason the QObject.sender() works as well now) , >> but the one thing that still doesn't work with the new button is >> when I try to set a drag action like this: >> drag.start( Qt.CopyAction ) >> >> The drag action for the "new button" is in line 192 which doesn't >> seem to have any effect when dragging the button, while the one >> for the "old button" in line 27 works as expected and turns the >> mouse cursor into a little plus sign when dragging. >> >> What am I missing? >> >> >> Cheers, >> frank >> >> >> On 3/05/12 7:11 PM, Frank Rueter | OHUfx wrote: >>> Indeed, my apologies. >>> Will strip out the relevant parts and post an example soon. >>> >>> On 3/05/12 7:06 PM, Srini Kommoori wrote: >>>> Code snippet would help in understanding what you are asking. >>>> >>>> For most of my cases this works great: >>>> pushButton.clicked.connect(findFile) >>>> >>>> If you are trying to implement your own style >>>> button, QPushButton also takes icon. >>>> >>>> >>>> On Wed, May 2, 2012 at 11:23 PM, Frank Rueter | OHUfx >>>> > wrote: >>>> >>>> Any takers? >>>> I'm kinda stuck with this one unfortunately. >>>> >>>> >>>> On 2/05/12 6:35 PM, Frank Rueter | OHUfx wrote: >>>> > sorry for the monolog, but I just realised I will also >>>> have to make my >>>> > custom button play along when it's parent's self.sender() >>>> function is >>>> > called as well. >>>> > The QSignalSender doesn't quite seem to be the right >>>> thing to use as it >>>> > is implemented in the parent's constructor, however I >>>> would like to >>>> > implement this in the button's class if possible to make >>>> it self contained. >>>> > >>>> > How would I go about that? >>>> > >>>> > Cheers, >>>> > frank >>>> > >>>> > On 2/05/12 5:44 PM, Frank Rueter | OHUfx wrote: >>>> >> Hi, >>>> >> >>>> >> to follow up my previous mail, I also just noticed that >>>> I have to >>>> >> re-implement the QAbstractButton's setDown() function to >>>> make sure my >>>> >> custom button doesn't stay down after a drag&drop action. >>>> >> >>>> >> With a QPushButton I simply called self.setDown( False ) >>>> from inside the >>>> >> mouseMoveEvent, so now I'm trying to understand how to >>>> recreate this >>>> >> behaviour when using a custom paintEvent(). >>>> >> >>>> >> Any tips would be greatly appreciated. >>>> >> >>>> >> Cheers, >>>> >> frank >>>> >> _______________________________________________ >>>> >> 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 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 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 > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From vasure at gmail.com Fri May 4 08:56:24 2012 From: vasure at gmail.com (Srini Kommoori) Date: Thu, 3 May 2012 23:56:24 -0700 Subject: [PySide] Fwd: re-implementing setDown() for custom button In-Reply-To: References: <4FA0C9A9.5000909@ohufx.com> <4FA0D5B5.20604@ohufx.com> <4FA22478.9030007@ohufx.com> <4FA22FB1.6050306@ohufx.com> <4FA243D9.9000601@ohufx.com> <4FA24717.4060307@ohufx.com> <4FA2EB5C.8080604@ohufx.com> <4FA2F6C5.3@ohufx.com> Message-ID: Forgot to cc the list. May be useful for archival purposes. ---------- Forwarded message ---------- From: Srini Kommoori Date: Thu, May 3, 2012 at 11:50 PM Subject: Re: [PySide] re-implementing setDown() for custom button To: Frank Rueter | OHUfx Frank, Here is a simple working example. https://gist.github.com/2592661 I tried to debug your snippet but it was taking too long to see what is happening with the new button in your example. May be check with the working example that I have posted and compare with what you are doing in the new button. Hope that helps. -Srini On Thu, May 3, 2012 at 2:21 PM, Frank Rueter | OHUfx wrote: > Thanks. I'm just trying to see what I could do better, as I'm very new to > QT. > So you reckon it would just be better to use the dragMoveEvent virtual > function instead of the dragEnterEvent? > > I can't see a difference between the two in my code even though they > should make the button itself receive drag events, right?! So I guess I > should see the plus icon after all like I do with the defaul tbutton, but I > don't. > > I must still be missing something obvious. > > Cheers, > frank > > > On 4/05/12 9:15 AM, Srini Kommoori wrote: > > Sorry Safari auto correction making it illegible. > > What I meant was, I would prefer to have dragMoveEvent > with event.setDropAction(QtCore.Qt.CopyAction). > > You can ignore the comment as you have custom implemented the whole > pushbutton and its behavior. > > -Srini > > On Thu, May 3, 2012 at 1:32 PM, Frank Rueter | OHUfx wrote: > >> Hi Srini, >> >> >> >>(I would prefer to do have dragMoveEvent) >> I'm not quite sure what you mean. Can you elaborate? >> >> I did what you suggested but can't see any change (maybe because I'm on >> osx?). >> >> Cheers, >> frank >> >> >> On 3/05/12 9:49 PM, Srini Kommoori wrote: >> >> Thats great. >> >> One thing I observed is: for your custom button, if you want a >> nice/native copy icon to appear - set following in your dragEnterEvent(I >> would prefer to do have dragMoveEvent) >> >> event.setDropAction(QtCore.Qt.CopyAction) >> event.accept() >> >> -Srini >> >> On Thu, May 3, 2012 at 1:51 AM, Frank Rueter | OHUfx wrote: >> >>> DOH. Minutes after posting this I realised that the little plus on the >>> cursor only shows up on a "droppable" area, which my custom button is not >>> (and the default one is I guess). >>> In other words I just solved all the problem myself by trying to strip >>> it all down for a test to post. >>> Sorry, will try and be a bit more accountable next time. >>> >>> >>> >>> >>> On 3/05/12 8:37 PM, Frank Rueter | OHUfx wrote: >>> >>> Here is some example code that I ripped out of my main code: >>> http://pastebin.com/jXZXXebW >>> >>> It compares a sub classed QPushButton with one I painted from scratch. >>> >>> I was actually able to solve some of the problems in the process (and >>> for some reason the QObject.sender() works as well now) , but the one thing >>> that still doesn't work with the new button is when I try to set a drag >>> action like this: >>> drag.start( Qt.CopyAction ) >>> >>> The drag action for the "new button" is in line 192 which doesn't seem >>> to have any effect when dragging the button, while the one for the "old >>> button" in line 27 works as expected and turns the mouse cursor into a >>> little plus sign when dragging. >>> >>> What am I missing? >>> >>> >>> Cheers, >>> frank >>> >>> >>> On 3/05/12 7:11 PM, Frank Rueter | OHUfx wrote: >>> >>> Indeed, my apologies. >>> Will strip out the relevant parts and post an example soon. >>> >>> On 3/05/12 7:06 PM, Srini Kommoori wrote: >>> >>> Code snippet would help in understanding what you are asking. >>> >>> For most of my cases this works great: >>> pushButton.clicked.connect(findFile) >>> >>> If you are trying to implement your own style button, QPushButton also >>> takes icon. >>> >>> >>> On Wed, May 2, 2012 at 11:23 PM, Frank Rueter | OHUfx wrote: >>> >>>> Any takers? >>>> I'm kinda stuck with this one unfortunately. >>>> >>>> >>>> On 2/05/12 6:35 PM, Frank Rueter | OHUfx wrote: >>>> > sorry for the monolog, but I just realised I will also have to make my >>>> > custom button play along when it's parent's self.sender() function is >>>> > called as well. >>>> > The QSignalSender doesn't quite seem to be the right thing to use as >>>> it >>>> > is implemented in the parent's constructor, however I would like to >>>> > implement this in the button's class if possible to make it self >>>> contained. >>>> > >>>> > How would I go about that? >>>> > >>>> > Cheers, >>>> > frank >>>> > >>>> > On 2/05/12 5:44 PM, Frank Rueter | OHUfx wrote: >>>> >> Hi, >>>> >> >>>> >> to follow up my previous mail, I also just noticed that I have to >>>> >> re-implement the QAbstractButton's setDown() function to make sure my >>>> >> custom button doesn't stay down after a drag&drop action. >>>> >> >>>> >> With a QPushButton I simply called self.setDown( False ) from inside >>>> the >>>> >> mouseMoveEvent, so now I'm trying to understand how to recreate this >>>> >> behaviour when using a custom paintEvent(). >>>> >> >>>> >> Any tips would be greatly appreciated. >>>> >> >>>> >> Cheers, >>>> >> frank >>>> >> _______________________________________________ >>>> >> 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 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 >>> >>> >>> >>> _______________________________________________ >>> PySide mailing listPySide at qt-project.orghttp://lists.qt-project.org/mailman/listinfo/pyside >>> >>> >>> _______________________________________________ >>> PySide mailing list >>> PySide at qt-project.org >>> http://lists.qt-project.org/mailman/listinfo/pyside >>> >>> >> > -------------- next part -------------- An HTML attachment was scrubbed... URL: From frank at ohufx.com Fri May 4 09:19:07 2012 From: frank at ohufx.com (Frank Rueter | OHUfx) Date: Fri, 04 May 2012 19:19:07 +1200 Subject: [PySide] Fwd: re-implementing setDown() for custom button In-Reply-To: References: <4FA0C9A9.5000909@ohufx.com> <4FA0D5B5.20604@ohufx.com> <4FA22478.9030007@ohufx.com> <4FA22FB1.6050306@ohufx.com> <4FA243D9.9000601@ohufx.com> <4FA24717.4060307@ohufx.com> <4FA2EB5C.8080604@ohufx.com> <4FA2F6C5.3@ohufx.com> Message-ID: <4FA382EB.1080408@ohufx.com> oops, indeed. this was my reply: Hm, I just realised that none of my drag events seem to have much of an effect, and the actual drag behaviour is inside the mouseMoveEvent(), which seems wrong, but works. I will have a closer look again, but this is very confusing. On 4/05/12 6:56 PM, Srini Kommoori wrote: > Forgot to cc the list. May be useful for archival purposes. > > ---------- Forwarded message ---------- > From: *Srini Kommoori* > > Date: Thu, May 3, 2012 at 11:50 PM > Subject: Re: [PySide] re-implementing setDown() for custom button > To: Frank Rueter | OHUfx > > > > Frank, > > Here is a simple working example. > > https://gist.github.com/2592661 > > I tried to debug your snippet but it was taking too long to see what > is happening with the new button in your example. > > May be check with the working example that I have posted and compare > with what you are doing in the new button. > > Hope that helps. > > -Srini > > > On Thu, May 3, 2012 at 2:21 PM, Frank Rueter | OHUfx > wrote: > > Thanks. I'm just trying to see what I could do better, as I'm very > new to QT. > So you reckon it would just be better to use the dragMoveEvent > virtual function instead of the dragEnterEvent? > > I can't see a difference between the two in my code even though > they should make the button itself receive drag events, right?! So > I guess I should see the plus icon after all like I do with the > defaul tbutton, but I don't. > > I must still be missing something obvious. > > Cheers, > frank > > > On 4/05/12 9:15 AM, Srini Kommoori wrote: >> Sorry Safari auto correction making it illegible. >> >> What I meant was, I would prefer to have dragMoveEvent >> with event.setDropAction(QtCore.Qt.CopyAction). >> >> You can ignore the comment as you have custom implemented the >> whole pushbutton and its behavior. >> >> -Srini >> >> On Thu, May 3, 2012 at 1:32 PM, Frank Rueter | OHUfx >> > wrote: >> >> Hi Srini, >> >> >> >>(I would prefer to do have dragMoveEvent) >> I'm not quite sure what you mean. Can you elaborate? >> >> I did what you suggested but can't see any change (maybe >> because I'm on osx?). >> >> Cheers, >> frank >> >> >> On 3/05/12 9:49 PM, Srini Kommoori wrote: >>> Thats great. >>> >>> One thing I observed is: for your custom button, if you want >>> a nice/native copy icon to appear - set following in your >>> dragEnterEvent(I would prefer to do have dragMoveEvent) >>> >>> event.setDropAction(QtCore.Qt.CopyAction) >>> event.accept() >>> >>> -Srini >>> >>> On Thu, May 3, 2012 at 1:51 AM, Frank Rueter | OHUfx >>> > wrote: >>> >>> DOH. Minutes after posting this I realised that the >>> little plus on the cursor only shows up on a "droppable" >>> area, which my custom button is not (and the default one >>> is I guess). >>> In other words I just solved all the problem myself by >>> trying to strip it all down for a test to post. >>> Sorry, will try and be a bit more accountable next time. >>> >>> >>> >>> >>> On 3/05/12 8:37 PM, Frank Rueter | OHUfx wrote: >>>> Here is some example code that I ripped out of my main >>>> code: >>>> http://pastebin.com/jXZXXebW >>>> >>>> It compares a sub classed QPushButton with one I >>>> painted from scratch. >>>> >>>> I was actually able to solve some of the problems in >>>> the process (and for some reason the QObject.sender() >>>> works as well now) , but the one thing that still >>>> doesn't work with the new button is when I try to set a >>>> drag action like this: >>>> drag.start( Qt.CopyAction ) >>>> >>>> The drag action for the "new button" is in line 192 >>>> which doesn't seem to have any effect when dragging the >>>> button, while the one for the "old button" in line 27 >>>> works as expected and turns the mouse cursor into a >>>> little plus sign when dragging. >>>> >>>> What am I missing? >>>> >>>> >>>> Cheers, >>>> frank >>>> >>>> >>>> On 3/05/12 7:11 PM, Frank Rueter | OHUfx wrote: >>>>> Indeed, my apologies. >>>>> Will strip out the relevant parts and post an example >>>>> soon. >>>>> >>>>> On 3/05/12 7:06 PM, Srini Kommoori wrote: >>>>>> Code snippet would help in understanding what you are >>>>>> asking. >>>>>> >>>>>> For most of my cases this works great: >>>>>> pushButton.clicked.connect(findFile) >>>>>> >>>>>> If you are trying to implement your own style >>>>>> button, QPushButton also takes icon. >>>>>> >>>>>> >>>>>> On Wed, May 2, 2012 at 11:23 PM, Frank Rueter | OHUfx >>>>>> > wrote: >>>>>> >>>>>> Any takers? >>>>>> I'm kinda stuck with this one unfortunately. >>>>>> >>>>>> >>>>>> On 2/05/12 6:35 PM, Frank Rueter | OHUfx wrote: >>>>>> > sorry for the monolog, but I just realised I >>>>>> will also have to make my >>>>>> > custom button play along when it's parent's >>>>>> self.sender() function is >>>>>> > called as well. >>>>>> > The QSignalSender doesn't quite seem to be the >>>>>> right thing to use as it >>>>>> > is implemented in the parent's constructor, >>>>>> however I would like to >>>>>> > implement this in the button's class if >>>>>> possible to make it self contained. >>>>>> > >>>>>> > How would I go about that? >>>>>> > >>>>>> > Cheers, >>>>>> > frank >>>>>> > >>>>>> > On 2/05/12 5:44 PM, Frank Rueter | OHUfx wrote: >>>>>> >> Hi, >>>>>> >> >>>>>> >> to follow up my previous mail, I also just >>>>>> noticed that I have to >>>>>> >> re-implement the QAbstractButton's setDown() >>>>>> function to make sure my >>>>>> >> custom button doesn't stay down after a >>>>>> drag&drop action. >>>>>> >> >>>>>> >> With a QPushButton I simply called >>>>>> self.setDown( False ) from inside the >>>>>> >> mouseMoveEvent, so now I'm trying to >>>>>> understand how to recreate this >>>>>> >> behaviour when using a custom paintEvent(). >>>>>> >> >>>>>> >> Any tips would be greatly appreciated. >>>>>> >> >>>>>> >> Cheers, >>>>>> >> frank >>>>>> >> _______________________________________________ >>>>>> >> 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 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 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 list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside -------------- next part -------------- An HTML attachment was scrubbed... URL: From tommaso.vinci at polytechnique.fr Fri May 4 09:27:25 2012 From: tommaso.vinci at polytechnique.fr (Tommaso Vinci) Date: Fri, 04 May 2012 09:27:25 +0200 Subject: [PySide] supehybrids Message-ID: <20120504092725.19056oc1ibafzjks@webmail.enseignement.polytechnique.fr> Dear all, (I'm not sure you received this email, I had some problem with gmail forwarding, sorry for double posting) I'm trying do do exactly what superhybrids tutorial (http://lynxline.com/superhybrids-part-2-now-qt-pyside/) does, but I get in trouble at a certain point: in the build.sh, after running generatorruner, it does a qmake in the hybridpy directory and it complains by WARNING: Failure to find: PyHybrid/mainwindow_wrapper.cpp and then it can't compile. I don't understand the whole generatorruner stuff but I think that that file should have been generated at that moment I've uploaded here : http://dl.free.fr/pMld77LUj the .zip stuff (I had to modify a bit 'cause I'm on osx...) and here is just the generatorrunner command: cd hybridpy QTGUI_INC=/Library/Frameworks/QtGui.framework/Versions/4/Headers QTCORE_INC=/Library/Frameworks/QtCore.framework/Versions/4/Headers QTTYPESYSTEM=/usr/share/PySide/typesystems generatorrunner --generator-set=/usr/lib/generatorrunner/shiboken_generator.dylib \ --include-paths=../hybrid:$QTCORE_INC:$QTGUI_INC:/usr/include \ --typesystem-paths=../data:$QTTYPESYSTEM \ --output-directory=. \ ../data/global.h \ ../data/typesystem.xml \ --enable-pyside-extensions \ --debug-level=full did anyone succeed in doing this tutorial? many many thanks tommaso ---------------------------------------------------------------- This message was sent using X-WebMail From anomalous_underdog at yahoo.com Sun May 6 10:59:09 2012 From: anomalous_underdog at yahoo.com (anomalous underdog) Date: Sun, 6 May 2012 01:59:09 -0700 (PDT) Subject: [PySide] How to uninstall pyside on Mac OS X? Message-ID: <1336294749.24641.YahooMailNeo@web46108.mail.sp1.yahoo.com> I'm using Mac OS X 10.6.8, and I installed pyside-1.1.0-qt48-py27apple.pkg, I want to remove it. How do I do that? -------------- next part -------------- An HTML attachment was scrubbed... URL: From ssorgatem at gmail.com Tue May 8 19:52:17 2012 From: ssorgatem at gmail.com (=?UTF-8?Q?Adri=C3=A0_Cereto_Massagu=C3=A9?=) Date: Tue, 8 May 2012 10:52:17 -0700 (PDT) Subject: [PySide] Building PySide for Android In-Reply-To: <409292.2008.1336499327369.JavaMail.geo-discussion-forums@vbez20> References: <7303564.14.1336488126383.JavaMail.geo-discussion-forums@ynjc16> <409292.2008.1336499327369.JavaMail.geo-discussion-forums@vbez20> Message-ID: <26051837.2816.1336499537101.JavaMail.geo-discussion-forums@yncd3> Hi again, I've managed to get PySide/QtCore.so built, (typesystem_core_android.xml did not disappear, after all) but now it's QtGui who's not wanting to build. Here's the output: [ 20%] Building CXX object > PySide/QtGui/CMakeFiles/QtGui.dir/PySide/QtGui/qcursor_wrapper.cpp.o > /home/ssorgatem/pyside-android/BuildScripts/pyside/build/PySide/QtGui/PySide/QtGui/qcursor_wrapper.cpp: > In function 'int Sbk_QCursor_Init(PyObject*, PyObject*, PyObject*)': > /home/ssorgatem/pyside-android/BuildScripts/pyside/build/PySide/QtGui/PySide/QtGui/qcursor_wrapper.cpp:158:45: > error: invalid conversion from 'Qt::HANDLE {aka void*}' to 'const char* > const*' [-fpermissive] > /home/ssorgatem/necessitas/Android/Qt/480/armeabi/include/QtGui/qpixmap.h:80:5: > error: initializing argument 1 of 'QPixmap::QPixmap(const char* const*)' > [-fpermissive] > /home/ssorgatem/pyside-android/BuildScripts/pyside/build/PySide/QtGui/PySide/QtGui/qcursor_wrapper.cpp: > In function 'void PySide_QtCore_Qt_HANDLE_PythonToCpp_QCursor(PyObject*, > void*)': > /home/ssorgatem/pyside-android/BuildScripts/pyside/build/PySide/QtGui/PySide/QtGui/qcursor_wrapper.cpp:895:44: > error: invalid conversion from 'Qt::HANDLE {aka void*}' to 'const char* > const*' [-fpermissive] > /home/ssorgatem/necessitas/Android/Qt/480/armeabi/include/QtGui/qpixmap.h:80:5: > error: initializing argument 1 of 'QPixmap::QPixmap(const char* const*)' > [-fpermissive] > make[2]: *** > [PySide/QtGui/CMakeFiles/QtGui.dir/PySide/QtGui/qcursor_wrapper.cpp.o] > Error 1 > make[1]: *** [PySide/QtGui/CMakeFiles/QtGui.dir/all] Error 2 > make: *** [all] Error 2 > I have no clue about what to do next. Any help would be greatly appreciated. Thank you, Adrià El dimarts 8 de maig de 2012 16:42:06 UTC+2, Adrià Cereto Massagué va escriure: > > Hi, > > I've been attempting to build PySide 1.1.1 for android, with python2.7 > from https://github.com/kivy/python-for-android > and using Necessita's bundled NDK r6b with Linaro's toolchain 4.6.2. > > I'm using adapted versions of Thomas Perl's patches. > > So far, I've been able to build libshiboken, but i has been built as > libshiboken.so while PySide looks for libshiboken-python27.so. > A symlink solved that. > Then PySide would complain about missing shiboken binary... then a symlink > against the host machine shiboken binary solved that too. > And now I'm stuck trying to generate the bindings for QtCore: > > Linking CXX shared library libpyside.so > [ 3%] Built target pyside > [ 4%] Running generator for QtCore... > Fatal error: line=22, column=71, message=Failed to parse: > 'typesystem_core_android.xml' > > Cannot parse file: > /home/ssorgatem/pyside-android/BuildScripts/pyside/build/PySide/QtCore/typesystem_core.xmlmake[2]: > *** [PySide/QtCore/PySide/QtCore/qabstracteventdispatcher_wrapper.cpp] > Error 1 > make[1]: *** [PySide/QtCore/CMakeFiles/QtCore.dir/all] Error 2 > make: *** [all] Error 2 > > The relevant contents on typesystem_core.xml are: > > >> >> >> >> > > But typesystem_core_android.xml does not exist. If I write it by hand it > disappears again, so it's being removed, and > therefore I guess it should be generated in some way... but how? > Any ideas? > > Thank you, > > Adrià > > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From mairas at iki.fi Wed May 9 17:41:00 2012 From: mairas at iki.fi (Matti Airas) Date: Wed, 9 May 2012 18:41:00 +0300 Subject: [PySide] PySide website, old Bugzilla Message-ID: Hi list, I've still been trying to arrange the move of the PySide website to Qt Project. It might be that the only thing we get is a redirect from www.pyside.org is to the Wiki main page (which would need to be adjusted accordingly). I'm a bit tired of haggling that, however, so I hope that's all right. Related to the website migration: the old Bugzilla is hosted at the virtual machine that's going to go away in 6 weeks or so. IMHO, the old bug reports have a great historic value and should optimally be preserved? Would someone be willing to host that? I'd be happy to provide a tarball of the Bugzilla directory + a dump of the MySQL database (scrubbed clean of usernames and passwords) if someone would be willing to host them. Cheers, ma. From techtonik at gmail.com Wed May 9 18:08:36 2012 From: techtonik at gmail.com (anatoly techtonik) Date: Wed, 9 May 2012 19:08:36 +0300 Subject: [PySide] PySide website, old Bugzilla In-Reply-To: References: Message-ID: On Wed, May 9, 2012 at 6:41 PM, Matti Airas wrote: > > I've still been trying to arrange the move of the PySide website to Qt > Project. It might be that the only thing we get is a redirect from > www.pyside.org is to the Wiki main page (which would need to be > adjusted accordingly). I'm a bit tired of haggling that, however, so I > hope that's all right. If it is solely an HTML content, we can host it on GitHub - http://help.github.com/pages/ > Related to the website migration: the old Bugzilla is hosted at the > virtual machine that's going to go away in 6 weeks or so. IMHO, the > old bug reports have a great historic value and should optimally be > preserved? Would someone be willing to host that? I'd be happy to > provide a tarball of the Bugzilla directory + a dump of the MySQL > database (scrubbed clean of usernames and passwords) if someone would > be willing to host them. Can you upload it to Google Drive from your own account and share with some people, so it won't be lost at least? Maybe somebody eventually find a way to export Qt-related bugs to Jira from Bugzilla and sync upstream status with them? Like launchpad does. -- anatoly t. From hugo.lima at openbossa.org Wed May 9 19:35:52 2012 From: hugo.lima at openbossa.org (Hugo Parente Lima) Date: Wed, 09 May 2012 14:35:52 -0300 Subject: [PySide] PySide website, old Bugzilla In-Reply-To: References: Message-ID: <2266321.qM4WJ7Cybe@hugodesktop> On Wednesday, May 09, 2012 07:08:36 PM anatoly techtonik wrote: > On Wed, May 9, 2012 at 6:41 PM, Matti Airas wrote: > > I've still been trying to arrange the move of the PySide website to Qt > > Project. It might be that the only thing we get is a redirect from > > www.pyside.org is to the Wiki main page (which would need to be > > adjusted accordingly). I'm a bit tired of haggling that, however, so I > > hope that's all right. > > If it is solely an HTML content, we can host it on GitHub - > http://help.github.com/pages/ yes, would be nice if we could turn the current bugzilla into static pages and put them on github pages for reference. I'm using github pages for my pet projects and they work like a charm! Maybe this can be achieved just using wget on bugzilla and stripping out the form fields to have a nice result =]. > > Related to the website migration: the old Bugzilla is hosted at the > > virtual machine that's going to go away in 6 weeks or so. IMHO, the > > old bug reports have a great historic value and should optimally be > > preserved? Would someone be willing to host that? I'd be happy to > > provide a tarball of the Bugzilla directory + a dump of the MySQL > > database (scrubbed clean of usernames and passwords) if someone would > > be willing to host them. > > Can you upload it to Google Drive from your own account and share with > some people, so it won't be lost at least? Maybe somebody eventually > find a way to export Qt-related bugs to Jira from Bugzilla and sync > upstream status with them? Like launchpad does. > -- > anatoly t. > _______________________________________________ > PySide mailing list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part. URL: From vasure at gmail.com Wed May 9 20:11:56 2012 From: vasure at gmail.com (Srini) Date: Wed, 9 May 2012 11:11:56 -0700 Subject: [PySide] PySide website, old Bugzilla In-Reply-To: <2266321.qM4WJ7Cybe@hugodesktop> References: <2266321.qM4WJ7Cybe@hugodesktop> Message-ID: Matti, I will go-ahead and convert bugzilla pages into HTML pages - a simple script would work. So are we going to host everything else on wiki? What about docs? -Srini On May 9, 2012, at 10:35 AM, Hugo Parente Lima wrote: > On Wednesday, May 09, 2012 07:08:36 PM anatoly techtonik wrote: >> On Wed, May 9, 2012 at 6:41 PM, Matti Airas wrote: >>> I've still been trying to arrange the move of the PySide website to Qt >>> Project. It might be that the only thing we get is a redirect from >>> www.pyside.org is to the Wiki main page (which would need to be >>> adjusted accordingly). I'm a bit tired of haggling that, however, so I >>> hope that's all right. >> >> If it is solely an HTML content, we can host it on GitHub - >> http://help.github.com/pages/ > > yes, would be nice if we could turn the current bugzilla into static pages and > put them on github pages for reference. I'm using github pages for my pet > projects and they work like a charm! > > Maybe this can be achieved just using wget on bugzilla and stripping out the > form fields to have a nice result =]. > >>> Related to the website migration: the old Bugzilla is hosted at the >>> virtual machine that's going to go away in 6 weeks or so. IMHO, the >>> old bug reports have a great historic value and should optimally be >>> preserved? Would someone be willing to host that? I'd be happy to >>> provide a tarball of the Bugzilla directory + a dump of the MySQL >>> database (scrubbed clean of usernames and passwords) if someone would >>> be willing to host them. >> >> Can you upload it to Google Drive from your own account and share with >> some people, so it won't be lost at least? Maybe somebody eventually >> find a way to export Qt-related bugs to Jira from Bugzilla and sync >> upstream status with them? Like launchpad does. >> -- >> anatoly t. >> _______________________________________________ >> 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 From hugo.lima at openbossa.org Wed May 9 20:17:21 2012 From: hugo.lima at openbossa.org (Hugo Parente Lima) Date: Wed, 09 May 2012 15:17:21 -0300 Subject: [PySide] PySide website, old Bugzilla In-Reply-To: References: <2266321.qM4WJ7Cybe@hugodesktop> Message-ID: <2494556.N4Z90BUQpD@hugodesktop> On Wednesday, May 09, 2012 11:11:56 AM Srini wrote: > Matti, I will go-ahead and convert bugzilla pages into HTML pages - a simple > script would work. > > So are we going to host everything else on wiki? What about docs? The docs can be placed on github pages too, but would be nice to have github repositories automatically sync'ed with the ones on gerrit like on gitorious. > -Srini > > On May 9, 2012, at 10:35 AM, Hugo Parente Lima wrote: > > On Wednesday, May 09, 2012 07:08:36 PM anatoly techtonik wrote: > >> On Wed, May 9, 2012 at 6:41 PM, Matti Airas wrote: > >>> I've still been trying to arrange the move of the PySide website to Qt > >>> Project. It might be that the only thing we get is a redirect from > >>> www.pyside.org is to the Wiki main page (which would need to be > >>> adjusted accordingly). I'm a bit tired of haggling that, however, so I > >>> hope that's all right. > >> > >> If it is solely an HTML content, we can host it on GitHub - > >> http://help.github.com/pages/ > > > > yes, would be nice if we could turn the current bugzilla into static pages > > and put them on github pages for reference. I'm using github pages for my > > pet projects and they work like a charm! > > > > Maybe this can be achieved just using wget on bugzilla and stripping out > > the form fields to have a nice result =]. > > > >>> Related to the website migration: the old Bugzilla is hosted at the > >>> virtual machine that's going to go away in 6 weeks or so. IMHO, the > >>> old bug reports have a great historic value and should optimally be > >>> preserved? Would someone be willing to host that? I'd be happy to > >>> provide a tarball of the Bugzilla directory + a dump of the MySQL > >>> database (scrubbed clean of usernames and passwords) if someone would > >>> be willing to host them. > >> > >> Can you upload it to Google Drive from your own account and share with > >> some people, so it won't be lost at least? Maybe somebody eventually > >> find a way to export Qt-related bugs to Jira from Bugzilla and sync > >> upstream status with them? Like launchpad does. > >> -- > >> anatoly t. > >> _______________________________________________ > >> 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 -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part. URL: From techtonik at gmail.com Wed May 9 20:24:22 2012 From: techtonik at gmail.com (anatoly techtonik) Date: Wed, 9 May 2012 21:24:22 +0300 Subject: [PySide] PySide website, old Bugzilla In-Reply-To: <2494556.N4Z90BUQpD@hugodesktop> References: <2266321.qM4WJ7Cybe@hugodesktop> <2494556.N4Z90BUQpD@hugodesktop> Message-ID: On Wed, May 9, 2012 at 9:17 PM, Hugo Parente Lima wrote: > On Wednesday, May 09, 2012 11:11:56 AM Srini wrote: >> Matti, I will go-ahead and convert bugzilla pages into HTML pages - a simple >> script would work. >> >> So are we going to host everything else on wiki? What about docs? > > The docs can be placed on github pages too, but would be nice to have github > repositories automatically sync'ed with the ones on gerrit like on gitorious. Well, I was thinking only about web part - not about Bugzilla in particular. As for documentation pages, where is the machine that does repository sync? -- anatoly t. From hugo.lima at openbossa.org Wed May 9 22:08:24 2012 From: hugo.lima at openbossa.org (Hugo Parente Lima) Date: Wed, 09 May 2012 17:08:24 -0300 Subject: [PySide] PySide website, old Bugzilla In-Reply-To: References: <2494556.N4Z90BUQpD@hugodesktop> Message-ID: <3190885.2NbcHgPhKV@hugodesktop> On Wednesday, May 09, 2012 09:24:22 PM anatoly techtonik wrote: > On Wed, May 9, 2012 at 9:17 PM, Hugo Parente Lima > > wrote: > > On Wednesday, May 09, 2012 11:11:56 AM Srini wrote: > >> Matti, I will go-ahead and convert bugzilla pages into HTML pages - a > >> simple script would work. > >> > >> So are we going to host everything else on wiki? What about docs? > > > > The docs can be placed on github pages too, but would be nice to have > > github repositories automatically sync'ed with the ones on gerrit like on > > gitorious. > Well, I was thinking only about web part - not about Bugzilla in > particular. As for documentation pages, where is the machine that does > repository sync? The bugzilla can be considered as documentation too, many unit tests are just "bug_XXX.py", and when they fail due to a regression the natural way to understand the issue is to go to bugzilla first to read the bug comments/description. > -- > anatoly t. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part. URL: From vasure at gmail.com Thu May 10 02:13:29 2012 From: vasure at gmail.com (Srini Kommoori) Date: Wed, 9 May 2012 17:13:29 -0700 Subject: [PySide] PySide website, old Bugzilla In-Reply-To: <3190885.2NbcHgPhKV@hugodesktop> References: <2494556.N4Z90BUQpD@hugodesktop> <3190885.2NbcHgPhKV@hugodesktop> Message-ID: Here is the archived bugzilla db http://srinikom.github.com/pyside-bz-archive/ All the scripts and html pages are on master branch at: https://github.com/srinikom/pyside-bz-archive We could do something similar for the whole documentation and other static pages. Could you guys point me to the doc generating scripts/steps and I will try to automate it. thanks, -Srini On Wed, May 9, 2012 at 1:08 PM, Hugo Parente Lima wrote: > On Wednesday, May 09, 2012 09:24:22 PM anatoly techtonik wrote: > > On Wed, May 9, 2012 at 9:17 PM, Hugo Parente Lima > > > > wrote: > > > On Wednesday, May 09, 2012 11:11:56 AM Srini wrote: > > >> Matti, I will go-ahead and convert bugzilla pages into HTML pages - a > > >> simple script would work. > > >> > > >> So are we going to host everything else on wiki? What about docs? > > > > > > The docs can be placed on github pages too, but would be nice to have > > > github repositories automatically sync'ed with the ones on gerrit like > on > > > gitorious. > > Well, I was thinking only about web part - not about Bugzilla in > > particular. As for documentation pages, where is the machine that does > > repository sync? > > The bugzilla can be considered as documentation too, many unit tests are > just > "bug_XXX.py", and when they fail due to a regression the natural way to > understand the issue is to go to bugzilla first to read the bug > comments/description. > > > -- > > anatoly t. -------------- next part -------------- An HTML attachment was scrubbed... URL: From mairas at iki.fi Thu May 10 08:16:36 2012 From: mairas at iki.fi (Matti Airas) Date: Thu, 10 May 2012 09:16:36 +0300 Subject: [PySide] PySide website, old Bugzilla In-Reply-To: References: <2494556.N4Z90BUQpD@hugodesktop> <3190885.2NbcHgPhKV@hugodesktop> Message-ID: Hi, On 10 May 2012 03:13, Srini Kommoori wrote: > Here is the archived bugzilla > db http://srinikom.github.com/pyside-bz-archive/ > All the scripts and html pages are on master branch > at: https://github.com/srinikom/pyside-bz-archive I'm sorry if I was a bit terse in my previous email. While a static mirror definitely is useful, I believe a full copy with complete search capability would be much preferable still. That's why I was proposing handing the database contents to anyone volunteering. > We could do something similar for the whole documentation and other static > pages. The plan was to provide the documentation at doc-snapshot.qt-project.org, and the binaries via releases.qt-project.org. This integration provides PySide much more status and visibility than a hobby project setup at github, and in my opinion is much preferable. Also, for the main site, it should be possible to have PySide more prominently visible on the qt-project.org site itself, and therefore the separate site wouldn't bring that much added benefit. > Could you guys point me to the doc generating scripts/steps and I will try > to automate it. That would be very good - but for doc-snapshot.qt-project.org. :-) ma. From vasure at gmail.com Thu May 10 09:57:02 2012 From: vasure at gmail.com (Srini Kommoori) Date: Thu, 10 May 2012 00:57:02 -0700 Subject: [PySide] PySide website, old Bugzilla In-Reply-To: References: <2494556.N4Z90BUQpD@hugodesktop> <3190885.2NbcHgPhKV@hugodesktop> Message-ID: >mirror definitely is useful, I believe a full copy with complete >search capability would be much preferable still. We could add google custom search box and remove all the dead links from static pages to make sure whoever is looking at the archive doesn't get confused and also get the search feature. > proposing handing the database contents to anyone volunteering. That is the problem. If we find someone to host archive bugzilla, then we wouldn't have this discussion. Calling github for hobby projects is little weird. For reference: https://github.com/torvalds/linux / redis/mongo/django Could you please let us know what is the final plan for everything? Do you think VM with wordpress is the best option? I could sponsor that but for long term, keeping everything simple and automated is the best option for the project to be successful. thanks, -Srini On Wed, May 9, 2012 at 11:16 PM, Matti Airas wrote: > Hi, > > On 10 May 2012 03:13, Srini Kommoori wrote: > > Here is the archived bugzilla > > db http://srinikom.github.com/pyside-bz-archive/ > > All the scripts and html pages are on master branch > > at: https://github.com/srinikom/pyside-bz-archive > > I'm sorry if I was a bit terse in my previous email. While a static > mirror definitely is useful, I believe a full copy with complete > search capability would be much preferable still. That's why I was > proposing handing the database contents to anyone volunteering. > > > We could do something similar for the whole documentation and other > static > > pages. > > The plan was to provide the documentation at > doc-snapshot.qt-project.org, and the binaries via > releases.qt-project.org. This integration provides PySide much more > status and visibility than a hobby project setup at github, and in my > opinion is much preferable. > > Also, for the main site, it should be possible to have PySide more > prominently visible on the qt-project.org site itself, and therefore > the separate site wouldn't bring that much added benefit. > > > Could you guys point me to the doc generating scripts/steps and I will > try > > to automate it. > > That would be very good - but for doc-snapshot.qt-project.org. :-) > > ma. > -------------- next part -------------- An HTML attachment was scrubbed... URL: From mairas at iki.fi Thu May 10 11:18:53 2012 From: mairas at iki.fi (Matti Airas) Date: Thu, 10 May 2012 12:18:53 +0300 Subject: [PySide] PySide website, old Bugzilla In-Reply-To: References: <2494556.N4Z90BUQpD@hugodesktop> <3190885.2NbcHgPhKV@hugodesktop> Message-ID: On 10 May 2012 10:57, Srini Kommoori wrote: >> proposing handing the database contents to anyone volunteering. > > That is the problem. If we find someone to host archive bugzilla, then we > wouldn't have this discussion. That's why I asked for that in the first place. :-) > Calling github for hobby projects is little weird. For > reference: https://github.com/torvalds/linux / redis/mongo/django Sorry - maybe I've just grown sensitive to the mentions of Github in this context. :-) But more generally, now that PySide is a Qt add-on and part of the larger Qt Project, I don't think it's a good idea to always have the first reaction of moving things *away* from Qt. It'd be a bit like working for an Apache project and always trying to move the facilities to, say, Ubuntu's Launchpad. I know I myself presented the idea of having a separate VM, but at that point I hadn't realized that PySide could have such an integral presence on qt-project.org. > Could you please let us know what is the final plan for everything? Do you > think VM with wordpress is the best option? I could sponsor that but for > long term, keeping everything simple and automated is the best option for > the project to be successful. Don't know about any final plans, but the current idea is to host the high-traffic static content (documentation and binaries) directly at the Qt Project, side-by-side with the other Qt Project resources. The little content on the WordPress website (basically, the Support and About pages) would need to be moved to the Wiki. The wiki homepage would need to be slightly revised accordingly to more easily provide an overview of the project. Then, PySide would also be presented somehow on the main Qt Project navigation hierarchy. The current news blog on PySide.org would need to be exported to some open blog provider such as blogger.com, but there we could try to use the opportunity to open the blog and use it as a more active conduit for all kinds of things related to PySide. I hope this clarifies the things a bit. Cheers, ma. From vasure at gmail.com Thu May 10 12:40:52 2012 From: vasure at gmail.com (Srini Kommoori) Date: Thu, 10 May 2012 03:40:52 -0700 Subject: [PySide] PySide website, old Bugzilla In-Reply-To: References: <2494556.N4Z90BUQpD@hugodesktop> <3190885.2NbcHgPhKV@hugodesktop> Message-ID: I will take the ownership in creating the blogger site. I will go-ahead and retain the current pyside.org style as much as possible. Will that work? As no one is volunteering, what are the plans for bugzilla archive? For the main site(docs and static content) is it going to be just the DNS remap? Are we going to maintain the pyside.org style or qt-project.org? thanks, -Srini On Thu, May 10, 2012 at 2:18 AM, Matti Airas wrote: > On 10 May 2012 10:57, Srini Kommoori wrote: > > >> proposing handing the database contents to anyone volunteering. > > > > That is the problem. If we find someone to host archive bugzilla, then we > > wouldn't have this discussion. > > That's why I asked for that in the first place. :-) > > > Calling github for hobby projects is little weird. For > > reference: https://github.com/torvalds/linux / redis/mongo/django > > Sorry - maybe I've just grown sensitive to the mentions of Github in > this context. :-) But more generally, now that PySide is a Qt add-on > and part of the larger Qt Project, I don't think it's a good idea to > always have the first reaction of moving things *away* from Qt. It'd > be a bit like working for an Apache project and always trying to move > the facilities to, say, Ubuntu's Launchpad. > > I know I myself presented the idea of having a separate VM, but at > that point I hadn't realized that PySide could have such an integral > presence on qt-project.org. > > > Could you please let us know what is the final plan for everything? Do > you > > think VM with wordpress is the best option? I could sponsor that but for > > long term, keeping everything simple and automated is the best option for > > the project to be successful. > > Don't know about any final plans, but the current idea is to host the > high-traffic static content (documentation and binaries) directly at > the Qt Project, side-by-side with the other Qt Project resources. The > little content on the WordPress website (basically, the Support and > About pages) would need to be moved to the Wiki. The wiki homepage > would need to be slightly revised accordingly to more easily provide > an overview of the project. Then, PySide would also be presented > somehow on the main Qt Project navigation hierarchy. > > The current news blog on PySide.org would need to be exported to some > open blog provider such as blogger.com, but there we could try to use > the opportunity to open the blog and use it as a more active conduit > for all kinds of things related to PySide. > > I hope this clarifies the things a bit. > > Cheers, > > ma. > -------------- next part -------------- An HTML attachment was scrubbed... URL: From frank at ohufx.com Thu May 10 14:15:21 2012 From: frank at ohufx.com (Frank Rueter | OHUfx) Date: Fri, 11 May 2012 00:15:21 +1200 Subject: [PySide] animated fade In-Reply-To: <4F8C086C.9060506@ohufx.com> References: <4F8C086C.9060506@ohufx.com> Message-ID: <4FABB159.5010204@ohufx.com> Hi, I'm just trying to re-jig this example code to create an animated StackedWidget: http://pastebin.com/eTDpxiKM Currently the above creates two classes, one "FaderWidget" class and one "StackedWidget" class. The two work together to cross dissolve between whatever widgets are passed into the FaderWidget's constructor. I am trying to re-write this to not have to pass the child widgets into the FaderWidget's constructor, but to use methods like "addWidget" to allow me to pass the children to an instance of the FaderWidget. In other words, I'm trying to subclass QStackedWidget and give it the cross dissolve animation when setCurrentIndex is called on it, therefore simplifying the way the widget works and not having to deal with two classes. Here is my attempt so far, but the problem is that the paintEvent is called to soon, i.e. before the child widgets are available to draw a pixmap of them. http://pastebin.com/LQaCSWDU Does anybody have an idea or comments on how to do this best? Maybe I should aim for a layout that does this instead (subclassing QStackedLayout)? Any help would be greatly appreciated. Cheers, frank On 4/16/12 11:54 PM, Frank Rueter | OHUfx wrote: > Hi guys, > > I'm currently looking at an example that cross fades two widgets when > the respective button is pushed. > > I have made a slight change to the sample code so that there is a button > on each page that is meant to start the cross fade animation back to the > other page (and therefore fade out itself). Looking at the terminal > output, I can see the animation running and printing the incremental > value changes, but the pages don't visibly animate when using the new > buttons. > > http://pastebin.com/eTDpxiKM > > When I use the old buttons at the bottom of the widget (the ones that > aren't part of the animated widgets), it all works as expected. Both > sets of buttons are hooked up to the same slot. > > What am I doing wrong? > > Cheers, > frank > > _______________________________________________ > PySide mailing list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside From blair at orcaware.com Fri May 11 03:06:17 2012 From: blair at orcaware.com (Blair Zajac) Date: Thu, 10 May 2012 18:06:17 -0700 Subject: [PySide] Binding unbound signals (need __get__) Message-ID: <4FAC6609.1000509@orcaware.com> We have some Python decorators that takes signals and we'd like to be able to convert a unbound signal into a bound signal. I don't see a way to do this in PySide, as in PyQt4 __get__ works to bind an unbound signal. From my reading, normal python usage lets you bind any method using __get__: class Bar(object): def method(self): pass bar = Bar() unbound_method = Bar.method bound_method = Bar.method.__get__(bar) Without __get__, the following doesn't work. Is there a way to do this in PySide? Thanks, Blair from PyQt4 import QtCore QtCore.Signal = QtCore.pyqtSignal QtCore.Slot = QtCore.pyqtSlot def emit_upon_success(signal): def f_(f): def f__(self): result = f(self) s = signal.__get__(self) s.emit() return result return f__ return f_ class Foo(QtCore.QObject): SIG = QtCore.Signal() @emit_upon_success(SIG) def do_something(self): pass foo = Foo() foo.do_something() From gds at mrxtech.com.au Mon May 14 03:46:21 2012 From: gds at mrxtech.com.au (Gerald Storer) Date: Mon, 14 May 2012 09:46:21 +0800 Subject: [PySide] PySide website, old Bugzilla In-Reply-To: References: Message-ID: <4FB063ED.6030907@mrxtech.com.au> Not that I have the time to do it, but is there any reason this isn't being considered? http://confluence.atlassian.com/display/JIRA044/Importing+Data+from+Bugzilla There is already a JIRA instance for Pyside and importing over the top of it could be difficult. Maybe it would be sensible to create a second "Pyside-archive" instance. At least all the bug reports will be in the same place then. Getting from Bugzilla 4.0 to JIRA 4.4 might be a bit tricky but I think there is a version of the importer that will do it: https://plugins.atlassian.com/plugins/com.atlassian.jira.plugins.jira-importers-plugin/version/36 Gerald On 9/05/2012 11:41 PM, Matti Airas wrote: > Hi list, > > I've still been trying to arrange the move of the PySide website to Qt > Project. It might be that the only thing we get is a redirect from > www.pyside.org is to the Wiki main page (which would need to be > adjusted accordingly). I'm a bit tired of haggling that, however, so I > hope that's all right. > > Related to the website migration: the old Bugzilla is hosted at the > virtual machine that's going to go away in 6 weeks or so. IMHO, the > old bug reports have a great historic value and should optimally be > preserved? Would someone be willing to host that? I'd be happy to > provide a tarball of the Bugzilla directory + a dump of the MySQL > database (scrubbed clean of usernames and passwords) if someone would > be willing to host them. > > Cheers, > > ma. > _______________________________________________ > PySide mailing list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside -- Gerald Storer | MRX Technologies +61 8 9227 4529 | gds at mrxtech.com.au | http://www.mrxtech.com.au From frank at ohufx.com Tue May 15 10:35:20 2012 From: frank at ohufx.com (Frank Rueter | OHUfx) Date: Tue, 15 May 2012 20:35:20 +1200 Subject: [PySide] QPushButton with text under icon Message-ID: <4FB21548.8000409@ohufx.com> Hi all, I thought this onw was going to be simple but I guess I was wrong: All Im' trying to do is draw the text of a QPushButton underneath it's icon. Is this possible? I can't find the answer and ended up using QToolButton instead, but that doesn't render the button as "pushed" when it's - well - pushed. This is what I'm after (not sure if the image will come through on the list): Any pointers would be great! Thanks, frank -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: image/jpeg Size: 15379 bytes Desc: not available URL: From techtonik at gmail.com Tue May 15 15:29:00 2012 From: techtonik at gmail.com (anatoly techtonik) Date: Tue, 15 May 2012 16:29:00 +0300 Subject: [PySide] Top priority issue half fixed in 1.1.1 Message-ID: Hi. With the release of PySide 1.1.1 the major segfault with cursors was fixed. However, the issue it still actual as setOverrideCursor() prints a lot of warnings on Linux (Fedora 16, Gnome 3). https://bugreports.qt-project.org/browse/PYSIDE-22 X Error: BadCursor (invalid Cursor parameter) 6 Major opcode: 2 (X_ChangeWindowAttributes) Resource id: 0x3 A test case is available in bug report. Would be nice if somebody knowledgeable can take a look and explain what happens. I am afraid there is no explanation from the Python world and it's very annoying. -- anatoly t. From mairas at iki.fi Tue May 15 19:49:25 2012 From: mairas at iki.fi (Matti Airas) Date: Tue, 15 May 2012 20:49:25 +0300 Subject: [PySide] PySide website, old Bugzilla In-Reply-To: <4FB063ED.6030907@mrxtech.com.au> References: <4FB063ED.6030907@mrxtech.com.au> Message-ID: On 14 May 2012 04:46, Gerald Storer wrote: > Not that I have the time to do it, but is there any reason this isn't > being considered? > http://confluence.atlassian.com/display/JIRA044/Importing+Data+from+Bugzilla Hi, It actually was considered, but probably not early enough. When I requested for the JIRA components to be created, it didn't occur to me that the bugs should be imported. Anatoly Techtonik later proposed the import to be done and made a request to the Qt admins, but I guess the request has been buried under more pressing issues (from their POV, that is). I'd guess in the short term it's going to be easier to just mirror the current Bugzilla instance somewhere... Cheers, ma. From frank at ohufx.com Tue May 15 23:11:17 2012 From: frank at ohufx.com (Frank Rueter | OHUfx) Date: Wed, 16 May 2012 09:11:17 +1200 Subject: [PySide] pyside user list Message-ID: <4FB2C675.8010401@ohufx.com> Hi everyone, comparing what you guys have been posting here to my user questions over the last few weeks makes me believe that I have probably joined the wrong group to get support whilst learning and using PySide. Can anybody recommend a pyside user group where I could get help with my problems without boring you guys to death ;) ? I was looking at the google group but that only seems to be a mirror of this list here. Thanking you in advance. Cheers, frank From a.richi at bluewin.ch Wed May 16 07:29:37 2012 From: a.richi at bluewin.ch (Aaron Richiger) Date: Wed, 16 May 2012 07:29:37 +0200 Subject: [PySide] pyside user list In-Reply-To: <4FB2C675.8010401@ohufx.com> References: <4FB2C675.8010401@ohufx.com> Message-ID: <4FB33B41.9030604@bluewin.ch> Hello Frank, dear list... I think, you are currently at the right place, afaik, there is no other list. It's no problem to post user questions here, I think it's important for a project like PySide to have a good list where Qt/PySide beginners may ask some questions. I also thought sometimes about splitting up the list in two parts, the user side and the Shiboken-CoreDev-etc. part, but today, it's still all the same. As soon as I'm a little bit less stressed, I will try to answer questions again, but at the moment, there are other things to do unfortunately;-) Have a nice day! Aaron From frank at ohufx.com Wed May 16 07:56:31 2012 From: frank at ohufx.com (Frank Rueter | OHUfx) Date: Wed, 16 May 2012 17:56:31 +1200 Subject: [PySide] pyside user list In-Reply-To: <4FB33B41.9030604@bluewin.ch> References: <4FB2C675.8010401@ohufx.com> <4FB33B41.9030604@bluewin.ch> Message-ID: <4FB3418F.6060500@ohufx.com> Hi Aaron, thanks for confirming that I'm in the right spot. I totally appreciate people not having time for constantly solving other people's problems. I was just beginning to wonder after many of my posts went completely unanswered. I certainly don't intend to annoy people with my beginners questions but sometimes it's hard to find the right info in which case a user list is heaven sent. Anyway, I appreciate all the help I have received here already, just wanted to make sure that I'm not barking up the wrong tree. Cheers, frank On 16/05/12 5:29 PM, Aaron Richiger wrote: > Hello Frank, dear list... > > I think, you are currently at the right place, afaik, there is no other > list. It's no problem to post user questions here, I think it's > important for a project like PySide to have a good list where Qt/PySide > beginners may ask some questions. > I also thought sometimes about splitting up the list in two parts, the > user side and the Shiboken-CoreDev-etc. part, but today, it's still all > the same. > As soon as I'm a little bit less stressed, I will try to answer > questions again, but at the moment, there are other things to do > unfortunately;-) > > Have a nice day! > Aaron > _______________________________________________ > PySide mailing list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside From antwebid at gmail.com Wed May 16 08:28:30 2012 From: antwebid at gmail.com (Laurence Anthony) Date: Wed, 16 May 2012 15:28:30 +0900 Subject: [PySide] pyside user list In-Reply-To: <4FB3418F.6060500@ohufx.com> References: <4FB2C675.8010401@ohufx.com> <4FB33B41.9030604@bluewin.ch> <4FB3418F.6060500@ohufx.com> Message-ID: I'm also happy to hear that user questions can be asked here. I'm one of those ghost members who just reads what's happening with interest. But, I plan to ask (and answer) questions when possible. Laurence. -------------- next part -------------- An HTML attachment was scrubbed... URL: From nathanjsmith at gmail.com Wed May 16 15:05:26 2012 From: nathanjsmith at gmail.com (Nathan Smith) Date: Wed, 16 May 2012 08:05:26 -0500 Subject: [PySide] pyside user list In-Reply-To: <4FB3418F.6060500@ohufx.com> References: <4FB2C675.8010401@ohufx.com> <4FB33B41.9030604@bluewin.ch> <4FB3418F.6060500@ohufx.com> Message-ID: Another place to try is stackoverflow.com. There is a pyside tag that you can put onto your questions. The nice thing about the similarity between PySide and PyQt is that even if nobody knows the answer to PySide directly, the PyQt answer just might work. Stackoverflow is most appropriate for bite sized concrete problems. Nathan On Wed, May 16, 2012 at 12:56 AM, Frank Rueter | OHUfx wrote: > Hi Aaron, > > thanks for confirming that I'm in the right spot. > I totally appreciate people not having time for constantly solving other > people's problems. I was just beginning to wonder after many of my posts > went completely unanswered. > I certainly don't intend to annoy people with my beginners questions but > sometimes it's hard to find the right info in which case a user list is > heaven sent. > > Anyway, I appreciate all the help I have received here already, just > wanted to make sure that I'm not barking up the wrong tree. > > Cheers, > frank > > > On 16/05/12 5:29 PM, Aaron Richiger wrote: > > Hello Frank, dear list... > > > > I think, you are currently at the right place, afaik, there is no other > > list. It's no problem to post user questions here, I think it's > > important for a project like PySide to have a good list where Qt/PySide > > beginners may ask some questions. > > I also thought sometimes about splitting up the list in two parts, the > > user side and the Shiboken-CoreDev-etc. part, but today, it's still all > > the same. > > As soon as I'm a little bit less stressed, I will try to answer > > questions again, but at the moment, there are other things to do > > unfortunately;-) > > > > Have a nice day! > > Aaron > > _______________________________________________ > > 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 > -------------- next part -------------- An HTML attachment was scrubbed... URL: From antwebid at gmail.com Wed May 16 15:37:09 2012 From: antwebid at gmail.com (Laurence Anthony) Date: Wed, 16 May 2012 22:37:09 +0900 Subject: [PySide] pyside user list In-Reply-To: References: <4FB2C675.8010401@ohufx.com> <4FB33B41.9030604@bluewin.ch> <4FB3418F.6060500@ohufx.com> Message-ID: Yes, stackoverflow is my main site for computer related issues. There are a lot of pyside questions and answers there already. Laurence. -------------- next part -------------- An HTML attachment was scrubbed... URL: From frank at ohufx.com Wed May 16 23:45:01 2012 From: frank at ohufx.com (Frank Rueter | OHUfx) Date: Thu, 17 May 2012 09:45:01 +1200 Subject: [PySide] pyside user list In-Reply-To: References: <4FB2C675.8010401@ohufx.com> <4FB33B41.9030604@bluewin.ch> <4FB3418F.6060500@ohufx.com> Message-ID: <4FB41FDD.5000108@ohufx.com> cool, thanks. I have actually found heaps of info on Stackoverflow already, just be reading up on other people's question, so yes, will join there. Cheers, frank On 5/17/12 1:05 AM, Nathan Smith wrote: > Another place to try is stackoverflow.com . > There is a pyside tag that you can put onto your questions. The nice > thing about the similarity between PySide and PyQt is that even if > nobody knows the answer to PySide directly, the PyQt answer just might > work. Stackoverflow is most appropriate for bite sized concrete > problems. > > Nathan > > On Wed, May 16, 2012 at 12:56 AM, Frank Rueter | OHUfx > > wrote: > > Hi Aaron, > > thanks for confirming that I'm in the right spot. > I totally appreciate people not having time for constantly solving > other > people's problems. I was just beginning to wonder after many of my > posts > went completely unanswered. > I certainly don't intend to annoy people with my beginners > questions but > sometimes it's hard to find the right info in which case a user > list is > heaven sent. > > Anyway, I appreciate all the help I have received here already, just > wanted to make sure that I'm not barking up the wrong tree. > > Cheers, > frank > > > On 16/05/12 5:29 PM, Aaron Richiger wrote: > > Hello Frank, dear list... > > > > I think, you are currently at the right place, afaik, there is > no other > > list. It's no problem to post user questions here, I think it's > > important for a project like PySide to have a good list where > Qt/PySide > > beginners may ask some questions. > > I also thought sometimes about splitting up the list in two > parts, the > > user side and the Shiboken-CoreDev-etc. part, but today, it's > still all > > the same. > > As soon as I'm a little bit less stressed, I will try to answer > > questions again, but at the moment, there are other things to do > > unfortunately;-) > > > > Have a nice day! > > Aaron > > _______________________________________________ > > 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 > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From p.indrakumar at gmail.com Thu May 17 09:18:35 2012 From: p.indrakumar at gmail.com (Indra kumar) Date: Thu, 17 May 2012 00:18:35 -0700 (PDT) Subject: [PySide] How to use animated SVG files using SvgRenderer Message-ID: <75610.219.1337239115919.JavaMail.geo-discussion-forums@pbboq8> Hi, I am trying to load svg file using SvgRenderer class. I am able to load the svg file properly. But how do we use those animated effects in the window. for ex:- I have one animated rect in svg file as shown below. So can anyone please send me one example how to use that animated effect using this class. Thanks, Indra -------------- next part -------------- An HTML attachment was scrubbed... URL: From backup.rlacko at gmail.com Fri May 18 23:47:48 2012 From: backup.rlacko at gmail.com (Roman Lacko) Date: Fri, 18 May 2012 23:47:48 +0200 Subject: [PySide] PySide 1.1.1 Windows builds Message-ID: Hi, I have prepared windows builds of PySide 1.1.1 for Pyhon 2.7 and 3.2, 32bit and 64bit. Builds will be available on my google Drive [1], until the PySide site is not moved to new place. For that reason, installing with easy_install is not available, You ned to download the package and install manually. You can allways let me know if you have better idea where to put the builds. Whats new in packages: - OpenSSL DLLs included and installed out-of-the-box (now you can use webkit with ssl) - typesystems, - PySide and Shibken c++ headers - complete examples (from pyside gitorious repo) Whats new in build scripts [2]: - build now creates nice folder structure for every configuration. For example after completed building with python 2.7 32bit release and debug, you get: ..\setuptools\build\py2.7-qt4.7.4-32bit-debug\ ..\setuptools\build\py2.7-qt4.7.4-32bit-release\ ..\setuptools\install\py2.7-qt4.7.4-32bit-debug\ ..\setuptools\install\py2.7-qt4.7.4-32bit-release\ ..\setuptools\packages\PySide-1.1.1qt474.win-amd64-py2.7.exe ..\setuptools\packages\PySide-1.1.1qt474dbg.win-amd64-py2.7.exe - debug build support - refactoring and preparation for building on linux (not completed) [1] https://docs.google.com/open?id=0B0aOk3P0ndoLUlhZSDUwWGU5MjQ [2] https://github.com/PySide/packaging/tree/master/setuptools From paul at fxtech.com Sun May 20 22:56:22 2012 From: paul at fxtech.com (Paul Miller) Date: Sun, 20 May 2012 15:56:22 -0500 Subject: [PySide] PySide 1.1.1 Windows builds (Roman Lacko) In-Reply-To: References: Message-ID: <4FB95A76.7070604@fxtech.com> > I have prepared windows builds of PySide 1.1.1 for Pyhon 2.7 and 3.2, > 32bit and 64bit. Roman - I've still never been able to get the PySide build process working on my Windows machine. What would it take to adjust your toolchain to spit out a version for an older version of Qt? (say, 4.5.3) From jch at jch.com Mon May 21 06:23:52 2012 From: jch at jch.com (YON - Jan C. Hardenbergh) Date: Mon, 21 May 2012 00:23:52 -0400 Subject: [PySide] Test message - seems like this moved to google groups Message-ID: <84B9FD13-F6ED-45E4-81D1-6DF251C7C3AE@jch.com> YON - Jan C. Hardenbergh <> www.jch.com <> Pixelsmith Jeff Lehman's five virtues: a love for complexity, a patient spirit, a will to communicate, a sense of humor, and an optimistic heart. From backup.rlacko at gmail.com Mon May 21 08:34:05 2012 From: backup.rlacko at gmail.com (Roman Lacko) Date: Mon, 21 May 2012 08:34:05 +0200 Subject: [PySide] PySide 1.1.1 Windows builds (Roman Lacko) In-Reply-To: <4FB95A76.7070604@fxtech.com> References: <4FB95A76.7070604@fxtech.com> Message-ID: Hi, 2012/5/20 Paul Miller : >> I have prepared windows builds of PySide 1.1.1 for Pyhon 2.7 and 3.2, >> 32bit and 64bit. > > Roman - I've still never been able to get the PySide build process > working on my Windows machine. > > What would it take to adjust your toolchain to spit out a version for an > older version of Qt? (say, 4.5.3) what version of python You need to compile with ? I will try to build PySide with Qt 4.5.3 and try to fix the build errors Regards Roman > > _______________________________________________ > PySide mailing list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside From hugo.lima at openbossa.org Mon May 21 14:48:37 2012 From: hugo.lima at openbossa.org (Hugo Parente Lima) Date: Mon, 21 May 2012 09:48:37 -0300 Subject: [PySide] PySide 1.1.1 Windows builds (Roman Lacko) In-Reply-To: References: <4FB95A76.7070604@fxtech.com> Message-ID: <1503876.FVIhVCQ2aJ@hugodesktop> On Monday, May 21, 2012 08:34:05 AM Roman Lacko wrote: > Hi, > > 2012/5/20 Paul Miller : > >> I have prepared windows builds of PySide 1.1.1 for Pyhon 2.7 and 3.2, > >> 32bit and 64bit. > > > > Roman - I've still never been able to get the PySide build process > > working on my Windows machine. > > > > What would it take to adjust your toolchain to spit out a version for an > > older version of Qt? (say, 4.5.3) > > what version of python You need to compile with ? > I will try to build PySide with Qt 4.5.3 and try to fix the build errors Better try with a newer version like 4.7 or 4.8, btw some compiler fixes for MSVC were landed last month. > Regards > Roman > > > _______________________________________________ > > 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 -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part. URL: From hugo.lima at openbossa.org Mon May 21 15:06:40 2012 From: hugo.lima at openbossa.org (Hugo Parente Lima) Date: Mon, 21 May 2012 10:06:40 -0300 Subject: [PySide] PySide 1.1.1 Windows builds In-Reply-To: References: Message-ID: <2960254.yKBiThiJ43@hugodesktop> On Friday, May 18, 2012 11:47:48 PM Roman Lacko wrote: > Hi, > > I have prepared windows builds of PySide 1.1.1 for Pyhon 2.7 and 3.2, > 32bit and 64bit. > Builds will be available on my google Drive [1], until the PySide site > is not moved to new place. > For that reason, installing with easy_install is not available, You > ned to download the package and install manually. > You can allways let me know if you have better idea where to put the builds. > > Whats new in packages: > > - OpenSSL DLLs included and installed out-of-the-box (now you can use > webkit with ssl) > - typesystems, > - PySide and Shibken c++ headers > - complete examples (from pyside gitorious repo) > > Whats new in build scripts [2]: > > - build now creates nice folder structure for every configuration. For > example after completed building with python 2.7 32bit release and > debug, you get: > ..\setuptools\build\py2.7-qt4.7.4-32bit-debug\ > ..\setuptools\build\py2.7-qt4.7.4-32bit-release\ > ..\setuptools\install\py2.7-qt4.7.4-32bit-debug\ > ..\setuptools\install\py2.7-qt4.7.4-32bit-release\ > ..\setuptools\packages\PySide-1.1.1qt474.win-amd64-py2.7.exe > ..\setuptools\packages\PySide-1.1.1qt474dbg.win-amd64-py2.7.exe > - debug build support > - refactoring and preparation for building on linux (not completed) Great job Roman!! > > [1] https://docs.google.com/open?id=0B0aOk3P0ndoLUlhZSDUwWGU5MjQ > [2] https://github.com/PySide/packaging/tree/master/setuptools > _______________________________________________ > PySide mailing list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part. URL: From ltanure at gmail.com Mon May 21 15:22:16 2012 From: ltanure at gmail.com (Lucas Tanure) Date: Mon, 21 May 2012 10:22:16 -0300 Subject: [PySide] PySide 1.1.1 for Arch Linux. Message-ID: Hi, I'm a user of Arch Linux and pyside. So, I would like to create the python2-pyside for Aur . Because pyside package has been flagged out of date. May I? Thanks Lucas A. Tanure Alves Skype : lucas.tanure +55 (19) 88176559 -------------- next part -------------- An HTML attachment was scrubbed... URL: From hugo.lima at openbossa.org Mon May 21 15:28:44 2012 From: hugo.lima at openbossa.org (Hugo Parente Lima) Date: Mon, 21 May 2012 10:28:44 -0300 Subject: [PySide] PySide 1.1.1 for Arch Linux. In-Reply-To: References: Message-ID: <1431545.FXt3O9lvXF@hugodesktop> On Monday, May 21, 2012 10:22:16 AM Lucas Tanure wrote: > Hi, > > I'm a user of Arch Linux and pyside. So, I would like to create the > python2-pyside for Aur . > Because pyside package has > been flagged out of date. > May I? Yes you can, but better to first ask the original package author about this to avoid duplication of work. > Thanks > > Lucas A. Tanure Alves > Skype : lucas.tanure > +55 (19) 88176559 -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part. URL: From paulo.alcantara at openbossa.org Mon May 21 15:46:40 2012 From: paulo.alcantara at openbossa.org (paulo alcantara) Date: Mon, 21 May 2012 10:46:40 -0300 Subject: [PySide] PySide 1.1.1 Windows builds In-Reply-To: References: Message-ID: Hi Roman, On Fri, May 18, 2012 at 6:47 PM, Roman Lacko wrote: > Hi, > > I have prepared windows builds of PySide 1.1.1 for Pyhon 2.7 and 3.2, > 32bit and 64bit. > Builds will be available on my google Drive [1], until the PySide site > is not moved to new place. > For that reason, installing with easy_install is not available, You > ned to download the package and install manually. > You can allways let me know if you have better idea where to put the builds. > > Whats new in packages: > > - OpenSSL DLLs included and installed out-of-the-box (now you can use > webkit with ssl) > - typesystems, > - PySide and Shibken c++ headers > - complete examples (from pyside gitorious repo) > > Whats new in build scripts [2]: > > - build now creates nice folder structure for every configuration. For > example after completed building with python 2.7 32bit release and > debug, you get: >    ..\setuptools\build\py2.7-qt4.7.4-32bit-debug\ >    ..\setuptools\build\py2.7-qt4.7.4-32bit-release\ >    ..\setuptools\install\py2.7-qt4.7.4-32bit-debug\ >    ..\setuptools\install\py2.7-qt4.7.4-32bit-release\ >    ..\setuptools\packages\PySide-1.1.1qt474.win-amd64-py2.7.exe >    ..\setuptools\packages\PySide-1.1.1qt474dbg.win-amd64-py2.7.exe > - debug build support > - refactoring and preparation for building on linux (not completed) > > > [1] https://docs.google.com/open?id=0B0aOk3P0ndoLUlhZSDUwWGU5MjQ > [2] https://github.com/PySide/packaging/tree/master/setuptools > _______________________________________________ > PySide mailing list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside Great Work! Thanks, -Paulo From paul at fxtech.com Mon May 21 17:36:29 2012 From: paul at fxtech.com (Paul Miller) Date: Mon, 21 May 2012 10:36:29 -0500 Subject: [PySide] PySide 1.1.1 Windows builds (Roman Lacko) In-Reply-To: References: <4FB95A76.7070604@fxtech.com> Message-ID: <4FBA60FD.1070508@fxtech.com> On 5/21/2012 1:34 AM, Roman Lacko wrote: > 2012/5/20 Paul Miller: >>> I have prepared windows builds of PySide 1.1.1 for Pyhon 2.7 and 3.2, >>> 32bit and 64bit. >> >> Roman - I've still never been able to get the PySide build process >> working on my Windows machine. >> >> What would it take to adjust your toolchain to spit out a version for an >> older version of Qt? (say, 4.5.3) > > what version of python You need to compile with ? > I will try to build PySide with Qt 4.5.3 and try to fix the build errors I'm still using 4.5.3 on Windows because we know it works with our application. I tried moving up to 4.8.1 and we started experiencing random crashes on one user's machine (unfortunately, not my dev machine). It's a complex application, and almost every version of Qt that we've moved up to has broken something in some subtle way. For this reason, we tend to stick with a version we know works. If you can easily build for different versions (x64), or help me figure it out, I'd be grateful. From backup.rlacko at gmail.com Tue May 22 08:54:53 2012 From: backup.rlacko at gmail.com (Roman Lacko) Date: Tue, 22 May 2012 08:54:53 +0200 Subject: [PySide] PySide 1.1.1 Windows builds (Roman Lacko) In-Reply-To: <4FBA60FD.1070508@fxtech.com> References: <4FB95A76.7070604@fxtech.com> <4FBA60FD.1070508@fxtech.com> Message-ID: Hi Paul, 2012/5/21 Paul Miller : > On 5/21/2012 1:34 AM, Roman Lacko wrote: >> >> 2012/5/20 Paul Miller: >>>> >>>> I have prepared windows builds of PySide 1.1.1 for Pyhon 2.7 and 3.2, >>>> 32bit and 64bit. >>> >>> >>> Roman - I've still never been able to get the PySide build process >>> working on my Windows machine. >>> >>> What would it take to adjust your toolchain to spit out a version for an >>> older version of Qt? (say, 4.5.3) >> >> >> what version of python You need to compile with ? >> I will try to build PySide with Qt 4.5.3 and try to fix the build errors > > > I'm still using 4.5.3 on Windows because we know it works with our > application. > > I tried moving up to 4.8.1 and we started experiencing random crashes on one > user's machine (unfortunately, not my dev machine). It's a complex > application, and almost every version of Qt that we've moved up to has > broken something in some subtle way. For this reason, we tend to stick with > a version we know works. > > If you can easily build for different versions (x64), or help me figure it > out, I'd be grateful. I'm getting compilation errors with Qt 4.5.3, but it's not bug in build scripts. The build fails when building PySide module, Shiboken compiles without errors. At this time I'm trying to fix Qt 4.8.1 build... Regards Roman From paul at fxtech.com Tue May 22 15:16:57 2012 From: paul at fxtech.com (Paul Miller) Date: Tue, 22 May 2012 08:16:57 -0500 Subject: [PySide] PySide 1.1.1 Windows builds (Roman Lacko) In-Reply-To: References: <4FB95A76.7070604@fxtech.com> <4FBA60FD.1070508@fxtech.com> Message-ID: <4FBB91C9.3040500@fxtech.com> On 5/22/2012 1:54 AM, Roman Lacko wrote: > Hi Paul, > > 2012/5/21 Paul Miller: >> On 5/21/2012 1:34 AM, Roman Lacko wrote: >>> >>> 2012/5/20 Paul Miller: >>>>> >>>>> I have prepared windows builds of PySide 1.1.1 for Pyhon 2.7 and 3.2, >>>>> 32bit and 64bit. >>>> >>>> >>>> Roman - I've still never been able to get the PySide build process >>>> working on my Windows machine. >>>> >>>> What would it take to adjust your toolchain to spit out a version for an >>>> older version of Qt? (say, 4.5.3) >>> >>> >>> what version of python You need to compile with ? >>> I will try to build PySide with Qt 4.5.3 and try to fix the build errors >> >> >> I'm still using 4.5.3 on Windows because we know it works with our >> application. >> >> I tried moving up to 4.8.1 and we started experiencing random crashes on one >> user's machine (unfortunately, not my dev machine). It's a complex >> application, and almost every version of Qt that we've moved up to has >> broken something in some subtle way. For this reason, we tend to stick with >> a version we know works. >> >> If you can easily build for different versions (x64), or help me figure it >> out, I'd be grateful. > > I'm getting compilation errors with Qt 4.5.3, but it's not bug in > build scripts. The build fails when building PySide module, Shiboken > compiles without errors. > At this time I'm trying to fix Qt 4.8.1 build... Roman - that's okay. I'm doing some tests now to see if we can safely move up to Qt 4.7.3. Thanks for your efforts. From stef.mientki at gmail.com Tue May 22 21:39:53 2012 From: stef.mientki at gmail.com (Stef Mientki) Date: Tue, 22 May 2012 21:39:53 +0200 Subject: [PySide] PySide 1.1.1 Windows builds In-Reply-To: References: Message-ID: <4FBBEB89.4080307@gmail.com> hello, any change there will be a (32 bits) windows version for python 2.6 ? thanks, Stef On 18-05-2012 23:47, Roman Lacko wrote: > Hi, > > I have prepared windows builds of PySide 1.1.1 for Pyhon 2.7 and 3.2, > 32bit and 64bit. > Builds will be available on my google Drive [1], until the PySide site > is not moved to new place. > For that reason, installing with easy_install is not available, You > ned to download the package and install manually. > You can allways let me know if you have better idea where to put the builds. > > Whats new in packages: > > - OpenSSL DLLs included and installed out-of-the-box (now you can use > webkit with ssl) > - typesystems, > - PySide and Shibken c++ headers > - complete examples (from pyside gitorious repo) > > Whats new in build scripts [2]: > > - build now creates nice folder structure for every configuration. For > example after completed building with python 2.7 32bit release and > debug, you get: > ..\setuptools\build\py2.7-qt4.7.4-32bit-debug\ > ..\setuptools\build\py2.7-qt4.7.4-32bit-release\ > ..\setuptools\install\py2.7-qt4.7.4-32bit-debug\ > ..\setuptools\install\py2.7-qt4.7.4-32bit-release\ > ..\setuptools\packages\PySide-1.1.1qt474.win-amd64-py2.7.exe > ..\setuptools\packages\PySide-1.1.1qt474dbg.win-amd64-py2.7.exe > - debug build support > - refactoring and preparation for building on linux (not completed) > > > [1] https://docs.google.com/open?id=0B0aOk3P0ndoLUlhZSDUwWGU5MjQ > [2] https://github.com/PySide/packaging/tree/master/setuptools > _______________________________________________ > PySide mailing list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside From backup.rlacko at gmail.com Tue May 22 23:13:13 2012 From: backup.rlacko at gmail.com (Roman Lacko) Date: Tue, 22 May 2012 23:13:13 +0200 Subject: [PySide] PySide 1.1.1 Windows builds In-Reply-To: <4FBBEB89.4080307@gmail.com> References: <4FBBEB89.4080307@gmail.com> Message-ID: Hi, 2012/5/22 Stef Mientki : > > hello, > any change there will be a (32 bits) windows version for python 2.6 ? > thanks, > Stef I will build them tomorrow, just for You :) R. > > On 18-05-2012 23:47, Roman Lacko wrote: >> Hi, >> >> I have prepared windows builds of PySide 1.1.1 for Pyhon 2.7 and 3.2, >> 32bit and 64bit. >> Builds will be available on my google Drive [1], until the PySide site >> is not moved to new place. >> For that reason, installing with easy_install is not available, You >> ned to download the package and install manually. >> You can allways let me know if you have better idea where to put the builds. >> >> Whats new in packages: >> >> - OpenSSL DLLs included and installed out-of-the-box (now you can use >> webkit with ssl) >> - typesystems, >> - PySide and Shibken c++ headers >> - complete examples (from pyside gitorious repo) >> >> Whats new in build scripts [2]: >> >> - build now creates nice folder structure for every configuration. For >> example after completed building with python 2.7 32bit release and >> debug, you get: >>      ..\setuptools\build\py2.7-qt4.7.4-32bit-debug\ >>      ..\setuptools\build\py2.7-qt4.7.4-32bit-release\ >>      ..\setuptools\install\py2.7-qt4.7.4-32bit-debug\ >>      ..\setuptools\install\py2.7-qt4.7.4-32bit-release\ >>      ..\setuptools\packages\PySide-1.1.1qt474.win-amd64-py2.7.exe >>      ..\setuptools\packages\PySide-1.1.1qt474dbg.win-amd64-py2.7.exe >> - debug build support >> - refactoring and preparation for building on linux (not completed) >> >> >> [1] https://docs.google.com/open?id=0B0aOk3P0ndoLUlhZSDUwWGU5MjQ >> [2] https://github.com/PySide/packaging/tree/master/setuptools >> _______________________________________________ >> 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 From paul at fxtech.com Tue May 22 23:14:48 2012 From: paul at fxtech.com (Paul Miller) Date: Tue, 22 May 2012 16:14:48 -0500 Subject: [PySide] Accessing C++ QObject subclass from script? Message-ID: <4FBC01C8.6020503@fxtech.com> I got PySide 1.1 working in my hybrid app (thanks Roman for your help!) using Qt 4.7.4 and Python 2.7. I can create new Qt widgets and use my existing C++ QApplication and it seems to work fine. Now I would like to expose some of my internal QObjects that have already been created by the time the Python interpreter starts up. For my existing data model interface to Python, I've hand-created some Python bindings for my classes. What do I need to do to return a pointer to one of my created QObjects back into PySide, so I can connect signals to it? *I can't use an auto-type-generator thingy like in the superhybrids article*. From backup.rlacko at gmail.com Wed May 23 09:34:12 2012 From: backup.rlacko at gmail.com (Roman Lacko) Date: Wed, 23 May 2012 09:34:12 +0200 Subject: [PySide] Fwd: PySide 1.1.1 Windows builds In-Reply-To: References: Message-ID: Hi, New package for Python 2.6 32bit uploaded to my Goggle Drive Share [1] Regards Roman [1] https://docs.google.com/open?id=0B0aOk3P0ndoLUlhZSDUwWGU5MjQ ---------- Forwarded message ---------- From: Roman Lacko Date: 2012/5/18 Subject: PySide 1.1.1 Windows builds To: pyside at qt-project.org Hi, I have prepared windows builds of PySide 1.1.1 for Pyhon 2.7 and 3.2, 32bit and 64bit. Builds will be available on my google Drive [1], until the PySide site is not moved to new place. For that reason, installing with easy_install is not available, You ned to download the package and install manually. You can allways let me know if you have better idea where to put the builds. Whats new in packages: - OpenSSL DLLs included and installed out-of-the-box (now you can use webkit with ssl) - typesystems, - PySide and Shibken c++ headers - complete examples (from pyside gitorious repo) Whats new in build scripts [2]: - build now creates nice folder structure for every configuration. For example after completed building with python 2.7 32bit release and debug, you get:    ..\setuptools\build\py2.7-qt4.7.4-32bit-debug\    ..\setuptools\build\py2.7-qt4.7.4-32bit-release\    ..\setuptools\install\py2.7-qt4.7.4-32bit-debug\    ..\setuptools\install\py2.7-qt4.7.4-32bit-release\    ..\setuptools\packages\PySide-1.1.1qt474.win-amd64-py2.7.exe    ..\setuptools\packages\PySide-1.1.1qt474dbg.win-amd64-py2.7.exe - debug build support - refactoring and preparation for building on linux (not completed) [1] https://docs.google.com/open?id=0B0aOk3P0ndoLUlhZSDUwWGU5MjQ [2] https://github.com/PySide/packaging/tree/master/setuptools From paul at fxtech.com Wed May 23 16:28:15 2012 From: paul at fxtech.com (Paul Miller) Date: Wed, 23 May 2012 09:28:15 -0500 Subject: [PySide] Getting existing QMainWindow from hybrid app Message-ID: <4FBCF3FF.80106@fxtech.com> I am trying to add a dock widget to my application's QMainWindow, which is being created in the C++ part of my hybrid application. But there doesn't appear to be any way to get the QApplication.activeWindow() as a QMainWindow - it always just comes back as a QWidget, and therefore has no "addDockWidget" method. Is there a way to "cast" the resulting top-level QWidget to a QMainWindow so I can call it's methods, from the Python side? (something similar to this was asked back in 2010, and considered a bug, but there is no resolution: ) If not, is there a tutorial for adding just enough glue to my C++ app so I can expose my own QObject-derived objects to the Python side? From techtonik at gmail.com Wed May 23 20:27:11 2012 From: techtonik at gmail.com (anatoly techtonik) Date: Wed, 23 May 2012 21:27:11 +0300 Subject: [PySide] QComboBox + focusOutEvent == double focus?? Message-ID: I am trying to make an editable ComboBox that saves its value when focus is lost (for example, when user clicks a button on the same page). I've defined focusOutEvent() for it, it seems to be called ok, but focus marker never leaves the widget. It fact it duplicates. Any hints what's going on that how to fix it? from PySide.QtGui import * from PySide.QtCore import Qt, QEvent class MyComboBox(QComboBox): def __init__(self): QComboBox.__init__(self) def event(self, event): if event.type() == QEvent.KeyPress and event.key() == Qt.Key_Return: self.addItem(self.currentText()) return QComboBox.event(self, event) def focusOutEvent(self, event): self.addItem(self.currentText()) return event class Widget(QWidget): def __init__(self, parent=None): super(Widget, self).__init__(parent) combo = MyComboBox() combo.setEditable(True) combo.addItems(['One', 'Two', 'Three']) lineedit = QLineEdit() layout = QVBoxLayout() layout.addWidget(combo) layout.addWidget(lineedit) self.setLayout(layout) app = QApplication([]) widget = Widget() widget.show() app.exec_() -- anatoly t. From stef.mientki at gmail.com Wed May 23 21:57:37 2012 From: stef.mientki at gmail.com (Stef Mientki) Date: Wed, 23 May 2012 21:57:37 +0200 Subject: [PySide] Fwd: PySide 1.1.1 Windows builds In-Reply-To: References: Message-ID: <4FBD4131.9090006@gmail.com> what a great service, THANKS VERY MUCH Laacko !! cheers, Stef On 23-05-2012 09:34, Roman Lacko wrote: > Hi, > New package for Python 2.6 32bit uploaded to my Goggle Drive Share [1] > > Regards > Roman > > [1] https://docs.google.com/open?id=0B0aOk3P0ndoLUlhZSDUwWGU5MjQ > > ---------- Forwarded message ---------- > From: Roman Lacko > Date: 2012/5/18 > Subject: PySide 1.1.1 Windows builds > To: pyside at qt-project.org > > > Hi, > > I have prepared windows builds of PySide 1.1.1 for Pyhon 2.7 and 3.2, > 32bit and 64bit. > Builds will be available on my google Drive [1], until the PySide site > is not moved to new place. > For that reason, installing with easy_install is not available, You > ned to download the package and install manually. > You can allways let me know if you have better idea where to put the builds. > > Whats new in packages: > > - OpenSSL DLLs included and installed out-of-the-box (now you can use > webkit with ssl) > - typesystems, > - PySide and Shibken c++ headers > - complete examples (from pyside gitorious repo) > > Whats new in build scripts [2]: > > - build now creates nice folder structure for every configuration. For > example after completed building with python 2.7 32bit release and > debug, you get: > ..\setuptools\build\py2.7-qt4.7.4-32bit-debug\ > ..\setuptools\build\py2.7-qt4.7.4-32bit-release\ > ..\setuptools\install\py2.7-qt4.7.4-32bit-debug\ > ..\setuptools\install\py2.7-qt4.7.4-32bit-release\ > ..\setuptools\packages\PySide-1.1.1qt474.win-amd64-py2.7.exe > ..\setuptools\packages\PySide-1.1.1qt474dbg.win-amd64-py2.7.exe > - debug build support > - refactoring and preparation for building on linux (not completed) > > > [1] https://docs.google.com/open?id=0B0aOk3P0ndoLUlhZSDUwWGU5MjQ > [2] https://github.com/PySide/packaging/tree/master/setuptools From stef.mientki at gmail.com Wed May 23 22:00:23 2012 From: stef.mientki at gmail.com (Stef Mientki) Date: Wed, 23 May 2012 22:00:23 +0200 Subject: [PySide] Fwd: PySide 1.1.1 Windows builds In-Reply-To: <4FBD4131.9090006@gmail.com> References: <4FBD4131.9090006@gmail.com> Message-ID: <4FBD41D7.1020105@gmail.com> Sorry Roman, in the excitement I used the wrong part of your name, and even misspelled it ;-( cheers, Stef On 23-05-2012 21:57, Stef Mientki wrote: > what a great service, > THANKS VERY MUCH Laacko !! > cheers, > Stef > > On 23-05-2012 09:34, Roman Lacko wrote: >> Hi, >> New package for Python 2.6 32bit uploaded to my Goggle Drive Share [1] >> >> Regards >> Roman >> >> [1] https://docs.google.com/open?id=0B0aOk3P0ndoLUlhZSDUwWGU5MjQ >> >> ---------- Forwarded message ---------- >> From: Roman Lacko >> Date: 2012/5/18 >> Subject: PySide 1.1.1 Windows builds >> To: pyside at qt-project.org >> >> >> Hi, >> >> I have prepared windows builds of PySide 1.1.1 for Pyhon 2.7 and 3.2, >> 32bit and 64bit. >> Builds will be available on my google Drive [1], until the PySide site >> is not moved to new place. >> For that reason, installing with easy_install is not available, You >> ned to download the package and install manually. >> You can allways let me know if you have better idea where to put the builds. >> >> Whats new in packages: >> >> - OpenSSL DLLs included and installed out-of-the-box (now you can use >> webkit with ssl) >> - typesystems, >> - PySide and Shibken c++ headers >> - complete examples (from pyside gitorious repo) >> >> Whats new in build scripts [2]: >> >> - build now creates nice folder structure for every configuration. For >> example after completed building with python 2.7 32bit release and >> debug, you get: >> ..\setuptools\build\py2.7-qt4.7.4-32bit-debug\ >> ..\setuptools\build\py2.7-qt4.7.4-32bit-release\ >> ..\setuptools\install\py2.7-qt4.7.4-32bit-debug\ >> ..\setuptools\install\py2.7-qt4.7.4-32bit-release\ >> ..\setuptools\packages\PySide-1.1.1qt474.win-amd64-py2.7.exe >> ..\setuptools\packages\PySide-1.1.1qt474dbg.win-amd64-py2.7.exe >> - debug build support >> - refactoring and preparation for building on linux (not completed) >> >> >> [1] https://docs.google.com/open?id=0B0aOk3P0ndoLUlhZSDUwWGU5MjQ >> [2] https://github.com/PySide/packaging/tree/master/setuptools > From backup.rlacko at gmail.com Wed May 23 22:01:44 2012 From: backup.rlacko at gmail.com (Roman Lacko) Date: Wed, 23 May 2012 22:01:44 +0200 Subject: [PySide] Fwd: PySide 1.1.1 Windows builds In-Reply-To: <4FBD41D7.1020105@gmail.com> References: <4FBD4131.9090006@gmail.com> <4FBD41D7.1020105@gmail.com> Message-ID: ok, It was funny :) 2012/5/23 Stef Mientki : > Sorry Roman, > in the excitement I used the wrong part of your name, > and even misspelled it ;-( > cheers, > Stef > > On 23-05-2012 21:57, Stef Mientki wrote: >> >> what a great service, >> THANKS VERY MUCH Laacko !! >> cheers, >> Stef >> >> On 23-05-2012 09:34, Roman Lacko wrote: >>> >>> Hi, >>> New package for Python 2.6 32bit uploaded to my Goggle Drive Share [1] >>> >>> Regards >>> Roman >>> >>> [1] https://docs.google.com/open?id=0B0aOk3P0ndoLUlhZSDUwWGU5MjQ >>> >>> ---------- Forwarded message ---------- >>> From: Roman Lacko >>> Date: 2012/5/18 >>> Subject: PySide 1.1.1 Windows builds >>> To: pyside at qt-project.org >>> >>> >>> Hi, >>> >>> I have prepared windows builds of PySide 1.1.1 for Pyhon 2.7 and 3.2, >>> 32bit and 64bit. >>> Builds will be available on my google Drive [1], until the PySide site >>> is not moved to new place. >>> For that reason, installing with easy_install is not available, You >>> ned to download the package and install manually. >>> You can allways let me know if you have better idea where to put the >>> builds. >>> >>> Whats new in packages: >>> >>> - OpenSSL DLLs included and installed out-of-the-box (now you can use >>> webkit with ssl) >>> - typesystems, >>> - PySide and Shibken c++ headers >>> - complete examples (from pyside gitorious repo) >>> >>> Whats new in build scripts [2]: >>> >>> - build now creates nice folder structure for every configuration. For >>> example after completed building with python 2.7 32bit release and >>> debug, you get: >>>    ..\setuptools\build\py2.7-qt4.7.4-32bit-debug\ >>>    ..\setuptools\build\py2.7-qt4.7.4-32bit-release\ >>>    ..\setuptools\install\py2.7-qt4.7.4-32bit-debug\ >>>    ..\setuptools\install\py2.7-qt4.7.4-32bit-release\ >>>    ..\setuptools\packages\PySide-1.1.1qt474.win-amd64-py2.7.exe >>>    ..\setuptools\packages\PySide-1.1.1qt474dbg.win-amd64-py2.7.exe >>> - debug build support >>> - refactoring and preparation for building on linux (not completed) >>> >>> >>> [1] https://docs.google.com/open?id=0B0aOk3P0ndoLUlhZSDUwWGU5MjQ >>> [2] https://github.com/PySide/packaging/tree/master/setuptools >> >> > From js.augustyn at gmail.com Wed May 23 22:21:14 2012 From: js.augustyn at gmail.com (Jason Augustyn) Date: Wed, 23 May 2012 16:21:14 -0400 Subject: [PySide] IPython Mac 10.7 QApplication instance already exists Message-ID: Hi all, I am new to using PySide and have been working through the novice tutorials, which are great. One problem - I cannot re-run a script without exiting the Python shell altogether. Re-running gives the RuntimeError: A QApplication instance already exists. I have searched for a solution and found an old bug tracker ticket from someone with a similar problem. The suggested solution was to upgrade to PySide 1.0.4 and the ticket was marked as invalid. Obviously that doesn;t work, as I'm running the newest version of PySide and still have the problem. Any suggestions? I am using IPython on a Mac OSX 10.7. Thanks! Jason -------------- next part -------------- An HTML attachment was scrubbed... URL: From nathanjsmith at gmail.com Wed May 23 23:03:56 2012 From: nathanjsmith at gmail.com (Nathan Smith) Date: Wed, 23 May 2012 16:03:56 -0500 Subject: [PySide] IPython Mac 10.7 QApplication instance already exists In-Reply-To: References: Message-ID: Hi Jason, PySide only supports creating a single persistent instance of QApplication. This singleton exists forever and cannot be deleted. If you have a function that tries to create a QApplication, then I recommend placing that creation inside of a try/except block: try: app = QtGui.QApplication(sys.argv) except RuntimeError: app = QtCore.QCoreApplication.instance() This will either create a new QApplication instance if one does not exist, or return the current QApplication instance if one already does exist. Nathan On Wed, May 23, 2012 at 3:21 PM, Jason Augustyn wrote: > Hi all, I am new to using PySide and have been working through the novice > tutorials, which are great. One problem - I cannot re-run a script without > exiting the Python shell altogether. Re-running gives the RuntimeError: A > QApplication instance already exists. > > I have searched for a solution and found an old bug tracker ticket from > someone with a similar problem. The suggested solution was to upgrade to > PySide 1.0.4 and the ticket was marked as invalid. Obviously that doesn;t > work, as I'm running the newest version of PySide and still have the > problem. Any suggestions? I am using IPython on a Mac OSX 10.7. > > Thanks! > > Jason > > _______________________________________________ > PySide mailing list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From js.augustyn at gmail.com Wed May 23 23:19:06 2012 From: js.augustyn at gmail.com (Jason Augustyn) Date: Wed, 23 May 2012 17:19:06 -0400 Subject: [PySide] IPython Mac 10.7 QApplication instance already exists In-Reply-To: References: Message-ID: Thanks, Nathan, after poking around a bit more I found a similar solution: app = QApplication.instance() if app == None: app = QApplication(sys.argv) This works! On Wed, May 23, 2012 at 5:03 PM, Nathan Smith wrote: > Hi Jason, > > PySide only supports creating a single persistent instance of > QApplication. This singleton exists forever and cannot be deleted. If you > have a function that tries to create a QApplication, then I recommend > placing that creation inside of a try/except block: > > try: > app = QtGui.QApplication(sys.argv) > except RuntimeError: > app = QtCore.QCoreApplication.instance() > > > This will either create a new QApplication instance if one does not exist, > or return the current QApplication instance if one already does exist. > > Nathan > > On Wed, May 23, 2012 at 3:21 PM, Jason Augustyn wrote: > >> Hi all, I am new to using PySide and have been working through the novice >> tutorials, which are great. One problem - I cannot re-run a script without >> exiting the Python shell altogether. Re-running gives the RuntimeError: A >> QApplication instance already exists. >> >> I have searched for a solution and found an old bug tracker ticket from >> someone with a similar problem. The suggested solution was to upgrade to >> PySide 1.0.4 and the ticket was marked as invalid. Obviously that doesn;t >> work, as I'm running the newest version of PySide and still have the >> problem. Any suggestions? I am using IPython on a Mac OSX 10.7. >> >> Thanks! >> >> Jason >> >> _______________________________________________ >> PySide mailing list >> PySide at qt-project.org >> http://lists.qt-project.org/mailman/listinfo/pyside >> >> > -------------- next part -------------- An HTML attachment was scrubbed... URL: From blair at orcaware.com Wed May 23 23:26:12 2012 From: blair at orcaware.com (Blair Zajac) Date: Wed, 23 May 2012 14:26:12 -0700 Subject: [PySide] IPython Mac 10.7 QApplication instance already exists In-Reply-To: References: Message-ID: <4FBD55F4.5060608@orcaware.com> According to PEP-8, don't compare using == to singletons: http://www.python.org/dev/peps/pep-0008/#programming-recommendations """Comparisons to singletons like None should always be done with is or is not, never the equality operators.""" app = QApplication.instance() if app is None: app = QApplication(sys.argv) Blair On 05/23/2012 02:19 PM, Jason Augustyn wrote: > Thanks, Nathan, after poking around a bit more I found a similar solution: > > app = QApplication.instance() > if app == None: > app = QApplication(sys.argv) > > This works! > > > > On Wed, May 23, 2012 at 5:03 PM, Nathan Smith > wrote: > > Hi Jason, > > PySide only supports creating a single persistent instance of > QApplication. This singleton exists forever and cannot be deleted. > If you have a function that tries to create a QApplication, then I > recommend placing that creation inside of a try/except block: > > try: > app = QtGui.QApplication(sys.argv) > except RuntimeError: > app = QtCore.QCoreApplication.instance() > > > This will either create a new QApplication instance if one does not > exist, or return the current QApplication instance if one already > does exist. > > Nathan > > On Wed, May 23, 2012 at 3:21 PM, Jason Augustyn > > wrote: > > Hi all, I am new to using PySide and have been working through > the novice tutorials, which are great. One problem - I cannot > re-run a script without exiting the Python shell altogether. > Re-running gives the RuntimeError: A QApplication instance > already exists. > > I have searched for a solution and found an old bug tracker > ticket from someone with a similar problem. The suggested > solution was to upgrade to PySide 1.0.4 and the ticket was > marked as invalid. Obviously that doesn;t work, as I'm running > the newest version of PySide and still have the problem. Any > suggestions? I am using IPython on a Mac OSX 10.7. > > Thanks! > > Jason > > _______________________________________________ > 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 From js.augustyn at gmail.com Thu May 24 00:56:44 2012 From: js.augustyn at gmail.com (Jason Augustyn) Date: Wed, 23 May 2012 18:56:44 -0400 Subject: [PySide] IPython Mac 10.7 QApplication instance already exists In-Reply-To: <4FBD55F4.5060608@orcaware.com> References: <4FBD55F4.5060608@orcaware.com> Message-ID: Thanks for the catch, Blair On Wed, May 23, 2012 at 5:26 PM, Blair Zajac wrote: > > > According to PEP-8, don't compare using == to singletons: > > http://www.python.org/dev/**peps/pep-0008/#programming-**recommendations > > """Comparisons to singletons like None should always be done with is or is > not, never the equality operators.""" > > app = QApplication.instance() > if app is None: > app = QApplication(sys.argv) > > > Blair > > > On 05/23/2012 02:19 PM, Jason Augustyn wrote: > >> Thanks, Nathan, after poking around a bit more I found a similar solution: >> >> app = QApplication.instance() >> if app == None: >> app = QApplication(sys.argv) >> >> This works! >> >> >> >> On Wed, May 23, 2012 at 5:03 PM, Nathan Smith > > wrote: >> >> Hi Jason, >> >> PySide only supports creating a single persistent instance of >> QApplication. This singleton exists forever and cannot be deleted. >> If you have a function that tries to create a QApplication, then I >> recommend placing that creation inside of a try/except block: >> >> try: >> app = QtGui.QApplication(sys.argv) >> except RuntimeError: >> app = QtCore.QCoreApplication.**instance() >> >> >> This will either create a new QApplication instance if one does not >> exist, or return the current QApplication instance if one already >> does exist. >> >> Nathan >> >> On Wed, May 23, 2012 at 3:21 PM, Jason Augustyn >> **> wrote: >> >> Hi all, I am new to using PySide and have been working through >> the novice tutorials, which are great. One problem - I cannot >> re-run a script without exiting the Python shell altogether. >> Re-running gives the RuntimeError: A QApplication instance >> already exists. >> >> I have searched for a solution and found an old bug tracker >> ticket from someone with a similar problem. The suggested >> solution was to upgrade to PySide 1.0.4 and the ticket was >> marked as invalid. Obviously that doesn;t work, as I'm running >> the newest version of PySide and still have the problem. Any >> suggestions? I am using IPython on a Mac OSX 10.7. >> >> Thanks! >> >> Jason >> >> ______________________________**_________________ >> 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 >> > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From alex.gronholm at nextday.fi Thu May 24 22:03:33 2012 From: alex.gronholm at nextday.fi (=?ISO-8859-1?Q?Alex_Gr=F6nholm?=) Date: Thu, 24 May 2012 23:03:33 +0300 Subject: [PySide] PySide 1.1.1 Windows builds In-Reply-To: References: Message-ID: <4FBE9415.2000207@nextday.fi> 19.05.2012 00:47, Roman Lacko kirjoitti: > Hi, > > I have prepared windows builds of PySide 1.1.1 for Pyhon 2.7 and 3.2, > 32bit and 64bit. > Builds will be available on my google Drive [1], until the PySide site > is not moved to new place. > For that reason, installing with easy_install is not available, You > ned to download the package and install manually. > You can allways let me know if you have better idea where to put the builds. How about the PyPI, where they are supposed be uploaded to? > > Whats new in packages: > > - OpenSSL DLLs included and installed out-of-the-box (now you can use > webkit with ssl) > - typesystems, > - PySide and Shibken c++ headers > - complete examples (from pyside gitorious repo) > > Whats new in build scripts [2]: > > - build now creates nice folder structure for every configuration. For > example after completed building with python 2.7 32bit release and > debug, you get: > ..\setuptools\build\py2.7-qt4.7.4-32bit-debug\ > ..\setuptools\build\py2.7-qt4.7.4-32bit-release\ > ..\setuptools\install\py2.7-qt4.7.4-32bit-debug\ > ..\setuptools\install\py2.7-qt4.7.4-32bit-release\ > ..\setuptools\packages\PySide-1.1.1qt474.win-amd64-py2.7.exe > ..\setuptools\packages\PySide-1.1.1qt474dbg.win-amd64-py2.7.exe > - debug build support > - refactoring and preparation for building on linux (not completed) > > > [1] https://docs.google.com/open?id=0B0aOk3P0ndoLUlhZSDUwWGU5MjQ > [2] https://github.com/PySide/packaging/tree/master/setuptools > _______________________________________________ > PySide mailing list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside From techtonik at gmail.com Fri May 25 00:45:51 2012 From: techtonik at gmail.com (anatoly techtonik) Date: Thu, 24 May 2012 15:45:51 -0700 (PDT) Subject: [PySide] Group is closed, subscribe to http://lists.qt-project.org/mailman/listinfo/pyside Message-ID: Dear subscribers of PySide Google Group, As you may have noticed Google rolled out new groups UI, which does not support mirroring external email archives. PySide group was such searchable mirror for PySide development mailing list. Starting from 27th of April messages posted to mailing list were not sent to the group and I suspect the opposite is also true. So, please subscribe to the main list. This group is now *read-only*, and will quickly outdate. For a searchable web archive go to Gmane . Thanks everybody for participation. See you there. ;) -------------- next part -------------- An HTML attachment was scrubbed... URL: From backup.rlacko at gmail.com Fri May 25 12:54:34 2012 From: backup.rlacko at gmail.com (Roman Lacko) Date: Fri, 25 May 2012 12:54:34 +0200 Subject: [PySide] PySide 1.1.1 Windows builds In-Reply-To: <4FBE9415.2000207@nextday.fi> References: <4FBE9415.2000207@nextday.fi> Message-ID: Hi, 2012/5/24 Alex Grönholm : > 19.05.2012 00:47, Roman Lacko kirjoitti: >> Hi, >> >> I have prepared windows builds of PySide 1.1.1 for Pyhon 2.7 and 3.2, >> 32bit and 64bit. >> Builds will be available on my google Drive [1], until the PySide site >> is not moved to new place. >> For that reason, installing with easy_install is not available, You >> ned to download the package and install manually. >> You can allways let me know if you have better idea where to put the builds. > How about the PyPI, where they are supposed be uploaded to? It will be uploaded when the PySide site move to new location is completed BTW, I'm developing new version of PySide setup script fully compatible with distutils, that will be usable also on linux (binaries will be built inplace), the development can be tracked here: https://github.com/lck/pyside-dist Regards Roman >> >> Whats new in packages: >> >> - OpenSSL DLLs included and installed out-of-the-box (now you can use >> webkit with ssl) >> - typesystems, >> - PySide and Shibken c++ headers >> - complete examples (from pyside gitorious repo) >> >> Whats new in build scripts [2]: >> >> - build now creates nice folder structure for every configuration. For >> example after completed building with python 2.7 32bit release and >> debug, you get: >>      ..\setuptools\build\py2.7-qt4.7.4-32bit-debug\ >>      ..\setuptools\build\py2.7-qt4.7.4-32bit-release\ >>      ..\setuptools\install\py2.7-qt4.7.4-32bit-debug\ >>      ..\setuptools\install\py2.7-qt4.7.4-32bit-release\ >>      ..\setuptools\packages\PySide-1.1.1qt474.win-amd64-py2.7.exe >>      ..\setuptools\packages\PySide-1.1.1qt474dbg.win-amd64-py2.7.exe >> - debug build support >> - refactoring and preparation for building on linux (not completed) >> >> >> [1] https://docs.google.com/open?id=0B0aOk3P0ndoLUlhZSDUwWGU5MjQ >> [2] https://github.com/PySide/packaging/tree/master/setuptools >> _______________________________________________ >> 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 From aj at dneg.com Fri May 25 22:59:26 2012 From: aj at dneg.com (Alexander Jones) Date: Fri, 25 May 2012 21:59:26 +0100 Subject: [PySide] Accessing C++ QObject subclass from script? In-Reply-To: <4FBC01C8.6020503@fxtech.com> References: <4FBC01C8.6020503@fxtech.com> Message-ID: Something like this? http://www.pyside.org/docs/shiboken/shibokenmodule.html#shiboken.wrapInstance On 22 May 2012 22:14, Paul Miller wrote: > I got PySide 1.1 working in my hybrid app (thanks Roman for your help!) > using Qt 4.7.4 and Python 2.7. I can create new Qt widgets and use my > existing C++ QApplication and it seems to work fine. > > Now I would like to expose some of my internal QObjects that have > already been created by the time the Python interpreter starts up. > > For my existing data model interface to Python, I've hand-created some > Python bindings for my classes. What do I need to do to return a pointer > to one of my created QObjects back into PySide, so I can connect signals > to it? > > *I can't use an auto-type-generator thingy like in the superhybrids > article*. > > _______________________________________________ > PySide mailing list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside > -- Alexander Jones Double Negative R&D www.dneg.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From paul at fxtech.com Fri May 25 23:09:24 2012 From: paul at fxtech.com (Paul Miller) Date: Fri, 25 May 2012 16:09:24 -0500 Subject: [PySide] Accessing C++ QObject subclass from script? In-Reply-To: References: <4FBC01C8.6020503@fxtech.com> Message-ID: <4FBFF504.9080601@fxtech.com> On 5/25/2012 3:59 PM, Alexander Jones wrote: > Something like this? > > http://www.pyside.org/docs/shiboken/shibokenmodule.html#shiboken.wrapInstance Yeah I saw that but it wasn't apparent how I was to use this from within my C++ application. Is there a tutorial or example for how to use this? > > On 22 May 2012 22:14, Paul Miller > wrote: > > I got PySide 1.1 working in my hybrid app (thanks Roman for your help!) > using Qt 4.7.4 and Python 2.7. I can create new Qt widgets and use my > existing C++ QApplication and it seems to work fine. > > Now I would like to expose some of my internal QObjects that have > already been created by the time the Python interpreter starts up. > > For my existing data model interface to Python, I've hand-created some > Python bindings for my classes. What do I need to do to return a pointer > to one of my created QObjects back into PySide, so I can connect signals > to it? > > *I can't use an auto-type-generator thingy like in the superhybrids > article*. > > _______________________________________________ > PySide mailing list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside > > > > > -- > Alexander Jones > Double Negative R&D > www.dneg.com > From schampailler at skynet.be Sat May 26 09:10:47 2012 From: schampailler at skynet.be (Stefan Champailler) Date: Sat, 26 May 2012 09:10:47 +0200 Subject: [PySide] PySide 1.1.1 Windows builds In-Reply-To: References: <4FBE9415.2000207@nextday.fi> Message-ID: <4FC081F7.8070008@skynet.be> That's great news ! I really depend on those. I'm currently trying to set up a PC with Windows so that I can build PySide too. BTW, is building PySide on Windows hard ? I mean, on linux, it's rather easy and I'd like to know if it's as easy on Win. stF On 05/25/2012 12:54 PM, Roman Lacko wrote: > Hi, > > 2012/5/24 Alex Grönholm: >> 19.05.2012 00:47, Roman Lacko kirjoitti: >>> Hi, >>> >>> I have prepared windows builds of PySide 1.1.1 for Pyhon 2.7 and 3.2, >>> 32bit and 64bit. >>> Builds will be available on my google Drive [1], until the PySide site >>> is not moved to new place. >>> For that reason, installing with easy_install is not available, You >>> ned to download the package and install manually. >>> You can allways let me know if you have better idea where to put the builds. >> How about the PyPI, where they are supposed be uploaded to? > > It will be uploaded when the PySide site move to new location is completed > > BTW, I'm developing new version of PySide setup script fully > compatible with distutils, that will be usable also on linux (binaries > will be built inplace), the development can be tracked here: > https://github.com/lck/pyside-dist > > Regards > Roman > >>> >>> Whats new in packages: >>> >>> - OpenSSL DLLs included and installed out-of-the-box (now you can use >>> webkit with ssl) >>> - typesystems, >>> - PySide and Shibken c++ headers >>> - complete examples (from pyside gitorious repo) >>> >>> Whats new in build scripts [2]: >>> >>> - build now creates nice folder structure for every configuration. For >>> example after completed building with python 2.7 32bit release and >>> debug, you get: >>> ..\setuptools\build\py2.7-qt4.7.4-32bit-debug\ >>> ..\setuptools\build\py2.7-qt4.7.4-32bit-release\ >>> ..\setuptools\install\py2.7-qt4.7.4-32bit-debug\ >>> ..\setuptools\install\py2.7-qt4.7.4-32bit-release\ >>> ..\setuptools\packages\PySide-1.1.1qt474.win-amd64-py2.7.exe >>> ..\setuptools\packages\PySide-1.1.1qt474dbg.win-amd64-py2.7.exe >>> - debug build support >>> - refactoring and preparation for building on linux (not completed) >>> >>> >>> [1] https://docs.google.com/open?id=0B0aOk3P0ndoLUlhZSDUwWGU5MjQ >>> [2] https://github.com/PySide/packaging/tree/master/setuptools >>> _______________________________________________ >>> 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 list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside > -- Timeo Danaos et dona ferentes From a.richi at bluewin.ch Sat May 26 13:49:18 2012 From: a.richi at bluewin.ch (Aaron Richiger) Date: Sat, 26 May 2012 13:49:18 +0200 Subject: [PySide] PySide with pypy In-Reply-To: References: <4FBD55F4.5060608@orcaware.com> Message-ID: <4FC0C33E.6040008@bluewin.ch> Dear list! I tried checking performance improvements using PySide with pypy, but failed installing PySide in my pypy site-packages... Did anybody have success using PySide with pypy on Ubuntu? The usual "sudo apt-get install python-pyside" does not work, because it installs it to the default python2.7 dist-packages directory. Is there a way to specify the target site-packages you want to use later (pypy site-packages in my case)? Thanks for any help! Aaron From lunaryorn at googlemail.com Sat May 26 19:18:21 2012 From: lunaryorn at googlemail.com (Sebastian Wiesner) Date: Sat, 26 May 2012 19:18:21 +0200 Subject: [PySide] PySide with pypy In-Reply-To: <4FC0C33E.6040008@bluewin.ch> References: <4FBD55F4.5060608@orcaware.com> <4FC0C33E.6040008@bluewin.ch> Message-ID: 2012/5/26 Aaron Richiger : > Dear list! > > I tried checking performance improvements using PySide with pypy, but > failed installing PySide in my pypy site-packages... Did anybody have > success using PySide with pypy on Ubuntu? The usual "sudo apt-get > install python-pyside" does not work, because it installs it to the > default python2.7 dist-packages directory. Is there a way to specify the > target site-packages you want to use later (pypy site-packages in my case)? PySide is unlikely to work with PyPy because PyPy only partially implements the CPython extension API. If you want to try nonetheless, you need to manually compile PySide against PyPy which will be somewhat difficult and cumbersome because the CMake build scripts of PySide will not detect PyPy out of the box, so you'll need a lot of tweaking. From frank at ohufx.com Sat May 26 22:34:58 2012 From: frank at ohufx.com (Frank Rueter | OHUfx) Date: Sun, 27 May 2012 08:34:58 +1200 Subject: [PySide] image format for best performance Message-ID: <4FC13E72.1020406@ohufx.com> Hi all, is there a recommended image format to use to get best performance out of QT? I've got some animated widgets (sliding docks etc.) that are populated with several small icons (currently 8bit pngs). Since the animation is a bit jerky I will have to go hunting for ways top optimise the code, and thought the image format for the icons would be a good place to start. Any experience and/or recommendations for this? Cheers, frank From sebastian at risefx.com Sun May 27 12:24:45 2012 From: sebastian at risefx.com (Sebastian Elsner) Date: Sun, 27 May 2012 12:24:45 +0200 Subject: [PySide] image format for best performance In-Reply-To: <4FC13E72.1020406@ohufx.com> References: <4FC13E72.1020406@ohufx.com> Message-ID: <4FC200ED.3060607@risefx.com> I did not notice a speed difference in formats, but since you are bound to use QPixmaps or QIcons (dont use Qimage/qpicture, its for manipulation of image data), which you pass your icon path to, and thus instanciate and read the actual data, I am always using a simple cache algorithm which would remember the whole pixmap after loading the data. You can do this in a separate dict with itemname:QPixmap once per item/icon or if you are using come kind of itemmodel in the item used. So at least in most cases you dont have to load your icon data twice. For exmaple in a TreeView or TableView this greatly improves scrolling speed. Am 26.05.2012 22:34, schrieb Frank Rueter | OHUfx: > Hi all, > > is there a recommended image format to use to get best performance out > of QT? > I've got some animated widgets (sliding docks etc.) that are populated > with several small icons (currently 8bit pngs). > Since the animation is a bit jerky I will have to go hunting for ways > top optimise the code, and thought the image format for the icons would > be a good place to start. > > Any experience and/or recommendations for this? > > Cheers, > frank > > _______________________________________________ > PySide mailing list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside From frank at ohufx.com Sun May 27 12:31:42 2012 From: frank at ohufx.com (Frank Rueter | OHUfx) Date: Sun, 27 May 2012 22:31:42 +1200 Subject: [PySide] image format for best performance In-Reply-To: <4FC200ED.3060607@risefx.com> References: <4FC13E72.1020406@ohufx.com> <4FC200ED.3060607@risefx.com> Message-ID: <4FC2028E.9010502@ohufx.com> Cool, thanks for the tips Sebastian, will have a go at these! Cheers, frank On 27/05/12 10:24 PM, Sebastian Elsner wrote: > I did not notice a speed difference in formats, but since you are bound > to use QPixmaps or QIcons (dont use Qimage/qpicture, its for > manipulation of image data), which you pass your icon path to, and thus > instanciate and read the actual data, I am always using a simple cache > algorithm which would remember the whole pixmap after loading the data. > You can do this in a separate dict with itemname:QPixmap once per > item/icon or if you are using come kind of itemmodel in the item used. > So at least in most cases you dont have to load your icon data twice. > For exmaple in a TreeView or TableView this greatly improves scrolling > speed. > > > Am 26.05.2012 22:34, schrieb Frank Rueter | OHUfx: >> Hi all, >> >> is there a recommended image format to use to get best performance out >> of QT? >> I've got some animated widgets (sliding docks etc.) that are populated >> with several small icons (currently 8bit pngs). >> Since the animation is a bit jerky I will have to go hunting for ways >> top optimise the code, and thought the image format for the icons would >> be a good place to start. >> >> Any experience and/or recommendations for this? >> >> Cheers, >> frank >> >> _______________________________________________ >> PySide mailing list >> PySide at qt-project.org >> http://lists.qt-project.org/mailman/listinfo/pyside From alex.gronholm at nextday.fi Sun May 27 14:29:52 2012 From: alex.gronholm at nextday.fi (=?ISO-8859-1?Q?Alex_Gr=F6nholm?=) Date: Sun, 27 May 2012 15:29:52 +0300 Subject: [PySide] PySide 1.1.1 Windows builds In-Reply-To: References: <4FBE9415.2000207@nextday.fi> Message-ID: <4FC21E40.3080107@nextday.fi> 25.05.2012 13:54, Roman Lacko kirjoitti: > Hi, > > 2012/5/24 Alex Grönholm: >> 19.05.2012 00:47, Roman Lacko kirjoitti: >>> Hi, >>> >>> I have prepared windows builds of PySide 1.1.1 for Pyhon 2.7 and 3.2, >>> 32bit and 64bit. >>> Builds will be available on my google Drive [1], until the PySide site >>> is not moved to new place. >>> For that reason, installing with easy_install is not available, You >>> ned to download the package and install manually. >>> You can allways let me know if you have better idea where to put the builds. >> How about the PyPI, where they are supposed be uploaded to? > It will be uploaded when the PySide site move to new location is completed Ok, but I don't quite understand why this hinges on something seemingly irrelevant like the move of some web site to another location. > > BTW, I'm developing new version of PySide setup script fully > compatible with distutils, that will be usable also on linux (binaries > will be built inplace), the development can be tracked here: > https://github.com/lck/pyside-dist > > Regards > Roman > >>> Whats new in packages: >>> >>> - OpenSSL DLLs included and installed out-of-the-box (now you can use >>> webkit with ssl) >>> - typesystems, >>> - PySide and Shibken c++ headers >>> - complete examples (from pyside gitorious repo) >>> >>> Whats new in build scripts [2]: >>> >>> - build now creates nice folder structure for every configuration. For >>> example after completed building with python 2.7 32bit release and >>> debug, you get: >>> ..\setuptools\build\py2.7-qt4.7.4-32bit-debug\ >>> ..\setuptools\build\py2.7-qt4.7.4-32bit-release\ >>> ..\setuptools\install\py2.7-qt4.7.4-32bit-debug\ >>> ..\setuptools\install\py2.7-qt4.7.4-32bit-release\ >>> ..\setuptools\packages\PySide-1.1.1qt474.win-amd64-py2.7.exe >>> ..\setuptools\packages\PySide-1.1.1qt474dbg.win-amd64-py2.7.exe >>> - debug build support >>> - refactoring and preparation for building on linux (not completed) >>> >>> >>> [1] https://docs.google.com/open?id=0B0aOk3P0ndoLUlhZSDUwWGU5MjQ >>> [2] https://github.com/PySide/packaging/tree/master/setuptools >>> _______________________________________________ >>> 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 From af.stackless at gmail.com Sun May 27 21:25:37 2012 From: af.stackless at gmail.com (Andy Frances) Date: Sun, 27 May 2012 15:25:37 -0400 Subject: [PySide] Questions about Pyside Assistant Message-ID: Hi Folks: I am new to N9 development. I am using Pyside Assistant to run the Pyside Mobility sample application showcoordinates.py. This is a simple console Python application. However I need aegis to give it permission. I have a few conceptual problems. 1) When I generate a showcoordinates direction with psa init, I get a showcoordinate template. Do I replace it with my copy of showcoordinates.py? 2) What does my setup.py file look like? I am also new to setup.py. 3) Does the application get a default application icon (the generated files imply this)? 4) Should the application be placed in /opt? (right now it is going in the same directory as where it resides on my host machine!) Any help would be appreciated. Cheers, Andrew -------------- next part -------------- An HTML attachment was scrubbed... URL: From th.perl at gmail.com Sun May 27 22:01:14 2012 From: th.perl at gmail.com (Thomas Perl) Date: Sun, 27 May 2012 22:01:14 +0200 Subject: [PySide] Questions about Pyside Assistant In-Reply-To: References: Message-ID: Hi Andy, 2012/5/27 Andy Frances : > I am new to N9 development. I am using Pyside Assistant to run the Pyside > Mobility sample application showcoordinates.py. This is a simple console > Python application. However I need aegis to give it permission. I have a few > conceptual problems. > >   1) When I generate a showcoordinates direction with psa init, I get a >   showcoordinate template. Do I replace it with my copy of > showcoordinates.py? Yes, you can overwrite the "showcoordinates" file in the generated directory with your showcoordinates.py file (make sure to remove the ".py" at the end, otherwise you would have to update the path to it in at least setup.py and the .desktop file). >   2) What does my setup.py file look like? I am also new to setup.py. For some general information on what setup.py does in the Python world, see this document: http://docs.python.org/distutils/setupscript.html The generated setup.py file should be okay - you can edit the description there if you want and make sure that the maintainer name and e-mail is okay. The "scripts=" tells which script is used as "main" application script. >   3) Does the application get a default application icon (the generated > files >    imply this)? In the generated folder there should be a showcoordinates.png file that you can replace with your own (80x80 PNG) icon. To create an icon, you can go to this URL and download the "Nokia Icon Toolkit" .zip file which includes an SVG template for the squircle with the right colors and guide lines, etc..: http://www.developer.nokia.com/Resources/Library/Design_and_UX/#!designing-for-nokia-platforms/designing-for-meego-12-harmattan/meego-launcher-icon-templates/meego-inkscape-launcher-icon-1.html >   4) Should the application be placed in /opt? (right now it is going in the >   same directory as where it resides on my host machine!) If you are targetting just your own device or apps.formeego.org, the default setup is okay. If you want to publish your app in Nokia Store (via Ovi Publish), you have to make sure that most files (except for the .desktop file, which has to go into /usr/share/applications) are installed into /opt. There are some blogs that talk about this, e.g.: http://techtransit.blogspot.com/2012/04/n9-trick-to-get-pyside-assistant-to.html http://techtransit.blogspot.com/2012/05/more-hints-on-publishing-pyside-qml.html HTH :) Thomas From af.stackless at gmail.com Sun May 27 22:26:44 2012 From: af.stackless at gmail.com (Andy Frances) Date: Sun, 27 May 2012 16:26:44 -0400 Subject: [PySide] Questions about Pyside Assistant In-Reply-To: References: Message-ID: Hi Thomas: Thanks for the reply. I still have some questions On Sun, May 27, 2012 at 4:01 PM, Thomas Perl wrote: Hi Andy, > > > 2) What does my setup.py file look like? I am also new to setup.py. > > For some general information on what setup.py does in the Python > world, see this document: > http://docs.python.org/distutils/setupscript.html > Yes I am reading up on setup.py. > > The generated setup.py file should be okay - you can edit the > description there if you want and make sure that the maintainer name > and e-mail is okay. The "scripts=" tells which script is used as > "main" application script. > The default is installing the main programme (showcoordinates) in the home/andrew directory of my N9 (which is the same as my home machine, a Ubuntu machine)? How do I change this to go in say, /opt on the N9? On the Joshua King blog, he talks about setting sys.prefix. However it is unclear to me where this is set? I am assuming setup.py. I know from using setup.py, I can give setup.py command line options like prefix. > 3) Does the application get a default application icon (the generated > > files > > imply this)? > > In the generated folder there should be a showcoordinates.png file > that you can replace with your own (80x80 PNG) icon. To create an > icon, you can go to this URL and download the "Nokia Icon Toolkit" > .zip file which includes an SVG template for the squircle with the > right colors and guide lines, etc..: > Thanks. On a related note, I don't see an icon generated for my application (yes yes it is a command line programme). Another question: my aegis manifest is empty. I am assuming it is my responsible to populate it accordingly? To put everything into context, for now I want the ability to run command line programmes that need the security properly set. I thought that Pyside Assistant would take care of the Aegis framework for me. I am happy to invoke the programme from the command line or from pressing an icon. Cheers, Andrew -------------- next part -------------- An HTML attachment was scrubbed... URL: From techtonik at gmail.com Mon May 28 08:09:56 2012 From: techtonik at gmail.com (anatoly techtonik) Date: Mon, 28 May 2012 09:09:56 +0300 Subject: [PySide] PySide 1.1.1 Windows builds In-Reply-To: References: <4FBE9415.2000207@nextday.fi> Message-ID: On Fri, May 25, 2012 at 1:54 PM, Roman Lacko wrote: > > 2012/5/24 Alex Grönholm : >> 19.05.2012 00:47, Roman Lacko kirjoitti: >>> >>> I have prepared windows builds of PySide 1.1.1 for Pyhon 2.7 and 3.2, >>> 32bit and 64bit. >>> Builds will be available on my google Drive [1], until the PySide site >>> is not moved to new place. >>> For that reason, installing with easy_install is not available, You >>> ned to download the package and install manually. >>> You can allways let me know if you have better idea where to put the builds. >> How about the PyPI, where they are supposed be uploaded to? > > It will be uploaded when the PySide site move to new location is completed What's the progress with move? We can already move it to GitHub pages. > BTW, I'm developing new version of PySide setup script fully > compatible with distutils, that will be usable also on linux (binaries > will be built inplace), the development can be tracked here: > https://github.com/lck/pyside-dist How does it relate to existing BuildScripts? From backup.rlacko at gmail.com Mon May 28 08:47:44 2012 From: backup.rlacko at gmail.com (Roman Lacko) Date: Mon, 28 May 2012 08:47:44 +0200 Subject: [PySide] PySide 1.1.1 Windows builds In-Reply-To: <4FC081F7.8070008@skynet.be> References: <4FBE9415.2000207@nextday.fi> <4FC081F7.8070008@skynet.be> Message-ID: Hi Stefan, 2012/5/26 Stefan Champailler : > That's great news ! I really depend on those. > I'm currently trying to set up a PC with Windows so that I can build > PySide too. > > BTW, is building PySide on Windows hard ? I mean, on linux, it's rather > easy and I'd like to know if it's as easy on Win. > With my build scripts It's simple and easy :-) You just need some prerequisities installed on your system. Please follow the build instructions here: http://qt-project.org/wiki/Building_PySide_on_Windows Regards Roman > stF > > > > On 05/25/2012 12:54 PM, Roman Lacko wrote: >> Hi, >> >> 2012/5/24 Alex Grönholm: >>> 19.05.2012 00:47, Roman Lacko kirjoitti: >>>> Hi, >>>> >>>> I have prepared windows builds of PySide 1.1.1 for Pyhon 2.7 and 3.2, >>>> 32bit and 64bit. >>>> Builds will be available on my google Drive [1], until the PySide site >>>> is not moved to new place. >>>> For that reason, installing with easy_install is not available, You >>>> ned to download the package and install manually. >>>> You can allways let me know if you have better idea where to put the builds. >>> How about the PyPI, where they are supposed be uploaded to? >> >> It will be uploaded when the PySide site move to new location is completed >> >> BTW, I'm developing new version of PySide setup script fully >> compatible with distutils, that will be usable also on linux (binaries >> will be built inplace), the development can be tracked here: >> https://github.com/lck/pyside-dist >> >> Regards >> Roman >> >>>> >>>> Whats new in packages: >>>> >>>> - OpenSSL DLLs included and installed out-of-the-box (now you can use >>>> webkit with ssl) >>>> - typesystems, >>>> - PySide and Shibken c++ headers >>>> - complete examples (from pyside gitorious repo) >>>> >>>> Whats new in build scripts [2]: >>>> >>>> - build now creates nice folder structure for every configuration. For >>>> example after completed building with python 2.7 32bit release and >>>> debug, you get: >>>>       ..\setuptools\build\py2.7-qt4.7.4-32bit-debug\ >>>>       ..\setuptools\build\py2.7-qt4.7.4-32bit-release\ >>>>       ..\setuptools\install\py2.7-qt4.7.4-32bit-debug\ >>>>       ..\setuptools\install\py2.7-qt4.7.4-32bit-release\ >>>>       ..\setuptools\packages\PySide-1.1.1qt474.win-amd64-py2.7.exe >>>>       ..\setuptools\packages\PySide-1.1.1qt474dbg.win-amd64-py2.7.exe >>>> - debug build support >>>> - refactoring and preparation for building on linux (not completed) >>>> >>>> >>>> [1] https://docs.google.com/open?id=0B0aOk3P0ndoLUlhZSDUwWGU5MjQ >>>> [2] https://github.com/PySide/packaging/tree/master/setuptools >>>> _______________________________________________ >>>> 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 list >> PySide at qt-project.org >> http://lists.qt-project.org/mailman/listinfo/pyside >> > > -- > Timeo Danaos et dona ferentes > _______________________________________________ > PySide mailing list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside From backup.rlacko at gmail.com Mon May 28 10:07:08 2012 From: backup.rlacko at gmail.com (Roman Lacko) Date: Mon, 28 May 2012 10:07:08 +0200 Subject: [PySide] PySide 1.1.1 Windows builds In-Reply-To: References: <4FBE9415.2000207@nextday.fi> Message-ID: Hi Anatoly and list, 2012/5/28 anatoly techtonik : > On Fri, May 25, 2012 at 1:54 PM, Roman Lacko wrote: >> >> 2012/5/24 Alex Grönholm : >>> 19.05.2012 00:47, Roman Lacko kirjoitti: >>>> >>>> I have prepared windows builds of PySide 1.1.1 for Pyhon 2.7 and 3.2, >>>> 32bit and 64bit. >>>> Builds will be available on my google Drive [1], until the PySide site >>>> is not moved to new place. >>>> For that reason, installing with easy_install is not available, You >>>> ned to download the package and install manually. >>>> You can allways let me know if you have better idea where to put the builds. >>> How about the PyPI, where they are supposed be uploaded to? >> >> It will be uploaded when the PySide site move to new location is completed > > What's the progress with move? We can already move it to GitHub pages. GitHub will not work with easy_install, because GitHub downloads page uses https and easy_install does not work very well with https. Another problem is that PySide project is using free edition of GitHub and there is a size limit (200 Mb or so). One windows package has cca 40 Mb and there are 6 packages (for py 2.6, 2.7, 3.2 - 32 and 64 bit), this is cca 240 Mb for each release. I will try to contact Qt project administrators if we can't upload binaries to Qt Project site. > >> BTW, I'm developing new version of PySide setup script fully >> compatible with distutils, that will be usable also on linux (binaries >> will be built inplace), the development can be tracked here: >> https://github.com/lck/pyside-dist > > How does it relate to existing BuildScripts? If completed, it should replace the existing build scripts and will enable to install PySide with pip. Regards Roman From mairas at iki.fi Mon May 28 10:53:52 2012 From: mairas at iki.fi (Matti Airas) Date: Mon, 28 May 2012 11:53:52 +0300 Subject: [PySide] PySide 1.1.1 Windows builds In-Reply-To: References: <4FBE9415.2000207@nextday.fi> Message-ID: On 28 May 2012 11:07, Roman Lacko wrote: > I will try to contact Qt project administrators if we can't upload > binaries to Qt Project site. The proposal by the Qt Project people was to use their facilities for the binary (and source) distribution. The files would probably be linked here: http://qt-project.org/downloads and the actual file distribution would be from releases.qt-project.org. I've been planning to push the whole website migration thing forward but have been a bit tied up by other issues. I'll try to return to that within a couple of days, but meanwhile, Roman, I'll send you the names of the Qt maintainers - it's probably the easiest if you could ask them directly. Cheers, ma. From yedderson at myopera.com Mon May 28 14:13:50 2012 From: yedderson at myopera.com (Yedderson) Date: Mon, 28 May 2012 13:13:50 +0100 Subject: [PySide] Hello Message-ID: Hello Community ! My name's Hassen, I'm new to the list, I just want to say hello and tell you I'm pretty impressed by the way you work ! I'm still initiating with PySide, I did this little application[1] so far to get some hands-on with it (it was a nice experience) and now I'm trying to get things done with Qt Quick.. the mail-list archive was also pleasant to read :) you guys rocks ! H. [1] https://github.com/yedderson/PyBooklet From th.perl at gmail.com Mon May 28 15:50:09 2012 From: th.perl at gmail.com (Thomas Perl) Date: Mon, 28 May 2012 15:50:09 +0200 Subject: [PySide] Questions about Pyside Assistant In-Reply-To: References: Message-ID: Hi Andy, 2012/5/27 Andy Frances : > On Sun, May 27, 2012 at 4:01 PM, Thomas Perl wrote: >> The generated setup.py file should be okay - you can edit the >> description there if you want and make sure that the maintainer name >> and e-mail is okay. The "scripts=" tells which script is used as >> "main" application script. > > The default is installing the main programme (showcoordinates) in the > home/andrew directory of my N9 > (which is the same as my home machine, a Ubuntu machine)? How do I change > this to go in say, /opt > on the N9? No, the default for PySide Assistant should be to install into /usr. What did you change from the defaults? For most of my projects, I'm not using PySide Assistant directly, but do the packaging via debhelper and friends. I do understand that PSA is easier for people to use. My installation schema is usually the following: /opt//bin/ - the main program /opt//lib/ - python library modules go here (add this to sys.path) /opt//qml/ - QML files and related stuff (e.g. images for the UI) /opt//... - other files that the program needs (resources, etc..) /opt//.png - the icon /usr/share/applications/.desktop - the desktop file > On the Joshua King blog, he talks about setting sys.prefix. However it is > unclear to me where this is set? > I am assuming setup.py. I know from using setup.py, I can give setup.py > command line options like prefix. You can get all accepted options for the "install" comamnd via "python setup.py --help install". It should be --prefix (python setup.py --prefix=/opt/mypackage/ install). >> In the generated folder there should be a showcoordinates.png file >> that you can replace with your own (80x80 PNG) icon. To create an >> icon, you can go to this URL and download the "Nokia Icon Toolkit" >> .zip file which includes an SVG template for the squircle with the >> right colors and guide lines, etc..: > > Thanks. On a related note, I don't see an icon generated for my application > (yes yes it is a command line programme). Which version of PySide Assistant are you using? I've tried it with the current Git master branch HEAD (https://github.com/PySide/PySideAssistant) and a "psa init" did add the .png file there. > Another question: my aegis manifest is empty. I am assuming it is my > responsible to populate it accordingly? Yes, although you only need it if you use some APIs. A list of which APIs need which tokens can be found here: http://harmattan-dev.nokia.com/docs/library/html/guide/html/Developer_Library_Developing_for_Harmattan_Harmattan_security_Security_guide_Harmattan_APIs_that_require_credentials.html > To put everything into context, for now I want the ability to run command > line programmes that need the security properly > set. I thought that Pyside Assistant would take care of the Aegis framework > for me. I am happy to invoke the programme from the > command line or from pressing an icon. I've created a mini-package about a week ago that has permissions for the calendar (because I wanted to play with the Qt Mobility Organizer APIs interactively to see how I can get the data). The binary package is here: http://thp.io/2012/maemo/python-organizer-console_1.0.0_all.deb You can get the source package with the "dget" utility from here: http://thp.io/2012/maemo/python-organizer-console_1.0.0.dsc It shows, among other things: - how to use only debhelper and no PSA - how to create a script that gives an interactive console - how an .aegis file could look like (in debian/) If the PSA patches that make it install into /opt are not merged soon, and there is demand for an easier solution, I might just publish a short guide that shows how to use debhelper 7 and some makefile/setup.py magic to create packaging for a simple Harmattan Python app from scratch (but I would love for someone to get PSA into a "Store ready" state, because that's what PSA is all about IMHO ;). HTH :) Thomas From hugo.lima at openbossa.org Mon May 28 19:51:34 2012 From: hugo.lima at openbossa.org (Hugo Parente Lima) Date: Mon, 28 May 2012 14:51:34 -0300 Subject: [PySide] Accessing C++ QObject subclass from script? In-Reply-To: <4FBFF504.9080601@fxtech.com> References: <4FBC01C8.6020503@fxtech.com> <4FBFF504.9080601@fxtech.com> Message-ID: <1443042.EPbW4qZaeO@hugodesktop> On Friday, May 25, 2012 04:09:24 PM Paul Miller wrote: > On 5/25/2012 3:59 PM, Alexander Jones wrote: > > Something like this? > > > > http://www.pyside.org/docs/shiboken/shibokenmodule.html#shiboken.wrapInsta > > nce > Yeah I saw that but it wasn't apparent how I was to use this from within > my C++ application. Is there a tutorial or example for how to use this? Maybe the unit test code could help you: https://github.com/PySide/Shiboken/blob/master/tests/shibokenmodule/module_test.py Regards > > On 22 May 2012 22:14, Paul Miller > > > > wrote: > > I got PySide 1.1 working in my hybrid app (thanks Roman for your > > help!) > > using Qt 4.7.4 and Python 2.7. I can create new Qt widgets and use my > > existing C++ QApplication and it seems to work fine. > > > > Now I would like to expose some of my internal QObjects that have > > already been created by the time the Python interpreter starts up. > > > > For my existing data model interface to Python, I've hand-created some > > Python bindings for my classes. What do I need to do to return a > > pointer > > to one of my created QObjects back into PySide, so I can connect > > signals > > to it? > > > > *I can't use an auto-type-generator thingy like in the superhybrids > > article*. > > > > _______________________________________________ > > PySide mailing list > > PySide at qt-project.org > > http://lists.qt-project.org/mailman/listinfo/pyside > > > > -- > > Alexander Jones > > Double Negative R&D > > www.dneg.com > > _______________________________________________ > PySide mailing list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part. URL: From yann.lanthony at gmail.com Mon May 28 21:04:07 2012 From: yann.lanthony at gmail.com (Yann Lanthony) Date: Mon, 28 May 2012 21:04:07 +0200 Subject: [PySide] Shiboken - Documentation support when binding a C++ library Message-ID: Hi everyone, I am new to this list, so first of all thanks for your work ! PySide and shiboken are such great tools. Actually, I had already created a post in the PySide googlegroups but if I get it right, here is a better place to ask questions :) So, I am currently using shiboken to do the python binding of my C++ Qt-based project. I am starting to get really nice results, as I'm now able to end up with a .pyd file that allows me to call my C++ functions within a python interpreter. However, I have trouble understanding how I can also retrieve my C++ doxygen-style documentation as python docstrings, to be able to print functions doc at runtime with help(myFunction). I am working on Windows (64bits) and I have compiled shiboken (and pyside) with docstring support (well, I linked libxml2 and libxslt which seem to be the 2 documentation related libraries). Is it at least possible to do something like that with the current version of shiboken (1.1.1) ? Many thanks, Yann -------------- next part -------------- An HTML attachment was scrubbed... URL: From backup.rlacko at gmail.com Tue May 29 14:26:44 2012 From: backup.rlacko at gmail.com (Roman Lacko) Date: Tue, 29 May 2012 14:26:44 +0200 Subject: [PySide] PySide 1.1.1 Windows builds In-Reply-To: References: <4FBE9415.2000207@nextday.fi> Message-ID: Hi list, The PySide version 1.1.1 is registerd on PyPI and You can now install packages with easy_install (read install instructions here [1]). Thanks to Matti and guys from Qt OSS department, we can use new location for PySide packages [2] Also the new build scripts development is going well [3] and I plan to release the scripts soon. [1] http://qt-project.org/wiki/PySide_Binaries_Windows [3] http://releases.qt-project.org/pyside/1.1.1/ [4] https://github.com/lck/pyside-dist Regards Roman 2012/5/28 Matti Airas : > On 28 May 2012 11:07, Roman Lacko wrote: > >> I will try to contact Qt project administrators if we can't upload >> binaries to Qt Project site. > > The proposal by the Qt Project people was to use their facilities for > the binary (and source) distribution. The files would probably be > linked here: > > http://qt-project.org/downloads > > and the actual file distribution would be from releases.qt-project.org. > > I've been planning to push the whole website migration thing forward > but have been a bit tied up by other issues. I'll try to return to > that within a couple of days, but meanwhile, Roman, I'll send you the > names of the Qt maintainers - it's probably the easiest if you could > ask them directly. > > Cheers, > > ma. From mairas at iki.fi Wed May 30 13:13:59 2012 From: mairas at iki.fi (Matti Airas) Date: Wed, 30 May 2012 14:13:59 +0300 Subject: [PySide] Restructured the wiki landing page Message-ID: Hi, Anticipating the rampdown of the original website, I shoveled a bit of the information to the wiki, and while at it, also restructured the wiki landing page: http://qt-project.org/wiki/Category:LanguageBindings::PySide Feel free to modify further. :-) Remaining bits to do: - Have the static documentation moved to http://qt-project/doc - Have the original Bugzilla archived somewhere. The Bugzilla issue was discussed already earlier. It can be either done as an HTML mirror or a full Bugzilla mirror. I'd prefer the latter because of the better search functionality (but then again, maybe a Google special search would do just as well). HTML mirror can of course be performed by anyone; if you want to have a go at replicating the actual Bugzilla, I'd be happy to provide the MySQL dumps for the purpose (sans the emails and passwords). Cheers, ma. From nathanjsmith at gmail.com Wed May 30 16:37:15 2012 From: nathanjsmith at gmail.com (Nathan Smith) Date: Wed, 30 May 2012 09:37:15 -0500 Subject: [PySide] Bug with new hash table feature Message-ID: Hi all, I uncovered a bug in the new hash feature of Shiboken in the repositories. I'll submit a bug report when the system comes back up, but I'm interested in actually submitting a patch. The new hash function doesn't check to see if the provided object is valid before calculating the hash value. This can be seen in shiboken/libshiboken/basewrapper.cpp at line 754 and demonstrated using the following code: >>> from PySide import QtCore >>> import shiboken >>> a = QtCore.QObject() >>> b = QtCore.QObject(a) >>> shiboken.isValid(a) True >>> shiboken.isValid(b) True >>> del a >>> shiboken.isValid(b) False >>> hash(b) Segmentation Fault I made the following change in the code, but it didn't do what I expected: if (!isValid(pyObj)) { return 0L; } isValid raises a RuntimeException in addition to returning false, but I never see that exception in Python. So, I have two questions: 1. Is there some other code that should be checking isValid before the call ever gets to the C++ hash() function? 2. How do you correctly raise an exception from within Shiboken? Nathan -------------- next part -------------- An HTML attachment was scrubbed... URL: From jcummings2 at users.sf.net Wed May 30 18:33:54 2012 From: jcummings2 at users.sf.net (John Cummings) Date: Wed, 30 May 2012 11:33:54 -0500 Subject: [PySide] Cannot pass enum by value Message-ID: <4FC64BF2.8000605@users.sf.net> I uncovered a bug with the latest Shiboken when passing an enum by reference. Like Nathan, I have a bug report ready to go when the system comes back up. In previous versions, you could pass an enum by reference to a method like so: class Val { enum ValEnum { One, Other }; ValEnum oneOrTheOtherEnumValue(ValEnum& enumValue) { return enumValue == One ? Other : One; } } You get compilation errors with version 1.1.1 and with HEAD. The generated code is trying to use a pointer instead of reference. Previous version of Shiboken worked just fine (specifically, 1.0.7 worked, I'm not sure about other versions until 1.1.1). Unlike Nathan, I don't have a patch ready. The workaround is obviously to pass an enum by value not by reference (which is admittedly a bit overkill anyway). If someone knows of a better solution, I'm willing to try it. John Cummings From jcummings2 at users.sf.net Wed May 30 19:46:25 2012 From: jcummings2 at users.sf.net (John Cummings) Date: Wed, 30 May 2012 12:46:25 -0500 Subject: [PySide] Cannot pass enum by value In-Reply-To: <4FC64BF2.8000605@users.sf.net> References: <4FC64BF2.8000605@users.sf.net> Message-ID: <4FC65CF1.6010501@users.sf.net> On 05/30/2012 11:33 AM, John Cummings wrote: > I uncovered a bug with the latest Shiboken when passing an enum by reference. > Like Nathan, I have a bug report ready to go when the system comes back up. > > In previous versions, you could pass an enum by reference to a method like so: > > class Val > { > enum ValEnum { One, Other }; > ValEnum oneOrTheOtherEnumValue(ValEnum& enumValue) { return enumValue == One ? > Other : One; } > } > > You get compilation errors with version 1.1.1 and with HEAD. The generated code > is trying to use a pointer instead of reference. Previous version of Shiboken > worked just fine (specifically, 1.0.7 worked, I'm not sure about other versions > until 1.1.1). > > Unlike Nathan, I don't have a patch ready. The workaround is obviously to pass > an enum by value not by reference (which is admittedly a bit overkill anyway). > If someone knows of a better solution, I'm willing to try it. > > John Cummings Just submitted the bug as PYSIDE-71. From hugo.lima at openbossa.org Wed May 30 20:09:18 2012 From: hugo.lima at openbossa.org (Hugo Parente Lima) Date: Wed, 30 May 2012 15:09:18 -0300 Subject: [PySide] Cannot pass enum by value In-Reply-To: <4FC64BF2.8000605@users.sf.net> References: <4FC64BF2.8000605@users.sf.net> Message-ID: <3835909.RiKXCFUrRb@hugodesktop> On Wednesday, May 30, 2012 11:33:54 AM John Cummings wrote: > I uncovered a bug with the latest Shiboken when passing an enum by > reference. Like Nathan, I have a bug report ready to go when the system > comes back up. > > In previous versions, you could pass an enum by reference to a method like > so: > > class Val > { > enum ValEnum { One, Other }; > ValEnum oneOrTheOtherEnumValue(ValEnum& enumValue) { return enumValue == One > ? Other : One; } > } > > You get compilation errors with version 1.1.1 and with HEAD. The generated > code is trying to use a pointer instead of reference. Previous version of > Shiboken worked just fine (specifically, 1.0.7 worked, I'm not sure about > other versions until 1.1.1). > > Unlike Nathan, I don't have a patch ready. The workaround is obviously to > pass an enum by value not by reference (which is admittedly a bit overkill > anyway). If someone knows of a better solution, I'm willing to try it. It can be passed by value unless it's a const reference, otherwise the following code wont work. enum ValEnum { One, Other }; void oneOrTheOtherEnumValue(ValEnum& enumValue) { enumValue = Other; } ValEnum a = One; oneOrTheOtherEnumValue(a); // a should be equal to Other. This is why Shiboken tries to use a pointer, but the code seems wrong anyway :-/. > John Cummings > _______________________________________________ > PySide mailing list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part. URL: From hugo.lima at openbossa.org Wed May 30 20:12:39 2012 From: hugo.lima at openbossa.org (Hugo Parente Lima) Date: Wed, 30 May 2012 15:12:39 -0300 Subject: [PySide] Bug with new hash table feature In-Reply-To: References: Message-ID: <1705919.z2CsdCvgXW@hugodesktop> On Wednesday, May 30, 2012 09:37:15 AM Nathan Smith wrote: > Hi all, > > I uncovered a bug in the new hash feature of Shiboken in the repositories. > I'll submit a bug report when the system comes back up, but I'm interested > in actually submitting a patch. > > The new hash function doesn't check to see if the provided object is valid > before calculating the hash value. This can be seen > in shiboken/libshiboken/basewrapper.cpp at line 754 and demonstrated using > > the following code: > >>> from PySide import QtCore > >>> import shiboken > >>> a = QtCore.QObject() > >>> b = QtCore.QObject(a) > >>> shiboken.isValid(a) > > True > > >>> shiboken.isValid(b) > > True > > >>> del a > >>> shiboken.isValid(b) > > False > > >>> hash(b) > > Segmentation Fault > > > I made the following change in the code, but it didn't do what I expected: > > if (!isValid(pyObj)) { > return 0L; > } > > > isValid raises a RuntimeException in addition to returning false, but I > never see that exception in Python. So, I have two questions: > > 1. Is there some other code that should be checking isValid before the > call ever gets to the C++ hash() function? > 2. How do you correctly raise an exception from within Shiboken? You raise an exception using the Python C API. Use the PyErr_* functions, e.g. PyErr_SetString > Nathan -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part. URL: From nathanjsmith at gmail.com Wed May 30 20:31:48 2012 From: nathanjsmith at gmail.com (Nathan Smith) Date: Wed, 30 May 2012 13:31:48 -0500 Subject: [PySide] Bug with new hash table feature In-Reply-To: <1705919.z2CsdCvgXW@hugodesktop> References: <1705919.z2CsdCvgXW@hugodesktop> Message-ID: Hi Hugo, Thanks for getting back to me. I thought that was the case, and the PyErr_format function is called within the isValid(PyObject *pyObj) function, but no exception is ever raised in Python. The following is called from within the body of isValid(PyObject *pyObj): PyErr_Format(PyExc_RuntimeError, "Internal C++ object (%s) already deleted.", pyObj->ob_type->tp_name); My modified hash function *always* returns even when called on invalid objects. The updated code is shown below for reference: Py_hash_t hash(PyObject* pyObj) { assert(Shiboken::Object::checkType(pyObj)); // begin changes by Nathan if (!isValid(pyObj)) { return 0L; } // end changes by Nathan return reinterpret_cast(reinterpret_cast(pyObj)->d->cptr[0]); } // Unchanged bool isValid(PyObject* pyObj) { if (!pyObj || pyObj == Py_None || Py_TYPE(pyObj->ob_type) != &SbkObjectType_Type) { return true; } SbkObjectPrivate* priv = reinterpret_cast(pyObj)->d; if (!priv->cppObjectCreated && isUserType(pyObj)) { PyErr_Format(PyExc_RuntimeError, "'__init__' method of object's base class (%s) not called.", pyObj->ob_type->tp_name); return false; } if (!priv->validCppObject) { PyErr_Format(PyExc_RuntimeError, "Internal C++ object (%s) already deleted.", pyObj->ob_type->tp_name); return false; } return true; } When passed an invalid object, this function just returns 0 rather than raising a RuntimeError. Why isn't a runtime error raised? Nathan On Wed, May 30, 2012 at 1:12 PM, Hugo Parente Lima wrote: > On Wednesday, May 30, 2012 09:37:15 AM Nathan Smith wrote: > > Hi all, > > > > I uncovered a bug in the new hash feature of Shiboken in the > repositories. > > I'll submit a bug report when the system comes back up, but I'm > interested > > in actually submitting a patch. > > > > The new hash function doesn't check to see if the provided object is > valid > > before calculating the hash value. This can be seen > > in shiboken/libshiboken/basewrapper.cpp at line 754 and demonstrated > using > > > > the following code: > > >>> from PySide import QtCore > > >>> import shiboken > > >>> a = QtCore.QObject() > > >>> b = QtCore.QObject(a) > > >>> shiboken.isValid(a) > > > > True > > > > >>> shiboken.isValid(b) > > > > True > > > > >>> del a > > >>> shiboken.isValid(b) > > > > False > > > > >>> hash(b) > > > > Segmentation Fault > > > > > > I made the following change in the code, but it didn't do what I > expected: > > > > if (!isValid(pyObj)) { > > return 0L; > > } > > > > > > isValid raises a RuntimeException in addition to returning false, but I > > never see that exception in Python. So, I have two questions: > > > > 1. Is there some other code that should be checking isValid before the > > call ever gets to the C++ hash() function? > > 2. How do you correctly raise an exception from within Shiboken? > > You raise an exception using the Python C API. > > Use the PyErr_* functions, e.g. PyErr_SetString > > > Nathan -------------- next part -------------- An HTML attachment was scrubbed... URL: From jpe at wingware.com Wed May 30 20:38:13 2012 From: jpe at wingware.com (John Ehresman) Date: Wed, 30 May 2012 14:38:13 -0400 Subject: [PySide] Bug with new hash table feature In-Reply-To: References: <1705919.z2CsdCvgXW@hugodesktop> Message-ID: <4FC66915.7080601@wingware.com> On 5/30/12 2:31 PM, Nathan Smith wrote: > Hi Hugo, > > Thanks for getting back to me. I thought that was the case, and the > PyErr_format function is called within the isValid(PyObject *pyObj) > function, but no exception is ever raised in Python. Is -1 returned iff an exception is set? -1 indicates an exception; see the discussion of tp_hash in http://docs.python.org/c-api/typeobj.html John From nathanjsmith at gmail.com Wed May 30 20:46:41 2012 From: nathanjsmith at gmail.com (Nathan Smith) Date: Wed, 30 May 2012 13:46:41 -0500 Subject: [PySide] Bug with new hash table feature In-Reply-To: <4FC66915.7080601@wingware.com> References: <1705919.z2CsdCvgXW@hugodesktop> <4FC66915.7080601@wingware.com> Message-ID: Thanks John! I was returning 0L, but changed it to -1L. The hash function now raises a RuntimeError when called on an invalid object. I've opened bug PYSIDE-72 for this and will attach an updated patch to it using the -1L return. Nathan On Wed, May 30, 2012 at 1:38 PM, John Ehresman wrote: > On 5/30/12 2:31 PM, Nathan Smith wrote: > >> Hi Hugo, >> >> Thanks for getting back to me. I thought that was the case, and the >> PyErr_format function is called within the isValid(PyObject *pyObj) >> function, but no exception is ever raised in Python. >> > > Is -1 returned iff an exception is set? -1 indicates an exception; see > the discussion of tp_hash in http://docs.python.org/c-api/**typeobj.html > > John > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From jcummings2 at users.sf.net Wed May 30 20:50:39 2012 From: jcummings2 at users.sf.net (John Cummings) Date: Wed, 30 May 2012 13:50:39 -0500 Subject: [PySide] Cannot pass enum by reference In-Reply-To: <3835909.RiKXCFUrRb@hugodesktop> References: <4FC64BF2.8000605@users.sf.net> <3835909.RiKXCFUrRb@hugodesktop> Message-ID: <4FC66BFF.2010108@users.sf.net> On 05/30/2012 01:09 PM, Hugo Parente Lima wrote: > > It can be passed by value unless it's a const reference, otherwise the > following code wont work. > > enum ValEnum { One, Other }; > void oneOrTheOtherEnumValue(ValEnum& enumValue) { > enumValue = Other; > } > > ValEnum a = One; > oneOrTheOtherEnumValue(a); > // a should be equal to Other. > > This is why Shiboken tries to use a pointer, but the code seems wrong anyway > :-/. > The subject of the thread was incorrect before. Sorry for the confusion. Your statement "It can be passed by value unless it's a const reference" is circular. Passing by value is by definition not passing by reference. Perhaps I don't understand your statement. It also doesn't matter if the reference is const or not in this case, the current Shiboken still does not create a correct wrapper. Old versions did. John Cummings From nathanjsmith at gmail.com Wed May 30 21:19:07 2012 From: nathanjsmith at gmail.com (Nathan Smith) Date: Wed, 30 May 2012 14:19:07 -0500 Subject: [PySide] Bug with new hash table feature In-Reply-To: <4FC66CA7.8060903@wingware.com> References: <1705919.z2CsdCvgXW@hugodesktop> <4FC66915.7080601@wingware.com> <4FC66CA7.8060903@wingware.com> Message-ID: There had better never be an address at -1 or the hash function won't work right. The Python documentation ( http://docs.python.org/c-api/typeobj.html#PyTypeObject.tp_hash) says that hash should only return -1 if there was an exception set. If it's possible to have an address -1, then we need a different hash function than just returning address. We may want to do this differently anyway because the current hash approach is incompatible with weakref.WeakSet. >>> from PySide import QtCore >>> import weakref >>> objs = weakref.WeakSet() >>> a = QtCore.QObject() >>> b = QtCore.QObject(a) >>> objs.update([a,b]) >>> del a >>> objs.remove(b) Traceback (most recent call last): File "", line 1, in File "/usr/lib64/python2.7/_weakrefset.py", line 105, in remove self.data.remove(ref(item)) RuntimeError: Internal C++ object (PySide.QtCore.QObject) already deleted. I use weaksets extensively in my application to hold weak references to Qt objects. In the example above, when 'a' is deleted its reference is removed from the WeakSet automatically. b points at an invalid object, but its reference count remains high so it isn't removed from the WeakSet. When I try to remove b from the WeakSet manually (objs.remove(b)), a RuntimeError is raised because weak references apparently rely on the hashing function. So now a weak reference to 'b' is stuck in the WeakSet * forever*. Can we use the address of the Shiboken object as the hash value? That remains valid so long as there are references to the object, even after the object itself has been deleted in C++ land. Nathan On Wed, May 30, 2012 at 1:53 PM, John Ehresman wrote: > On 5/30/12 2:46 PM, Nathan Smith wrote: > >> Thanks John! I was returning 0L, but changed it to -1L. The hash >> function now raises a RuntimeError when called on an invalid object. >> I've opened bug PYSIDE-72 for this and will attach an updated patch to >> it using the -1L return. >> > > Took a peek at it -- you might want to test the return value in the non > exception case and return -2 if it's -1. I don't know if it's guaranteed > that an address can never be the bit value of -1. > > John > -------------- next part -------------- An HTML attachment was scrubbed... URL: From jpe at wingware.com Wed May 30 21:22:24 2012 From: jpe at wingware.com (John Ehresman) Date: Wed, 30 May 2012 15:22:24 -0400 Subject: [PySide] Is there a reason PySideSignalInstance source isn't incref'd? Message-ID: <4FC67370.108@wingware.com> I'm getting a segfault when I run the following: w.selectionModel().selectionChanged.connect(callback) with updated sources from git. I can work around it with m = w.selectionModel() m.selectionChanged.connect(callback) Stepping through the code in pysidesignal.cpp, it looks like d->source of a PySideSignalInstance object isn't incref'd so when the temporary ref returned by selectionModel is decref'd, the source is pointing at a invalid object. The function that initializes a PySideSignalInstance is instanceInitialize Oddly this only happens in a larger program and not in a simple test case, but I don't see how source can be assumed to be valid without increfing it. Thanks, John From jpe at wingware.com Wed May 30 21:35:08 2012 From: jpe at wingware.com (John Ehresman) Date: Wed, 30 May 2012 15:35:08 -0400 Subject: [PySide] Bug with new hash table feature In-Reply-To: References: <1705919.z2CsdCvgXW@hugodesktop> <4FC66915.7080601@wingware.com> <4FC66CA7.8060903@wingware.com> Message-ID: <4FC6766C.9010909@wingware.com> On 5/30/12 3:19 PM, Nathan Smith wrote: > There had better never be an address at -1 or the hash function won't > work right. The Python documentation > (http://docs.python.org/c-api/typeobj.html#PyTypeObject.tp_hash) says > that hash should only return -1 if there was an exception set. If it's > possible to have an address -1, then we need a different hash function > than just returning address. I'd simply return -2 if the value is -1 just to be on the safe side. > Can we use the address of the Shiboken object as the hash value? That > remains valid so long as there are references to the object, even after > the object itself has been deleted in C++ land. This works if there can only be one wrapper at a time for a given QObject. I don't know if this is the case. John From hugo.lima at openbossa.org Wed May 30 22:25:25 2012 From: hugo.lima at openbossa.org (Hugo Parente Lima) Date: Wed, 30 May 2012 17:25:25 -0300 Subject: [PySide] Bug with new hash table feature In-Reply-To: References: <4FC66915.7080601@wingware.com> Message-ID: <1972097.NJuGYJfl7A@hugodesktop> On Wednesday, May 30, 2012 01:46:41 PM Nathan Smith wrote: > Thanks John! I was returning 0L, but changed it to -1L. The hash function > now raises a RuntimeError when called on an invalid object. I've opened > bug PYSIDE-72 for this and will attach an updated patch to it using the -1L > return. Create a change on gerrit to push your patch, will be better =] http://qt-project.org/wiki/Gerrit-Introduction http://qt-project.org/wiki/Setting-up-Gerrit Just ask if you need help to push the patch to gerrit > Nathan > > On Wed, May 30, 2012 at 1:38 PM, John Ehresman wrote: > > On 5/30/12 2:31 PM, Nathan Smith wrote: > >> Hi Hugo, > >> > >> Thanks for getting back to me. I thought that was the case, and the > >> PyErr_format function is called within the isValid(PyObject *pyObj) > >> function, but no exception is ever raised in Python. > > > > Is -1 returned iff an exception is set? -1 indicates an exception; see > > the discussion of tp_hash in > > http://docs.python.org/c-api/**typeobj.html > typeobj.html> > > > > John -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part. URL: From hugo.lima at openbossa.org Wed May 30 22:32:29 2012 From: hugo.lima at openbossa.org (Hugo Parente Lima) Date: Wed, 30 May 2012 17:32:29 -0300 Subject: [PySide] Cannot pass enum by reference In-Reply-To: <4FC66BFF.2010108@users.sf.net> References: <4FC64BF2.8000605@users.sf.net> <3835909.RiKXCFUrRb@hugodesktop> <4FC66BFF.2010108@users.sf.net> Message-ID: <4522604.LZWIf3tM2f@hugodesktop> On Wednesday, May 30, 2012 01:50:39 PM John Cummings wrote: > On 05/30/2012 01:09 PM, Hugo Parente Lima wrote: > > It can be passed by value unless it's a const reference, otherwise the > > following code wont work. > > > > enum ValEnum { One, Other }; > > void oneOrTheOtherEnumValue(ValEnum& enumValue) { > > > > enumValue = Other; > > > > } > > > > ValEnum a = One; > > oneOrTheOtherEnumValue(a); > > // a should be equal to Other. > > > > This is why Shiboken tries to use a pointer, but the code seems wrong > > anyway> > > :-/. > > The subject of the thread was incorrect before. Sorry for the confusion. > > Your statement "It can be passed by value unless it's a const reference" is > circular. Passing by value is by definition not passing by reference. > Perhaps I don't understand your statement. What I meant was: If the library function being bound has a const reference parameter, we can internally copy this value around and this will not affect the API of the bound library. But if the reference isn't const, the generated code can't pass this parameter by value in any place because it'll change the API behavior, i.e. if the function changes the parameter, the change wont be visible after the function call. This is why Shiboken uses pointers on non-const references, but as you noticed it's buggy and generating broken code. > It also doesn't matter if the reference is const or not in this case, the > current Shiboken still does not create a correct wrapper. Old versions did. > > John Cummings > _______________________________________________ > PySide mailing list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part. URL: From alex.gronholm at nextday.fi Thu May 31 14:05:22 2012 From: alex.gronholm at nextday.fi (=?ISO-8859-1?Q?Alex_Gr=F6nholm?=) Date: Thu, 31 May 2012 15:05:22 +0300 Subject: [PySide] PySide 1.1.1 Windows builds In-Reply-To: References: <4FBE9415.2000207@nextday.fi> Message-ID: <4FC75E82.3080500@nextday.fi> 28.05.2012 11:07, Roman Lacko kirjoitti: > Hi Anatoly and list, > > 2012/5/28 anatoly techtonik: >> On Fri, May 25, 2012 at 1:54 PM, Roman Lacko wrote: >>> 2012/5/24 Alex Grönholm: >>>> 19.05.2012 00:47, Roman Lacko kirjoitti: >>>>> I have prepared windows builds of PySide 1.1.1 for Pyhon 2.7 and 3.2, >>>>> 32bit and 64bit. >>>>> Builds will be available on my google Drive [1], until the PySide site >>>>> is not moved to new place. >>>>> For that reason, installing with easy_install is not available, You >>>>> ned to download the package and install manually. >>>>> You can allways let me know if you have better idea where to put the builds. >>>> How about the PyPI, where they are supposed be uploaded to? >>> It will be uploaded when the PySide site move to new location is completed >> What's the progress with move? We can already move it to GitHub pages. > GitHub will not work with easy_install, because GitHub downloads page > uses https and easy_install does not work very well with https. > Another problem is that PySide project is using free edition of GitHub > and there is a size limit (200 Mb or so). One windows package has cca > 40 Mb and there are 6 packages (for py 2.6, 2.7, 3.2 - 32 and 64 bit), > this is cca 240 Mb for each release. > > I will try to contact Qt project administrators if we can't upload > binaries to Qt Project site. Please tell me -- what is the problem with uploading them to PyPI itself?? It seems to me that hasn't even been considered here! > >>> BTW, I'm developing new version of PySide setup script fully >>> compatible with distutils, that will be usable also on linux (binaries >>> will be built inplace), the development can be tracked here: >>> https://github.com/lck/pyside-dist >> How does it relate to existing BuildScripts? > If completed, it should replace the existing build scripts and will > enable to install PySide with pip. > > Regards > Roman From techtonik at gmail.com Thu May 31 14:46:20 2012 From: techtonik at gmail.com (anatoly techtonik) Date: Thu, 31 May 2012 15:46:20 +0300 Subject: [PySide] PySide 1.1.1 Windows builds In-Reply-To: <4FC75E82.3080500@nextday.fi> References: <4FBE9415.2000207@nextday.fi> <4FC75E82.3080500@nextday.fi> Message-ID: On Thu, May 31, 2012 at 3:05 PM, Alex Grönholm wrote: > > Please tell me -- what is the problem with uploading them to PyPI > itself?? It seems to me that hasn't even been considered here! In addition PyPI hosts a public download counter. Still I wonder how pythonpackages managed to count downloads - http://pythonpackages.com/package/pyside?tab=downloaded From backup.rlacko at gmail.com Thu May 31 15:05:33 2012 From: backup.rlacko at gmail.com (Roman Lacko) Date: Thu, 31 May 2012 15:05:33 +0200 Subject: [PySide] PySide 1.1.1 Windows builds In-Reply-To: References: <4FBE9415.2000207@nextday.fi> <4FC75E82.3080500@nextday.fi> Message-ID: Hi, 2012/5/31 anatoly techtonik : > On Thu, May 31, 2012 at 3:05 PM, Alex Grönholm wrote: >> >> Please tell me -- what is the problem with uploading them to PyPI >> itself?? It seems to me that hasn't even been considered here! > > In addition PyPI hosts a public download counter. Still I wonder how > pythonpackages managed to count downloads - > http://pythonpackages.com/package/pyside?tab=downloaded > _______________________________________________ > PySide mailing list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside PySide is registerd on PyPI with customized download url, packages are not downloaded directly from PyPI server but from that customized url [1]. That is way You dont see any files on PyPI directly. Setuptools/PIP handles this download urls transparently for You when installing via [easy_install PySide]. If You have problems installing PySide on Windows via easy_install just let me know. Regards Roman [1] http://releases.qt-project.org/pyside/1.1.1/ From backup.rlacko at gmail.com Thu May 31 15:10:00 2012 From: backup.rlacko at gmail.com (Roman Lacko) Date: Thu, 31 May 2012 15:10:00 +0200 Subject: [PySide] PySide 1.1.1 Windows builds In-Reply-To: <4FC75E82.3080500@nextday.fi> References: <4FBE9415.2000207@nextday.fi> <4FC75E82.3080500@nextday.fi> Message-ID: 2012/5/31 Alex Grönholm : > 28.05.2012 11:07, Roman Lacko kirjoitti: >> Hi Anatoly and list, >> >> 2012/5/28 anatoly techtonik: >>> On Fri, May 25, 2012 at 1:54 PM, Roman Lacko  wrote: >>>> 2012/5/24 Alex Grönholm: >>>>> 19.05.2012 00:47, Roman Lacko kirjoitti: >>>>>> I have prepared windows builds of PySide 1.1.1 for Pyhon 2.7 and 3.2, >>>>>> 32bit and 64bit. >>>>>> Builds will be available on my google Drive [1], until the PySide site >>>>>> is not moved to new place. >>>>>> For that reason, installing with easy_install is not available, You >>>>>> ned to download the package and install manually. >>>>>> You can allways let me know if you have better idea where to put the builds. >>>>> How about the PyPI, where they are supposed be uploaded to? >>>> It will be uploaded when the PySide site move to new location is completed >>> What's the progress with move? We can already move it to GitHub pages. >> GitHub will not work with easy_install, because GitHub downloads page >> uses https and easy_install does not work very well with https. >> Another problem is that PySide project is using free edition of GitHub >> and there is a size limit  (200 Mb or so). One windows package has cca >> 40 Mb and there are 6 packages (for py 2.6, 2.7, 3.2 - 32 and 64 bit), >> this is cca 240 Mb for each release. >> >> I will try to contact Qt project administrators if we can't upload >> binaries to Qt Project site. > Please tell me -- what is the problem with uploading them to PyPI > itself?? It seems to me that hasn't even been considered here! ...There is just one problem, that PyPI has size limit 20 Mb for one file, PySide packages have 40Mb+ For that reason we use custom download url and You dont see any packages directli on PyPI... >> >>>> BTW, I'm developing new version of PySide setup script fully >>>> compatible with distutils, that will be usable also on linux (binaries >>>> will be built inplace), the development can be tracked here: >>>> https://github.com/lck/pyside-dist >>> How does it relate to existing BuildScripts? >> If completed, it should replace the existing build scripts and will >> enable to install PySide with pip. >> >> Regards >> Roman > > _______________________________________________ > PySide mailing list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside From hugo.lima at openbossa.org Thu May 31 15:44:20 2012 From: hugo.lima at openbossa.org (Hugo Parente Lima) Date: Thu, 31 May 2012 10:44:20 -0300 Subject: [PySide] Is there a reason PySideSignalInstance source isn't incref'd? In-Reply-To: <4FC67370.108@wingware.com> References: <4FC67370.108@wingware.com> Message-ID: <9989556.3hZ014XOlW@hugodesktop> On Wednesday, May 30, 2012 03:22:24 PM John Ehresman wrote: > I'm getting a segfault when I run the following: > > w.selectionModel().selectionChanged.connect(callback) > > with updated sources from git. I can work around it with > > m = w.selectionModel() > m.selectionChanged.connect(callback) > > Stepping through the code in pysidesignal.cpp, it looks like d->source > of a PySideSignalInstance object isn't incref'd so when the temporary > ref returned by selectionModel is decref'd, the source is pointing at a > invalid object. The function that initializes a PySideSignalInstance is > instanceInitialize When we delete an QObject we want to have all their connections disconnected, e.g. o = QObject() o.connect(...) del o # The object must be destroyed and all connections disconnected. But if the connection holds a reference to the sender this will never happen, IIRC we currently hold a weakref. With w.selectionModel() the thing is different, because the object returned is owned by C++, so when you delete m or m goes out of scope the QSelectionModel shouldn't be deleted despite of not existing a single Python object pointing to it. > Oddly this only happens in a larger program and not in a simple test > case, but I don't see how source can be assumed to be valid without > increfing it. This need further investigation, file a bug, after all, it's a bug and the program shouldn't crash there. > Thanks, > > John > _______________________________________________ > PySide mailing list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part. URL: From jpe at wingware.com Thu May 31 16:27:59 2012 From: jpe at wingware.com (John Ehresman) Date: Thu, 31 May 2012 10:27:59 -0400 Subject: [PySide] Is there a reason PySideSignalInstance source isn't incref'd? In-Reply-To: <9989556.3hZ014XOlW@hugodesktop> References: <4FC67370.108@wingware.com> <9989556.3hZ014XOlW@hugodesktop> Message-ID: <4FC77FEF.4050106@wingware.com> On 5/31/12 9:44 AM, Hugo Parente Lima wrote: >> Stepping through the code in pysidesignal.cpp, it looks like d->source >> of a PySideSignalInstance object isn't incref'd so when the temporary >> ref returned by selectionModel is decref'd, the source is pointing at a >> invalid object. The function that initializes a PySideSignalInstance is >> instanceInitialize > > When we delete an QObject we want to have all their connections disconnected, > e.g. > > o = QObject() > o.connect(...) > del o > # The object must be destroyed and all connections disconnected. > > But if the connection holds a reference to the sender this will never happen, > IIRC we currently hold a weakref. I think PySideSignalInstance is used for the usually short-lived object created for the obj.signalname attribute. It has connect, disconnect, and emit methods. Possibly a weakref should be used here for the case of: obj = get_existing_object() signal = obj.destroyed del obj signal.connect(callback) but I don't see great harm in keeping the PyObject* that was bound to obj alive. John From alex.gronholm at nextday.fi Thu May 31 19:19:30 2012 From: alex.gronholm at nextday.fi (=?ISO-8859-1?Q?Alex_Gr=F6nholm?=) Date: Thu, 31 May 2012 20:19:30 +0300 Subject: [PySide] PySide 1.1.1 Windows builds In-Reply-To: References: <4FBE9415.2000207@nextday.fi> <4FC75E82.3080500@nextday.fi> Message-ID: <4FC7A822.9010200@nextday.fi> 31.05.2012 16:05, Roman Lacko kirjoitti: > Hi, > > 2012/5/31 anatoly techtonik: >> On Thu, May 31, 2012 at 3:05 PM, Alex Grönholm wrote: >>> Please tell me -- what is the problem with uploading them to PyPI >>> itself?? It seems to me that hasn't even been considered here! >> In addition PyPI hosts a public download counter. Still I wonder how >> pythonpackages managed to count downloads - >> http://pythonpackages.com/package/pyside?tab=downloaded >> _______________________________________________ >> PySide mailing list >> PySide at qt-project.org >> http://lists.qt-project.org/mailman/listinfo/pyside > PySide is registerd on PyPI with customized download url, packages are > not downloaded directly from PyPI server but from that customized url > [1]. I know how it works. In my opinion, PyPI should hide any packages that don't offer direct downloads from PyPI itself. Why is this even being discussed instead of just uploading said packages to PyPI? It should be a standard part of the release procedure. PyPI has a fair number of mirrors you can use if the primary one is down. However, if an external website goes down and no files are hosted in PyPI, you're SOL. > That is way You dont see any files on PyPI directly. > Setuptools/PIP handles this download urls transparently for You when > installing via [easy_install PySide]. > > If You have problems installing PySide on Windows via easy_install > just let me know. > > Regards > Roman > > [1] http://releases.qt-project.org/pyside/1.1.1/ From backup.rlacko at gmail.com Thu May 31 21:20:57 2012 From: backup.rlacko at gmail.com (Roman Lacko) Date: Thu, 31 May 2012 21:20:57 +0200 Subject: [PySide] PySide 1.1.1 Windows builds In-Reply-To: <4FC7A822.9010200@nextday.fi> References: <4FBE9415.2000207@nextday.fi> <4FC75E82.3080500@nextday.fi> <4FC7A822.9010200@nextday.fi> Message-ID: 2012/5/31, Alex Grönholm : > 31.05.2012 16:05, Roman Lacko kirjoitti: >> Hi, >> >> 2012/5/31 anatoly techtonik: >>> On Thu, May 31, 2012 at 3:05 PM, Alex Grönholm >>> wrote: >>>> Please tell me -- what is the problem with uploading them to PyPI >>>> itself?? It seems to me that hasn't even been considered here! >>> In addition PyPI hosts a public download counter. Still I wonder how >>> pythonpackages managed to count downloads - >>> http://pythonpackages.com/package/pyside?tab=downloaded >>> _______________________________________________ >>> PySide mailing list >>> PySide at qt-project.org >>> http://lists.qt-project.org/mailman/listinfo/pyside >> PySide is registerd on PyPI with customized download url, packages are >> not downloaded directly from PyPI server but from that customized url >> [1]. > I know how it works. In my opinion, PyPI should hide any packages that > don't offer direct downloads from PyPI itself. > Why is this even being discussed instead of just uploading said packages > to PyPI? It should be a standard part of the release procedure. > PyPI has a fair number of mirrors you can use if the primary one is > down. However, if an external website goes down and no files are hosted > in PyPI, you're SOL. Well, If You know how it works then You should know about 20Mb upload limit to PyPI! >> That is way You dont see any files on PyPI directly. >> Setuptools/PIP handles this download urls transparently for You when >> installing via [easy_install PySide]. >> >> If You have problems installing PySide on Windows via easy_install >> just let me know. >> >> Regards >> Roman >> >> [1] http://releases.qt-project.org/pyside/1.1.1/ > > _______________________________________________ > PySide mailing list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside > From jcummings2 at users.sf.net Thu May 31 21:30:21 2012 From: jcummings2 at users.sf.net (John Cummings) Date: Thu, 31 May 2012 14:30:21 -0500 Subject: [PySide] Typedef template containers Message-ID: <4FC7C6CD.9000505@users.sf.net> I am sending this message to the list in case it can save anyone else hours of frustration. If someone knows a better way to handle typedef'd template containers as arguments, please let me know. If you have a typedef template container (Qt-based, STL-based or your own custom one) and functions (or methods) such as: QVector FooType; FooType returnFoo(); void callWithFoo(FooType foo); Your typesystem XML file will need to have the following lines: Note the difference between returning a typedef'd container and passing one as an argument. Namely, Shiboken is fine with returning the typedef directly since you don't have to specify it in the typesystem XML file. However, if you are passing a typedef'd container into a function (or method), you must specify the underlying type (and convert '<' to "lt;" and '>' to "gt;"). If you try to use "FooType" instead of specifying the underlying type, you will get a warning message from Shiboken along the lines of: "Global function 'callWithFoo(FooType foo)' is specified in typesystem, but not defined. This could potentially lead to compilation errors." It might compile but callWithFoo will not be accessible from python. Adding various "object-type" or "value-type" tags won't help either. You will also need a "container-type" conversion tag. If you are using QVector, then you can just point at the existing conversion logic by adding the following to your typesystem XML file: You can also look at those two files as an example for handling other container types. As I said, I hope this saves someone else some time. It would be a nice feature if Shiboken could understand the typedefs for you instead of you having to manually make the substitution. John Cummings -------------- next part -------------- An HTML attachment was scrubbed... URL: