kicad-developers team mailing list archive

[Miguel Angel Ajo Pelayo <miguelangel@xxxxxxx>]

I get a crash too, sightly different, it's like if the GetParametersPageName was unable to build the passing parameters, hmmm….

Thread 0 Crashed:: Dispatch queue: com.apple.main-thread
0   pcbnew.bin                    	0x00000001062ef38c PYTHON_FOOTPRINT_WIZARD::GetParameterPageName(int) + 140
1   pcbnew.bin                    	0x0000000105e664d7 FOOTPRINT_WIZARD_FRAME::ReCreatePageList() + 199
2   pcbnew.bin                    	0x0000000105e643d8 FOOTPRINT_WIZARD_FRAME::SelectFootprintWizard() + 456
3   pcbnew.bin                    	0x0000000105e66288 FOOTPRINT_WIZARD_FRAME::FOOTPRINT_WIZARD_FRAME(FOOTPRINT_EDIT_FRAME*, wxSemaphore*, long) + 5480
4   pcbnew.bin                    	0x0000000105e64d0d FOOTPRINT_WIZARD_FRAME::FOOTPRINT_WIZARD_FRAME(FOOTPRINT_EDIT_FRAME*, wxSemaphore*, long) + 45
5   pcbnew.bin                    	0x0000000105eee068 FOOTPRINT_EDIT_FRAME::Process_Special_Functions(wxCommandEvent&) + 3288
6   libwx_osx_cocoau-2.9.dylib    	0x0000000106f2ce35 wxEvtHandler::ProcessEventIfMatchesId(wxEventTableEntryBase const&, wxEvtHandler*, wxEvent&) + 93


and this on the console:

calling GetParameterPageName()
Exception: GetParameterPageName() takes exactly 2 arguments (1 given)
         : <NULL>
./pcbnew.app/Contents/MacOS/pcbnew:


------

Dick, I think you reformatted the "(i)" etc inside commas , that are used to build the python object parameters, we cannot
change the spaces etc inside those lines.

 I'm reverting    PyObject* arglist = Py_BuildValue( "( i )", aPage );  to  PyObject* arglist = Py_BuildValue( "(i)", aPage );  (and all the others), and re-testing.

May be your class-locker works, and that's the crash brian is getting now :)


Miguel Angel Ajo
http://www.nbee.es
+34911407752
skype: ajoajoajo

On 11/03/2013, at 22:26, Brian Sidebotham <brian.sidebotham@xxxxxxxxx> wrote:

> 
> On 11 March 2013 08:12, Dick Hollenbeck <dick@xxxxxxxxxxx> wrote:
> > I will commit something in a few minutes.
> >
> > Brian if you can test that please, then you are da man.
> 
> Its very late here, it was rushed, but I checked in something that might help.
> 
> in rev 3994.
> 
> Hi Dick,
> 
> rev 3994 still crashes at the same place - Py_DECREF( arglist ) Line 177 in pcbnew_footprint_wizards.cpp rev 3994.
> 
> I will look into it and see what's up - in principle the rev changes look entirely sane (and really neat!). I'll get to look at it real soon, but not tonight unfortunately. I'll carry on with gdb and read some more about the GIL to see what's up.
> 
> Thanks for your quick turn-around.
> 
> Brian.