+- Forums (https://www.vdrift.net/Forum)
+-- Forum: Community (https://www.vdrift.net/Forum/forumdisplay.php?fid=3)
+--- Forum: Cars & Tracks (https://www.vdrift.net/Forum/forumdisplay.php?fid=11)
+--- Thread: PyGTK track editor idea (/showthread.php?tid=517)
- thelusiv - 05-05-2007
Alright, I also added an Object Properties dialog to the Glade project. It's not used in the code yet, but I checked it in anyway (track editor repository, pygtk track editor branch, revision 60).
Edit: Here's a screenshot tell me what you think:
![[Image: 0_object_properties.jpg]](http://vdrift.net/Gallery/mediaobjects/disp/0/0_object_properties.jpg)
- thelusiv - 05-05-2007
rookie1, could you tell me where you got the info on how to set up the wrapper? I'd like to set up some C++ test routines for it, sometimes when I try to load a track it segfaults on me...I think it's due to a problem in the shared object that I'm not able to debug from python. For instance it segfaults when trying to load hockenheim or monaco, but lagunaseca works fine. Any ideas?
edit: I just went through all of them (at least the ones from last version) and the only ones that worked were lagunaseca and zandvoort, which happen to be our two smallest tracks...although I'm not sure the problem is related to size, it's possible.
- rookie1 - 05-05-2007
If you want to know how the wrapping works, take a look at Boost.Python tutorial here.
I'm not sure what the problem is. On my machine (main mem 512M, video ram 64M), I can load most of the tracks, including hockenheim, monaco or nuerburgring.
I do get segfault when loading road_atlanta or lemans due to missing texture. I've traced this to a line in textures.cpp,
"cerr << "error loading texture file \"" << file << "\"..." << endl;"
However if I change 'cerr' to 'cout', the segfault does not occur. Haven't figure out why yet.
You can use gdb to debug if you want. To do this, just use 'gdb --args python vdrift-track-editor.py'. Then debug as per normal, like obtaining back trace, etc.
Regarding the object property dialog, is it possible to put it somewhere in the main window instead of a new dialog box? Maybe we can replace the message window at bottom with this? Or split the object window into 2 parts, 1 for object list, another for object property? So when a user click an object, its properties will be automatically displayed there.
- thelusiv - 05-07-2007
When I load hockenheim I get the following backtrace. Not sure I understand it...
Code:
Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread -1210124624 (LWP 16931)]
0xb66461ba in std::basic_string<char, std::char_traits<char>, std::allocator<char> >::basic_string () from /usr/lib/libstdc++.so.6
(gdb) bt
#0 0xb66461ba in std::basic_string<char, std::char_traits<char>, std::allocator<char> >::basic_string () from /usr/lib/libstdc++.so.6
#1 0xb5e5dc77 in OBJECTNODE::GetName (this=0xb6143b50) at include/objects.h:57
#2 0xb5e5f7d6 in boost::python::detail::invoke<boost::python::to_python_value<std::string const&>, std::string (OBJECTNODE::*)(), boost::python::arg_from_python<OBJECTNODE&> > (rc=@0xbffc80ee, f=@0x82debd4,
tc=@0xbffc80e0) at /usr/include/boost/python/detail/invoke.hpp:88
#3 0xb5e5f8f5 in boost::python::detail::caller_arity<1u>::impl<std::string (OBJECTNODE::*)(), boost::python::default_call_policies, boost::mpl::vector2<std::string, OBJECTNODE&> >::operator() (this=0x82debd4,
args_=0xb64fd42c) at /usr/include/boost/python/detail/caller.hpp:199
#4 0xb5e5f950 in boost::python::objects::caller_py_function_impl<boost::python::detail::caller<std::string (OBJECTNODE::*)(), boost::python::default_call_policies, boost::mpl::vector2<std::string, OBJECTNODE&> > >::operator() (this=0x82debd0, args=0xb64fd42c, kw=0x0) at /usr/include/boost/python/object/py_function.hpp:38
#5 0xb5a066d6 in boost::python::objects::function::call () from /usr/lib/libboost_python-gcc-mt-1_33_1.so.1.33.1
#6 0x082debd0 in ?? ()
#7 0xb64fd42c in ?? ()
#8 0x00000000 in ?? ()I think you're right on the object properties being in the main window. I did a little work on the Glade setup to do this and checked it in. It still needs work however. If you take a look at it you'll see what I mean.
I'm not sure if I want to get rid of the message window, but it needs to be smaller...and easier to hide.
- rookie1 - 05-07-2007
That's weird. I don't have this segfault when loading hockenheim. Can you print out the string value of model name in OBJECTNODE::GetName()?
- rookie1 - 05-18-2007
Hi thelusiv, any progress on the track editor? Didn't have time to work on it myself.
- alanic - 05-30-2007
rookie1 Wrote:"cerr << "error loading texture file \"" << file << "\"..." << endl;"
I'm not related to this community and I landed here with a google search about my problem. This looks like a Boost.Python problem to me, even though I'm not entirely sure that you guys are using Boost.Python at all, but the posts make me think so. Here are some related posts about this:
http://mail.datenhain.de/pipermail/libavg-users/2007-March/000085.html
http://thread.gmane.org/gmane.comp.python.c++/11443/focus=11444
I'm having this on Gentoo Linux with a laptop with intel 855 graphics card. Here are versions of everything I tried these with:
gcc 4.1.2
python 2.4 and 2.3
mesa 6.5.2 (this is where the opengl library comes from)
boost 1.33.1 and 1.34.0
And here is the bug report I just filed:
http://svn.boost.org/trac/boost/ticket/1018
- thelusiv - 05-30-2007
That sounds like exactly the problem we're having. Thanks for submitting the bug report, let us know if there are any updates about this.
Oh yeah, and yay for Google. 8)