visualpython-users

Re: [Visualpython-users] problems with ipython -pylab -p scipy

[Bruce S. <Bru...@nc...>]

Here's a thought. I don't see how, but maybe somehow there's a conflict 
between the old numeric module installed by the standard VPython and the 
new numpy module you already had installed. You might try temporarily 
uninstalling numpy to see whether that makes a difference. Or you might 
uninstall the standard VPython and instead install the beta version (see 
"Recent developments" at vpython.org). The beta version (which 
unfortunately has serious bugs on Windows) uses numpy.

Bruce Sherwood

Ryan Krauss wrote:
> I am trying to use Vpython with ipython running with the scipy options
> -pylab -p scipy.  I am getting this error message:
> 
> In [1]: import visual
> ---------------------------------------------------------------------------
> <type 'exceptions.ImportError'>           Traceback (most recent call last)
> 
> C:\Documents and Settings\rkrauss\<ipython console> in <module>()
> 
> C:\Python25\lib\site-packages\visual\__init__.py in <module>()
>      13
>      14 # Don't try this at home!
> ---> 15 import array_backend
>      16 # The following is intended to be equivalent to:
>      17 # >>> from array_backend.backend[0] import *
> 
> C:\Python25\lib\site-packages\visual\array_backend.py in <module>()
>       2
>       3 # I am performing the import statements within functions to prevent poll
> uting the global
>       4 # namespace
>       5 def _try_init_numeric():
>       6     try:
> 
> <type 'exceptions.ImportError'>: DLL load failed: Invalid access to memory locat
> ion.
> 
> 
> I saw another thread with this error message but with no answer.  Is
> there a solution to this problem?  I am running Windows XP and Python
> 2.5.  Until I ran the VPython installer, this computer had only numpy
> installed.
> 
> Thanks,
> 
> Ryan
> 
> -------------------------------------------------------------------------
> This SF.net email is sponsored by DB2 Express
> Download DB2 Express C - the FREE version of DB2 express and take
> control of your XML. No limits. Just data. Click to get it now.
> http://sourceforge.net/powerbar/db2/
> _______________________________________________
> Visualpython-users mailing list
> Vis...@li...
> https://lists.sourceforge.net/lists/listinfo/visualpython-users

[Visualpython-users] problems with ipython -pylab -p scipy

[Ryan K. <rya...@gm...>]

I am trying to use Vpython with ipython running with the scipy options
-pylab -p scipy.  I am getting this error message:

In [1]: import visual
---------------------------------------------------------------------------
<type 'exceptions.ImportError'>           Traceback (most recent call last)

C:\Documents and Settings\rkrauss\<ipython console> in <module>()

C:\Python25\lib\site-packages\visual\__init__.py in <module>()
     13
     14 # Don't try this at home!
---> 15 import array_backend
     16 # The following is intended to be equivalent to:
     17 # >>> from array_backend.backend[0] import *

C:\Python25\lib\site-packages\visual\array_backend.py in <module>()
      2
      3 # I am performing the import statements within functions to prevent poll
uting the global
      4 # namespace
      5 def _try_init_numeric():
      6     try:

<type 'exceptions.ImportError'>: DLL load failed: Invalid access to memory locat
ion.


I saw another thread with this error message but with no answer.  Is
there a solution to this problem?  I am running Windows XP and Python
2.5.  Until I ran the VPython installer, this computer had only numpy
installed.

Thanks,

Ryan

[Visualpython-users] Problems in Running Some Vpython Scripts

[Lorenzo I. <lor...@gm...>]

Dear All,
I am fairly new to Vpython (and not a terribly experienced Python
programmer in general).
I am running Debian testing on my box and I installed Vpython from
Debian's repositories.
Now, I first tried running some online scripts. Some of them run fine,
but with other ones I get obscure (to me) error messages.

For instance when I try running the script at:

http://physics.syr.edu/courses/PHY307/LABS/lab7.percolation.py

This is what I get.

$ python lab10_BKearthquake.py
Traceback (most recent call last):
  File "lab10_BKearthquake.py", line 17, in ?
    background=color.white, foreground=color.blue)
  File "/usr/lib/python2.4/site-packages/visual/graph.py", line 192, in __init__
    currentdisplay.select()
RuntimeError: boost::bad_weak_ptr

Anybody has experienced the same problem?
Many thanks

Lorenzo

Re: [Visualpython-users] Still no progress with the Beta version of visual python

[David H. <dha...@ta...>]

I have done some further testing.  Here are the example
scripts that successfully run (all of the others are failing):

convex.py
doublependulum.py
faces_cone.py
faces_heightfield.py
glinfo.py
gyro.py
gyro2.py
hanoi.py
labels.py
lathe.py
lorenz.py
texttest.py
texturetest.py (although I see no
  patterned textures -- transparency seems to work.  It
  also seems to hang on exit).
toroid_drag.py
toroid.py
wave.py

Most of the ones that fail take seg-faults.  A few hang with
no redisplay.  Almost all of them produce lots of warnings
to standard output like this:

(<unknown>:25310): Gtk-WARNING **: Mixing deprecated and non-deprecated 
GtkToolbar API is not allowed
(<unknown>:25310): Gtk-WARNING **: Mixing deprecated and non-deprecated 
GtkToolbar API is not allowed
(<unknown>:25310): Gtk-WARNING **: Mixing deprecated and non-deprecated 
GtkToolbar API is not allowed
(<unknown>:25310): Gtk-WARNING **: Mixing deprecated and non-deprecated 
GtkToolbar API is not allowed

So it is not completely DOA.  There are 36 examples in the
examples directory and 16 of them are passing.  Of course, my
particular program is one of the ones that is failing :-(

David Harrison wrote:
>
> It appears that other standard examples (like stonehenge) don't
> work either.  Is it possible that boost 1.34 is not compatible
> with visual-beta15?  The backtrace from stonehenge is:
>
> #0  PyDict_GetItem (op=0xd122f0, key=0x0) at Objects/dictobject.c:569
> #1  0x0000002a9c8b777a in boost::python::objects::function::call ()
>   from 
> /home/dharrison/wares/rhe4_a64/lib/libboost_python-gcc34-1_34.so.1.34.0
> #2  0x0000002a9c8bb305 in 
> boost::detail::function::void_function_ref_invoker0<boost::python::objects::(anonymous 
> namespace)::bind_return, void>::invoke ()
>   from 
> /home/dharrison/wares/rhe4_a64/lib/libboost_python-gcc34-1_34.so.1.34.0
> #3  0x0000002a9c8c2167 in boost::function0<void, 
> std::allocator<boost::function_base> >::operator() ()
>   from 
> /home/dharrison/wares/rhe4_a64/lib/libboost_python-gcc34-1_34.so.1.34.0
> #4  0x0000002a9c8c1c58 in 
> boost::python::detail::exception_handler::operator()
>    ()
>   from 
> /home/dharrison/wares/rhe4_a64/lib/libboost_python-gcc34-1_34.so.1.34.0
> #5  0x0000002a99b90dbf in 
> boost::detail::function::function_obj_invoker2<boost::_bi::bind_t<bool, 
> boost::python::detail::translate_exception<std::runtime_error, void 
> (*)(std::runtime_error)>, boost::_bi::list3<boost::arg<1> (*)(), 
> boost::arg<2> (*)(), boost::_bi::value<void (*)(std::runtime_error)> > 
> >, bool, boost::python::detail::exception_handler const&, 
> boost::function0<void, std::allocator<boost::function_base> > 
> const&>::invoke (function_obj_ptr=Variable "function_obj_ptr" is not 
> available.
> )
>    at 
> /home/dharrison/wares/rhe4_a64/include/boost-1_34/boost/python/detail/translate_exception.hpp:45 
>
> #6  0x0000002a9c8c1fe7 in boost::function2<bool, 
> boost::python::detail::exception_handler const&, 
> boost::function0<void, std::allocator<boost::function_base> >const&, 
> std::allocator<boost::function_base> >::operator() ()
>   from 
> /home/dharrison/wares/rhe4_a64/lib/libboost_python-gcc34-1_34.so.1.34.0
> #7  0x0000002a9c8c1c48 in 
> boost::python::detail::exception_handler::operator()
>    ()
>   from 
> /home/dharrison/wares/rhe4_a64/lib/libboost_python-gcc34-1_34.so.1.34.0
> #8  0x0000002a99b90ccf in 
> boost::detail::function::function_obj_invoker2<boost::_bi::bind_t<bool, 
> boost::python::detail::translate_exception<std::invalid_argument, void 
> (*)(std::invalid_argument)>, boost::_bi::list3<boost::arg<1> (*)(), 
> boost::arg<2> (*)(), boost::_bi::value<void 
> (*)(std::invalid_argument)> > >, 
> bool,boost::python::detail::exception_handler const&, 
> boost::function0<void, std::allocator<boost::function_base> > 
> const&>::invoke (function_obj_ptr=Variable "function_obj_ptr" is not 
> available.
> )
>    at 
> /home/dharrison/wares/rhe4_a64/include/boost-1_34/boost/python/detail/translate_exception.hpp:45 
>
> #9  0x0000002a9c8c1fe7 in boost::function2<bool, 
> boost::python::detail::exception_handler const&, 
> boost::function0<void, std::allocator<boost::function_base> >const&, 
> std::allocator<boost::function_base> >::operator() ()
>   from 
> /home/dharrison/wares/rhe4_a64/lib/libboost_python-gcc34-1_34.so.1.34.0
> #10 0x0000002a9c8c1c48 in 
> boost::python::detail::exception_handler::operator()
>    ()
>   from 
> /home/dharrison/wares/rhe4_a64/lib/libboost_python-gcc34-1_34.so.1.34.0
> #11 0x0000002a99b90baf in 
> boost::detail::function::function_obj_invoker2<boost::_bi::bind_t<bool, 
> boost::python::detail::translate_exception<std::out_of_range,void 
> (*)(std::out_of_range)>, boost::_bi::list3<boost::arg<1> (*)(), 
> boost::arg<2> (*)(), boost::_bi::value<void (*)(std::out_of_range)> > 
> >, bool, boost::python::detail::exception_handler const&, 
> boost::function0<void, std::allocator<boost::function_base> > 
> const&>::invoke (function_obj_ptr=Variable "function_obj_ptr" is not 
> available.
> )
>    at 
> /home/dharrison/wares/rhe4_a64/include/boost-1_34/boost/python/detail/translate_exception.hpp:45 
>
> #12 0x0000002a9c8c1fe7 in boost::function2<bool, 
> boost::python::detail::exception_handler const&, 
> boost::function0<void, std::allocator<boost::function_base> >const&, 
> std::allocator<boost::function_base> >::operator() ()
>   from 
> /home/dharrison/wares/rhe4_a64/lib/libboost_python-gcc34-1_34.so.1.34.0
> #13 0x0000002a9c8c1c7b in boost::python::handle_exception_impl ()
>   from 
> /home/dharrison/wares/rhe4_a64/lib/libboost_python-gcc34-1_34.so.1.34.0
> #14 0x0000002a9c8bb219 in 
> boost::python::handle_exception<boost::python::objects::(anonymous 
> namespace)::bind_return> ()
>   from 
> /home/dharrison/wares/rhe4_a64/lib/libboost_python-gcc34-1_34.so.1.34.0
> #15 0x0000002a9c8b9a1e in 
> boost::python::objects::function::add_to_namespace ()
>   from 
> /home/dharrison/wares/rhe4_a64/lib/libboost_python-gcc34-1_34.so.1.34.0
> #16 0x00000000004194c3 in PyObject_Call (func=Variable "func" is not 
> available.
> ) at Objects/abstract.c:1860
>
>
> David Harrison wrote:
>>
>> Well, to get around the latest issue, I went ahead and downloaded
>> a modern libboost (1.34) and compiled it as well.  There are some issues
>> in getting visual to use this libboost.  It appears it now counts on
>> libboost being available from the compiler.  But the boost provided
>> for my compiler, g++ (GCC) 3.4.6 20060404 (Red Hat 3.4.6-3),
>> doesn't appear to link.  I managed to force it to use a local libboost
>> by modifying the Makefiles.  After that, everything finally compiled,
>> linked, and installed.  However, a simple test program dies with
>> the following backtrace.  Any ideas?
>>
>>       David
>>
>> #0  0x0000002a9a3ae1b5 in cvisual::icososphere::subdivide 
>> (this=0x40a01090,
>>    span=Variable "span" is not available.
>> ) at ./core/util/icososphere.cpp:137
>> #1  0x0000002a9a3aea0f in icososphere (this=0x40a01090, 
>> depth=Variable "depth" is not available.
>> )
>>    at 
>> /home/dharrison/wares/rhe4_a64/include/boost-1_34/boost/shared_array.hpp:93 
>>
>> #2  0x0000002a9a2e685c in cvisual::sphere::create_cache ()
>>    at 
>> /usr/lib/gcc/x86_64-redhat-linux/3.4.6/../../../../include/c++/3.4.6/bits/stl_vector.h:557 
>>
>> #3  0x0000002a9a2e437e in cvisual::renderable::refresh_cache 
>> (this=0x8b0890,
>>    geometry=Variable "geometry" is not available.
>> ) at ./core/renderable.cpp:122
>> #4  0x0000002a9a2cca96 in cvisual::display_kernel::draw (this=0x8ca6a0,
>>    scene_geometry=@0x40a01690, whicheye=Variable "whicheye" is not 
>> available.
>> )
>>    at 
>> /home/dharrison/wares/rhe4_a64/include/boost-1_34/boost/iterator/iterator_facade.hpp:644 
>>
>> #5  0x0000002a9a2cfc04 in cvisual::display_kernel::render_scene 
>> (this=0x8ca6a0)
>>    at ./core/display_kernel.cpp:791
>> #6  0x0000002a9a3c1840 in cvisual::render_surface::on_expose_event 
>> (this=Variable "this" is not available.
>> )
>>    at ./gtk2/render_surface.cpp:237
>> #7  0x0000002a9b071e46 in Gtk::Widget_Class::expose_event_callback (
>>    self=0x8f4030, p0=0x40a01e50) at widget.cc:4494
>> #8  0x0000002a9bae3fa8 in _gtk_marshal_BOOLEAN__BOXED (closure=0x9316b0,
>>    return_value=0x40a01990, n_param_values=Variable "n_param_values" 
>> is not available.
>> ) at gtkmarshalers.c:84
>> #9  0x0000002a9c5dbd39 in IA__g_closure_invoke (closure=0x9316b0,
>>    return_value=0x40a01990, n_param_values=2, param_values=0x40a01b20,
>>    invocation_hint=0x40a019c0) at gclosure.c:490
>> #10 0x0000002a9c5effcc in signal_emit_unlocked_R (node=0x931720, 
>> detail=0,
>>    instance=0x8f4030, emission_return=0x40a01aa0,
>>    instance_and_params=0x40a01b20) at gsignal.c:2478
>> #11 0x0000002a9c5f08fa in IA__g_signal_emit_valist (instance=0x8f4030,
>>    signal_id=Variable "signal_id" is not available.
>> ) at gsignal.c:2209
>> #12 0x0000002a9c5f0f13 in IA__g_signal_emit (instance=Variable 
>> "instance" is not available.
>> ) at gsignal.c:2243
>> #13 0x0000002a9bbebcc0 in gtk_widget_event_internal (widget=0x8f4030,
>>    event=0x40a01e50) at gtkwidget.c:3915
>> #14 0x0000002a9bae2821 in IA__gtk_main_do_event (event=0x40a01e50)
>>    at gtkmain.c:1537
>> #15 0x0000002a9be70712 in gdk_window_process_updates_internal 
>> (window=0x8fbe10)
>>    at gdkwindow.c:2338
>> #16 0x0000002a9be707db in IA__gdk_window_process_all_updates ()
>>    at gdkwindow.c:2401
>> #17 0x0000002a9be70848 in gdk_window_update_idle (data=Variable 
>> "data" is not available.
>> ) at gdkwindow.c:2259
>> #18 0x0000002a9cb6f147 in IA__g_main_context_dispatch (context=0x8ff410)
>>    at gmain.c:2045
>> #19 0x0000002a9cb70b78 in g_main_context_iterate (context=0x8ff410,
>>    block=2037111104, dispatch=1, self=Variable "self" is not available.
>> ) at gmain.c:2677
>> #20 0x0000002a9cb70eda in IA__g_main_loop_run (loop=0x92b5f0) at 
>> gmain.c:2881
>> #21 0x0000002a9bae1b81 in IA__gtk_main () at gtkmain.c:1154
>> #22 0x0000002a9a3c769d in cvisual::gui_main::run (this=0x8cf010)
>>    at ./gtk2/display.cpp:547
>> #23 0x0000002a9a3cad71 in cvisual::gui_main::thread_proc ()
>>    at ./gtk2/display.cpp:567
>> #24 0x0000002a9cd1c167 in boost::function0<void, 
>> std::allocator<boost::function_base> >::operator() ()
>>   from 
>> /home/dharrison/wares/rhe4_a64/lib/libboost_python-gcc34-1_34.so.1.34.0
>> #25 0x0000002a9ce37427 in boost::thread_group::join_all ()
>>   from 
>> /home/dharrison/wares/rhe4_a64/lib/libboost_thread-gcc34-mt-1_34.so.1.34.0 
>>
>>
>>
>>
>> David Harrison wrote:
>>> I got past the problem in num_util.cpp: it looks like a simple typo.
>>> Line 234 in that file should read:
>>>
>>>   npy_intp* dims_ptr = PyArray_DIMS(arr.ptr());
>>>
>>> I then found that visual inherits the linking problem from
>>> gtkglextmm-1.2.0.  It tries to link against libGL twice: once
>>> in the right place /usr/lib64/libGL.so and once in the wrong
>>> place: /usr/lib/libGL.so.  A hack to its generated libtool
>>> script gets around that problem.
>>>
>>> But then it finally threw me a very significant curve:
>>>
>>> /usr/bin/ld: 
>>> /usr/lib/gcc/x86_64-redhat-linux/3.4.6/../../../../lib64/libboost_python.a(numeric.o): 
>>> relocation R_X86_64_32 against `a local symbol' can not be used when 
>>> making a shared object; recompile with -fPIC
>>> /usr/lib/gcc/x86_64-redhat-linux/3.4.6/../../../../lib64/libboost_python.a: 
>>> could not read symbols: Bad value
>>>
>>> It appears it is trying to link against a non-shared version of 
>>> libboost_python.
>>> I will try to hack around this one too (there appears to be a .so for
>>> libboost_python) but I think I am going to have to edit the final 
>>> link line.
>>>
>>>        David
>>>
>>> Bruce Sherwood wrote:
>>>  
>>>> Alas, I have no experience with nor access to a 64-bit environment, 
>>>> and it does sound like that's the issue. I'm surprised that there 
>>>> aren't standard 64-bit packages for GTK, which after all is a 
>>>> pretty standard library. Is there anyone reading this list with 
>>>> 64-bit experience?
>>>>
>>>> Bruce Sherwood
>>>>
>>>> David Harrison wrote:
>>>>   
>>>>> At Bruce's suggestion, I downloaded visual-4.beta15
>>>>> and tried installing it on my AMD 64-bit machine
>>>>> running RedHat Enterprise 4.  It uses numpy and
>>>>> that is good.  But it now has a very major dependency
>>>>> on GTK and its libraries.  I spent about 4 hours today
>>>>> compiling an ever increasing number of dependent
>>>>> libraries to try to get things going.  Just for grins,
>>>>> take a look at what libraries are needed:
>>>>>
>>>>> 4 atk-1.19.3/     4 gtk+-2.10.13/     4 libsigc++-1.2.7/  4 
>>>>> cairo-1.4.8/    4 gtkglext-1.2.0/   4 libsigc++-2.0.17/
>>>>> 4 cairomm-1.2.4/  4 gtkglextmm-1.2.0/   4 gtkmm-2.10.10/    4 
>>>>> numpy-1.0.3/     4 glib-2.12.9/    4 gtkmm-2.2.12/     4 
>>>>> pango-1.17.3/     4 visual-4.beta15/
>>>>> 4 glibmm-2.12.10/ 4 libglademm-2.4.2/ 4 Python-2.5.1/
>>>>>
>>>>> Many of these are tricky to install because they are either
>>>>> not very friendly to 64-bit or to installs in non-standard
>>>>> locations.  After a fair bit of hacking, I managed to get
>>>>> all of these installed (watch out for gtkglextmm -- it has some
>>>>> sort of wild problem attempting to link against libGL -- libtool
>>>>> for it had to be seriously hacked).
>>>>>
>>>>> After everything was installed, visual-4.beta15 finally stopped
>>>>> complaining about dependent libraries and compiled.  But
>>>>> it looks like it has a problem:
>>>>>
>>>>> steiner.visual-4.beta15: make
>>>>> Making all in site-packages/visual
>>>>> make[1]: Entering directory 
>>>>> `/home/dharrison/wares/rhe4_a64/src/visual-4.beta15/site-packages/visual' 
>>>>>
>>>>> make[1]: Nothing to be done for `all'.
>>>>> make[1]: Leaving directory 
>>>>> `/home/dharrison/wares/rhe4_a64/src/visual-4.beta15/site-packages/visual' 
>>>>>
>>>>> Making all in src
>>>>> make[1]: Entering directory 
>>>>> `/home/dharrison/wares/rhe4_a64/src/visual-4.beta15/src'
>>>>> This is a quiet Makefile. If make exits with an error, check 
>>>>> src/build.log to see the complete error message(s). In the event 
>>>>> of an error that you cannot debug, please send a message to 
>>>>> vis...@li..., including the files 
>>>>> config.log and src/build.log, requesting assistance.
>>>>> Compiling ./core/arrow.cpp ...
>>>>> Compiling ./core/axial.cpp ...
>>>>> Compiling ./core/box.cpp ...
>>>>> Compiling ./core/cone.cpp ...
>>>>> Compiling ./core/cylinder.cpp ...
>>>>> Compiling ./core/display_kernel.cpp ...
>>>>> Compiling ./core/ellipsoid.cpp ...
>>>>> Compiling ./core/frame.cpp ...
>>>>> Compiling ./core/label.cpp ...
>>>>> Compiling ./core/mouseobject.cpp ...
>>>>> Compiling ./core/primitive.cpp ...
>>>>> Compiling ./core/pyramid.cpp ...
>>>>> Compiling ./core/rectangular.cpp ...
>>>>> Compiling ./core/renderable.cpp ...
>>>>> Compiling ./core/ring.cpp ...
>>>>> Compiling ./core/sphere.cpp ...
>>>>> Compiling ./python/convex.cpp ...
>>>>> Compiling ./python/curve.cpp ...
>>>>> Compiling ./python/cvisualmodule.cpp ...
>>>>> Compiling ./python/faces.cpp ...
>>>>> Compiling ./python/num_util.cpp ...
>>>>> make[1]: *** [num_util.lo] Error 1
>>>>> make[1]: Leaving directory 
>>>>> `/home/dharrison/wares/rhe4_a64/src/visual-4.beta15/src'
>>>>> make: *** [all-recursive] Error 1
>>>>>
>>>>> I looked at the build log and here is what it says:
>>>>>
>>>>>  g++ -I/home/dharrison/wares/rhe4_a64/include/python2.5 
>>>>> -I/home/dharrison/wares/rhe4_a64/lib/python2.5/site-packages/numpy/core/include 
>>>>> -DHAVE_CONFIG_H -I../include -I../include 
>>>>> -I/home/dharrison/wares/rhe4_a64/include/gtkglextmm-1.2 
>>>>> -I/home/dharrison/wares/rhe4_a64/lib/gtkglextmm-1.2/include 
>>>>> -I/home/dharrison/wares/rhe4_a64/include/gtkglext-1.0 
>>>>> -I/home/dharrison/wares/rhe4_a64/include/gtkmm-2.4 
>>>>> -I/home/dharrison/wares/rhe4_a64/lib/gtkmm-2.4/include 
>>>>> -I/home/dharrison/wares/rhe4_a64/lib/gtkglext-1.0/include 
>>>>> -I/usr/X11R6/include 
>>>>> -I/home/dharrison/wares/rhe4_a64/include/gdkmm-2.4 
>>>>> -I/home/dharrison/wares/rhe4_a64/lib/gdkmm-2.4/include 
>>>>> -I/home/dharrison/wares/rhe4_a64/include/pangomm-1.4 
>>>>> -I/home/dharrison/wares/rhe4_a64/include/gtk-2.0 
>>>>> -I/home/dharrison/wares/rhe4_a64/lib/gtk-2.0/include 
>>>>> -I/home/dharrison/wares/rhe4_a64/include/pango-1.0 
>>>>> -I/home/dharrison/wares/rhe4_a64/include/glib-2.0 
>>>>> -I/home/dharrison/wares/rhe4_a64/lib/glib-2.0/include 
>>>>> -I/home/dharrison/wares/rhe4_a64/include/cairo 
>>>>> -I/home/dharrison/wares/rhe4_a64/include/glibmm-2.4 
>>>>> -I/home/dharrison/wares/rhe4_a64/lib/glibmm-2.4/include 
>>>>> -I/home/dharrison/wares/rhe4_a64/include/cairomm-1.0 
>>>>> -I/home/dharrison/wares/rhe4_a64/include/sigc++-2.0 
>>>>> -I/home/dharrison/wares/rhe4_a64/lib/sigc++-2.0/include 
>>>>> -I/home/dharrison/wares/rhe4_a64/include/atk-1.0 
>>>>> -I/home/dharrison/wares/rhe4_a64/include/atkmm-1.6 
>>>>> -I/usr/include/freetype2 
>>>>> -I/home/dharrison/wares/rhe4_a64/include/libglademm-2.4 
>>>>> -I/home/dharrison/wares/rhe4_a64/lib/libglademm-2.4/include 
>>>>> -I/usr/include/libpng12 -I/usr/include/libglade-2.0 
>>>>> -I/usr/include/libxml2 -I../include/gtk2 -pthread 
>>>>> -I/home/dharrison/wares/rhe4_a64/include/glib-2.0 
>>>>> -I/home/dharrison/wares/rhe4_a64/lib/glib-2.0/include 
>>>>> -I/home/dharrison/wares/rhe4_a64/include/python2.5 
>>>>> -I/home/dharrison/wares/rhe4_a64/lib/python2.5/site-packages/numpy/core/include 
>>>>> -fpic -DPIC -g -O2 -ftemplate-depth-120 -MMD -MF num_util.d -MT 
>>>>> "num_util.d num_util.lo" -c ./python/num_util.cpp  -fPIC -DPIC -o 
>>>>> .libs/num_util.o
>>>>> ./python/num_util.cpp: In function `std::vector<npy_intp, 
>>>>> std::allocator<npy_intp> > 
>>>>> cvisual::python::shape(boost::python::numeric::array)':
>>>>> ./python/num_util.cpp:234: error: cannot convert `npy_intp*' to 
>>>>> `int*' in initialization
>>>>>
>>>>> I will take a look see and see if this is something that can be 
>>>>> fixed. Has anyone else
>>>>> managed to get the beta to compile from ground zero on 64-bit 
>>>>> linux?  So far,
>>>>> I think I have spent about 1.5 days just on the installation.  I hope
>>>>> this is not typical.
>>>>>
>>>>>        David
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> ------------------------------------------------------------------------- 
>>>>>
>>>>> This SF.net email is sponsored by DB2 Express
>>>>> Download DB2 Express C - the FREE version of DB2 express and take
>>>>> control of your XML. No limits. Just data. Click to get it now.
>>>>> http://sourceforge.net/powerbar/db2/
>>>>> _______________________________________________
>>>>> Visualpython-users mailing list
>>>>> Vis...@li...
>>>>> https://lists.sourceforge.net/lists/listinfo/visualpython-users
>>>>>       
>>>
>>>
>>> ------------------------------------------------------------------------- 
>>>
>>> This SF.net email is sponsored by DB2 Express
>>> Download DB2 Express C - the FREE version of DB2 express and take
>>> control of your XML. No limits. Just data. Click to get it now.
>>> http://sourceforge.net/powerbar/db2/
>>> _______________________________________________
>>> Visualpython-users mailing list
>>> Vis...@li...
>>> https://lists.sourceforge.net/lists/listinfo/visualpython-users
>>>
>>>   
>>
>>
>
>

Re: [Visualpython-users] Still no progress with the Beta version of visual python

[David H. <dha...@ta...>]

It appears that other standard examples (like stonehenge) don't
work either.  Is it possible that boost 1.34 is not compatible
with visual-beta15?  The backtrace from stonehenge is:

#0  PyDict_GetItem (op=0xd122f0, key=0x0) at Objects/dictobject.c:569
#1  0x0000002a9c8b777a in boost::python::objects::function::call ()
   from 
/home/dharrison/wares/rhe4_a64/lib/libboost_python-gcc34-1_34.so.1.34.0
#2  0x0000002a9c8bb305 in 
boost::detail::function::void_function_ref_invoker0<boost::python::objects::(anonymous 
namespace)::bind_return, void>::invoke ()
   from 
/home/dharrison/wares/rhe4_a64/lib/libboost_python-gcc34-1_34.so.1.34.0
#3  0x0000002a9c8c2167 in boost::function0<void, 
std::allocator<boost::function_base> >::operator() ()
   from 
/home/dharrison/wares/rhe4_a64/lib/libboost_python-gcc34-1_34.so.1.34.0
#4  0x0000002a9c8c1c58 in 
boost::python::detail::exception_handler::operator()
    ()
   from 
/home/dharrison/wares/rhe4_a64/lib/libboost_python-gcc34-1_34.so.1.34.0
#5  0x0000002a99b90dbf in 
boost::detail::function::function_obj_invoker2<boost::_bi::bind_t<bool, 
boost::python::detail::translate_exception<std::runtime_error, void 
(*)(std::runtime_error)>, boost::_bi::list3<boost::arg<1> (*)(), 
boost::arg<2> (*)(), boost::_bi::value<void (*)(std::runtime_error)> > 
 >, bool, boost::python::detail::exception_handler const&, 
boost::function0<void, std::allocator<boost::function_base> > 
const&>::invoke (function_obj_ptr=Variable "function_obj_ptr" is not 
available.
)
    at 
/home/dharrison/wares/rhe4_a64/include/boost-1_34/boost/python/detail/translate_exception.hpp:45
#6  0x0000002a9c8c1fe7 in boost::function2<bool, 
boost::python::detail::exception_handler const&, boost::function0<void, 
std::allocator<boost::function_base> >const&, 
std::allocator<boost::function_base> >::operator() ()
   from 
/home/dharrison/wares/rhe4_a64/lib/libboost_python-gcc34-1_34.so.1.34.0
#7  0x0000002a9c8c1c48 in 
boost::python::detail::exception_handler::operator()
    ()
   from 
/home/dharrison/wares/rhe4_a64/lib/libboost_python-gcc34-1_34.so.1.34.0
#8  0x0000002a99b90ccf in 
boost::detail::function::function_obj_invoker2<boost::_bi::bind_t<bool, 
boost::python::detail::translate_exception<std::invalid_argument, void 
(*)(std::invalid_argument)>, boost::_bi::list3<boost::arg<1> (*)(), 
boost::arg<2> (*)(), boost::_bi::value<void (*)(std::invalid_argument)> 
 > >, bool,boost::python::detail::exception_handler const&, 
boost::function0<void, std::allocator<boost::function_base> > 
const&>::invoke (function_obj_ptr=Variable "function_obj_ptr" is not 
available.
)
    at 
/home/dharrison/wares/rhe4_a64/include/boost-1_34/boost/python/detail/translate_exception.hpp:45
#9  0x0000002a9c8c1fe7 in boost::function2<bool, 
boost::python::detail::exception_handler const&, boost::function0<void, 
std::allocator<boost::function_base> >const&, 
std::allocator<boost::function_base> >::operator() ()
   from 
/home/dharrison/wares/rhe4_a64/lib/libboost_python-gcc34-1_34.so.1.34.0
#10 0x0000002a9c8c1c48 in 
boost::python::detail::exception_handler::operator()
    ()
   from 
/home/dharrison/wares/rhe4_a64/lib/libboost_python-gcc34-1_34.so.1.34.0
#11 0x0000002a99b90baf in 
boost::detail::function::function_obj_invoker2<boost::_bi::bind_t<bool, 
boost::python::detail::translate_exception<std::out_of_range,void 
(*)(std::out_of_range)>, boost::_bi::list3<boost::arg<1> (*)(), 
boost::arg<2> (*)(), boost::_bi::value<void (*)(std::out_of_range)> > >, 
bool, boost::python::detail::exception_handler const&, 
boost::function0<void, std::allocator<boost::function_base> > 
const&>::invoke (function_obj_ptr=Variable "function_obj_ptr" is not 
available.
)
    at 
/home/dharrison/wares/rhe4_a64/include/boost-1_34/boost/python/detail/translate_exception.hpp:45
#12 0x0000002a9c8c1fe7 in boost::function2<bool, 
boost::python::detail::exception_handler const&, boost::function0<void, 
std::allocator<boost::function_base> >const&, 
std::allocator<boost::function_base> >::operator() ()
   from 
/home/dharrison/wares/rhe4_a64/lib/libboost_python-gcc34-1_34.so.1.34.0
#13 0x0000002a9c8c1c7b in boost::python::handle_exception_impl ()
   from 
/home/dharrison/wares/rhe4_a64/lib/libboost_python-gcc34-1_34.so.1.34.0
#14 0x0000002a9c8bb219 in 
boost::python::handle_exception<boost::python::objects::(anonymous 
namespace)::bind_return> ()
   from 
/home/dharrison/wares/rhe4_a64/lib/libboost_python-gcc34-1_34.so.1.34.0
#15 0x0000002a9c8b9a1e in 
boost::python::objects::function::add_to_namespace ()
   from 
/home/dharrison/wares/rhe4_a64/lib/libboost_python-gcc34-1_34.so.1.34.0
#16 0x00000000004194c3 in PyObject_Call (func=Variable "func" is not 
available.
) at Objects/abstract.c:1860


David Harrison wrote:
>
> Well, to get around the latest issue, I went ahead and downloaded
> a modern libboost (1.34) and compiled it as well.  There are some issues
> in getting visual to use this libboost.  It appears it now counts on
> libboost being available from the compiler.  But the boost provided
> for my compiler, g++ (GCC) 3.4.6 20060404 (Red Hat 3.4.6-3),
> doesn't appear to link.  I managed to force it to use a local libboost
> by modifying the Makefiles.  After that, everything finally compiled,
> linked, and installed.  However, a simple test program dies with
> the following backtrace.  Any ideas?
>
>       David
>
> #0  0x0000002a9a3ae1b5 in cvisual::icososphere::subdivide 
> (this=0x40a01090,
>    span=Variable "span" is not available.
> ) at ./core/util/icososphere.cpp:137
> #1  0x0000002a9a3aea0f in icososphere (this=0x40a01090, depth=Variable 
> "depth" is not available.
> )
>    at 
> /home/dharrison/wares/rhe4_a64/include/boost-1_34/boost/shared_array.hpp:93 
>
> #2  0x0000002a9a2e685c in cvisual::sphere::create_cache ()
>    at 
> /usr/lib/gcc/x86_64-redhat-linux/3.4.6/../../../../include/c++/3.4.6/bits/stl_vector.h:557 
>
> #3  0x0000002a9a2e437e in cvisual::renderable::refresh_cache 
> (this=0x8b0890,
>    geometry=Variable "geometry" is not available.
> ) at ./core/renderable.cpp:122
> #4  0x0000002a9a2cca96 in cvisual::display_kernel::draw (this=0x8ca6a0,
>    scene_geometry=@0x40a01690, whicheye=Variable "whicheye" is not 
> available.
> )
>    at 
> /home/dharrison/wares/rhe4_a64/include/boost-1_34/boost/iterator/iterator_facade.hpp:644 
>
> #5  0x0000002a9a2cfc04 in cvisual::display_kernel::render_scene 
> (this=0x8ca6a0)
>    at ./core/display_kernel.cpp:791
> #6  0x0000002a9a3c1840 in cvisual::render_surface::on_expose_event 
> (this=Variable "this" is not available.
> )
>    at ./gtk2/render_surface.cpp:237
> #7  0x0000002a9b071e46 in Gtk::Widget_Class::expose_event_callback (
>    self=0x8f4030, p0=0x40a01e50) at widget.cc:4494
> #8  0x0000002a9bae3fa8 in _gtk_marshal_BOOLEAN__BOXED (closure=0x9316b0,
>    return_value=0x40a01990, n_param_values=Variable "n_param_values" 
> is not available.
> ) at gtkmarshalers.c:84
> #9  0x0000002a9c5dbd39 in IA__g_closure_invoke (closure=0x9316b0,
>    return_value=0x40a01990, n_param_values=2, param_values=0x40a01b20,
>    invocation_hint=0x40a019c0) at gclosure.c:490
> #10 0x0000002a9c5effcc in signal_emit_unlocked_R (node=0x931720, 
> detail=0,
>    instance=0x8f4030, emission_return=0x40a01aa0,
>    instance_and_params=0x40a01b20) at gsignal.c:2478
> #11 0x0000002a9c5f08fa in IA__g_signal_emit_valist (instance=0x8f4030,
>    signal_id=Variable "signal_id" is not available.
> ) at gsignal.c:2209
> #12 0x0000002a9c5f0f13 in IA__g_signal_emit (instance=Variable 
> "instance" is not available.
> ) at gsignal.c:2243
> #13 0x0000002a9bbebcc0 in gtk_widget_event_internal (widget=0x8f4030,
>    event=0x40a01e50) at gtkwidget.c:3915
> #14 0x0000002a9bae2821 in IA__gtk_main_do_event (event=0x40a01e50)
>    at gtkmain.c:1537
> #15 0x0000002a9be70712 in gdk_window_process_updates_internal 
> (window=0x8fbe10)
>    at gdkwindow.c:2338
> #16 0x0000002a9be707db in IA__gdk_window_process_all_updates ()
>    at gdkwindow.c:2401
> #17 0x0000002a9be70848 in gdk_window_update_idle (data=Variable "data" 
> is not available.
> ) at gdkwindow.c:2259
> #18 0x0000002a9cb6f147 in IA__g_main_context_dispatch (context=0x8ff410)
>    at gmain.c:2045
> #19 0x0000002a9cb70b78 in g_main_context_iterate (context=0x8ff410,
>    block=2037111104, dispatch=1, self=Variable "self" is not available.
> ) at gmain.c:2677
> #20 0x0000002a9cb70eda in IA__g_main_loop_run (loop=0x92b5f0) at 
> gmain.c:2881
> #21 0x0000002a9bae1b81 in IA__gtk_main () at gtkmain.c:1154
> #22 0x0000002a9a3c769d in cvisual::gui_main::run (this=0x8cf010)
>    at ./gtk2/display.cpp:547
> #23 0x0000002a9a3cad71 in cvisual::gui_main::thread_proc ()
>    at ./gtk2/display.cpp:567
> #24 0x0000002a9cd1c167 in boost::function0<void, 
> std::allocator<boost::function_base> >::operator() ()
>   from 
> /home/dharrison/wares/rhe4_a64/lib/libboost_python-gcc34-1_34.so.1.34.0
> #25 0x0000002a9ce37427 in boost::thread_group::join_all ()
>   from 
> /home/dharrison/wares/rhe4_a64/lib/libboost_thread-gcc34-mt-1_34.so.1.34.0 
>
>
>
>
> David Harrison wrote:
>> I got past the problem in num_util.cpp: it looks like a simple typo.
>> Line 234 in that file should read:
>>
>>   npy_intp* dims_ptr = PyArray_DIMS(arr.ptr());
>>
>> I then found that visual inherits the linking problem from
>> gtkglextmm-1.2.0.  It tries to link against libGL twice: once
>> in the right place /usr/lib64/libGL.so and once in the wrong
>> place: /usr/lib/libGL.so.  A hack to its generated libtool
>> script gets around that problem.
>>
>> But then it finally threw me a very significant curve:
>>
>> /usr/bin/ld: 
>> /usr/lib/gcc/x86_64-redhat-linux/3.4.6/../../../../lib64/libboost_python.a(numeric.o): 
>> relocation R_X86_64_32 against `a local symbol' can not be used when 
>> making a shared object; recompile with -fPIC
>> /usr/lib/gcc/x86_64-redhat-linux/3.4.6/../../../../lib64/libboost_python.a: 
>> could not read symbols: Bad value
>>
>> It appears it is trying to link against a non-shared version of 
>> libboost_python.
>> I will try to hack around this one too (there appears to be a .so for
>> libboost_python) but I think I am going to have to edit the final 
>> link line.
>>
>>        David
>>
>> Bruce Sherwood wrote:
>>  
>>> Alas, I have no experience with nor access to a 64-bit environment, 
>>> and it does sound like that's the issue. I'm surprised that there 
>>> aren't standard 64-bit packages for GTK, which after all is a pretty 
>>> standard library. Is there anyone reading this list with 64-bit 
>>> experience?
>>>
>>> Bruce Sherwood
>>>
>>> David Harrison wrote:
>>>    
>>>> At Bruce's suggestion, I downloaded visual-4.beta15
>>>> and tried installing it on my AMD 64-bit machine
>>>> running RedHat Enterprise 4.  It uses numpy and
>>>> that is good.  But it now has a very major dependency
>>>> on GTK and its libraries.  I spent about 4 hours today
>>>> compiling an ever increasing number of dependent
>>>> libraries to try to get things going.  Just for grins,
>>>> take a look at what libraries are needed:
>>>>
>>>> 4 atk-1.19.3/     4 gtk+-2.10.13/     4 libsigc++-1.2.7/  4 
>>>> cairo-1.4.8/    4 gtkglext-1.2.0/   4 libsigc++-2.0.17/
>>>> 4 cairomm-1.2.4/  4 gtkglextmm-1.2.0/   4 gtkmm-2.10.10/    4 
>>>> numpy-1.0.3/     4 glib-2.12.9/    4 gtkmm-2.2.12/     4 
>>>> pango-1.17.3/     4 visual-4.beta15/
>>>> 4 glibmm-2.12.10/ 4 libglademm-2.4.2/ 4 Python-2.5.1/
>>>>
>>>> Many of these are tricky to install because they are either
>>>> not very friendly to 64-bit or to installs in non-standard
>>>> locations.  After a fair bit of hacking, I managed to get
>>>> all of these installed (watch out for gtkglextmm -- it has some
>>>> sort of wild problem attempting to link against libGL -- libtool
>>>> for it had to be seriously hacked).
>>>>
>>>> After everything was installed, visual-4.beta15 finally stopped
>>>> complaining about dependent libraries and compiled.  But
>>>> it looks like it has a problem:
>>>>
>>>> steiner.visual-4.beta15: make
>>>> Making all in site-packages/visual
>>>> make[1]: Entering directory 
>>>> `/home/dharrison/wares/rhe4_a64/src/visual-4.beta15/site-packages/visual' 
>>>>
>>>> make[1]: Nothing to be done for `all'.
>>>> make[1]: Leaving directory 
>>>> `/home/dharrison/wares/rhe4_a64/src/visual-4.beta15/site-packages/visual' 
>>>>
>>>> Making all in src
>>>> make[1]: Entering directory 
>>>> `/home/dharrison/wares/rhe4_a64/src/visual-4.beta15/src'
>>>> This is a quiet Makefile. If make exits with an error, check 
>>>> src/build.log to see the complete error message(s). In the event of 
>>>> an error that you cannot debug, please send a message to 
>>>> vis...@li..., including the files 
>>>> config.log and src/build.log, requesting assistance.
>>>> Compiling ./core/arrow.cpp ...
>>>> Compiling ./core/axial.cpp ...
>>>> Compiling ./core/box.cpp ...
>>>> Compiling ./core/cone.cpp ...
>>>> Compiling ./core/cylinder.cpp ...
>>>> Compiling ./core/display_kernel.cpp ...
>>>> Compiling ./core/ellipsoid.cpp ...
>>>> Compiling ./core/frame.cpp ...
>>>> Compiling ./core/label.cpp ...
>>>> Compiling ./core/mouseobject.cpp ...
>>>> Compiling ./core/primitive.cpp ...
>>>> Compiling ./core/pyramid.cpp ...
>>>> Compiling ./core/rectangular.cpp ...
>>>> Compiling ./core/renderable.cpp ...
>>>> Compiling ./core/ring.cpp ...
>>>> Compiling ./core/sphere.cpp ...
>>>> Compiling ./python/convex.cpp ...
>>>> Compiling ./python/curve.cpp ...
>>>> Compiling ./python/cvisualmodule.cpp ...
>>>> Compiling ./python/faces.cpp ...
>>>> Compiling ./python/num_util.cpp ...
>>>> make[1]: *** [num_util.lo] Error 1
>>>> make[1]: Leaving directory 
>>>> `/home/dharrison/wares/rhe4_a64/src/visual-4.beta15/src'
>>>> make: *** [all-recursive] Error 1
>>>>
>>>> I looked at the build log and here is what it says:
>>>>
>>>>  g++ -I/home/dharrison/wares/rhe4_a64/include/python2.5 
>>>> -I/home/dharrison/wares/rhe4_a64/lib/python2.5/site-packages/numpy/core/include 
>>>> -DHAVE_CONFIG_H -I../include -I../include 
>>>> -I/home/dharrison/wares/rhe4_a64/include/gtkglextmm-1.2 
>>>> -I/home/dharrison/wares/rhe4_a64/lib/gtkglextmm-1.2/include 
>>>> -I/home/dharrison/wares/rhe4_a64/include/gtkglext-1.0 
>>>> -I/home/dharrison/wares/rhe4_a64/include/gtkmm-2.4 
>>>> -I/home/dharrison/wares/rhe4_a64/lib/gtkmm-2.4/include 
>>>> -I/home/dharrison/wares/rhe4_a64/lib/gtkglext-1.0/include 
>>>> -I/usr/X11R6/include 
>>>> -I/home/dharrison/wares/rhe4_a64/include/gdkmm-2.4 
>>>> -I/home/dharrison/wares/rhe4_a64/lib/gdkmm-2.4/include 
>>>> -I/home/dharrison/wares/rhe4_a64/include/pangomm-1.4 
>>>> -I/home/dharrison/wares/rhe4_a64/include/gtk-2.0 
>>>> -I/home/dharrison/wares/rhe4_a64/lib/gtk-2.0/include 
>>>> -I/home/dharrison/wares/rhe4_a64/include/pango-1.0 
>>>> -I/home/dharrison/wares/rhe4_a64/include/glib-2.0 
>>>> -I/home/dharrison/wares/rhe4_a64/lib/glib-2.0/include 
>>>> -I/home/dharrison/wares/rhe4_a64/include/cairo 
>>>> -I/home/dharrison/wares/rhe4_a64/include/glibmm-2.4 
>>>> -I/home/dharrison/wares/rhe4_a64/lib/glibmm-2.4/include 
>>>> -I/home/dharrison/wares/rhe4_a64/include/cairomm-1.0 
>>>> -I/home/dharrison/wares/rhe4_a64/include/sigc++-2.0 
>>>> -I/home/dharrison/wares/rhe4_a64/lib/sigc++-2.0/include 
>>>> -I/home/dharrison/wares/rhe4_a64/include/atk-1.0 
>>>> -I/home/dharrison/wares/rhe4_a64/include/atkmm-1.6 
>>>> -I/usr/include/freetype2 
>>>> -I/home/dharrison/wares/rhe4_a64/include/libglademm-2.4 
>>>> -I/home/dharrison/wares/rhe4_a64/lib/libglademm-2.4/include 
>>>> -I/usr/include/libpng12 -I/usr/include/libglade-2.0 
>>>> -I/usr/include/libxml2 -I../include/gtk2 -pthread 
>>>> -I/home/dharrison/wares/rhe4_a64/include/glib-2.0 
>>>> -I/home/dharrison/wares/rhe4_a64/lib/glib-2.0/include 
>>>> -I/home/dharrison/wares/rhe4_a64/include/python2.5 
>>>> -I/home/dharrison/wares/rhe4_a64/lib/python2.5/site-packages/numpy/core/include 
>>>> -fpic -DPIC -g -O2 -ftemplate-depth-120 -MMD -MF num_util.d -MT 
>>>> "num_util.d num_util.lo" -c ./python/num_util.cpp  -fPIC -DPIC -o 
>>>> .libs/num_util.o
>>>> ./python/num_util.cpp: In function `std::vector<npy_intp, 
>>>> std::allocator<npy_intp> > 
>>>> cvisual::python::shape(boost::python::numeric::array)':
>>>> ./python/num_util.cpp:234: error: cannot convert `npy_intp*' to 
>>>> `int*' in initialization
>>>>
>>>> I will take a look see and see if this is something that can be 
>>>> fixed. Has anyone else
>>>> managed to get the beta to compile from ground zero on 64-bit 
>>>> linux?  So far,
>>>> I think I have spent about 1.5 days just on the installation.  I hope
>>>> this is not typical.
>>>>
>>>>        David
>>>>
>>>>
>>>>
>>>>
>>>> ------------------------------------------------------------------------- 
>>>>
>>>> This SF.net email is sponsored by DB2 Express
>>>> Download DB2 Express C - the FREE version of DB2 express and take
>>>> control of your XML. No limits. Just data. Click to get it now.
>>>> http://sourceforge.net/powerbar/db2/
>>>> _______________________________________________
>>>> Visualpython-users mailing list
>>>> Vis...@li...
>>>> https://lists.sourceforge.net/lists/listinfo/visualpython-users
>>>>       
>>
>>
>> ------------------------------------------------------------------------- 
>>
>> This SF.net email is sponsored by DB2 Express
>> Download DB2 Express C - the FREE version of DB2 express and take
>> control of your XML. No limits. Just data. Click to get it now.
>> http://sourceforge.net/powerbar/db2/
>> _______________________________________________
>> Visualpython-users mailing list
>> Vis...@li...
>> https://lists.sourceforge.net/lists/listinfo/visualpython-users
>>
>>   
>
>

Re: [Visualpython-users] Still no progress with the Beta version of visual python

[David H. <dha...@ta...>]

Well, to get around the latest issue, I went ahead and downloaded
a modern libboost (1.34) and compiled it as well.  There are some issues
in getting visual to use this libboost.  It appears it now counts on
libboost being available from the compiler.  But the boost provided
for my compiler, g++ (GCC) 3.4.6 20060404 (Red Hat 3.4.6-3),
doesn't appear to link.  I managed to force it to use a local libboost
by modifying the Makefiles.  After that, everything finally compiled,
linked, and installed.  However, a simple test program dies with
the following backtrace.  Any ideas?

       David

#0  0x0000002a9a3ae1b5 in cvisual::icososphere::subdivide (this=0x40a01090,
    span=Variable "span" is not available.
) at ./core/util/icososphere.cpp:137
#1  0x0000002a9a3aea0f in icososphere (this=0x40a01090, depth=Variable 
"depth" is not available.
)
    at 
/home/dharrison/wares/rhe4_a64/include/boost-1_34/boost/shared_array.hpp:93
#2  0x0000002a9a2e685c in cvisual::sphere::create_cache ()
    at 
/usr/lib/gcc/x86_64-redhat-linux/3.4.6/../../../../include/c++/3.4.6/bits/stl_vector.h:557
#3  0x0000002a9a2e437e in cvisual::renderable::refresh_cache (this=0x8b0890,
    geometry=Variable "geometry" is not available.
) at ./core/renderable.cpp:122
#4  0x0000002a9a2cca96 in cvisual::display_kernel::draw (this=0x8ca6a0,
    scene_geometry=@0x40a01690, whicheye=Variable "whicheye" is not 
available.
)
    at 
/home/dharrison/wares/rhe4_a64/include/boost-1_34/boost/iterator/iterator_facade.hpp:644
#5  0x0000002a9a2cfc04 in cvisual::display_kernel::render_scene 
(this=0x8ca6a0)
    at ./core/display_kernel.cpp:791
#6  0x0000002a9a3c1840 in cvisual::render_surface::on_expose_event 
(this=Variable "this" is not available.
)
    at ./gtk2/render_surface.cpp:237
#7  0x0000002a9b071e46 in Gtk::Widget_Class::expose_event_callback (
    self=0x8f4030, p0=0x40a01e50) at widget.cc:4494
#8  0x0000002a9bae3fa8 in _gtk_marshal_BOOLEAN__BOXED (closure=0x9316b0,
    return_value=0x40a01990, n_param_values=Variable "n_param_values" is 
not available.
) at gtkmarshalers.c:84
#9  0x0000002a9c5dbd39 in IA__g_closure_invoke (closure=0x9316b0,
    return_value=0x40a01990, n_param_values=2, param_values=0x40a01b20,
    invocation_hint=0x40a019c0) at gclosure.c:490
#10 0x0000002a9c5effcc in signal_emit_unlocked_R (node=0x931720, detail=0,
    instance=0x8f4030, emission_return=0x40a01aa0,
    instance_and_params=0x40a01b20) at gsignal.c:2478
#11 0x0000002a9c5f08fa in IA__g_signal_emit_valist (instance=0x8f4030,
    signal_id=Variable "signal_id" is not available.
) at gsignal.c:2209
#12 0x0000002a9c5f0f13 in IA__g_signal_emit (instance=Variable 
"instance" is not available.
) at gsignal.c:2243
#13 0x0000002a9bbebcc0 in gtk_widget_event_internal (widget=0x8f4030,
    event=0x40a01e50) at gtkwidget.c:3915
#14 0x0000002a9bae2821 in IA__gtk_main_do_event (event=0x40a01e50)
    at gtkmain.c:1537
#15 0x0000002a9be70712 in gdk_window_process_updates_internal 
(window=0x8fbe10)
    at gdkwindow.c:2338
#16 0x0000002a9be707db in IA__gdk_window_process_all_updates ()
    at gdkwindow.c:2401
#17 0x0000002a9be70848 in gdk_window_update_idle (data=Variable "data" 
is not available.
) at gdkwindow.c:2259
#18 0x0000002a9cb6f147 in IA__g_main_context_dispatch (context=0x8ff410)
    at gmain.c:2045
#19 0x0000002a9cb70b78 in g_main_context_iterate (context=0x8ff410,
    block=2037111104, dispatch=1, self=Variable "self" is not available.
) at gmain.c:2677
#20 0x0000002a9cb70eda in IA__g_main_loop_run (loop=0x92b5f0) at 
gmain.c:2881
#21 0x0000002a9bae1b81 in IA__gtk_main () at gtkmain.c:1154
#22 0x0000002a9a3c769d in cvisual::gui_main::run (this=0x8cf010)
    at ./gtk2/display.cpp:547
#23 0x0000002a9a3cad71 in cvisual::gui_main::thread_proc ()
    at ./gtk2/display.cpp:567
#24 0x0000002a9cd1c167 in boost::function0<void, 
std::allocator<boost::function_base> >::operator() ()
   from 
/home/dharrison/wares/rhe4_a64/lib/libboost_python-gcc34-1_34.so.1.34.0
#25 0x0000002a9ce37427 in boost::thread_group::join_all ()
   from 
/home/dharrison/wares/rhe4_a64/lib/libboost_thread-gcc34-mt-1_34.so.1.34.0



David Harrison wrote:
> I got past the problem in num_util.cpp: it looks like a simple typo.
> Line 234 in that file should read:
>
>   npy_intp* dims_ptr = PyArray_DIMS(arr.ptr());
>
> I then found that visual inherits the linking problem from
> gtkglextmm-1.2.0.  It tries to link against libGL twice: once
> in the right place /usr/lib64/libGL.so and once in the wrong
> place: /usr/lib/libGL.so.  A hack to its generated libtool
> script gets around that problem.
>
> But then it finally threw me a very significant curve:
>
> /usr/bin/ld: 
> /usr/lib/gcc/x86_64-redhat-linux/3.4.6/../../../../lib64/libboost_python.a(numeric.o): 
> relocation R_X86_64_32 against `a local symbol' can not be used when 
> making a shared object; recompile with -fPIC
> /usr/lib/gcc/x86_64-redhat-linux/3.4.6/../../../../lib64/libboost_python.a: 
> could not read symbols: Bad value
>
> It appears it is trying to link against a non-shared version of 
> libboost_python.
> I will try to hack around this one too (there appears to be a .so for
> libboost_python) but I think I am going to have to edit the final link line.
>
>        David
>
> Bruce Sherwood wrote:
>   
>> Alas, I have no experience with nor access to a 64-bit environment, 
>> and it does sound like that's the issue. I'm surprised that there 
>> aren't standard 64-bit packages for GTK, which after all is a pretty 
>> standard library. Is there anyone reading this list with 64-bit 
>> experience?
>>
>> Bruce Sherwood
>>
>> David Harrison wrote:
>>     
>>> At Bruce's suggestion, I downloaded visual-4.beta15
>>> and tried installing it on my AMD 64-bit machine
>>> running RedHat Enterprise 4.  It uses numpy and
>>> that is good.  But it now has a very major dependency
>>> on GTK and its libraries.  I spent about 4 hours today
>>> compiling an ever increasing number of dependent
>>> libraries to try to get things going.  Just for grins,
>>> take a look at what libraries are needed:
>>>
>>> 4 atk-1.19.3/     4 gtk+-2.10.13/     4 libsigc++-1.2.7/  4 
>>> cairo-1.4.8/    4 gtkglext-1.2.0/   4 libsigc++-2.0.17/
>>> 4 cairomm-1.2.4/  4 gtkglextmm-1.2.0/   4 gtkmm-2.10.10/    4 
>>> numpy-1.0.3/     4 glib-2.12.9/    4 gtkmm-2.2.12/     4 
>>> pango-1.17.3/     4 visual-4.beta15/
>>> 4 glibmm-2.12.10/ 4 libglademm-2.4.2/ 4 Python-2.5.1/
>>>
>>> Many of these are tricky to install because they are either
>>> not very friendly to 64-bit or to installs in non-standard
>>> locations.  After a fair bit of hacking, I managed to get
>>> all of these installed (watch out for gtkglextmm -- it has some
>>> sort of wild problem attempting to link against libGL -- libtool
>>> for it had to be seriously hacked).
>>>
>>> After everything was installed, visual-4.beta15 finally stopped
>>> complaining about dependent libraries and compiled.  But
>>> it looks like it has a problem:
>>>
>>> steiner.visual-4.beta15: make
>>> Making all in site-packages/visual
>>> make[1]: Entering directory 
>>> `/home/dharrison/wares/rhe4_a64/src/visual-4.beta15/site-packages/visual' 
>>>
>>> make[1]: Nothing to be done for `all'.
>>> make[1]: Leaving directory 
>>> `/home/dharrison/wares/rhe4_a64/src/visual-4.beta15/site-packages/visual' 
>>>
>>> Making all in src
>>> make[1]: Entering directory 
>>> `/home/dharrison/wares/rhe4_a64/src/visual-4.beta15/src'
>>> This is a quiet Makefile. If make exits with an error, check 
>>> src/build.log to see the complete error message(s). In the event of 
>>> an error that you cannot debug, please send a message to 
>>> vis...@li..., including the files 
>>> config.log and src/build.log, requesting assistance.
>>> Compiling ./core/arrow.cpp ...
>>> Compiling ./core/axial.cpp ...
>>> Compiling ./core/box.cpp ...
>>> Compiling ./core/cone.cpp ...
>>> Compiling ./core/cylinder.cpp ...
>>> Compiling ./core/display_kernel.cpp ...
>>> Compiling ./core/ellipsoid.cpp ...
>>> Compiling ./core/frame.cpp ...
>>> Compiling ./core/label.cpp ...
>>> Compiling ./core/mouseobject.cpp ...
>>> Compiling ./core/primitive.cpp ...
>>> Compiling ./core/pyramid.cpp ...
>>> Compiling ./core/rectangular.cpp ...
>>> Compiling ./core/renderable.cpp ...
>>> Compiling ./core/ring.cpp ...
>>> Compiling ./core/sphere.cpp ...
>>> Compiling ./python/convex.cpp ...
>>> Compiling ./python/curve.cpp ...
>>> Compiling ./python/cvisualmodule.cpp ...
>>> Compiling ./python/faces.cpp ...
>>> Compiling ./python/num_util.cpp ...
>>> make[1]: *** [num_util.lo] Error 1
>>> make[1]: Leaving directory 
>>> `/home/dharrison/wares/rhe4_a64/src/visual-4.beta15/src'
>>> make: *** [all-recursive] Error 1
>>>
>>> I looked at the build log and here is what it says:
>>>
>>>  g++ -I/home/dharrison/wares/rhe4_a64/include/python2.5 
>>> -I/home/dharrison/wares/rhe4_a64/lib/python2.5/site-packages/numpy/core/include 
>>> -DHAVE_CONFIG_H -I../include -I../include 
>>> -I/home/dharrison/wares/rhe4_a64/include/gtkglextmm-1.2 
>>> -I/home/dharrison/wares/rhe4_a64/lib/gtkglextmm-1.2/include 
>>> -I/home/dharrison/wares/rhe4_a64/include/gtkglext-1.0 
>>> -I/home/dharrison/wares/rhe4_a64/include/gtkmm-2.4 
>>> -I/home/dharrison/wares/rhe4_a64/lib/gtkmm-2.4/include 
>>> -I/home/dharrison/wares/rhe4_a64/lib/gtkglext-1.0/include 
>>> -I/usr/X11R6/include 
>>> -I/home/dharrison/wares/rhe4_a64/include/gdkmm-2.4 
>>> -I/home/dharrison/wares/rhe4_a64/lib/gdkmm-2.4/include 
>>> -I/home/dharrison/wares/rhe4_a64/include/pangomm-1.4 
>>> -I/home/dharrison/wares/rhe4_a64/include/gtk-2.0 
>>> -I/home/dharrison/wares/rhe4_a64/lib/gtk-2.0/include 
>>> -I/home/dharrison/wares/rhe4_a64/include/pango-1.0 
>>> -I/home/dharrison/wares/rhe4_a64/include/glib-2.0 
>>> -I/home/dharrison/wares/rhe4_a64/lib/glib-2.0/include 
>>> -I/home/dharrison/wares/rhe4_a64/include/cairo 
>>> -I/home/dharrison/wares/rhe4_a64/include/glibmm-2.4 
>>> -I/home/dharrison/wares/rhe4_a64/lib/glibmm-2.4/include 
>>> -I/home/dharrison/wares/rhe4_a64/include/cairomm-1.0 
>>> -I/home/dharrison/wares/rhe4_a64/include/sigc++-2.0 
>>> -I/home/dharrison/wares/rhe4_a64/lib/sigc++-2.0/include 
>>> -I/home/dharrison/wares/rhe4_a64/include/atk-1.0 
>>> -I/home/dharrison/wares/rhe4_a64/include/atkmm-1.6 
>>> -I/usr/include/freetype2 
>>> -I/home/dharrison/wares/rhe4_a64/include/libglademm-2.4 
>>> -I/home/dharrison/wares/rhe4_a64/lib/libglademm-2.4/include 
>>> -I/usr/include/libpng12 -I/usr/include/libglade-2.0 
>>> -I/usr/include/libxml2 -I../include/gtk2 -pthread 
>>> -I/home/dharrison/wares/rhe4_a64/include/glib-2.0 
>>> -I/home/dharrison/wares/rhe4_a64/lib/glib-2.0/include 
>>> -I/home/dharrison/wares/rhe4_a64/include/python2.5 
>>> -I/home/dharrison/wares/rhe4_a64/lib/python2.5/site-packages/numpy/core/include 
>>> -fpic -DPIC -g -O2 -ftemplate-depth-120 -MMD -MF num_util.d -MT 
>>> "num_util.d num_util.lo" -c ./python/num_util.cpp  -fPIC -DPIC -o 
>>> .libs/num_util.o
>>> ./python/num_util.cpp: In function `std::vector<npy_intp, 
>>> std::allocator<npy_intp> > 
>>> cvisual::python::shape(boost::python::numeric::array)':
>>> ./python/num_util.cpp:234: error: cannot convert `npy_intp*' to 
>>> `int*' in initialization
>>>
>>> I will take a look see and see if this is something that can be 
>>> fixed. Has anyone else
>>> managed to get the beta to compile from ground zero on 64-bit linux?  
>>> So far,
>>> I think I have spent about 1.5 days just on the installation.  I hope
>>> this is not typical.
>>>
>>>        David
>>>
>>>
>>>
>>>
>>> ------------------------------------------------------------------------- 
>>>
>>> This SF.net email is sponsored by DB2 Express
>>> Download DB2 Express C - the FREE version of DB2 express and take
>>> control of your XML. No limits. Just data. Click to get it now.
>>> http://sourceforge.net/powerbar/db2/
>>> _______________________________________________
>>> Visualpython-users mailing list
>>> Vis...@li...
>>> https://lists.sourceforge.net/lists/listinfo/visualpython-users
>>>       
>
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by DB2 Express
> Download DB2 Express C - the FREE version of DB2 express and take
> control of your XML. No limits. Just data. Click to get it now.
> http://sourceforge.net/powerbar/db2/
> _______________________________________________
> Visualpython-users mailing list
> Vis...@li...
> https://lists.sourceforge.net/lists/listinfo/visualpython-users
>
>   

Re: [Visualpython-users] Still no progress with the Beta version of visual python

[David H. <dha...@ta...>]

I got past the problem in num_util.cpp: it looks like a simple typo.
Line 234 in that file should read:

  npy_intp* dims_ptr = PyArray_DIMS(arr.ptr());

I then found that visual inherits the linking problem from
gtkglextmm-1.2.0.  It tries to link against libGL twice: once
in the right place /usr/lib64/libGL.so and once in the wrong
place: /usr/lib/libGL.so.  A hack to its generated libtool
script gets around that problem.

But then it finally threw me a very significant curve:

/usr/bin/ld: 
/usr/lib/gcc/x86_64-redhat-linux/3.4.6/../../../../lib64/libboost_python.a(numeric.o): 
relocation R_X86_64_32 against `a local symbol' can not be used when 
making a shared object; recompile with -fPIC
/usr/lib/gcc/x86_64-redhat-linux/3.4.6/../../../../lib64/libboost_python.a: 
could not read symbols: Bad value

It appears it is trying to link against a non-shared version of 
libboost_python.
I will try to hack around this one too (there appears to be a .so for
libboost_python) but I think I am going to have to edit the final link line.

       David

Bruce Sherwood wrote:
> Alas, I have no experience with nor access to a 64-bit environment, 
> and it does sound like that's the issue. I'm surprised that there 
> aren't standard 64-bit packages for GTK, which after all is a pretty 
> standard library. Is there anyone reading this list with 64-bit 
> experience?
>
> Bruce Sherwood
>
> David Harrison wrote:
>> At Bruce's suggestion, I downloaded visual-4.beta15
>> and tried installing it on my AMD 64-bit machine
>> running RedHat Enterprise 4.  It uses numpy and
>> that is good.  But it now has a very major dependency
>> on GTK and its libraries.  I spent about 4 hours today
>> compiling an ever increasing number of dependent
>> libraries to try to get things going.  Just for grins,
>> take a look at what libraries are needed:
>>
>> 4 atk-1.19.3/     4 gtk+-2.10.13/     4 libsigc++-1.2.7/  4 
>> cairo-1.4.8/    4 gtkglext-1.2.0/   4 libsigc++-2.0.17/
>> 4 cairomm-1.2.4/  4 gtkglextmm-1.2.0/   4 gtkmm-2.10.10/    4 
>> numpy-1.0.3/     4 glib-2.12.9/    4 gtkmm-2.2.12/     4 
>> pango-1.17.3/     4 visual-4.beta15/
>> 4 glibmm-2.12.10/ 4 libglademm-2.4.2/ 4 Python-2.5.1/
>>
>> Many of these are tricky to install because they are either
>> not very friendly to 64-bit or to installs in non-standard
>> locations.  After a fair bit of hacking, I managed to get
>> all of these installed (watch out for gtkglextmm -- it has some
>> sort of wild problem attempting to link against libGL -- libtool
>> for it had to be seriously hacked).
>>
>> After everything was installed, visual-4.beta15 finally stopped
>> complaining about dependent libraries and compiled.  But
>> it looks like it has a problem:
>>
>> steiner.visual-4.beta15: make
>> Making all in site-packages/visual
>> make[1]: Entering directory 
>> `/home/dharrison/wares/rhe4_a64/src/visual-4.beta15/site-packages/visual' 
>>
>> make[1]: Nothing to be done for `all'.
>> make[1]: Leaving directory 
>> `/home/dharrison/wares/rhe4_a64/src/visual-4.beta15/site-packages/visual' 
>>
>> Making all in src
>> make[1]: Entering directory 
>> `/home/dharrison/wares/rhe4_a64/src/visual-4.beta15/src'
>> This is a quiet Makefile. If make exits with an error, check 
>> src/build.log to see the complete error message(s). In the event of 
>> an error that you cannot debug, please send a message to 
>> vis...@li..., including the files 
>> config.log and src/build.log, requesting assistance.
>> Compiling ./core/arrow.cpp ...
>> Compiling ./core/axial.cpp ...
>> Compiling ./core/box.cpp ...
>> Compiling ./core/cone.cpp ...
>> Compiling ./core/cylinder.cpp ...
>> Compiling ./core/display_kernel.cpp ...
>> Compiling ./core/ellipsoid.cpp ...
>> Compiling ./core/frame.cpp ...
>> Compiling ./core/label.cpp ...
>> Compiling ./core/mouseobject.cpp ...
>> Compiling ./core/primitive.cpp ...
>> Compiling ./core/pyramid.cpp ...
>> Compiling ./core/rectangular.cpp ...
>> Compiling ./core/renderable.cpp ...
>> Compiling ./core/ring.cpp ...
>> Compiling ./core/sphere.cpp ...
>> Compiling ./python/convex.cpp ...
>> Compiling ./python/curve.cpp ...
>> Compiling ./python/cvisualmodule.cpp ...
>> Compiling ./python/faces.cpp ...
>> Compiling ./python/num_util.cpp ...
>> make[1]: *** [num_util.lo] Error 1
>> make[1]: Leaving directory 
>> `/home/dharrison/wares/rhe4_a64/src/visual-4.beta15/src'
>> make: *** [all-recursive] Error 1
>>
>> I looked at the build log and here is what it says:
>>
>>  g++ -I/home/dharrison/wares/rhe4_a64/include/python2.5 
>> -I/home/dharrison/wares/rhe4_a64/lib/python2.5/site-packages/numpy/core/include 
>> -DHAVE_CONFIG_H -I../include -I../include 
>> -I/home/dharrison/wares/rhe4_a64/include/gtkglextmm-1.2 
>> -I/home/dharrison/wares/rhe4_a64/lib/gtkglextmm-1.2/include 
>> -I/home/dharrison/wares/rhe4_a64/include/gtkglext-1.0 
>> -I/home/dharrison/wares/rhe4_a64/include/gtkmm-2.4 
>> -I/home/dharrison/wares/rhe4_a64/lib/gtkmm-2.4/include 
>> -I/home/dharrison/wares/rhe4_a64/lib/gtkglext-1.0/include 
>> -I/usr/X11R6/include 
>> -I/home/dharrison/wares/rhe4_a64/include/gdkmm-2.4 
>> -I/home/dharrison/wares/rhe4_a64/lib/gdkmm-2.4/include 
>> -I/home/dharrison/wares/rhe4_a64/include/pangomm-1.4 
>> -I/home/dharrison/wares/rhe4_a64/include/gtk-2.0 
>> -I/home/dharrison/wares/rhe4_a64/lib/gtk-2.0/include 
>> -I/home/dharrison/wares/rhe4_a64/include/pango-1.0 
>> -I/home/dharrison/wares/rhe4_a64/include/glib-2.0 
>> -I/home/dharrison/wares/rhe4_a64/lib/glib-2.0/include 
>> -I/home/dharrison/wares/rhe4_a64/include/cairo 
>> -I/home/dharrison/wares/rhe4_a64/include/glibmm-2.4 
>> -I/home/dharrison/wares/rhe4_a64/lib/glibmm-2.4/include 
>> -I/home/dharrison/wares/rhe4_a64/include/cairomm-1.0 
>> -I/home/dharrison/wares/rhe4_a64/include/sigc++-2.0 
>> -I/home/dharrison/wares/rhe4_a64/lib/sigc++-2.0/include 
>> -I/home/dharrison/wares/rhe4_a64/include/atk-1.0 
>> -I/home/dharrison/wares/rhe4_a64/include/atkmm-1.6 
>> -I/usr/include/freetype2 
>> -I/home/dharrison/wares/rhe4_a64/include/libglademm-2.4 
>> -I/home/dharrison/wares/rhe4_a64/lib/libglademm-2.4/include 
>> -I/usr/include/libpng12 -I/usr/include/libglade-2.0 
>> -I/usr/include/libxml2 -I../include/gtk2 -pthread 
>> -I/home/dharrison/wares/rhe4_a64/include/glib-2.0 
>> -I/home/dharrison/wares/rhe4_a64/lib/glib-2.0/include 
>> -I/home/dharrison/wares/rhe4_a64/include/python2.5 
>> -I/home/dharrison/wares/rhe4_a64/lib/python2.5/site-packages/numpy/core/include 
>> -fpic -DPIC -g -O2 -ftemplate-depth-120 -MMD -MF num_util.d -MT 
>> "num_util.d num_util.lo" -c ./python/num_util.cpp  -fPIC -DPIC -o 
>> .libs/num_util.o
>> ./python/num_util.cpp: In function `std::vector<npy_intp, 
>> std::allocator<npy_intp> > 
>> cvisual::python::shape(boost::python::numeric::array)':
>> ./python/num_util.cpp:234: error: cannot convert `npy_intp*' to 
>> `int*' in initialization
>>
>> I will take a look see and see if this is something that can be 
>> fixed. Has anyone else
>> managed to get the beta to compile from ground zero on 64-bit linux?  
>> So far,
>> I think I have spent about 1.5 days just on the installation.  I hope
>> this is not typical.
>>
>>        David
>>
>>
>>
>>
>> ------------------------------------------------------------------------- 
>>
>> This SF.net email is sponsored by DB2 Express
>> Download DB2 Express C - the FREE version of DB2 express and take
>> control of your XML. No limits. Just data. Click to get it now.
>> http://sourceforge.net/powerbar/db2/
>> _______________________________________________
>> Visualpython-users mailing list
>> Vis...@li...
>> https://lists.sourceforge.net/lists/listinfo/visualpython-users
>

Re: [Visualpython-users] Still no progress with the Beta version of visual python

[Bruce S. <Bru...@nc...>]

Alas, I have no experience with nor access to a 64-bit environment, and 
it does sound like that's the issue. I'm surprised that there aren't 
standard 64-bit packages for GTK, which after all is a pretty standard 
library. Is there anyone reading this list with 64-bit experience?

Bruce Sherwood

David Harrison wrote:
> At Bruce's suggestion, I downloaded visual-4.beta15
> and tried installing it on my AMD 64-bit machine
> running RedHat Enterprise 4.  It uses numpy and
> that is good.  But it now has a very major dependency
> on GTK and its libraries.  I spent about 4 hours today
> compiling an ever increasing number of dependent
> libraries to try to get things going.  Just for grins,
> take a look at what libraries are needed:
> 
> 4 atk-1.19.3/     
> 4 gtk+-2.10.13/     
> 4 libsigc++-1.2.7/  
> 4 cairo-1.4.8/    
> 4 gtkglext-1.2.0/   
> 4 libsigc++-2.0.17/
> 4 cairomm-1.2.4/  
> 4 gtkglextmm-1.2.0/   
> 4 gtkmm-2.10.10/    
> 4 numpy-1.0.3/     
> 4 glib-2.12.9/    
> 4 gtkmm-2.2.12/     
> 4 pango-1.17.3/     
> 4 visual-4.beta15/
> 4 glibmm-2.12.10/ 
> 4 libglademm-2.4.2/ 
> 4 Python-2.5.1/
> 
> Many of these are tricky to install because they are either
> not very friendly to 64-bit or to installs in non-standard
> locations.  After a fair bit of hacking, I managed to get
> all of these installed (watch out for gtkglextmm -- it has some
> sort of wild problem attempting to link against libGL -- libtool
> for it had to be seriously hacked).
> 
> After everything was installed, visual-4.beta15 finally stopped
> complaining about dependent libraries and compiled.  But
> it looks like it has a problem:
> 
> steiner.visual-4.beta15: make
> Making all in site-packages/visual
> make[1]: Entering directory 
> `/home/dharrison/wares/rhe4_a64/src/visual-4.beta15/site-packages/visual'
> make[1]: Nothing to be done for `all'.
> make[1]: Leaving directory 
> `/home/dharrison/wares/rhe4_a64/src/visual-4.beta15/site-packages/visual'
> Making all in src
> make[1]: Entering directory 
> `/home/dharrison/wares/rhe4_a64/src/visual-4.beta15/src'
> This is a quiet Makefile. If make exits with an error, check 
> src/build.log to see the complete error message(s). In the event of an 
> error that you cannot debug, please send a message to 
> vis...@li..., including the files config.log 
> and src/build.log, requesting assistance.
> Compiling ./core/arrow.cpp ...
> Compiling ./core/axial.cpp ...
> Compiling ./core/box.cpp ...
> Compiling ./core/cone.cpp ...
> Compiling ./core/cylinder.cpp ...
> Compiling ./core/display_kernel.cpp ...
> Compiling ./core/ellipsoid.cpp ...
> Compiling ./core/frame.cpp ...
> Compiling ./core/label.cpp ...
> Compiling ./core/mouseobject.cpp ...
> Compiling ./core/primitive.cpp ...
> Compiling ./core/pyramid.cpp ...
> Compiling ./core/rectangular.cpp ...
> Compiling ./core/renderable.cpp ...
> Compiling ./core/ring.cpp ...
> Compiling ./core/sphere.cpp ...
> Compiling ./python/convex.cpp ...
> Compiling ./python/curve.cpp ...
> Compiling ./python/cvisualmodule.cpp ...
> Compiling ./python/faces.cpp ...
> Compiling ./python/num_util.cpp ...
> make[1]: *** [num_util.lo] Error 1
> make[1]: Leaving directory 
> `/home/dharrison/wares/rhe4_a64/src/visual-4.beta15/src'
> make: *** [all-recursive] Error 1
> 
> I looked at the build log and here is what it says:
> 
>  g++ -I/home/dharrison/wares/rhe4_a64/include/python2.5 
> -I/home/dharrison/wares/rhe4_a64/lib/python2.5/site-packages/numpy/core/include 
> -DHAVE_CONFIG_H -I../include -I../include 
> -I/home/dharrison/wares/rhe4_a64/include/gtkglextmm-1.2 
> -I/home/dharrison/wares/rhe4_a64/lib/gtkglextmm-1.2/include 
> -I/home/dharrison/wares/rhe4_a64/include/gtkglext-1.0 
> -I/home/dharrison/wares/rhe4_a64/include/gtkmm-2.4 
> -I/home/dharrison/wares/rhe4_a64/lib/gtkmm-2.4/include 
> -I/home/dharrison/wares/rhe4_a64/lib/gtkglext-1.0/include 
> -I/usr/X11R6/include -I/home/dharrison/wares/rhe4_a64/include/gdkmm-2.4 
> -I/home/dharrison/wares/rhe4_a64/lib/gdkmm-2.4/include 
> -I/home/dharrison/wares/rhe4_a64/include/pangomm-1.4 
> -I/home/dharrison/wares/rhe4_a64/include/gtk-2.0 
> -I/home/dharrison/wares/rhe4_a64/lib/gtk-2.0/include 
> -I/home/dharrison/wares/rhe4_a64/include/pango-1.0 
> -I/home/dharrison/wares/rhe4_a64/include/glib-2.0 
> -I/home/dharrison/wares/rhe4_a64/lib/glib-2.0/include 
> -I/home/dharrison/wares/rhe4_a64/include/cairo 
> -I/home/dharrison/wares/rhe4_a64/include/glibmm-2.4 
> -I/home/dharrison/wares/rhe4_a64/lib/glibmm-2.4/include 
> -I/home/dharrison/wares/rhe4_a64/include/cairomm-1.0 
> -I/home/dharrison/wares/rhe4_a64/include/sigc++-2.0 
> -I/home/dharrison/wares/rhe4_a64/lib/sigc++-2.0/include 
> -I/home/dharrison/wares/rhe4_a64/include/atk-1.0 
> -I/home/dharrison/wares/rhe4_a64/include/atkmm-1.6 
> -I/usr/include/freetype2 
> -I/home/dharrison/wares/rhe4_a64/include/libglademm-2.4 
> -I/home/dharrison/wares/rhe4_a64/lib/libglademm-2.4/include 
> -I/usr/include/libpng12 -I/usr/include/libglade-2.0 
> -I/usr/include/libxml2 -I../include/gtk2 -pthread 
> -I/home/dharrison/wares/rhe4_a64/include/glib-2.0 
> -I/home/dharrison/wares/rhe4_a64/lib/glib-2.0/include 
> -I/home/dharrison/wares/rhe4_a64/include/python2.5 
> -I/home/dharrison/wares/rhe4_a64/lib/python2.5/site-packages/numpy/core/include 
> -fpic -DPIC -g -O2 -ftemplate-depth-120 -MMD -MF num_util.d -MT 
> "num_util.d num_util.lo" -c ./python/num_util.cpp  -fPIC -DPIC -o 
> .libs/num_util.o
> ./python/num_util.cpp: In function `std::vector<npy_intp, 
> std::allocator<npy_intp> > 
> cvisual::python::shape(boost::python::numeric::array)':
> ./python/num_util.cpp:234: error: cannot convert `npy_intp*' to `int*' 
> in initialization
> 
> I will take a look see and see if this is something that can be fixed. 
> Has anyone else
> managed to get the beta to compile from ground zero on 64-bit linux?  So 
> far,
> I think I have spent about 1.5 days just on the installation.  I hope
> this is not typical.
> 
>        David
> 
> 
> 
> 
> -------------------------------------------------------------------------
> This SF.net email is sponsored by DB2 Express
> Download DB2 Express C - the FREE version of DB2 express and take
> control of your XML. No limits. Just data. Click to get it now.
> http://sourceforge.net/powerbar/db2/
> _______________________________________________
> Visualpython-users mailing list
> Vis...@li...
> https://lists.sourceforge.net/lists/listinfo/visualpython-users

[Visualpython-users] Still no progress with the Beta version of visual python

[David H. <dha...@ta...>]

At Bruce's suggestion, I downloaded visual-4.beta15
and tried installing it on my AMD 64-bit machine
running RedHat Enterprise 4.  It uses numpy and
that is good.  But it now has a very major dependency
on GTK and its libraries.  I spent about 4 hours today
compiling an ever increasing number of dependent
libraries to try to get things going.  Just for grins,
take a look at what libraries are needed:

4 atk-1.19.3/     
4 gtk+-2.10.13/     
4 libsigc++-1.2.7/  
4 cairo-1.4.8/    
4 gtkglext-1.2.0/   
4 libsigc++-2.0.17/
4 cairomm-1.2.4/  
4 gtkglextmm-1.2.0/   
4 gtkmm-2.10.10/    
4 numpy-1.0.3/     
4 glib-2.12.9/    
4 gtkmm-2.2.12/     
4 pango-1.17.3/     
4 visual-4.beta15/
4 glibmm-2.12.10/ 
4 libglademm-2.4.2/ 
4 Python-2.5.1/

Many of these are tricky to install because they are either
not very friendly to 64-bit or to installs in non-standard
locations.  After a fair bit of hacking, I managed to get
all of these installed (watch out for gtkglextmm -- it has some
sort of wild problem attempting to link against libGL -- libtool
for it had to be seriously hacked).

After everything was installed, visual-4.beta15 finally stopped
complaining about dependent libraries and compiled.  But
it looks like it has a problem:

steiner.visual-4.beta15: make
Making all in site-packages/visual
make[1]: Entering directory 
`/home/dharrison/wares/rhe4_a64/src/visual-4.beta15/site-packages/visual'
make[1]: Nothing to be done for `all'.
make[1]: Leaving directory 
`/home/dharrison/wares/rhe4_a64/src/visual-4.beta15/site-packages/visual'
Making all in src
make[1]: Entering directory 
`/home/dharrison/wares/rhe4_a64/src/visual-4.beta15/src'
This is a quiet Makefile. If make exits with an error, check 
src/build.log to see the complete error message(s). In the event of an 
error that you cannot debug, please send a message to 
vis...@li..., including the files config.log 
and src/build.log, requesting assistance.
Compiling ./core/arrow.cpp ...
Compiling ./core/axial.cpp ...
Compiling ./core/box.cpp ...
Compiling ./core/cone.cpp ...
Compiling ./core/cylinder.cpp ...
Compiling ./core/display_kernel.cpp ...
Compiling ./core/ellipsoid.cpp ...
Compiling ./core/frame.cpp ...
Compiling ./core/label.cpp ...
Compiling ./core/mouseobject.cpp ...
Compiling ./core/primitive.cpp ...
Compiling ./core/pyramid.cpp ...
Compiling ./core/rectangular.cpp ...
Compiling ./core/renderable.cpp ...
Compiling ./core/ring.cpp ...
Compiling ./core/sphere.cpp ...
Compiling ./python/convex.cpp ...
Compiling ./python/curve.cpp ...
Compiling ./python/cvisualmodule.cpp ...
Compiling ./python/faces.cpp ...
Compiling ./python/num_util.cpp ...
make[1]: *** [num_util.lo] Error 1
make[1]: Leaving directory 
`/home/dharrison/wares/rhe4_a64/src/visual-4.beta15/src'
make: *** [all-recursive] Error 1

I looked at the build log and here is what it says:

 g++ -I/home/dharrison/wares/rhe4_a64/include/python2.5 
-I/home/dharrison/wares/rhe4_a64/lib/python2.5/site-packages/numpy/core/include 
-DHAVE_CONFIG_H -I../include -I../include 
-I/home/dharrison/wares/rhe4_a64/include/gtkglextmm-1.2 
-I/home/dharrison/wares/rhe4_a64/lib/gtkglextmm-1.2/include 
-I/home/dharrison/wares/rhe4_a64/include/gtkglext-1.0 
-I/home/dharrison/wares/rhe4_a64/include/gtkmm-2.4 
-I/home/dharrison/wares/rhe4_a64/lib/gtkmm-2.4/include 
-I/home/dharrison/wares/rhe4_a64/lib/gtkglext-1.0/include 
-I/usr/X11R6/include -I/home/dharrison/wares/rhe4_a64/include/gdkmm-2.4 
-I/home/dharrison/wares/rhe4_a64/lib/gdkmm-2.4/include 
-I/home/dharrison/wares/rhe4_a64/include/pangomm-1.4 
-I/home/dharrison/wares/rhe4_a64/include/gtk-2.0 
-I/home/dharrison/wares/rhe4_a64/lib/gtk-2.0/include 
-I/home/dharrison/wares/rhe4_a64/include/pango-1.0 
-I/home/dharrison/wares/rhe4_a64/include/glib-2.0 
-I/home/dharrison/wares/rhe4_a64/lib/glib-2.0/include 
-I/home/dharrison/wares/rhe4_a64/include/cairo 
-I/home/dharrison/wares/rhe4_a64/include/glibmm-2.4 
-I/home/dharrison/wares/rhe4_a64/lib/glibmm-2.4/include 
-I/home/dharrison/wares/rhe4_a64/include/cairomm-1.0 
-I/home/dharrison/wares/rhe4_a64/include/sigc++-2.0 
-I/home/dharrison/wares/rhe4_a64/lib/sigc++-2.0/include 
-I/home/dharrison/wares/rhe4_a64/include/atk-1.0 
-I/home/dharrison/wares/rhe4_a64/include/atkmm-1.6 
-I/usr/include/freetype2 
-I/home/dharrison/wares/rhe4_a64/include/libglademm-2.4 
-I/home/dharrison/wares/rhe4_a64/lib/libglademm-2.4/include 
-I/usr/include/libpng12 -I/usr/include/libglade-2.0 
-I/usr/include/libxml2 -I../include/gtk2 -pthread 
-I/home/dharrison/wares/rhe4_a64/include/glib-2.0 
-I/home/dharrison/wares/rhe4_a64/lib/glib-2.0/include 
-I/home/dharrison/wares/rhe4_a64/include/python2.5 
-I/home/dharrison/wares/rhe4_a64/lib/python2.5/site-packages/numpy/core/include 
-fpic -DPIC -g -O2 -ftemplate-depth-120 -MMD -MF num_util.d -MT 
"num_util.d num_util.lo" -c ./python/num_util.cpp  -fPIC -DPIC -o 
.libs/num_util.o
./python/num_util.cpp: In function `std::vector<npy_intp, 
std::allocator<npy_intp> > 
cvisual::python::shape(boost::python::numeric::array)':
./python/num_util.cpp:234: error: cannot convert `npy_intp*' to `int*' 
in initialization

I will take a look see and see if this is something that can be fixed. 
Has anyone else
managed to get the beta to compile from ground zero on 64-bit linux?  So 
far,
I think I have spent about 1.5 days just on the installation.  I hope
this is not typical.

       David

Re: [Visualpython-users] Getting current visual python to use Numpy

[Bruce S. <Bru...@nc...>]

The simplest solution at this moment is to use the beta version of 
Visual, which uses numpy and seems to work mostly fine on Linux (there 
are serious problems on Windows). Go to the "Recent developments" 
section of vpython.org. Sounds like I should put something about this 
new development on the regular Linux download page.

Bruce Sherwood

David Harrison wrote:
> We just upgraded some of our linux machines to Red Hat enterprise 4
> and I was going through the exercise of updating visual python.  I had
> a very bad experience installing the software initially a couple of
> months ago and I was hoping to see some improvement.
> 
> Unfortunately, it seems the installation experience is still very poor.
> Probably the number one issue is the Numpy vs. (Numeric or Numarray)
> issue.  Although the config and installation instructions say you
> should use Numeric or Numarray, neither of these software packages
> are supported anymore.  Downloading and installing the currently
> supported software, Numpy, does not work with the current visual
> python setup.
> 
> The claim from the Numpy author is that that its interface is
> close but not fully compatible.  So, just how bad would it be
> to update the current 3.2.9 so that it could use the Numpy
> interface?  Is this a huge amount of work or a small amount?
> If it is small, I might take a whack at it.  Just moving that one
> step forward would help.
> 
> That still leaves the hugely awful BLAS, etc stuff.  But at least
> for this round, I decided not to bother with it and let it use
> the slower defaults.
> 
>        David
> 
> 
> 
> -------------------------------------------------------------------------
> This SF.net email is sponsored by DB2 Express
> Download DB2 Express C - the FREE version of DB2 express and take
> control of your XML. No limits. Just data. Click to get it now.
> http://sourceforge.net/powerbar/db2/
> _______________________________________________
> Visualpython-users mailing list
> Vis...@li...
> https://lists.sourceforge.net/lists/listinfo/visualpython-users

[Visualpython-users] Getting current visual python to use Numpy

[David H. <dha...@ta...>]

We just upgraded some of our linux machines to Red Hat enterprise 4
and I was going through the exercise of updating visual python.  I had
a very bad experience installing the software initially a couple of
months ago and I was hoping to see some improvement.

Unfortunately, it seems the installation experience is still very poor.
Probably the number one issue is the Numpy vs. (Numeric or Numarray)
issue.  Although the config and installation instructions say you
should use Numeric or Numarray, neither of these software packages
are supported anymore.  Downloading and installing the currently
supported software, Numpy, does not work with the current visual
python setup.

The claim from the Numpy author is that that its interface is
close but not fully compatible.  So, just how bad would it be
to update the current 3.2.9 so that it could use the Numpy
interface?  Is this a huge amount of work or a small amount?
If it is small, I might take a whack at it.  Just moving that one
step forward would help.

That still leaves the hugely awful BLAS, etc stuff.  But at least
for this round, I decided not to bother with it and let it use
the slower defaults.

       David

[Visualpython-users] need VPython Mac icon

[Joe H. <hea...@gm...>]

Could I enlist someone's help to make one or more high quality OS X
icons for VPython? I have some ideas but don't have the graphics
experience or Photoshop.

[Visualpython-users] clickable VPython on the Mac

[Joe H. <hea...@gm...>]

At long last, I have finally managed to created a fully clickable
VPython on the Mac. Don't get too excited because I have *NOT* ported
visual to Aqua as that's beyond my capabilities. We just got twelve
brand new MacBooks at work to use in astronomy and physics classes and
I put Python 2.5.1 and VPython on them via Fink, compiling from
source. I used a tool called XDroplets
(http://www.rhythmiccanvas.com/software/xdroplets/index.html) that
encapsulates a script of most any type and makes it behave and look
like an OS X application bundle. The best part is that it allows for
dropping a source file onto an icon and having both X11 and IDLE
launch and the source file shows up in IDLE! So with this applet in
the OS X Dock, students can launch VPython by clicking on a single
icon. X11 is launched if it's not already running. Multiple source
files can be dropped and each will open into its own window. It's
simply fantastic! The other best part is that when students open an
existing file, it opens in their home directory; they no longer have
to navigate through the filesystem and possible store their programs
in an insecure place or overwrite existing demo files. The app is only
about 228KB in size so it's very small. Actually, it's nothing more
than an encapsulated shell script.

I'm awaiting permission to use the icon I chose from the PSF. As soon
as I get the okay from them, I will make the app available for
distribution.

Joe Heafner

Re: [Visualpython-users] raising/lowering windows("scenes")

[Gary P. <gar...@gm...>]

On 6/18/07, Bruce Sherwood <Bru...@nc...> wrote:
>
> marc desmarais wrote:



[...]


>
> VPython, which is the only term that has been used since sometime in the
> year 2000. It is interesting that despite the consistent use of
> "VPython" it is not uncommon to hear people use the name "Visual Python"
> which has not been used to refer to the project since 2000.



On the other hand, we still do
import visual

Usually someone asks about that. And I sometimes call it "visual python".
(I thought the two were synonymous.)    Perhaps when development recommences

import vpython

can be made to work also, with "visual" deprecated.

Re: [Visualpython-users] raising/lowering windows("scenes")

[Bruce S. <Bru...@nc...>]

marc desmarais wrote:
> First, congratulations on Vpython, you do great work.  I discovered 
> python 8 months ago, and Vpython 6 months ago, and I now use Vpython 
> at Boeing to communicate concepts.  I hope to contribute some examples 
> in the future ( I can send screen shots if you are interested, let me 
> know what email address to use).
Thanks! Please send them to Bru...@nc....
> For the GUI, I much prefer the look and feel of Tkinter (or better yet 
> wxpython) and I wish  you covered how best to bring those together. 
> I've had some success combining Tkinter and Vpython, but it took some 
> digging, and I'm not sure if the resulting jerkiness of the 3D 
> animation is my doing or just computer overloading.
There has been no serious attempt to build a real GUI for VPython, 
though there is the module "controls" that has been used to some extent. 
Presumably you found the example program that is installed with VPython 
that illustrates combining VPython with Tkinter, and which in fact uses 
Tkinter to change the color of a VPython object.
> I've recently added the ability to choose colors using the tinker 
> color chooser, but it always pops up behind the Vpython scene/window.  
> Is there a way to access the window manager to lower or raise a 
> scene/window)?  This function seems to be missing from the "display" 
> class. I use XP/Vista btw.
I faced a similar problem in the program "fields" available at 
http://www4.ncsu.edu/~rwchabay/mi/Public_Programs_VII.htm. See the 
functions filedialog and getsources.
> PS I'm now thinking of moving to wxpython for my 2D avionic panel & 
> HUD simulations, and would love to see some instructions on making 
> Vpython and Wxpython work together.
Question to the list: If anyone has relevant experience, could you 
contribute a sample program?
>
> PPS.  active state has an editor called visual python, which can 
> confuse the uninitiated that google for vpython btw.  I don't know 
> what you can do about that, other than sticking with vpthon (this 
> users' list for instance could be confused with active state's product 
> as it is called visualpython-users). Perhaps 3DPython would be better?
 From http://www.activestate.com/Products/Visual_Python: "Engineering 
support for Visual Python was discontinued as of December 15, 2005. 
Installation and configuration support is no longer available." Some 
history: David Scherer originally called VPython "Visual Python", and 
that name persists in sourceforge.net and the mailing list, due to the 
difficulty of changing the name for those purposes. But later, Active 
State started a project "Visual Python" in what seemed to be a 
collaboration with Microsoft, and it was decided to change the name to 
VPython, which is the only term that has been used since sometime in the 
year 2000. It is interesting that despite the consistent use of 
"VPython" it is not uncommon to hear people use the name "Visual Python" 
which has not been used to refer to the project since 2000.
>
> Marc Desmarais
> 5916 Bixby Village Dr. #95
> Long Beach, CA, 90803
>
> 707-540-5367 cell
>
> ------------------------------------------------------------------------
>

[Visualpython-users] raising/lowering windows("scenes")

[marc d. <mde...@ya...>]

First, congratulations on Vpython, you do great work.  I discovered python 8 months ago, and Vpython 6 months ago, and I now use Vpython at Boeing to communicate concepts.  I hope to contribute some examples in the future ( I can send screen shots if you are interested, let me know what email address to use).

For the GUI, I much prefer the look and feel of Tkinter (or better yet wxpython) and I wish  you covered how best to bring those together. I've had some success combining Tkinter and Vpython, but it took some digging, and I'm not sure if the resulting jerkiness of the 3D animation is my doing or just computer overloading.

I've recently added the ability to choose colors using the tinker color chooser, but it always pops up behind the Vpython scene/window.  Is there a way to access the window manager to lower or raise a scene/window)?  This function seems to be missing from the "display" class. I use XP/Vista btw.

PS I'm now thinking of moving to wxpython for my 2D avionic panel & HUD simulations, and would love to see some instructions on making Vpython and Wxpython work together.

PPS.  active state has an editor called visual python, which can confuse the uninitiated that google for vpython btw.  I don't know what you can do about that, other than sticking with vpthon (this users' list for instance could be confused with active state's product as it is called visualpython-users). Perhaps 3DPython would be better?


Marc Desmarais
5916 Bixby Village Dr. #95
Long Beach, CA, 90803
 
707-540-5367 cell
       
---------------------------------
Park yourself in front of a world of choices in alternative vehicles.
Visit the Yahoo! Auto Green Center.

Re: [Visualpython-users] how to visualize 3D surface?

[Bruce S. <Bru...@nc...>]

One thing you could do is simply use curve to superimpose some 
"longitude" and "latitude" lines in the gyro2.py program, so that the 
trajectory of the end of the gyroscope axis would visibly be on a 
spherical surface.

Bruce Sherwood

pri...@gm... wrote:
>
> yeh for real, that was a stupid mistake by myself by mixing up that arrange and linspace command. i think that was more an influence of my aggression that Vpython doesnt behave like I want it to do...;)
>
> nevertheless, I have tried that faces_heightfield.py examplle again, but still it doesnt draw my partial sphere. Bruce gave me the advice that the motiton of a gyro is also an axample prog in vpython. thats true and it also works good there with leaving the trail by the figure axis...I just thought with that partial sphere it gonna look a little more nice and should help to understand the motion of the top...okay, if this problem is too heavy...but I still have one or more week to solve this problem till I will speak about my project in front of some co-students.
>
> okay, if somebody has an idea...please reply. 
>
> good night
>
> -philipp varso
>
>   
>

Re: [Visualpython-users] how to visualize 3D surface?

[<pri...@gm...>]

-------- Original-Nachricht --------
Datum: Sat, 16 Jun 2007 11:11:48 -0400
Von: "Gary Pajer" <gar...@gm...>
An: "pri...@gm..." <pri...@gm...>, vis...@li...
Betreff: Re: [Visualpython-users] how to visualize 3D surface?

> I see two problems right away, and one future difficulty
> 
> [I'll continue to look at your problem later... I have other tasks to work
> on at the moment]
> 
> 1.)
> theta and phi aren't the same length.  It can be tricky to get arange() to
> produce arrays of the same length.   The linspace() function is usually
> preferred, but linspace is not part of plain python or visual.  It is part
> of numpy.  You have two choices.  Play around with arange() until the
> arrays
> are the same length. To help with that, add some print statements like
> 
> phi = arange(  ,  ,  )
> print phi.shape
> theta = arange( , , )
> print theta.shape
> 
> then play with the third argument of arange() until the shapes are the
> same.
> Your second option is to import numpy.  You may have to install it.  For
> installation instructions see
> http://numpy.scipy.org
> Then you will have linspace available:
> 
> >from numpy import *
> phi = linspace(  ,   ,  20)
> theta = linspace(  ,   ,  20)
> 
> will insure that both arrays have 20 elements.
> 
> Note:  You should have no problems with the syntax above.  However, it is
> considered poor python style to use   "import *"  as that form has the
> *potential* for overwriting previously written variable names.  Better:
> 
> import numpy as N        #  "N" can  be any valid python variable name
> phi = N.linspace(  ,   ,   )
> theta = N.linspace(  ,  ,   )
> 
> For short/simple programs it usually not an issue.  *BUT*  the current
> version of Vpython depends on an earlier version of numpy (Numeric).   I
> worry slightly that importing numpy over Numeric might cause a problem.  I
> may be being too cautious.
> 
> 2.)
> Mesh() is not a python function.  I think you probably already know that.
> Once you settle issue 1.) above, you should look at the example
> faces_cone.py for hints on how to proceed.  Numpy does have "mesh"
> functions
> that might do what you are trying to do ... if that approach is necessary.
> 
> Future difficulty:
> Bruce says that transparency works only in the Linux beta version.  I'm
> not
> sure how to deal with that.  One idea is to make your faces thin, and your
> curves thick so that they would be seen from top or bottom.  You would
> have
> to rotate the scene to see the top and the curve and the manifold.
> 
> -gary

yeh for real, that was a stupid mistake by myself by mixing up that arrange and linspace command. i think that was more an influence of my aggression that Vpython doesnt behave like I want it to do...;)

nevertheless, I have tried that faces_heightfield.py examplle again, but still it doesnt draw my partial sphere. Bruce gave me the advice that the motiton of a gyro is also an axample prog in vpython. thats true and it also works good there with leaving the trail by the figure axis...I just thought with that partial sphere it gonna look a little more nice and should help to understand the motion of the top...okay, if this problem is too heavy...but I still have one or more week to solve this problem till I will speak about my project in front of some co-students.

okay, if somebody has an idea...please reply. 

good night

-philipp varso

> On 6/16/07, pri...@gm... <pri...@gm...> wrote:
> >
> > -------- Original-Nachricht --------
> > Datum: Sat, 16 Jun 2007 09:35:35 -0400
> > Von: "Gary Pajer" <gar...@gm...>
> > An: "pri...@gm..." <pri...@gm...>
> > CC: vis...@li...
> > Betreff: Re: [Visualpython-users] how to visualize 3D surface?
> >
> > > If I understand correctly, you are drawing the path of the top of the
> > > sphere
> > > (using curve( ) ?)  But you would also like to show the underlying
> > > manifold,
> > > the part of the sphere between theta_min and theta_max.
> > >
> > > Are you on Windows or Linux?  I'll look more closely at your original
> > post
> > > now.
> > >
> > > How is the weather in Dresden?
> > >
> > > -gary
> > >
> > > On 6/16/07, pri...@gm... <pri...@gm...> wrote:
> > > >
> > > > well, I am trying to visualize the dynamics of a heavy symmetrical
> top
> > > > fixed in one point which is under the influence of gravitational
> force
> > > > (without a dissipative function...so the  energy is constant).
> > therefore
> > > I
> > > > have solved the Hamilton equations in python (ODEint). now I wanted
> to
> > > show
> > > > the dynamic of the figure axis of the top (which is the rotating
> > z-axis)
> > > > which should leaving a trail on a 2D-manifold. in this case it is a
> > part
> > > of
> > > > a sphere. The Euler angle Theta as a function of time is in this
> case
> > a
> > > > periodic function which oscillates between theta_max and theta_min.
> > > these
> > > > are the boundraries for my partial sphere.
> > > >
> > > > so I am depended on using Vpython if I want to draw this surface,
> > since
> > > i
> > > > want to show they dynamics of this top. i hope some of you
> understood
> > my
> > > > project...hopefully i will receive some help on this. thanks on
> that.
> > > >
> > > > greets from desden/germany
> > > >
> > > > -philipp
> > > > --
> > > > Psssst! Schon vom neuen GMX MultiMessenger gehört?
> > > > Der kanns mit allen: http://www.gmx.net/de/go/multimessenger
> > > >
> > > >
> > >
> >
> -------------------------------------------------------------------------
> > > > This SF.net email is sponsored by DB2 Express
> > > > Download DB2 Express C - the FREE version of DB2 express and take
> > > > control of your XML. No limits. Just data. Click to get it now.
> > > > http://sourceforge.net/powerbar/db2/
> > > > _______________________________________________
> > > > Visualpython-users mailing list
> > > > Vis...@li...
> > > > https://lists.sourceforge.net/lists/listinfo/visualpython-users
> > > >
> >
> > yes, you got it. i am using cruve(), to see which path the Euler Angle
> > Theta has taken (drawing the path of the figure axis of the top). this
> path
> > lies on a surface of a sphere between those theta_max and theta_min...so
> I
> > am trying to show the undrlying part of the sphere.
> >
> > i could plot a whole sphere, but then you arent able to see the rotating
> > top, just the path of Theta(time). So I wanted just to draw a part of
> the
> > sphere (2D-manifold)...thats it. I hope it won't be a problem that I am
> > working with windows at home, but at our university the system is linux.
> so
> > I've got an alternative...
> >
> > -philipp varso
> >
> > PS: the weather is fine, but i dont have too much of it since I am
> sitting
> > all day long learning for physics
> > --
> > Ist Ihr Browser Vista-kompatibel? Jetzt die neuesten
> > Browser-Versionen downloaden: http://www.gmx.net/de/go/browser
> >

-- 
GMX FreeMail: 1 GB Postfach, 5 E-Mail-Adressen, 10 Free SMS.
Alle Infos und kostenlose Anmeldung: http://www.gmx.net/de/go/freemail

Re: [Visualpython-users] how to visualize 3D surface?

[Bruce S. <Bru...@nc...>]

I didn't say quite that transparency only works in the beta version for 
Linux. Transparency DOES work in the beta version for Windows. Rather 
the issue is that the Windows beta version is buggy in various ways (not 
transparency), and this version is not ready for routine production use.

Bruce Sherwood

Gary Pajer wrote:

> 
> Future difficulty:
> Bruce says that transparency works only in the Linux beta version.  I'm 
> not sure how to deal with that.  One idea is to make your faces thin, 
> and your curves thick so that they would be seen from top or bottom.  
> You would have to rotate the scene to see the top and the curve and the 
> manifold.
> 
> -gary
> 

Re: [Visualpython-users] how to visualize 3D surface?

[Bruce S. <Bru...@nc...>]

Take a look at the program gyro2.py in the suite of example programs 
that is installed with VPython. It's a visualization of a precessing 
nutating top using Lagrangian mechanics, with a curve drawn to show the 
trajectory of the tip of the axle of the top.

Bruce Sherwood

pri...@gm... wrote:
> well, I am trying to visualize the dynamics of a heavy symmetrical top fixed in one point which is under the influence of gravitational force (without a dissipative function...so the  energy is constant). therefore I have solved the Hamilton equations in python (ODEint). now I wanted to show the dynamic of the figure axis of the top (which is the rotating z-axis) which should leaving a trail on a 2D-manifold. in this case it is a part of a sphere. The Euler angle Theta as a function of time is in this case a periodic function which oscillates between theta_max and theta_min. these are the boundraries for my partial sphere. 
> 
> so I am depended on using Vpython if I want to draw this surface, since i want to show they dynamics of this top. i hope some of you understood my project...hopefully i will receive some help on this. thanks on that.
> 
> greets from desden/germany
> 
> -philipp  

Re: [Visualpython-users] how to visualize 3D surface?

[Gary P. <gar...@gm...>]

I see two problems right away, and one future difficulty

[I'll continue to look at your problem later... I have other tasks to work
on at the moment]

1.)
theta and phi aren't the same length.  It can be tricky to get arange() to
produce arrays of the same length.   The linspace() function is usually
preferred, but linspace is not part of plain python or visual.  It is part
of numpy.  You have two choices.  Play around with arange() until the array=
s
are the same length. To help with that, add some print statements like

phi =3D arange(  ,  ,  )
print phi.shape
theta =3D arange( , , )
print theta.shape

then play with the third argument of arange() until the shapes are the same=
.
Your second option is to import numpy.  You may have to install it.  For
installation instructions see
http://numpy.scipy.org
Then you will have linspace available:

from numpy import *
phi =3D linspace(  ,   ,  20)
theta =3D linspace(  ,   ,  20)

will insure that both arrays have 20 elements.

Note:  You should have no problems with the syntax above.  However, it is
considered poor python style to use   "import *"  as that form has the
*potential* for overwriting previously written variable names.  Better:

import numpy as N        #  "N" can  be any valid python variable name
phi =3D N.linspace(  ,   ,   )
theta =3D N.linspace(  ,  ,   )

For short/simple programs it usually not an issue.  *BUT*  the current
version of Vpython depends on an earlier version of numpy (Numeric).   I
worry slightly that importing numpy over Numeric might cause a problem.  I
may be being too cautious.

2.)
Mesh() is not a python function.  I think you probably already know that.
Once you settle issue 1.) above, you should look at the example
faces_cone.py for hints on how to proceed.  Numpy does have "mesh" function=
s
that might do what you are trying to do ... if that approach is necessary.

Future difficulty:
Bruce says that transparency works only in the Linux beta version.  I'm not
sure how to deal with that.  One idea is to make your faces thin, and your
curves thick so that they would be seen from top or bottom.  You would have
to rotate the scene to see the top and the curve and the manifold.

-gary

On 6/16/07, pri...@gm... <pri...@gm...> wrote:
>
> -------- Original-Nachricht --------
> Datum: Sat, 16 Jun 2007 09:35:35 -0400
> Von: "Gary Pajer" <gar...@gm...>
> An: "pri...@gm..." <pri...@gm...>
> CC: vis...@li...
> Betreff: Re: [Visualpython-users] how to visualize 3D surface?
>
> > If I understand correctly, you are drawing the path of the top of the
> > sphere
> > (using curve( ) ?)  But you would also like to show the underlying
> > manifold,
> > the part of the sphere between theta_min and theta_max.
> >
> > Are you on Windows or Linux?  I'll look more closely at your original
> post
> > now.
> >
> > How is the weather in Dresden?
> >
> > -gary
> >
> > On 6/16/07, pri...@gm... <pri...@gm...> wrote:
> > >
> > > well, I am trying to visualize the dynamics of a heavy symmetrical to=
p
> > > fixed in one point which is under the influence of gravitational forc=
e
> > > (without a dissipative function...so the  energy is constant).
> therefore
> > I
> > > have solved the Hamilton equations in python (ODEint). now I wanted t=
o
> > show
> > > the dynamic of the figure axis of the top (which is the rotating
> z-axis)
> > > which should leaving a trail on a 2D-manifold. in this case it is a
> part
> > of
> > > a sphere. The Euler angle Theta as a function of time is in this case
> a
> > > periodic function which oscillates between theta_max and theta_min.
> > these
> > > are the boundraries for my partial sphere.
> > >
> > > so I am depended on using Vpython if I want to draw this surface,
> since
> > i
> > > want to show they dynamics of this top. i hope some of you understood
> my
> > > project...hopefully i will receive some help on this. thanks on that.
> > >
> > > greets from desden/germany
> > >
> > > -philipp
> > > --
> > > Psssst! Schon vom neuen GMX MultiMessenger geh=F6rt?
> > > Der kanns mit allen: http://www.gmx.net/de/go/multimessenger
> > >
> > >
> >
> -------------------------------------------------------------------------
> > > This SF.net email is sponsored by DB2 Express
> > > Download DB2 Express C - the FREE version of DB2 express and take
> > > control of your XML. No limits. Just data. Click to get it now.
> > > http://sourceforge.net/powerbar/db2/
> > > _______________________________________________
> > > Visualpython-users mailing list
> > > Vis...@li...
> > > https://lists.sourceforge.net/lists/listinfo/visualpython-users
> > >
>
> yes, you got it. i am using cruve(), to see which path the Euler Angle
> Theta has taken (drawing the path of the figure axis of the top). this pa=
th
> lies on a surface of a sphere between those theta_max and theta_min...so =
I
> am trying to show the undrlying part of the sphere.
>
> i could plot a whole sphere, but then you arent able to see the rotating
> top, just the path of Theta(time). So I wanted just to draw a part of the
> sphere (2D-manifold)...thats it. I hope it won't be a problem that I am
> working with windows at home, but at our university the system is linux. =
so
> I've got an alternative...
>
> -philipp varso
>
> PS: the weather is fine, but i dont have too much of it since I am sittin=
g
> all day long learning for physics
> --
> Ist Ihr Browser Vista-kompatibel? Jetzt die neuesten
> Browser-Versionen downloaden: http://www.gmx.net/de/go/browser
>

Re: [Visualpython-users] how to visualize 3D surface?

[Gary P. <gar...@gm...>]

 <vis...@li...>
If I understand correctly, you are drawing the path of the top of the spher=
e
(using curve( ) ?)  But you would also like to show the underlying manifold=
,
the part of the sphere between theta_min and theta_max.

Are you on Windows or Linux?  I'll look more closely at your original post
now.

How is the weather in Dresden?

-gary

On 6/16/07, pri...@gm... <pri...@gm...> wrote:
>
> well, I am trying to visualize the dynamics of a heavy symmetrical top
> fixed in one point which is under the influence of gravitational force
> (without a dissipative function...so the  energy is constant). therefore =
I
> have solved the Hamilton equations in python (ODEint). now I wanted to sh=
ow
> the dynamic of the figure axis of the top (which is the rotating z-axis)
> which should leaving a trail on a 2D-manifold. in this case it is a part =
of
> a sphere. The Euler angle Theta as a function of time is in this case a
> periodic function which oscillates between theta_max and theta_min. these
> are the boundraries for my partial sphere.
>
> so I am depended on using Vpython if I want to draw this surface, since i
> want to show they dynamics of this top. i hope some of you understood my
> project...hopefully i will receive some help on this. thanks on that.
>
> greets from desden/germany
>
> -philipp
> --
> Psssst! Schon vom neuen GMX MultiMessenger geh=F6rt?
> Der kanns mit allen: http://www.gmx.net/de/go/multimessenger
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by DB2 Express
> Download DB2 Express C - the FREE version of DB2 express and take
> control of your XML. No limits. Just data. Click to get it now.
> http://sourceforge.net/powerbar/db2/
> _______________________________________________
> Visualpython-users mailing list
> Vis...@li...
> https://lists.sourceforge.net/lists/listinfo/visualpython-users
>

[Visualpython-users] how to visualize 3D surface?

[<pri...@gm...>]

well, I am trying to visualize the dynamics of a heavy symmetrical top fixed in one point which is under the influence of gravitational force (without a dissipative function...so the  energy is constant). therefore I have solved the Hamilton equations in python (ODEint). now I wanted to show the dynamic of the figure axis of the top (which is the rotating z-axis) which should leaving a trail on a 2D-manifold. in this case it is a part of a sphere. The Euler angle Theta as a function of time is in this case a periodic function which oscillates between theta_max and theta_min. these are the boundraries for my partial sphere. 

so I am depended on using Vpython if I want to draw this surface, since i want to show they dynamics of this top. i hope some of you understood my project...hopefully i will receive some help on this. thanks on that.

greets from desden/germany

-philipp  
-- 
Psssst! Schon vom neuen GMX MultiMessenger gehört?
Der kanns mit allen: http://www.gmx.net/de/go/multimessenger

Re: [Visualpython-users] how to visualize 3D surface?

[Bruce S. <Bru...@nc...>]

Sorry, but I'm swamped with other work and can't help with Python 
coding. Maybe someone else will give you some help. The "frames not 
aligned" sounds like maybe the problem could be that you can't add x**2 
and y**2 because these two arrays don't have the same lengths.

Bruce Sherwood

pri...@gm... wrote:
> hi folks,
>
> I am new to VPython so this question may sound a little bit naive;
> I've got a question concerning the problem how to visiualize a 2D-manifold in Vypthon. In this special case, the parameters are Theta and Phi to just plot a part of the surface of a sphere, where the Radius is constant. An additional feature would be that the surface is something like transparent or it also could be a spherical grid. I hope my english is not too bad, so that some of you could understand what my problem currently is. I hope I'll receive any help soon.
>
> Thank you very much.
>
> greets from germany
>
> -Philipp
>
>   
>> You might study the example program faces_heightfield.py that comes with
>> VPython, as it creates a surface.
>>
>> The most general VPython object is "faces" which lets you display
>> arbitrary triangles and specify lighting at vertices. This low-level
>> object could be used to display a portion of a sphere. Or if you want to
>> display a portion of a sphere as a grid, use the "curve" object.
>>
>> There exists a beta version of VPython which lets you specify that an
>> object is transparent. See the "Recent developments" section of
>> vpython.org. Unfortunately the Windows version has some serious bugs
>> (see the documentation in the Recent developments section) and there is
>> no Macintosh version. The Linux version mostly works correctly.
>>
>> Bruce Sherwood
>>
>> pri...@gm... wrote:
>>     
>>> hi folks,
>>>
>>> I am new to VPython so this question may sound a little bit naive;
>>> I've got a question concerning the problem how to visiualize a
>>>       
>> 2D-manifold in Vypthon. In this special case, the parameters are Theta and Phi to
>> just plot a part of the surface of a sphere, where the Radius is constant. An
>> additional feature would be that the surface is something like transparent
>> or it also could be a spherical grid. I hope my english is not too bad, so
>> that some of you could understand what my problem currently is. I hope
>> I'll receive any help soon.
>>     
>>> Thank you very much.
>>>
>>> greets from germany
>>>
>>> -Philipp
>>>   
>>>       
>
> thx for the fast post. just like you recommanded, i have studied this example, and also tried to plot my surface. but this heightfiled_faces.py example comes a little bit complicated and it also doesn't do that what I expected. i need to add that I am new to python and I just wonder that it doesn't seem to be easy to plot a 3D-surface (just like in mathematica for example). nevertheless, I would truly like to understand this example programm, and also what I made wrong in useage; here are the differences to the example prog:
>
> ############################
> theta = arange(0,pi/2.0,2./20)
> phi= arange(0,2.0*pi,2./20)
> x = sin(theta)*cos(phi)
> y = sin(theta)*sin(phi)
>
> m = Mesh( x, sqrt(1.0 - (x**2.0 + y**2.0)), y)
> ############################
>
> when I try to run faces_heightfield.py with these modifications, "frames are not aligned" occurs. maybe somebody could help me with this problem, I would truly appreciate it....Thank you very much.
>
> -phil 
>
>
>   

[Visualpython-users] how to visualize 3D surface?

[<pri...@gm...>]

hi folks,

I am new to VPython so this question may sound a little bit naive;
I've got a question concerning the problem how to visiualize a 2D-manifold in Vypthon. In this special case, the parameters are Theta and Phi to just plot a part of the surface of a sphere, where the Radius is constant. An additional feature would be that the surface is something like transparent or it also could be a spherical grid. I hope my english is not too bad, so that some of you could understand what my problem currently is. I hope I'll receive any help soon.

Thank you very much.

greets from germany

-Philipp

> You might study the example program faces_heightfield.py that comes with
> VPython, as it creates a surface.
>
> The most general VPython object is "faces" which lets you display
> arbitrary triangles and specify lighting at vertices. This low-level
> object could be used to display a portion of a sphere. Or if you want to
> display a portion of a sphere as a grid, use the "curve" object.
>
> There exists a beta version of VPython which lets you specify that an
> object is transparent. See the "Recent developments" section of
> vpython.org. Unfortunately the Windows version has some serious bugs
> (see the documentation in the Recent developments section) and there is
> no Macintosh version. The Linux version mostly works correctly.
>
> Bruce Sherwood
>
> pri...@gm... wrote:
> > hi folks,
> >
> > I am new to VPython so this question may sound a little bit naive;
> > I've got a question concerning the problem how to visiualize a
> 2D-manifold in Vypthon. In this special case, the parameters are Theta and Phi to
> just plot a part of the surface of a sphere, where the Radius is constant. An
> additional feature would be that the surface is something like transparent
> or it also could be a spherical grid. I hope my english is not too bad, so
> that some of you could understand what my problem currently is. I hope
> I'll receive any help soon.
> >
> > Thank you very much.
> >
> > greets from germany
> >
> > -Philipp
> >   

thx for the fast post. just like you recommanded, i have studied this example, and also tried to plot my surface. but this heightfiled_faces.py example comes a little bit complicated and it also doesn't do that what I expected. i need to add that I am new to python and I just wonder that it doesn't seem to be easy to plot a 3D-surface (just like in mathematica for example). nevertheless, I would truly like to understand this example programm, and also what I made wrong in useage; here are the differences to the example prog:

############################
theta = arange(0,pi/2.0,2./20)
phi= arange(0,2.0*pi,2./20)
x = sin(theta)*cos(phi)
y = sin(theta)*sin(phi)

m = Mesh( x, sqrt(1.0 - (x**2.0 + y**2.0)), y)
############################

when I try to run faces_heightfield.py with these modifications, "frames are not aligned" occurs. maybe somebody could help me with this problem, I would truly appreciate it....Thank you very much.

-phil 


-- 
Psssst! Schon vom neuen GMX MultiMessenger gehört?
Der kanns mit allen: http://www.gmx.net/de/go/multimessenger