visualpython-users

Re: [Visualpython-users] destroying objects

[Jonathan B. <jbr...@ea...>]

On Sat, 2005-01-22 at 21:19 -0500, Jonathan Brandmeyer wrote:
> On Sat, 2005-01-22 at 20:50 -0200, Flavio Coelho wrote:

> > Unfortunately, there is a pretty significant memory leak in my
> > drawTickLabels function That I couldn't fix. Could it be originated
> > from the label object? 

> I am looking into the memory leak.
> 

Found.  It was an internal bug in Visual's label and font deletion code.
The fix will be in the next release.

Actually, I'm a little surprised that this hasn't affected anyone else
before this.  As far as I can tell, it has been in there since VPython's
original release!

Thank you for your useful reports and feedback.

-Jonathan

Re: [Visualpython-users] destroying objects

[Jonathan B. <jbr...@ea...>]

On Sat, 2005-01-22 at 20:50 -0200, Flavio Coelho wrote:
> Hi Johnathan,
> 
> While waiting for your response, I managed to fix most of the problems
> I had, including the keystrike catcher.
> 
> I posted the updated version of my stripchart module here:
> 
> http://www.procc.fiocruz.br:8080/procc/Members/flavio/codepy/stripch.py/file_view
> 
> Unfortunately, there is a pretty significant memory leak in my
> drawTickLabels function That I couldn't fix. Could it be originated
> from the label object? Have you had any issues concerning creating and
> destroying many labels continuously?
> 
> because of this leak the call to this function in the plot function is
> commented.
> 
> I would appreciate any ideas/ suggestions on this...

Creating new renderable objects is one of the slowest things you can do
in VPython, so I would suggest that after the initial creation of the
labels, you should keep track of which ones represent what, and just
update their content every update rather than destroy and recreate them.

I am looking into the memory leak.

-Jonathan

Re: [Visualpython-users] destroying objects

[Flavio C. <fcc...@gm...>]

Hi Johnathan,

While waiting for your response, I managed to fix most of the problems
I had, including the keystrike catcher.

I posted the updated version of my stripchart module here:

http://www.procc.fiocruz.br:8080/procc/Members/flavio/codepy/stripch.py/fil=
e_view

Unfortunately, there is a pretty significant memory leak in my
drawTickLabels function That I couldn't fix. Could it be originated
from the label object? Have you had any issues concerning creating and
destroying many labels continuously?

because of this leak the call to this function in the plot function is
commented.

I would appreciate any ideas/ suggestions on this...


On Fri, 21 Jan 2005 15:00:35 -0200, Flavio Coelho <fcc...@gm...> wrot=
e:
> Hi Johnathan,
>=20
> Thanks for your reply, it was very useful.
>=20
> I wrote a quick stripchart module for vpython (attached) that would be
> glad to contribute to vpython if you think it may be of use. It still
> need some polishing such as y-axis autoscaling plus adding more
> features.
>=20



--=20
Fl=E1vio Code=E7o Coelho
---------------------------
Great spirits have always found violent opposition from mediocrities. The
latter cannot understand it when a man does not thoughtlessly submit to
hereditary prejudices but honestly and courageously uses his intelligence.
        Albert Einstein

Re: [Visualpython-users] docstrings?

[Jonathan B. <jbr...@ea...>]

On Thu, 2005-01-20 at 19:33 -0200, Flavio Coelho wrote:
> Hi,
> 
> I am a new user of vpython, and I am very excited about its
> capabilities and its potential.
> 
> However, there is one little aspect  of the python source that stroke
> me as something that could be easily fixed, and would be of great help
> to users and would-be-developers alike:
> There is a total lack of docstrings in the whole source!! Besides
> being very unpythonic, it prevents users from getting help on a class
> or function.
> 
> I don't want this to sound like a complaint, but it is such an easy
> thing (for the developers) to do, that I had to make this suggestion.
> 
> Otherwise the package is great and unparalleled in the python world.
> 
> congratulations,

Are you using one of the versions in the 3.x series?  If so, I think you
will find that most of the functions have docstrings in them.
Unfortunately, Boost.Python does not support docstrings for properties
at this time.

-Jonathan

Re: [Visualpython-users] destroying objects

[Jonathan B. <jbr...@ea...>]

On Thu, 2005-01-20 at 21:13 -0200, Flavio Coelho wrote:
> Hello,
> 
> Is there a way to destroy objects (not just make them invisible)?
> removing them from a scene?

I'm not entirely sure what you are asking, but I'll give this one a
shot(gun;) anyway.  Python's memory management is a reference-counted
system.  When the reference count on an object drops to zero, it is
immediately deleted/destroyed/freed.  Internally, every visible
primitive object's owning display owns one reference to the primitive.
Making the object invisible is implemented by removing the object from
its parent display, and removes the display's reference.  This leaves
only the references that might exist in your program.  In general, every
means of accessing the object counts as a reference.  In the simplest
case, you only have a single local or global name that refers to the
object, and you just use the 'del' keyword to get rid of it, like this:

x = sphere()
x.visible = False
del x
# This step will raise an exception, since x does not exist:
x.pos = vector(1,0,0)

Or just assign a different object that name:
x = sphere()
x.q = 1.6e-19
x.visible = False
x = sphere()
print x.q # Error!, x.q does not exist, since this is a new object.

> another question is there a way to let a part of an object get out of
> sight? for example, suppose I have a curve object that keeps growing
> in a given direction. How  do i allow for the older parts of the curve
> to move out of the frame to avoid auto rescaling?

For the curve object, you can write your own special append() function
that performs a shift+replace rather than a true append, like this:

def append_nogrow( curve_object, new_pos):
	curve_object.pos[:-1] = curve_object.pos[1:]
	curve_object.pos[-1] = new_pos

HTH,
-Jonathan Brandmeyer

[Visualpython-users] destroying objects

[Flavio C. <fcc...@gm...>]

Hello,

Is there a way to destroy objects (not just make them invisible)?
removing them from a scene?

another question is there a way to let a part of an object get out of
sight? for example, suppose I have a curve object that keeps growing
in a given direction. How  do i allow for the older parts of the curve
to move out of the frame to avoid auto rescaling?

--=20
Fl=E1vio Code=E7o Coelho
---------------------------
Great spirits have always found violent opposition from mediocrities. The
latter cannot understand it when a man does not thoughtlessly submit to
hereditary prejudices but honestly and courageously uses his intelligence.
        Albert Einstein

[Visualpython-users] docstrings?

[Flavio C. <fcc...@gm...>]

Hi,

I am a new user of vpython, and I am very excited about its
capabilities and its potential.

However, there is one little aspect  of the python source that stroke
me as something that could be easily fixed, and would be of great help
to users and would-be-developers alike:
There is a total lack of docstrings in the whole source!! Besides
being very unpythonic, it prevents users from getting help on a class
or function.

I don't want this to sound like a complaint, but it is such an easy
thing (for the developers) to do, that I had to make this suggestion.

Otherwise the package is great and unparalleled in the python world.

congratulations,



--=20
Fl=E1vio Code=E7o Coelho
---------------------------
Great spirits have always found violent opposition from mediocrities. The
latter cannot understand it when a man does not thoughtlessly submit to
hereditary prejudices but honestly and courageously uses his intelligence.
        Albert Einstein

[Visualpython-users] closing the display

[Flavio C. <fcc...@gm...>]

Hi,

how do I nicely close a gdisplay window?

it doesn't seem to have a close() method...

does it have a destroyer we can call?

thanks...

--=20
Fl=E1vio Code=E7o Coelho
---------------------------
Great spirits have always found violent opposition from mediocrities. The
latter cannot understand it when a man does not thoughtlessly submit to
hereditary prejudices but honestly and courageously uses his intelligence.
        Albert Einstein

Re: [Visualpython-users] mice

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

If you're not a programmer, I don't see what information could be useful 
to you, since VPython is a programming environment. But in the VPython 
docs in the section on mouse interactions are specific examples of how 
to handle mouse clicks, and how to handle mouse drags. And among the 
demo programs the following use mouse interactions: colorsliders, 
crossproduct, drape, gyro, hanoi, planar, stonehenge, tictac, and 
toroid_drag.

Bruce Sherwood

John Brawley wrote:
> Please, can anyone tell me where to find a rather complete tutorial on
> getting the various mouse functions working in a program using VPython?
> I have my windows, my program's visuals run fine in the windows (I'm doing
> stereo).
> I have my controls window, with a few buttons and such.
> Now I need to know how to get myself in contact with the innards of my
> program, through the mouse, but all I can find are the VPython docs and
> tutorials (I'm _not_ a programmer...), which are not complete enough, and
> the few example programs easily available also do not give me enough info.
> I can figure all this out in a year or two....  ....but I haven't that much
> time (*grin*).
> 
> Can anyone shoot me a few URLs to good examples, tutorials, or instructions
> _specifically_ dealing with the VPython mouse functions?
> 
> Thanks!
> 
> peace
> JB
> jgb...@ch...
> http://tetrahedraverse.com
> NOTE! Charter is not blocking viruses,
> Therefore NO ATTACHMENTS, please;
> They will not be downloaded from the Charter mail server.
> __Prearrange__ any attachments, with me first.
> 
> 
> 
> -------------------------------------------------------
> The SF.Net email is sponsored by: Beat the post-holiday blues
> Get a FREE limited edition SourceForge.net t-shirt from ThinkGeek.
> It's fun and FREE -- well, almost....http://www.thinkgeek.com/sfshirt
> _______________________________________________
> Visualpython-users mailing list
> Vis...@li...
> https://lists.sourceforge.net/lists/listinfo/visualpython-users

[Visualpython-users] mice

[John B. <jb...@te...>]

Please, can anyone tell me where to find a rather complete tutorial on
getting the various mouse functions working in a program using VPython?
I have my windows, my program's visuals run fine in the windows (I'm doing
stereo).
I have my controls window, with a few buttons and such.
Now I need to know how to get myself in contact with the innards of my
program, through the mouse, but all I can find are the VPython docs and
tutorials (I'm _not_ a programmer...), which are not complete enough, and
the few example programs easily available also do not give me enough info.
I can figure all this out in a year or two....  ....but I haven't that much
time (*grin*).

Can anyone shoot me a few URLs to good examples, tutorials, or instructions
_specifically_ dealing with the VPython mouse functions?

Thanks!

peace
JB
jgb...@ch...
http://tetrahedraverse.com
NOTE! Charter is not blocking viruses,
Therefore NO ATTACHMENTS, please;
They will not be downloaded from the Charter mail server.
__Prearrange__ any attachments, with me first.

Re: [Visualpython-users] boost problem: visual-3.1.0/make

[Jonathan B. <jbr...@ea...>]

On Fri, 2005-01-14 at 08:49 -0800, Eric Ayars wrote:

> I installed boost with this command:
> sudo bjam  -sTOOLS=gcc -sPYTHON_ROOT=/usr -sPYTHON_VERSION=2.3 install
> The first file that isn't found, wrap_python.hpp, is located at 
> /usr/local/include/boost-1_32/boost/python/detail/wrap_python.hpp
> (and ~/boost_1_32_0/boost/python/detail/wrap_python.hpp).
> 
> Is there something else I need to pass visual-3.1.0/configure or make
> so that they know where to find these 'missing' files?

Yes.  Since  Boost.Python is installed under /usr/local, and that is not
on the default include or library paths, you must do the following:
set the CPPFLAGS environment variable to
-I/usr/local/include/boost_1_32_0

set the LDFLAGS environment variable to -L/usr/local/lib

Make a symbolic link in /usr/local/lib
from /usr/local/lib/libboost-python.so -> libboost-python-gcc-mt-1_32.so

Run configure again (otherwise using the same options as before)

Run make.

Note that in order to use the extension module at runtime, you will have
to add /usr/local/lib to LD_LIBRARY_PATH.

HTH,
-Jonathan Brandmeyer

[Visualpython-users] boost problem: visual-3.1.0/make

[Eric A. <Ay...@ma...>]

Good morning,

I'm having a bit of trouble installing visual on my slackware 
installation(s). I get through the
	visual-3.1.0/configure --prefix=/usr
step, but the next step,
	visual-3.1.0/make
gives me the following first error in src/build.log:
-------------------
  g++ -I/usr/include/python2.3 -DHAVE_CONFIG_H -I../include 
-I..//include -I/usr/in
clude/gtk-1.2 -I/usr/include/glib-1.2 -I/usr/lib/glib/include 
-I/usr/X11R6/include
  -I/usr/include/gtk-1.2 -I/usr/include/glib-1.2 -I/usr/lib/glib/include 
-I/usr/X11
R6/include -D_REENTRANT -I/usr/include/glib-1.2 -I/usr/lib/glib/include 
-I/usr/inc
lude/gtk-1.2 -I/usr/include/glib-1.2 -I/usr/lib/glib/include 
-I/usr/X11R6/include
-fpic -DPIC -g -O2 -ftemplate-depth-120 -g0 -c arrow.cpp  -fPIC -DPIC 
-o .libs/arr
ow.o
In file included from ../include/displaylist.h:8,
                  from ../include/prim.h:8,
                  from ../include/axial.h:8,
                  from ../include/arrow.h:7,
                  from arrow.cpp:4:
../include/cvisual.h:8:47: boost/python/detail/wrap_python.hpp: No such 
file or di
rectory
../include/cvisual.h:12:34: boost/python/tuple.hpp: No such file or 
directory
../include/cvisual.h:13:33: boost/python/list.hpp: No such file or 
directory
--------------------
This sort of message repeats numerous times, and then 'make' eventually 
gives up.
I'm guessing that the problem may be in my boost installation (this is 
the first reference to boost in build.log).

I installed boost with this command:
	sudo bjam  -sTOOLS=gcc -sPYTHON_ROOT=/usr -sPYTHON_VERSION=2.3 install
The first file that isn't found, wrap_python.hpp, is located at
	/usr/local/include/boost-1_32/boost/python/detail/wrap_python.hpp
(and ~/boost_1_32_0/boost/python/detail/wrap_python.hpp).

Is there something else I need to pass visual-3.1.0/configure or make 
so that they know where to find these 'missing' files?

Thanks for any assistance,
-EA
-----------------------------------------------------------------
Dr. Eric Ayars
Assistant Professor of Physics
California State University, Chico
ay...@ma...

Re: [Visualpython-users] Quantian

[Jonathan B. <jbr...@ea...>]

On Thu, 2005-01-13 at 09:01 -0500, Gary wrote:

> But vpython is missing.  I asked the developer, Dirk Eddelbuettel, about 
> adding it, and he replied
> "
> Sounds cool. Once again, packaged in Debian would my life a hell of a lot
> easier -- so could you file an 'rfp' bug against the virtual wnpp package?
> If you use the reportbug(1) tool, a stanza should appear that you can fill
> with some data we commonly supply for 'request for packages' bug.
> "

RFP: Request for Package.  A kind of bug that a Debian user files
against the WNPP (Work-Needing and Prospective Packages) package.  This
has already been done
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=112118, but little
action has been taken on it in years.

> This paragraph may as well be written in Greek.  I assume it's in 
> Debianese.  Can anyone interpret and comment on the feasibility of doing 
> this?  Is it a big deal?  I also assume one needs a Debian to prepare 
> the pacakge.  Volunteers? 

Floris Bruynooghe had produced an unofficial package of VPython for
Debian.  You can find various messages about them in the archives if you
search for her name.

HTH,
-Jonathan

[Visualpython-users] Quantian

[Gary <pa...@in...>]

Between semesters I have been looking at the Quantian scientific 
computing environment, "A Knoppix / Debian variant tailored to numerical 
and quantitative analysis."

Knoppix is a "live CD Linux", that is, it boots from a CD, and the 
entire OS is on the CD.   So it can run on any  x86 machine that can 
boot from a CD.   Your home directory and configuration (/etc) can be 
saved to a floppy or USB device.  The Quantian varient is loaded with 
just about every bit of technical software that you can imagine ...  
TeX,  Grace, xfig, TeXmacs, octave, maxima, scilab, ipython, matplotlib, 
etc etc.  Most of it I've never even heard of.  I've been very 
impressed.  It works really well.  The only downside is that you can't 
easily upgrade or install new software.  (You can do it, but the key 
word is 'easily'.)  That's the trade-off for a completely portable Linux 
scientific environment:  wait for the next version.

But vpython is missing.  I asked the developer, Dirk Eddelbuettel, about 
adding it, and he replied
"
Sounds cool. Once again, packaged in Debian would my life a hell of a lot
easier -- so could you file an 'rfp' bug against the virtual wnpp package?
If you use the reportbug(1) tool, a stanza should appear that you can fill
with some data we commonly supply for 'request for packages' bug.
"

This paragraph may as well be written in Greek.  I assume it's in 
Debianese.  Can anyone interpret and comment on the feasibility of doing 
this?  Is it a big deal?  I also assume one needs a Debian to prepare 
the pacakge.  Volunteers? 

some fine print about Quantian:
Knoppix fits on a CD, but Quantian is too big.  It's actually a "live 
DVD".  But if you, like me, can't burn DVDs there's a very easy 
work-around. 

There is a facility to make a permanent installation to hard drive.  
Then you can update using apt-get.  But the installer is rudimentary, 
and puts everything in one partition.  Not the best. 

It also appears to be possible to install to an external USB drive and 
boot directly from that if your bios supports it.   ( I've gotten Fedora 
to boot that way after a fair amount of work under the hood, but it 
appears that Knoppix has it built-in.  Not sure about that ... my (used 
/ ebay) external USB drive crashed as I was trying it out.)

-gary

vpython 3.1.0 (Was: Re: [Visualpython-users] numarray? Numpy? Slackware 10?)

[Floris B. <fb...@so...>]

Unrelated on this thread here, but anyway...

On Wed, Jan 12, 2005 at 09:49:06PM -0500, Jonathan Brandmeyer wrote:
> If you really want to build with just Numarray, you can do this with
> version 3.1.0 of Visual, which was the first release to support
> Numarray.  It can be found from the Sourceforge download page.

Hmm I didn't realise 3.1.0 was released for *NIX already, I must
have missed the announcement.  To make matters even worse I have
had a bad hardware crash and am thus (hopefully not for too long)
working on a machine not capable of compiling or running VPython.
This means I can't really produce a Debian package for version
3.1.0 for now.

What I remember from packaging the 3.1.0 where things still
needed to be changed in the build scripts (did you manage to get
them right?) is that it didn't need any changes to de debian/
subtree at all.  So I could just apply the old debian tree to it
but I don't want to advertise it publicly when I can't test it.
If you're really eager in getting 3.1.0 as a Debian package you
can just try and copy the debian/ tree yourself, it will be
equally untested.

I'll try to get remote access to a machine running debian where I
have enough administrative rights to build and install which *is*
capable to compile VPython over the weekend.  This means I still
won't be able to test the graphics but I will be able to test and
tweak the build.

Greetings
Floris

-- 
Debian GNU/Linux -- The power of freedom
www.debian.org | www.gnu.org | www.kernel.org

Re: [Visualpython-users] numarray? Numpy? Slackware 10?

[Jonathan B. <jbr...@ea...>]

On Wed, 2005-01-12 at 21:49 -0500, Jonathan Brandmeyer wrote:
> On Tue, 2005-01-11 at 23:16 -0800, Eric Ayars wrote:
> > Anyone here successfully install VPython on Slackware 10? So far I'm  
> > stuck at installing Numeric (23.7), which gives me an error about  
> > "cannot find -llapack" at the "python setup.py install" step.
> 
> First off, are you sure that there isn't a prebuilt binary package of
> Numeric in Slackware?  If you are sure that you need to install Numeric
> from source, install lapack (in Debian, it is called lapack-dev), or
> edit setup.py to use its own simple blas/lapack subroutines.
> 
> > I can install numarray fine, but visual-3.0.3/configure still wants  
> > numeric... Is it possible to run vpython with numarray instead of  
> > numeric? If so, how do I go about changing the configure file? Would  
> > more changes be required? (probably!)
> 
> If you really want to build with just Numarray, you can do this with
> version 3.1.0 of Visual, which was the first release to support
> Numarray.  It can be found from the Sourceforge download page.
> 
> > Or should I keep trying to get numeric installed? Any suggestions?  
> > Here's where the numeric install fails:
> 
> Other people have reported this issue to the Numeric developers, who
> should have this issue fixed in the next release.  You should probably
> file a bug report with them anyway, just as a reminder that this still
> needs fixing.

I just realized that that wasn't a well-structured reply.  The bug is
that the Numeric setup.py script assumes that you have LAPACK installed.
Numeric can be built without LAPACK, in which case it uses a built-in
set of functions that provide the same functionality, at the expense of
performance.  You can therefore either install the lapack library
(easier if a prebuilt package for it is in Slackware), or edit setup.py
to use Numeric's own versions of those functions (easier than building
lapack from source).

HTH,
-Jonathan

Re: [Visualpython-users] numarray? Numpy? Slackware 10?

[Jonathan B. <jbr...@ea...>]

On Tue, 2005-01-11 at 23:16 -0800, Eric Ayars wrote:
> Anyone here successfully install VPython on Slackware 10? So far I'm  
> stuck at installing Numeric (23.7), which gives me an error about  
> "cannot find -llapack" at the "python setup.py install" step.

First off, are you sure that there isn't a prebuilt binary package of
Numeric in Slackware?  If you are sure that you need to install Numeric
from source, install lapack (in Debian, it is called lapack-dev), or
edit setup.py to use its own simple blas/lapack subroutines.

> I can install numarray fine, but visual-3.0.3/configure still wants  
> numeric... Is it possible to run vpython with numarray instead of  
> numeric? If so, how do I go about changing the configure file? Would  
> more changes be required? (probably!)

If you really want to build with just Numarray, you can do this with
version 3.1.0 of Visual, which was the first release to support
Numarray.  It can be found from the Sourceforge download page.

> Or should I keep trying to get numeric installed? Any suggestions?  
> Here's where the numeric install fails:

Other people have reported this issue to the Numeric developers, who
should have this issue fixed in the next release.  You should probably
file a bug report with them anyway, just as a reminder that this still
needs fixing.

-Jonathan Brandmeyer

[Visualpython-users] numarray? Numpy? Slackware 10?

[Eric A. <Ay...@ma...>]

Anyone here successfully install VPython on Slackware 10? So far I'm  
stuck at installing Numeric (23.7), which gives me an error about  
"cannot find -llapack" at the "python setup.py install" step.

I can install numarray fine, but visual-3.0.3/configure still wants  
numeric... Is it possible to run vpython with numarray instead of  
numeric? If so, how do I go about changing the configure file? Would  
more changes be required? (probably!)

Or should I keep trying to get numeric installed? Any suggestions?  
Here's where the numeric install fails:
...
building 'lapack_lite' extension
gcc -pthread -shared build/temp.linux-i686-2.3/Src/lapack_litemodule.o
-L/usr/lib/atlas -llapack -lcblas -lf77blas -latlas -lg2c -o
build/lib.linux-i686-2.3/lapack_lite.so
/usr/lib/gcc-lib/i486-slackware-linux/3.3.4/../../../../i486-slackware- 
linux/bin/ld:
cannot find -llapack
collect2: ld returned 1 exit status
error: command 'gcc' failed with exit status 1

Thanks for any help you can offer.
-EA

-----------------------------------------------------------------
Dr. Eric Ayars
Assistant Professor of Physics
California State University, Chico
ay...@ma...

[Visualpython-users] Correction to Windows for Python 2.4

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

Jonathan Brandmeyer corrected a bug in the function for rotating an 
object, and there is a new experimental Windows version for Python 2.4.

Bruce Sherwood

Re: [Visualpython-users] Help: color - what did I do wrong?

[Jonathan B. <jbr...@ea...>]

On Sun, 2005-01-02 at 20:09 -0800, Toan T Nguyen wrote:
> > But I want to have different color for some faces. Can somebody help?
> > Thanks a lot.

When you make a bug report, please include a complete working example.

Try this patch.  You don't need to rebuild VPython to apply it, just
apply it to the files installed in the site-packages directory.  I've
already committed it to CVS.

-Jonathan Brandmeyer

Index: visual/primitives.py
===================================================================
RCS
file: /cvsroot/visualpython/vpython/site-packages/visual/primitives.py,v
retrieving revision 1.1
diff -u -r1.1 primitives.py
--- visual/primitives.py        22 Dec 2004 16:25:15 -0000      1.1
+++ visual/primitives.py        3 Jan 2005 17:30:37 -0000
@@ -31,6 +31,13 @@
                del keywords['frame']
        else:
                frame = None
+
+       # Add special handling for the faces object.
+       if displayobject.__class__ == faces:
+               if keywords.has_key('pos'):
+                       displayobject.pos = keywords['pos']
+                       del keywords['pos']
+
        # Some objects (like the curve and convex) need to have color
set before        # pos if color is a single tuple.
        if keywords.has_key('color'):
@@ -38,11 +45,6 @@
                del keywords['color']
        else:
                displayobject.color = display.foreground
-       # Add special handling for the faces object.
-       if displayobject.__class__ == faces:
-               if keywords.has_key('pos'):
-                       displayobject.pos = keywords['pos']
-                       del keywords['pos']

        # Assign all other properties
        for key, value in keywords.iteritems():

Re: [Visualpython-users] Help: color - what did I do wrong?

[Toan T N. <nt...@ph...>]

On Sunday 02 January 2005 19:45, Toan T Nguyen wrote:
> Hi, I'm using the face object and I keep getting this error message about
> color must be the same size as pos, but I don't know how to fix it. My
> script is like this:
>
> pos = zeros( (nface*3,3), Float)
> nor = pos
> col = pos
>
> ## assigning pos, nor, col here
> for i in range(nface)
> 	pos[i*3]=(p1x, p1y, p1z)
> 	nor[i*3]=(n1x, n1y, n1z)
> 	col[i*3]=color.red
> 	... # similarly for i*3+1, i*3+2
>
> faces(pos=pos, normal=nor, color=col)
>
>
> It gives me the following error
>
>   File "/usr/local/lib64/python2.3/site-packages/visual/primitives.py",
> line 301, in __init__
>     process_init_args_from_keyword_dictionary(self, keywords)
>   File "/usr/local/lib64/python2.3/site-packages/visual/primitives.py",
> line 37, in process_init_args_from_keyword_dictionary
>     displayobject.color =  keywords['color']
> ValueError: color must be the same size as pos.
>
>
> I tried
>
>  faces(pos=pos, normal=nor, color=col)

I meant 

faces(pos=pos, normal=nor, color=pos)


>
> It gives the same error. The only thing that works is
>
> faces(pos=pos, normal=nor, color=color.red)
>
> But I want to have different color for some faces. Can somebody help?
> Thanks a lot.
>
>
>
> -------------------------------------------------------
> The SF.Net email is sponsored by: Beat the post-holiday blues
> Get a FREE limited edition SourceForge.net t-shirt from ThinkGeek.
> It's fun and FREE -- well, almost....http://www.thinkgeek.com/sfshirt
> _______________________________________________
> Visualpython-users mailing list
> Vis...@li...
> https://lists.sourceforge.net/lists/listinfo/visualpython-users

[Visualpython-users] Help: color - what did I do wrong?

[Toan T N. <nt...@ph...>]

Hi, I'm using the face object and I keep getting this error message about 
color must be the same size as pos, but I don't know how to fix it. My script 
is like this:

pos = zeros( (nface*3,3), Float)
nor = pos
col = pos

## assigning pos, nor, col here 
for i in range(nface)
	pos[i*3]=(p1x, p1y, p1z)
	nor[i*3]=(n1x, n1y, n1z)
	col[i*3]=color.red
	... # similarly for i*3+1, i*3+2

faces(pos=pos, normal=nor, color=col)


It gives me the following error

  File "/usr/local/lib64/python2.3/site-packages/visual/primitives.py", line 
301, in __init__
    process_init_args_from_keyword_dictionary(self, keywords)
  File "/usr/local/lib64/python2.3/site-packages/visual/primitives.py", line 
37, in process_init_args_from_keyword_dictionary
    displayobject.color =  keywords['color']
ValueError: color must be the same size as pos.


I tried

 faces(pos=pos, normal=nor, color=col)

It gives the same error. The only thing that works is 

faces(pos=pos, normal=nor, color=color.red)

But I want to have different color for some faces. Can somebody help? Thanks a 
lot.

Re: [Visualpython-users] Experimental Windows version for Python 2.4

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

The Windows version available at vpython.org does not at the moment 
include/install Numarray, just Numeric. But if you install Numarray 
yourself, the behavior will be as Jonathan describes and has implemented.

Bruce Sherwood

Jonathan Brandmeyer wrote:

> On Tue, 2004-12-28 at 11:48 -0500, Bruce Sherwood wrote:
> 
>>Now available at vpython.org is an experimental version of VPython for 
>>Python 2.4 on Windows. The only known bug is that programs that use the 
>>"text" module to display block letters (including the demo programs 
>>stonehenge.py and texttest.py) fail on some but not all computers. We 
>>would appreciate feedback on this or other bugs you may find. This was 
>>built using the new Boost-based package created by Jonathan Brandmeyer.
>>
> 
> 
> 
> There is one additional change that this release has versus prior
> releases that used Boost.Python for its bindings.  This build supports
> and includes both Numeric 23.6 and Numarray 1.1.1.
> 
> By default, executing "from visual import *" will bring in Numeric, and
> not Numarray, and will use Numeric arrays to hold the data of curves,
> convex, and faces objects.  You can alter this behavior by calling the
> function set_array_backend() with either the string "numeric" or
> "numarray".  To be safe, you should only do this once at the beginning
> of your program, before creating any array-using objects.  If you are
> forcing the use of Numarray in this way, you must also run "from
> numarray import *".  It is not safe to change the backend after you have
> created a visual object that uses an array (even if its pos, color, or
> normal attribute appears to be empty).
> 
> Alternatively, you can edit site-packages/visual/__init__.py (lines
> 16-23) to change the preference for all applications on your particular
> installation.
> 
> I intend to provide support for Numeric as long as it is supported
> upstream and not too much of a burden on Visual.  Numarray has very good
> backwards compatibility with Numeric, and Visual's core works exactly
> the same with both.  However, there are some differences which you may
> see in your own programs (some of our demos are also affected).  So,
> Visual will prefer Numeric by default rather than Numarray, for maybe 6
> months to a year, depending on feedback from you and the development of
> Numeric/Numarray.
> 
> -Jonathan Brandmeyer
> 
> P.S.  The source package for Linux and OSX will be released soon (to be
> versioned 3.1.0).  I am working out a few kinks in the configuration and
> build control files.
> 
> 
> 
> -------------------------------------------------------
> SF email is sponsored by - The IT Product Guide
> Read honest & candid reviews on hundreds of IT Products from real users.
> Discover which products truly live up to the hype. Start reading now. 
> http://productguide.itmanagersjournal.com/
> _______________________________________________
> Visualpython-users mailing list
> Vis...@li...
> https://lists.sourceforge.net/lists/listinfo/visualpython-users

[Fwd: Re: [Visualpython-users] Experimental Windows version for Python 2.4]

[Jonathan B. <jbr...@ea...>]

This was intended to go the m-l.

-------- Forwarded Message --------
> From: Bruce Sherwood 
> To: Jonathan Brandmeyer 
> Subject: Re: [Visualpython-users] Experimental Windows version for
> Python 2.4
> Date: Wed, 29 Dec 2004 00:33:37 -0500
> The Windows version available at vpython.org does not at the moment 
> include/install Numarray, just Numeric. But if you install Numarray 
> yourself, the behavior will be as Jonathan describes and has implemented.
> 
> Bruce Sherwood

Re: [Visualpython-users] Experimental Windows version for Python 2.4

[Jonathan B. <jbr...@ea...>]

On Tue, 2004-12-28 at 11:48 -0500, Bruce Sherwood wrote:
> Now available at vpython.org is an experimental version of VPython for 
> Python 2.4 on Windows. The only known bug is that programs that use the 
> "text" module to display block letters (including the demo programs 
> stonehenge.py and texttest.py) fail on some but not all computers. We 
> would appreciate feedback on this or other bugs you may find. This was 
> built using the new Boost-based package created by Jonathan Brandmeyer.
> 


There is one additional change that this release has versus prior
releases that used Boost.Python for its bindings.  This build supports
and includes both Numeric 23.6 and Numarray 1.1.1.

By default, executing "from visual import *" will bring in Numeric, and
not Numarray, and will use Numeric arrays to hold the data of curves,
convex, and faces objects.  You can alter this behavior by calling the
function set_array_backend() with either the string "numeric" or
"numarray".  To be safe, you should only do this once at the beginning
of your program, before creating any array-using objects.  If you are
forcing the use of Numarray in this way, you must also run "from
numarray import *".  It is not safe to change the backend after you have
created a visual object that uses an array (even if its pos, color, or
normal attribute appears to be empty).

Alternatively, you can edit site-packages/visual/__init__.py (lines
16-23) to change the preference for all applications on your particular
installation.

I intend to provide support for Numeric as long as it is supported
upstream and not too much of a burden on Visual.  Numarray has very good
backwards compatibility with Numeric, and Visual's core works exactly
the same with both.  However, there are some differences which you may
see in your own programs (some of our demos are also affected).  So,
Visual will prefer Numeric by default rather than Numarray, for maybe 6
months to a year, depending on feedback from you and the development of
Numeric/Numarray.

-Jonathan Brandmeyer

P.S.  The source package for Linux and OSX will be released soon (to be
versioned 3.1.0).  I am working out a few kinks in the configuration and
build control files.