visualpython-users Mailing List for VPython (Page 158)

At http://www.physics.colostate.edu/users/gelfand/VPcompiled.tgz
you'll find Visual Python compiled for Woody/i386.  Move it
into /usr/lib/python2.2/site-packages and then tar xzf it.
(Maybe it should/could go into /usr/local/lib/python2.2/site-packages.)
If you get errors on importing visual then you'll know you're
missing some libraries.

Regards,
Martin Gelfand
Dept of Physics, Colorado State University

On Mon, 12 May 2003 ima...@gm... wrote:

> Hi everyone,=20
>=20
> as a search on this list indicated, there are some people=20
> who have installed VPython under Debian Woody=20
> which I also use.=20
> But opposite to some previous posters who found it=20
> to be quite easy I am a bit stuck, since ./configure=20
> complains about missing gtkglarea which=20
> I am sure is installed.=20
> (At least everything dselect shows about gtkglarea)
> In fact I installed every package listed in a=20
> previous post on this topic.=20
>=20
> I cannot append the output of ./configure ...=20
> right now since I'm not at home, but=20
> should it be necessary I can send it later.=20
>=20
> Bye
>=20
> --=20
> +++ GMX - Mail, Messaging & more  http://www.gmx.net +++
> Bitte l=E4cheln! Fotogalerie online mit GMX ohne eigene Homepage!
>=20
>=20
>=20
> -------------------------------------------------------
> Enterprise Linux Forum Conference & Expo, June 4-6, 2003, Santa Clara
> The only event dedicated to issues related to Linux enterprise solutions
> www.enterpriselinuxforum.com
>=20
> _______________________________________________
> Visualpython-users mailing list
> Vis...@li...
> https://lists.sourceforge.net/lists/listinfo/visualpython-users
>=20

Hi everyone, 

as a search on this list indicated, there are some people 
who have installed VPython under Debian Woody 
which I also use. 
But opposite to some previous posters who found it 
to be quite easy I am a bit stuck, since ./configure 
complains about missing gtkglarea which 
I am sure is installed. 
(At least everything dselect shows about gtkglarea)
In fact I installed every package listed in a 
previous post on this topic. 

I cannot append the output of ./configure ... 
right now since I'm not at home, but 
should it be necessary I can send it later. 

Bye

-- 
+++ GMX - Mail, Messaging & more  http://www.gmx.net +++
Bitte lächeln! Fotogalerie online mit GMX ohne eigene Homepage!

>
>
>hardware. An einai dynaton, na perimeneis gia na kanei boot to
kompiouter.
>8a eprepe na einai san to video, na patas ena koumpi kai na einai etoimo.
>(Shmeiwsh: o Amstrad CPC6128 kanei boot se 1sec. se auto 8a gyrisoume?)
>
>
E egw giati 3erw pws ta PC mporoun na mpoutaroun se 1sec;;;;; Mhpws
giati 3erw ti skata exoun balei mesa sto BIOS;;;; h mhpws giati 3erw gia
to linux bios me ton kernoula mesa sto BIOS;;;;;;
Ma epitelous ti skata exoun sto kefali tous aytoi ekei sthn microsoft
mallon prepei na alla3oun onoma se microbrain PC fuck up solutions :)
Ma kala einai toso amerikaniko ayto pou 8eloun na kanoun pou
adeiazw...prwta dhmiourgoume to problhma kai meta poulame thn ,lysh :)

On Sat, 2003-05-10 at 16:21, Gary Ruben wrote:
> Hi fellow VPython Users,
> A few times in my code I have wanted to show a trail behind some moving object.
> For example, you might want to see the path of the free end of the
> double pendulum in the doublependulum.py example. To do this, you could add the
> following code:
> 
> <code>
> trailLength = 10000 # no of points in trail
> trail = curve(color=(0.5,0.5,0.5))
> 
> # the existing while loop
> while 1:
> 	.
> 	. other code here
> 	.
>     posnew = frame2.pos+frame2.axis*L2
>     trail.append(pos=posnew)
> 	# Now keep the trail length finite
>     if len(trail.pos) > trailLength:
>         trail.pos = trail.pos[20:]
> 
>     t = t+dt
> 
> </code>
> 
> However, I have encountered a problem with doing this which I assume is to do
> with the curve being held in a Numeric array structure. If I understand Numeric
> correctly, trail.pos = trail.pos[20:] won't cause a copy or truncation, just a
> change to the offset to the start of the array, so the array/curve object will
> continue to grow despite only showing the last 10000 points (in this example).
> If a copy could be forced, the curve object would have to be reassociated with
> the new array object, so I don't think a copy is occurring. A copy would be
> very inefficient anyway.
> Actually, when I run this example, everything seems to work OK, but in another
> example I have, if I leave it running, the vpython window eventually locks up
> and the lockup is associated with the code I have shown. I'm running Win98.
> So, has anyone got a more efficient way to limit the length of a curve to some
> maximum? Do others ever do this sort of thing and if so, do they agree that it
> would be nice to have a property such as maxlength added to the curve object?
> 
> Gary

Under the current implementation the curve grows without bound, although
a maximum of 1024 points are shown on the screen. Once this limit is
exceeded only every N point is shown, N being (length/1024)+1
The memory model btw is that once the array limit of points is reached,
a new array twice the size is created and all the points are copied to
it. A maxlength property might be a solution but once you have reached
the limit you would take a performance hit as you would have to loop
though the points making point[N] = point[N+1] before appending the new
point. But then again this may be a solution.

As part of my Masters project I have created a new VPython object called
<points> These are sets of discrete points rendered as points on the
screen and are intended for use in the simulation of particle systems. A
problem I encountered during the implementation of this shape is similar
to yours. My problem is "What do with unnecesary points?". The obvious
answer is to delete them, but the points model was based upon the curve
object (without the connecting lines) and therefore didn't have a delete
method. So I added one. When a point is deleted I do what I described
above eg Loop from the deletion point to the end of the set replacing
the position and colour of point[N] with point[N+1], and modifying the
length attribute as well. It doesn't cause the same performance hit, as
points are appened and deleted at different time and rates, but could be
used to simulate exactly what you want to do eg

    posnew = frame2.pos+frame2.axis*L2
    if (len(trail.pos) == trailLength):
	trail.delete(0)
    trail.append(pos=posnew)

Hi,

 I'm a newbie at VPython and could use a hand.
 I have managed to compile VPython in a modified Mandrake 8.2
 using gcc-2.96. When I try to run the demos I get:

 ImportError: /usr/local/lib/python2.2/site-packages/cvisualmodule.so: undefined symbol: __ti3ios

 I do know that it menas that cvisualmodule.so was loaded but it refers to
 the symbol __ti3ios which is missing.
 
 Running ldd cvisualmodule.so I get:
  libstdc++.so.5 => /usr/local/lib/./libstdc++.so.5 (0x4034c000)
        libgtkgl.so.5 => /usr/lib/libgtkgl.so.5 (0x40417000)
        libGLU.so.1 => /usr/local/lib/./libGLU.so.1 (0x4041d000)
        libGL.so.1 => /usr/local/lib/./libGL.so.1 (0x404a7000)
        libgtk-1.2.so.0 => /usr/lib/libgtk-1.2.so.0 (0x40681000)
        libgdk-1.2.so.0 => /usr/lib/libgdk-1.2.so.0 (0x407c4000)
        libXi.so.6 => /usr/X11R6/lib/libXi.so.6 (0x407fd000)
        libXext.so.6 => /usr/X11R6/lib/libXext.so.6 (0x40805000)
        libX11.so.6 => /usr/X11R6/lib/libX11.so.6 (0x40815000)
        libm.so.6 => /lib/libm.so.6 (0x408dc000)
        libgthread-1.2.so.0 => /usr/lib/libgthread-1.2.so.0 (0x408fe000)
        libpthread.so.0 => /lib/libpthread.so.0 (0x40901000)
        libglib-1.2.so.0 => /usr/lib/libglib-1.2.so.0 (0x40917000)
        libc.so.6 => /lib/libc.so.6 (0x4093d000)
        libgcc_s.so.1 => /usr/local/lib/./libgcc_s.so.1 (0x40a7a000)
        libgmodule-1.2.so.0 => /usr/lib/libgmodule-1.2.so.0 (0x40a82000)
        libdl.so.2 => /lib/libdl.so.2 (0x40a85000)
        libSM.so.6 => /usr/X11R6/lib/libSM.so.6 (0x40a88000)
        libICE.so.6 => /usr/X11R6/lib/libICE.so.6 (0x40a91000)
        libXmu.so.6 => /usr/X11R6/lib/libXmu.so.6 (0x40aa8000)
        /lib/ld-linux.so.2 => /lib/ld-linux.so.2 (0x80000000)
        libXt.so.6 => /usr/X11R6/lib/libXt.so.6 (0x40ac0000)


 Does anybody know who's to blame and what can I do to fix it or
 at least to investigate it further.

 Please carbon-copy your answers to rs...@ac... because I'm not yet a
 memeber of this mailing-list.

 Thanks in advance
 Rod Senra

-- 
 Rodrigo Senra       <rs...@ac...>

 GPr Sistemas http://www.gpr.com.br
 PUC-Campinas http://docentes.puc-campinas.edu.br/ceatec/rodrigo
 IC - Unicamp http://www.ic.unicamp.br/~921234

Has anyone tried using VPython with the MacPython distribution found at
<http://homepages.cwi.nl/~jack/macpython.html> ?

I'd be interested in hearing about your experiences.

Cheers,
Joe Heafner

-----
The ONLY way to stop computer viruses is to STOP using Microsoft 
products, openly known as viral petri dishes. Why don't you all get 
this?

Hi fellow VPython Users,
A few times in my code I have wanted to show a trail behind some moving object.
For example, you might want to see the path of the free end of the
double pendulum in the doublependulum.py example. To do this, you could add the
following code:

<code>
trailLength = 10000 # no of points in trail
trail = curve(color=(0.5,0.5,0.5))

# the existing while loop
while 1:
	.
	. other code here
	.
    posnew = frame2.pos+frame2.axis*L2
    trail.append(pos=posnew)
	# Now keep the trail length finite
    if len(trail.pos) > trailLength:
        trail.pos = trail.pos[20:]

    t = t+dt

</code>

However, I have encountered a problem with doing this which I assume is to do
with the curve being held in a Numeric array structure. If I understand Numeric
correctly, trail.pos = trail.pos[20:] won't cause a copy or truncation, just a
change to the offset to the start of the array, so the array/curve object will
continue to grow despite only showing the last 10000 points (in this example).
If a copy could be forced, the curve object would have to be reassociated with
the new array object, so I don't think a copy is occurring. A copy would be
very inefficient anyway.
Actually, when I run this example, everything seems to work OK, but in another
example I have, if I leave it running, the vpython window eventually locks up
and the lockup is associated with the code I have shown. I'm running Win98.
So, has anyone got a more efficient way to limit the length of a curve to some
maximum? Do others ever do this sort of thing and if so, do they agree that it
would be nice to have a property such as maxlength added to the curve object?

Gary
-- 
_______________________________________________
Sign-up for your own FREE Personalized E-mail at Mail.com
http://www.mail.com/?sr=signup

As to the F1-help, that is probably an internal configuration problem
for the fink - installed python.  Maybe it expects Konqueror/Mozilla?

It seems mine also bombs out on Find and Replace, but with different
errors.

I'm working on a new makefile that works around a limitation of libtool
on OSX.  It seems that the libtool paradigm is that shared libraries
cannot link against static ones, EVEN IF THE PLATFORM SUPPORTS IT.  A
number of people have griped about this to the libtool dev team, to
which the responce seems to be, "Tough, that's our paradigm."  So, a
workaround is in progress.

Since more than one vendor uses an exclusively static libstdc++, libtool
is effectively broken on them for C++ work.  (MinGW on Win32, ppc-darwin
and some(all?) ppc-BSD are affected).

I still haven't knocked out the linker errors for the configure stage on
our OSX box.  The fink-users guys said it was an OS-version, X-version,
Dev-tools-version mismatch problem, but we haven't found the magic combo
yet.

IIRC, the new distro includes a Makefile.MaxOSX or somesuch in the
cvisual directory, which dates to the old installer.  If configure
passes on your system, you can try copying Makefile.MacOSX to Makefile
and running make in the cvisual directory.  I'm integrating portions of
that Makefile into a non-automake version of Makefile.in, so the basic
rules will hopefully work.  You may need to edit some configured
variables at the top of the file.

Final exams are coming over the next week, so I might not be able to
help troubleshoot right away.  By all means, fill out bug-reports and
patches to the Sourceforge project site, and I'll get to them soon.

Jonathan Brandmeyer

On Thu, 2003-05-01 at 22:45, Joe Heafner wrote:
> I'm using the 2002-07-22 version of Visual under Mac OS 10.2.5. I've 
> installed Python 2.2.2 and necessary libs via Fink.
> 
> Inside IDLEfork, when I hit F1 to bring up help, I get the following:
>      raise Error("could not locate runnable browser")
> Error: could not locate runnable browser
> 
> When I try any of the Find or Replace functions from the Edit menu, 
> they all bomb. For example, Find gives the following:

--snipped traceback--
> 
> Replace gives:

--snipped traceback--

> 
> Still no luck compiling Visual.
> 
> Cheers,
> Joe Heafner
> 
> -----
> I don't have a Lexus, but I have a Mac. Same thing.
> 
> 

I'm using the 2002-07-22 version of Visual under Mac OS 10.2.5. I've 
installed Python 2.2.2 and necessary libs via Fink.

Inside IDLEfork, when I hit F1 to bring up help, I get the following:

Traceback (most recent call last):
   File 
"/sw/src/root-python22-2.2.2-5/sw/lib/python2.2/lib-tk/Tkinter.py", 
line 1300, in __call__
     return apply(self.func, args)
   File "/sw/lib/python2.2/Tools/idle_VPython/EditorWindow.py", line 
329, in python_docs
     webbrowser.open(self.help_url)
   File "/sw/src/root-python22-2.2.2-5/sw/lib/python2.2/webbrowser.py", 
line 43, in open
     get().open(url, new, autoraise)
   File "/sw/src/root-python22-2.2.2-5/sw/lib/python2.2/webbrowser.py", 
line 38, in get
     raise Error("could not locate runnable browser")
Error: could not locate runnable browser

When I try any of the Find or Replace functions from the Edit menu, 
they all bomb. For example, Find gives the following:

Traceback (most recent call last):
   File 
"/sw/src/root-python22-2.2.2-5/sw/lib/python2.2/lib-tk/Tkinter.py", 
line 1300, in __call__
     return apply(self.func, args)
   File "/sw/lib/python2.2/Tools/idle_VPython/SearchBinding.py", line 
63, in find_event
     SearchDialog.find(self.editwin.text)
   File "/sw/lib/python2.2/Tools/idle_VPython/SearchDialog.py", line 14, 
in find
     return _setup(text).open(text)
   File "/sw/lib/python2.2/Tools/idle_VPython/SearchDialogBase.py", line 
22, in open
     self.ent.focus_set()
AttributeError: SearchDialog instance has no attribute 'ent'

Replace gives:

Traceback (most recent call last):
   File 
"/sw/src/root-python22-2.2.2-5/sw/lib/python2.2/lib-tk/Tkinter.py", 
line 1300, in __call__
     return apply(self.func, args)
   File "/sw/lib/python2.2/Tools/idle_VPython/SearchBinding.py", line 
83, in replace_event
     ReplaceDialog.replace(self.editwin.text)
   File "/sw/lib/python2.2/Tools/idle_VPython/ReplaceDialog.py", line 
16, in replace
     dialog.open(text)
   File "/sw/lib/python2.2/Tools/idle_VPython/ReplaceDialog.py", line 
28, in open
     SearchDialogBase.open(self, text)
   File "/sw/lib/python2.2/Tools/idle_VPython/SearchDialogBase.py", line 
22, in open
     self.ent.focus_set()
AttributeError: ReplaceDialog instance has no attribute 'ent'

Still no luck compiling Visual.

Cheers,
Joe Heafner

-----
I don't have a Lexus, but I have a Mac. Same thing.

Thank you Bruce,

This was exactly what I was looking for. I guess the documentation does not
really explain the "up" attribute/method, or I may have mist it.

Regards, and again many thanks for the very rapid response.

Claude Brunet

-----Original Message-----
From: Bruce Sherwood [mailto:bas...@un...]
Sent: Tuesday, April 29, 2003 8:20 PM
To: Brunet, Claude; vis...@li...
Subject: Re: [Visualpython-users] how do we access the state of an
object rotated along its own axi s?


The short answer is to look at my_frame.up.

A longer answer involves me being confused about an aspect of Visual 3D
graphics manipulations involving the scene.up direction, typically (0,1,0).

In your program you create your frame with its objects with frame axis along
(1,0,0), with the red arrow pointing in the (0,1,0) direction. Next you set
my_frame.axis = (0,1,0), and your red arrow now points in the (1,0,0)
direction, which seems a bit arbitrary to me. At this point if you print
my_frame.up it has the surprising value (0,1,0), just as it did before the
rotate, which seems odd to me. If you then make a small rotation about
(0,1,0), even a zero rotation, my_frame.up suddenly becomes (1,0,0)!
Thereafter my_frame.up rotates as you rotate the frame.

If on the other hand after creating the objects you execute
my_shape.rotate(axis=(0,0,1), angle=pi/2.), the red arrow points in the
(-1,0,0) direction and my_frame.up is also (-1,0,0). This seems more
understandable.

Another way of seeing the strangeness is that after my_frame.axis = (0,1,0)
your red arrow points to the right, but if you execute
my_frame.axis(0.01,1,0), your red arrow points to the left. My head aches.

Bruce Sherwood

----- Original Message ----- 
From: "Brunet, Claude" <Cla...@sp...>
To: <vis...@li...>
Sent: Tuesday, April 29, 2003 4:17 PM
Subject: [Visualpython-users] how do we access the state of an object
rotated along its own axi s?


> VPython group,
>
> I just started using VPython to test some ideas for a 3D project. Found it
> is a great environment to work with. I stumble accross the following
> problem:
>
> If one creates a frame and then rotates this object arround its own axis,
> how can we access the rotation state of that object since neither the
> ".axis" nor the ".pos" methods provide any usefull information? Obviously
> the state of the object is logged within VPython's internal since
successive
> rotate operations perform as expected.
>
> In the little example below, how can I get (which method) the actual
> direction of the arrow "pointer_1" (red) after any rotation operation to
> my_shape?
>
> Many thanks,
>
> Claude Brunet
> Manager Software & Ground Segment
> Canadian Space Agency
> tel  : (450) 926-4462
> Fax: (450) 926-4449
>
> # little example of my problem
>
> from visual import *
> scene.exit = 0
> scene.title = "rotation test"
> scene.height = 600
> scene.width = 600
> scene.range = (18,18,18)
> scene.background = color.white
> scene.center = (0,2,0)
> scene.visible=1
>
> for obj in scene.objects:
>    obj.visible=0
>
> # show the reference axis
> arrow(pos=(0,0,0), axis=(10,0,0), shaftwidth=.1)
> arrow(pos=(0,0,0), axis=(0,10,0), shaftwidth=.1)
> arrow(pos=(0,0,0), axis=(0,0,10), shaftwidth=.1)
>
> #define a frame object my_shape
>
> my_shape = frame(pos=(0.,0.,0.), axis=(1.,0.,0.))
> box_main=box(frame=my_shape,
>              pos=(0.,0.,0.), axis=(1.,0.,0.),
>              length=10., height=1., width=1., color=color.yellow)
> box(frame=my_shape,
>              pos=(4.5,1.,0.), axis=(1.,0.,0.),
>              length=1., height=1., width=1., color=color.green)
> pointer_1=arrow(frame=my_shape,pos=(5.1,-.5,0),axis=(0,3,0))
> pointer_1.color=(1,0,0)
>
> # align it along the Y axis
>
> my_shape.pos=(0,0,0)
> my_shape.axis=vector(0,1,0)
>
> twopi_over_360=2*3.1416/360.
>
> # rotate the frame object alomg the Y axis
> for i in range(0,360):
>    my_shape.rotate(angle=(twopi_over_360))
>    rate(30)
>
> # How can I know (which method to use) where the pointer_1 (red arrow)
> points after any rotation operation?
>
>
>
>
>
>
> -------------------------------------------------------
> This sf.net email is sponsored by:ThinkGeek
> Welcome to geek heaven.
> http://thinkgeek.com/sf
> _______________________________________________
> Visualpython-users mailing list
> Vis...@li...
> https://lists.sourceforge.net/lists/listinfo/visualpython-users
>

The short answer is to look at my_frame.up.

A longer answer involves me being confused about an aspect of Visual 3D
graphics manipulations involving the scene.up direction, typically (0,1,0).

In your program you create your frame with its objects with frame axis along
(1,0,0), with the red arrow pointing in the (0,1,0) direction. Next you set
my_frame.axis = (0,1,0), and your red arrow now points in the (1,0,0)
direction, which seems a bit arbitrary to me. At this point if you print
my_frame.up it has the surprising value (0,1,0), just as it did before the
rotate, which seems odd to me. If you then make a small rotation about
(0,1,0), even a zero rotation, my_frame.up suddenly becomes (1,0,0)!
Thereafter my_frame.up rotates as you rotate the frame.

If on the other hand after creating the objects you execute
my_shape.rotate(axis=(0,0,1), angle=pi/2.), the red arrow points in the
(-1,0,0) direction and my_frame.up is also (-1,0,0). This seems more
understandable.

Another way of seeing the strangeness is that after my_frame.axis = (0,1,0)
your red arrow points to the right, but if you execute
my_frame.axis(0.01,1,0), your red arrow points to the left. My head aches.

Bruce Sherwood

----- Original Message ----- 
From: "Brunet, Claude" <Cla...@sp...>
To: <vis...@li...>
Sent: Tuesday, April 29, 2003 4:17 PM
Subject: [Visualpython-users] how do we access the state of an object
rotated along its own axi s?


> VPython group,
>
> I just started using VPython to test some ideas for a 3D project. Found it
> is a great environment to work with. I stumble accross the following
> problem:
>
> If one creates a frame and then rotates this object arround its own axis,
> how can we access the rotation state of that object since neither the
> ".axis" nor the ".pos" methods provide any usefull information? Obviously
> the state of the object is logged within VPython's internal since
successive
> rotate operations perform as expected.
>
> In the little example below, how can I get (which method) the actual
> direction of the arrow "pointer_1" (red) after any rotation operation to
> my_shape?
>
> Many thanks,
>
> Claude Brunet
> Manager Software & Ground Segment
> Canadian Space Agency
> tel  : (450) 926-4462
> Fax: (450) 926-4449
>
> # little example of my problem
>
> from visual import *
> scene.exit = 0
> scene.title = "rotation test"
> scene.height = 600
> scene.width = 600
> scene.range = (18,18,18)
> scene.background = color.white
> scene.center = (0,2,0)
> scene.visible=1
>
> for obj in scene.objects:
>    obj.visible=0
>
> # show the reference axis
> arrow(pos=(0,0,0), axis=(10,0,0), shaftwidth=.1)
> arrow(pos=(0,0,0), axis=(0,10,0), shaftwidth=.1)
> arrow(pos=(0,0,0), axis=(0,0,10), shaftwidth=.1)
>
> #define a frame object my_shape
>
> my_shape = frame(pos=(0.,0.,0.), axis=(1.,0.,0.))
> box_main=box(frame=my_shape,
>              pos=(0.,0.,0.), axis=(1.,0.,0.),
>              length=10., height=1., width=1., color=color.yellow)
> box(frame=my_shape,
>              pos=(4.5,1.,0.), axis=(1.,0.,0.),
>              length=1., height=1., width=1., color=color.green)
> pointer_1=arrow(frame=my_shape,pos=(5.1,-.5,0),axis=(0,3,0))
> pointer_1.color=(1,0,0)
>
> # align it along the Y axis
>
> my_shape.pos=(0,0,0)
> my_shape.axis=vector(0,1,0)
>
> twopi_over_360=2*3.1416/360.
>
> # rotate the frame object alomg the Y axis
> for i in range(0,360):
>    my_shape.rotate(angle=(twopi_over_360))
>    rate(30)
>
> # How can I know (which method to use) where the pointer_1 (red arrow)
> points after any rotation operation?
>
>
>
>
>
>
> -------------------------------------------------------
> This sf.net email is sponsored by:ThinkGeek
> Welcome to geek heaven.
> http://thinkgeek.com/sf
> _______________________________________________
> Visualpython-users mailing list
> Vis...@li...
> https://lists.sourceforge.net/lists/listinfo/visualpython-users
>

VPython group,

I just started using VPython to test some ideas for a 3D project. Found it
is a great environment to work with. I stumble accross the following
problem:

If one creates a frame and then rotates this object arround its own axis,
how can we access the rotation state of that object since neither the
".axis" nor the ".pos" methods provide any usefull information? Obviously
the state of the object is logged within VPython's internal since successive
rotate operations perform as expected. 

In the little example below, how can I get (which method) the actual
direction of the arrow "pointer_1" (red) after any rotation operation to
my_shape?

Many thanks,

Claude Brunet
Manager Software & Ground Segment
Canadian Space Agency
tel  : (450) 926-4462
Fax: (450) 926-4449

# little example of my problem

from visual import *
scene.exit = 0         
scene.title = "rotation test"
scene.height = 600
scene.width = 600
scene.range = (18,18,18)
scene.background = color.white
scene.center = (0,2,0)
scene.visible=1

for obj in scene.objects:
   obj.visible=0

# show the reference axis
arrow(pos=(0,0,0), axis=(10,0,0), shaftwidth=.1)
arrow(pos=(0,0,0), axis=(0,10,0), shaftwidth=.1)
arrow(pos=(0,0,0), axis=(0,0,10), shaftwidth=.1)

#define a frame object my_shape

my_shape = frame(pos=(0.,0.,0.), axis=(1.,0.,0.))
box_main=box(frame=my_shape,
             pos=(0.,0.,0.), axis=(1.,0.,0.),
             length=10., height=1., width=1., color=color.yellow)
box(frame=my_shape,
             pos=(4.5,1.,0.), axis=(1.,0.,0.),
             length=1., height=1., width=1., color=color.green)
pointer_1=arrow(frame=my_shape,pos=(5.1,-.5,0),axis=(0,3,0))
pointer_1.color=(1,0,0)

# align it along the Y axis

my_shape.pos=(0,0,0)
my_shape.axis=vector(0,1,0)

twopi_over_360=2*3.1416/360.

# rotate the frame object alomg the Y axis
for i in range(0,360):
   my_shape.rotate(angle=(twopi_over_360))
   rate(30)

# How can I know (which method to use) where the pointer_1 (red arrow)
points after any rotation operation?

I have narrowed it down to what seems to be a problem with the version of
Numeric you run. On the machine I am having trouble with, it wants to find a
dotblas module that is imported, if necessary, in Numeric.py. And, py2exe is
correct in that it cannot be found anywhere. When I used py2exe with an
older Numeric install, it created the executable just fine. I'll inform
everyone of the solution when I find it, and if anyone recognizes what I am
talking about, please reply.

Thanks Arthur,
Thom

-----Original Message-----
From: Arthur [mailto:ajs...@op...]
Sent: Saturday, April 26, 2003 5:08 AM
To: Visual Python Users (E-mail)
Subject: Re: [Visualpython-users] VPython with py2exe




Arthur wrote:

> Was it necessary to change __init__.py to vpython.py?

That would be visual.py
                        ^^^^^




-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
_______________________________________________
Visualpython-users mailing list
Vis...@li...
https://lists.sourceforge.net/lists/listinfo/visualpython-users

Arthur wrote:

> Was it necessary to change __init__.py to vpython.py?

That would be visual.py
                        ^^^^^

IVES,THOM (HP-Boise,ex1) wrote:

> VPython-ers,
>  
> Are there any known problems using py2exe with VPython? I seem to be 
> having them. I am searching the web trying to find references to any

I have used it successfully.  Do not remember if I had to do some 
workaround stuff.

Was it necessary to change __init__.py to vpython.py?

Ran into the need to do that somewhere in my journeys.

At


http://www.dstoys.com/content/education/index_html/Visual%20Arts/Interactive


you will the VPython Tower of Hanoi demo as a Windows exe.

Looking inside it may give you a clue.

What is it you are running into?

Art

VPython-ers,
 
Are there any known problems using py2exe with VPython? I seem to be having
them. I am searching the web trying to find references to any.
 
Thanks,
Thom

 
 
  _____  

 
<http://www.hp.com/Redirect/gw/useng_welcome/logo/=http://welcome.hp.com/cou
ntry/us/eng/welcome.htm> HP logo	 
Thom Ives, Ph.D.

R&D Engineering Scientist	

11413 Chinden Blvd MS 400
Boise, ID.  83714

208.396.6880 Phone
208.396.3587 Fax
tho...@hp...  <mailto:tho...@hp...> 

Home:
5556 N Columbine Pl
Boise, ID 83713
208.412.3971 Phone
413.332.6558 Fax
tw...@ca... <mailto:tw...@ca...> 
  _____  

 
 

VPython Group,

Sorry to raise such a fuss. I forgot that our company is automatically
installing updates for OS fixes now, and after taking Bruce's advice (I had
only reinstalled when the problems first occurred - then, after Bruce's
e-mail, I uninstalled everything and reinstalled), everything worked fine. 

Thanks Bruce and Andy,
Thom

       |-----Original Message-----
       |From: Bruce Sherwood [mailto:bas...@un...]
       |Sent: Wednesday, April 23, 2003 2:45 PM
       |To: IVES,THOM (HP-Boise,ex1); Visual Python Users (E-mail)
       |Subject: Re: [Visualpython-users] Major Problems with 
       |Latest Windows
       |Release of VPython
       |
       |
       |I installed the latest VPython for Windows (prints 
       |2003-03-15 when you run),
       |and I can't reproduce any of these problems. All the 
       |demos and test routines
       |run fine. This is on Windows 2000.
       |
       |At first your very last test showed nothing, but that 
       |was simply because
       |there was a typo in your faces position list:
       |
       |VertexList = [(0,0,0),(1,0,0),(1,0,0)]
       |should be
       |VertexList = [(0,0,0),(1,0,0),(1,1,0)]
       |
       |Unless you've already tried it, the first thing to do 
       |might be to uninstall
       |Visual (and maybe even Python) and reinstall. It sounds 
       |like something is
       |corrupted on your machine, or in the download.
       |
       |Bruce Sherwood
       |
       |----- Original Message -----
       |From: "IVES,THOM (HP-Boise,ex1)" <tho...@hp...>
       |To: "Visual Python Users (E-mail)"
       |<vis...@li...>
       |Sent: Wednesday, April 23, 2003 3:46 PM
       |Subject: [Visualpython-users] Major Problems with 
       |Latest Windows Release of
       |VPython
       |
       |
       |> This is a BUG REPORT. I found these bugs running the 
       |Windows 2000 Pro OS.
       |> Please be patient as you read on.
       |>
       |> I was trying to get rid of some bugs in a new piece 
       |of code I am writing,
       |> which relies heavily on VPython. It was maturing 
       |nicely, but I was running
       |> into some problems with the scene.exit = 0 attributes 
       |of display (see here
       |> http://vpython.org/webdoc/visual/display.html
       |> <http://vpython.org/webdoc/visual/display.html> ), so 
       |I decided to try and
       |> upgrade my VPython, and things stopped working 
       |altogether, introducing
       |> problems that I have never experienced before. 
       |Thinking the problem might
       |be
       |> with me, I went back to some of my simplest 
       |applications and the demo
       |> programs and still noted problems. One of these 
       |aspects can be represented
       |> with the following small program modified from the 
       |cross product.py demo
       |> program below. If you run this code, it should 
       |produce a grid for the
       |plane
       |> that illustrates the visual cross product, but it 
       |does not. Please read
       |> further below this small program.
       |>
       |> ________________________ Test Program 1 - Modification of the
       |> CrossProduct.py Demo________________________
       |> from visual import *
       |> # demonstration of vector cross product
       |>
       |> print """
       |> Vector cross product: Red cross Green = Yellow
       |> Drag to change green vector
       |> Click to toggle fixed angle or fixed length
       |> """
       |>
       |> scene.title="Vector Cross Product"
       |> scene.width=600
       |> scene.height=600
       |> R = 0.15*4
       |> plane = curve(pos=[(0, -10, -10), (0, -10, 10), (0, 
       |10,10), (0, 10, -10),
       |>               (0, -10, -10), (0,-6,-10), (0,-6,10), 
       |(0, -2, 10), (0, -2,
       |> -10),
       |>               (0,2,-10), (0,2,10), (0,6,10), 
       |(0,6,-10),(0,10, -10),
       |>
       |>               (0,10,-6), (0,-10,-6), (0,-10,-2), 
       |(0,10,-2), (0,10,2),
       |>               (0,-10,2), (0,-10,6), (0,10,6)])
       |> ___________________________________End of Test Program
       |> 1__________________________________________
       |>
       |> So, I performed some other tests. This next small 
       |program is a slight
       |> modification of the graphtest.py demo. Try running 
       |it. Note that no axes
       |are
       |> drawn and that you get an error. If you raise the 
       |value of -70 to say
       |about
       |> -50, or lower the 70 to something lower, the errors 
       |will go away, but you
       |> still get a problem with no axes being drawn. In 
       |fact, with regard to the
       |> axes problem, even the simplest program where you try 
       |to use the curve
       |> object will not draw a curve on the screen in 3D.
       |>
       |> _________________________________ Modification of 
       |graphtest.py demo
       |> __________________________________
       |> from visual.graph import *
       |>
       |> # Using a graph-plotting module
       |> oscillation = gdisplay(xtitle='t', ytitle='Response')
       |>
       |> list = []
       |> for t in arange(-70, 70, 1):
       |>     pos=(t, 5.0*cos(-0.2*t)) # )
       |>     list.append(pos)
       |>
       |> funct1 = gcurve(pos = list, color=color.cyan)
       |>
       |> ________________________________ End of Test Program 2
       |> _____________________________________________
       |>
       |> Finally, I tried a simple test with the last feature 
       |I was struggling
       |with.
       |> First, if I try to run the faces_heightfield.py demo, 
       |it will generate an
       |> error and not draw anything. Second, when I run the 
       |faces_cone.py, I do
       |not
       |> get a cone. I get a cube, and the following message...
       |>
       |>
       |> Traceback (innermost last)
       |>   File 
       |"c:\python22\lib\site-packages\visual\demos\faces_cone.py", line
       |13,
       |> in ?
       |>     model = faces( pos = zeros( (N*3,3), Float ), frame = f )
       |> TypeError: Array can not be safely cast to required type
       |>
       |> Third, if I try to run the most simple face possible, 
       |a triangle, I get
       |> nothing drawn at all. Try the program below...
       |>
       |> ________________________________ Simple program to 
       |draw a Triangle Face
       |> _____________________________________
       |> from visual import *
       |>
       |> f = frame()
       |> #scene.background = color.black
       |> VertexList = [(0,0,0),(1,0,0),(1,0,0)]
       |> NormalList = [(0,0,1),(1,0,1),(1,0,1)]
       |>
       |> surface = faces(pos=VertexList, normal=NormalList, 
       |color=color.red)
       |> _________________________________ End of Test Program 3
       |> ____________________________________________________
       |>
       |> I think a download of the previous version of Visual 
       |should be made
       |> available until these problems can be fixed.
       |>
       |> Thom
       |>
       |>
       |>
       |>
       |>
       |>   _____
       |>
       |>
       |>
       |<http://www.hp.com/Redirect/gw/useng_welcome/logo/=http:
       |//welcome.hp.com/cou
       |> ntry/us/eng/welcome.htm> HP logo
       |> Thom Ives, Ph.D.
       |>
       |> R&D Engineering Scientist
       |>
       |> 11413 Chinden Blvd MS 400
       |> Boise, ID.  83714
       |>
       |> 208.396.6880 Phone
       |> 208.396.3587 Fax
       |> tho...@hp...  <mailto:tho...@hp...>
       |>
       |> Home:
       |> 5556 N Columbine Pl
       |> Boise, ID 83713
       |> 208.412.3971 Phone
       |> 413.332.6558 Fax
       |> tw...@ca... <mailto:tw...@ca...>
       |>   _____
       |>
       |>
       |>
       |>
       |

On Wed, 23 Apr 2003, Bruce Sherwood wrote:

> I installed the latest VPython for Windows (prints 2003-03-15 when you run),
> and I can't reproduce any of these problems. All the demos and test routines
> run fine. This is on Windows 2000.

As a second data point, I also couldn't reproduce any of these problems on
Windows 98 with the latest VPython.

-- 
    Andy Dougherty		dou...@la...

I installed the latest VPython for Windows (prints 2003-03-15 when you run),
and I can't reproduce any of these problems. All the demos and test routines
run fine. This is on Windows 2000.

At first your very last test showed nothing, but that was simply because
there was a typo in your faces position list:

VertexList = [(0,0,0),(1,0,0),(1,0,0)]
should be
VertexList = [(0,0,0),(1,0,0),(1,1,0)]

Unless you've already tried it, the first thing to do might be to uninstall
Visual (and maybe even Python) and reinstall. It sounds like something is
corrupted on your machine, or in the download.

Bruce Sherwood

----- Original Message -----
From: "IVES,THOM (HP-Boise,ex1)" <tho...@hp...>
To: "Visual Python Users (E-mail)"
<vis...@li...>
Sent: Wednesday, April 23, 2003 3:46 PM
Subject: [Visualpython-users] Major Problems with Latest Windows Release of
VPython


> This is a BUG REPORT. I found these bugs running the Windows 2000 Pro OS.
> Please be patient as you read on.
>
> I was trying to get rid of some bugs in a new piece of code I am writing,
> which relies heavily on VPython. It was maturing nicely, but I was running
> into some problems with the scene.exit = 0 attributes of display (see here
> http://vpython.org/webdoc/visual/display.html
> <http://vpython.org/webdoc/visual/display.html> ), so I decided to try and
> upgrade my VPython, and things stopped working altogether, introducing
> problems that I have never experienced before. Thinking the problem might
be
> with me, I went back to some of my simplest applications and the demo
> programs and still noted problems. One of these aspects can be represented
> with the following small program modified from the cross product.py demo
> program below. If you run this code, it should produce a grid for the
plane
> that illustrates the visual cross product, but it does not. Please read
> further below this small program.
>
> ________________________ Test Program 1 - Modification of the
> CrossProduct.py Demo________________________
> from visual import *
> # demonstration of vector cross product
>
> print """
> Vector cross product: Red cross Green = Yellow
> Drag to change green vector
> Click to toggle fixed angle or fixed length
> """
>
> scene.title="Vector Cross Product"
> scene.width=600
> scene.height=600
> R = 0.15*4
> plane = curve(pos=[(0, -10, -10), (0, -10, 10), (0, 10,10), (0, 10, -10),
>               (0, -10, -10), (0,-6,-10), (0,-6,10), (0, -2, 10), (0, -2,
> -10),
>               (0,2,-10), (0,2,10), (0,6,10), (0,6,-10),(0,10, -10),
>
>               (0,10,-6), (0,-10,-6), (0,-10,-2), (0,10,-2), (0,10,2),
>               (0,-10,2), (0,-10,6), (0,10,6)])
> ___________________________________End of Test Program
> 1__________________________________________
>
> So, I performed some other tests. This next small program is a slight
> modification of the graphtest.py demo. Try running it. Note that no axes
are
> drawn and that you get an error. If you raise the value of -70 to say
about
> -50, or lower the 70 to something lower, the errors will go away, but you
> still get a problem with no axes being drawn. In fact, with regard to the
> axes problem, even the simplest program where you try to use the curve
> object will not draw a curve on the screen in 3D.
>
> _________________________________ Modification of graphtest.py demo
> __________________________________
> from visual.graph import *
>
> # Using a graph-plotting module
> oscillation = gdisplay(xtitle='t', ytitle='Response')
>
> list = []
> for t in arange(-70, 70, 1):
>     pos=(t, 5.0*cos(-0.2*t)) # )
>     list.append(pos)
>
> funct1 = gcurve(pos = list, color=color.cyan)
>
> ________________________________ End of Test Program 2
> _____________________________________________
>
> Finally, I tried a simple test with the last feature I was struggling
with.
> First, if I try to run the faces_heightfield.py demo, it will generate an
> error and not draw anything. Second, when I run the faces_cone.py, I do
not
> get a cone. I get a cube, and the following message...
>
>
> Traceback (innermost last)
>   File "c:\python22\lib\site-packages\visual\demos\faces_cone.py", line
13,
> in ?
>     model = faces( pos = zeros( (N*3,3), Float ), frame = f )
> TypeError: Array can not be safely cast to required type
>
> Third, if I try to run the most simple face possible, a triangle, I get
> nothing drawn at all. Try the program below...
>
> ________________________________ Simple program to draw a Triangle Face
> _____________________________________
> from visual import *
>
> f = frame()
> #scene.background = color.black
> VertexList = [(0,0,0),(1,0,0),(1,0,0)]
> NormalList = [(0,0,1),(1,0,1),(1,0,1)]
>
> surface = faces(pos=VertexList, normal=NormalList, color=color.red)
> _________________________________ End of Test Program 3
> ____________________________________________________
>
> I think a download of the previous version of Visual should be made
> available until these problems can be fixed.
>
> Thom
>
>
>
>
>
>   _____
>
>
>
<http://www.hp.com/Redirect/gw/useng_welcome/logo/=http://welcome.hp.com/cou
> ntry/us/eng/welcome.htm> HP logo
> Thom Ives, Ph.D.
>
> R&D Engineering Scientist
>
> 11413 Chinden Blvd MS 400
> Boise, ID.  83714
>
> 208.396.6880 Phone
> 208.396.3587 Fax
> tho...@hp...  <mailto:tho...@hp...>
>
> Home:
> 5556 N Columbine Pl
> Boise, ID 83713
> 208.412.3971 Phone
> 413.332.6558 Fax
> tw...@ca... <mailto:tw...@ca...>
>   _____
>
>
>
>

This is a BUG REPORT. I found these bugs running the Windows 2000 Pro OS.
Please be patient as you read on.
 
I was trying to get rid of some bugs in a new piece of code I am writing,
which relies heavily on VPython. It was maturing nicely, but I was running
into some problems with the scene.exit = 0 attributes of display (see here
http://vpython.org/webdoc/visual/display.html
<http://vpython.org/webdoc/visual/display.html> ), so I decided to try and
upgrade my VPython, and things stopped working altogether, introducing
problems that I have never experienced before. Thinking the problem might be
with me, I went back to some of my simplest applications and the demo
programs and still noted problems. One of these aspects can be represented
with the following small program modified from the cross product.py demo
program below. If you run this code, it should produce a grid for the plane
that illustrates the visual cross product, but it does not. Please read
further below this small program.
 
________________________ Test Program 1 - Modification of the
CrossProduct.py Demo________________________
from visual import *
# demonstration of vector cross product
 
print """
Vector cross product: Red cross Green = Yellow
Drag to change green vector
Click to toggle fixed angle or fixed length
"""
 
scene.title="Vector Cross Product"
scene.width=600
scene.height=600
R = 0.15*4
plane = curve(pos=[(0, -10, -10), (0, -10, 10), (0, 10,10), (0, 10, -10),
              (0, -10, -10), (0,-6,-10), (0,-6,10), (0, -2, 10), (0, -2,
-10),
              (0,2,-10), (0,2,10), (0,6,10), (0,6,-10),(0,10, -10),

              (0,10,-6), (0,-10,-6), (0,-10,-2), (0,10,-2), (0,10,2),
              (0,-10,2), (0,-10,6), (0,10,6)])
___________________________________End of Test Program
1__________________________________________
 
So, I performed some other tests. This next small program is a slight
modification of the graphtest.py demo. Try running it. Note that no axes are
drawn and that you get an error. If you raise the value of -70 to say about
-50, or lower the 70 to something lower, the errors will go away, but you
still get a problem with no axes being drawn. In fact, with regard to the
axes problem, even the simplest program where you try to use the curve
object will not draw a curve on the screen in 3D.

_________________________________ Modification of graphtest.py demo
__________________________________
from visual.graph import *
 
# Using a graph-plotting module
oscillation = gdisplay(xtitle='t', ytitle='Response')
 
list = []
for t in arange(-70, 70, 1):
    pos=(t, 5.0*cos(-0.2*t)) # )
    list.append(pos)
 
funct1 = gcurve(pos = list, color=color.cyan)
 
________________________________ End of Test Program 2
_____________________________________________
 
Finally, I tried a simple test with the last feature I was struggling with.
First, if I try to run the faces_heightfield.py demo, it will generate an
error and not draw anything. Second, when I run the faces_cone.py, I do not
get a cone. I get a cube, and the following message...
 

Traceback (innermost last)
  File "c:\python22\lib\site-packages\visual\demos\faces_cone.py", line 13,
in ?
    model = faces( pos = zeros( (N*3,3), Float ), frame = f )
TypeError: Array can not be safely cast to required type

Third, if I try to run the most simple face possible, a triangle, I get
nothing drawn at all. Try the program below...
 
________________________________ Simple program to draw a Triangle Face
_____________________________________
from visual import *
 
f = frame()
#scene.background = color.black
VertexList = [(0,0,0),(1,0,0),(1,0,0)]
NormalList = [(0,0,1),(1,0,1),(1,0,1)]
 
surface = faces(pos=VertexList, normal=NormalList, color=color.red)
_________________________________ End of Test Program 3
____________________________________________________
 
I think a download of the previous version of Visual should be made
available until these problems can be fixed.
 
Thom

 
 
 
 
  _____  

 
<http://www.hp.com/Redirect/gw/useng_welcome/logo/=http://welcome.hp.com/cou
ntry/us/eng/welcome.htm> HP logo	 
Thom Ives, Ph.D.

R&D Engineering Scientist	

11413 Chinden Blvd MS 400
Boise, ID.  83714

208.396.6880 Phone
208.396.3587 Fax
tho...@hp...  <mailto:tho...@hp...> 

Home:
5556 N Columbine Pl
Boise, ID 83713
208.412.3971 Phone
413.332.6558 Fax
tw...@ca... <mailto:tw...@ca...> 
  _____  

 
 

On Wed, 2003-04-16 at 05:21, Shaun Press wrote:
> Bruce Sherwood wrote:
> 
> >Many thanks for the detailed feedback -- it's just the kind of data we need
> >to improve the installer. One thing seems quite odd: You couldn't install
> >gtkglarea, which is needed, yet you eventually managed to compile anyway. I
> >wouldn't have thought that mesa would include the missing piece....
> >
> >Bruce Sherwood
> >
> >  
> >
> To make it clearer. I installed gtkglarea both using apt-get and also by 
> downloading it as a seperate deb package and installing it using dpkg -i
> The problem was the configure script was still claiming it hadn't beeen 
> installed. But on investigation of the configure file I think that this 
> error message may be misleading and instead was triggered by mesa not 
> being installed. Installing the mesa packages enables the configure 
> script to complete normally. This may indicate a problem in the 
> configure script rather than packages really being missing.


I had the wrong required packages listed.  They are:
gtkglarea5-dev
python2.2-dev
python2.2-numeric

The gtkglarea5 package should automatically bring in a GL library if you
still need one, from one of 5 choices, all of which include libGL and
libGLU.

The configure test for GtkGLarea is the one provided by the library
author, and consists of link tests for libGL + libGLU or libMesaGL +
libMesaGLU if it cannot find libGL (without and with pthreads for
each).  Can you send me the log file that included the errors?

-Jonathan Brandmeyer 

http://packages.debian.org/stable/libs/gtkglarea5.html

Bruce Sherwood wrote:

>Many thanks for the detailed feedback -- it's just the kind of data we need
>to improve the installer. One thing seems quite odd: You couldn't install
>gtkglarea, which is needed, yet you eventually managed to compile anyway. I
>wouldn't have thought that mesa would include the missing piece....
>
>Bruce Sherwood
>
>  
>
To make it clearer. I installed gtkglarea both using apt-get and also by 
downloading it as a seperate deb package and installing it using dpkg -i
The problem was the configure script was still claiming it hadn't beeen 
installed. But on investigation of the configure file I think that this 
error message may be misleading and instead was triggered by mesa not 
being installed. Installing the mesa packages enables the configure 
script to complete normally. This may indicate a problem in the 
configure script rather than packages really being missing.

gtkglarea is not supplied by the libgtkgl5 package but by the
the packages gtkglarea5 and gtkglarea5-dev.

Installation under debian woody is pretty simple:  see
the detailed instructions I gave in an email to this list some
months back.  I've then installed to other debian woody machines by simply
copying by cvisualmodule.so and the files in the visual directory 
from one machine to the next, after installing apt-getting gtkglarea5.
Then it's not even necessary to apt-get all the various '-dev' packages.

Martin Gelfand
Dept of Physics, Colorado State
>===== Original Message From "Bruce Sherwood" <bas...@un...> =====
>Many thanks for the detailed feedback -- it's just the kind of data we need
>to improve the installer. One thing seems quite odd: You couldn't install
>gtkglarea, which is needed, yet you eventually managed to compile anyway. I
>wouldn't have thought that mesa would include the missing piece....
>
>Bruce Sherwood
>
>----- Original Message -----
>From: "Shaun Press" <Sha...@an...>
>To: "vpusers" <vis...@li...>
>Sent: Tuesday, April 15, 2003 11:23 AM
>Subject: [Visualpython-users] New installer under Debian
>
>
>> I have recently shifted jobs and have a debian install on my desktop. I
>> have't been able to get VPython working on my machine so I was happy to
>> try the new installer this morning. This is what I discovered.
>>
>> I am running debian with a 2.4.19 kernel on a Dell Optiplex GX260
>> Downloaded the package OK. Unzipped and untarred it fine.
>> First important problem apt-get install libgtkgl5 does not work as the
>> package doesn't exist. Attempts at finding other install servers that
>> might have it were unsuccesful. No matter, I just plowed ahead.
>> ./configure --prefix=/usr/local blew up when looking for gtk+ > 1.0
>> This was caused by the missing gtk-config program
>> Needed to load the gtk1.2lib-dev package which I found using dselect
>> Then it complained about gtkglarea not being installed, so I ran dselect
>> and installed every package connected with gtkglarea
>> Still complained about gtkglarea not being installed
>> Needed to install mesa support, which I would have assumed came with the
>> standard debian install but clearly did not in my case.
>> Again I used dselect to find a mesa package
>> ./configure --prefix=/usr/local now worked fine
>> make install then ran without a hitch, if you ignore all the deprecated
>> warning messages
>> Tested that idle ran and that little balls bounced around. Woo Hoo!
>>
>>
>>
>> -------------------------------------------------------
>> This sf.net email is sponsored by:ThinkGeek
>> Welcome to geek heaven.
>> http://thinkgeek.com/sf
>> _______________________________________________
>> Visualpython-users mailing list
>> Vis...@li...
>> https://lists.sourceforge.net/lists/listinfo/visualpython-users
>>
>>
>
>
>
>-------------------------------------------------------
>This sf.net email is sponsored by:ThinkGeek
>Welcome to geek heaven.
>http://thinkgeek.com/sf
>_______________________________________________
>Visualpython-users mailing list
>Vis...@li...
>https://lists.sourceforge.net/lists/listinfo/visualpython-users

Many thanks for the detailed feedback -- it's just the kind of data we need
to improve the installer. One thing seems quite odd: You couldn't install
gtkglarea, which is needed, yet you eventually managed to compile anyway. I
wouldn't have thought that mesa would include the missing piece....

Bruce Sherwood

----- Original Message -----
From: "Shaun Press" <Sha...@an...>
To: "vpusers" <vis...@li...>
Sent: Tuesday, April 15, 2003 11:23 AM
Subject: [Visualpython-users] New installer under Debian


> I have recently shifted jobs and have a debian install on my desktop. I
> have't been able to get VPython working on my machine so I was happy to
> try the new installer this morning. This is what I discovered.
>
> I am running debian with a 2.4.19 kernel on a Dell Optiplex GX260
> Downloaded the package OK. Unzipped and untarred it fine.
> First important problem apt-get install libgtkgl5 does not work as the
> package doesn't exist. Attempts at finding other install servers that
> might have it were unsuccesful. No matter, I just plowed ahead.
> ./configure --prefix=/usr/local blew up when looking for gtk+ > 1.0
> This was caused by the missing gtk-config program
> Needed to load the gtk1.2lib-dev package which I found using dselect
> Then it complained about gtkglarea not being installed, so I ran dselect
> and installed every package connected with gtkglarea
> Still complained about gtkglarea not being installed
> Needed to install mesa support, which I would have assumed came with the
> standard debian install but clearly did not in my case.
> Again I used dselect to find a mesa package
> ./configure --prefix=/usr/local now worked fine
> make install then ran without a hitch, if you ignore all the deprecated
> warning messages
> Tested that idle ran and that little balls bounced around. Woo Hoo!
>
>
>
> -------------------------------------------------------
> This sf.net email is sponsored by:ThinkGeek
> Welcome to geek heaven.
> http://thinkgeek.com/sf
> _______________________________________________
> Visualpython-users mailing list
> Vis...@li...
> https://lists.sourceforge.net/lists/listinfo/visualpython-users
>
>

I have recently shifted jobs and have a debian install on my desktop. I 
have't been able to get VPython working on my machine so I was happy to 
try the new installer this morning. This is what I discovered.

I am running debian with a 2.4.19 kernel on a Dell Optiplex GX260
Downloaded the package OK. Unzipped and untarred it fine.
First important problem apt-get install libgtkgl5 does not work as the 
package doesn't exist. Attempts at finding other install servers that 
might have it were unsuccesful. No matter, I just plowed ahead.
./configure --prefix=/usr/local blew up when looking for gtk+ > 1.0
This was caused by the missing gtk-config program
Needed to load the gtk1.2lib-dev package which I found using dselect
Then it complained about gtkglarea not being installed, so I ran dselect 
and installed every package connected with gtkglarea
Still complained about gtkglarea not being installed
Needed to install mesa support, which I would have assumed came with the 
standard debian install but clearly did not in my case.
Again I used dselect to find a mesa package
./configure --prefix=/usr/local now worked fine
make install then ran without a hitch, if you ignore all the deprecated 
warning messages
Tested that idle ran and that little balls bounced around. Woo Hoo!