visualpython-users

Re: [Visualpython-users] Visual and Python 3.x

[Scott D. D. <Sco...@Ac...>]

Bruce Sherwood wrote:
> I occasionally do searches to see what the status of numpy is for Python 3.x, 
> since without numpy it's not possible to build Visual for Python 3.x. What I 
> read today is that it is unlikely that numpy will be ported to Python 3.x until 
> late 2010.
> 
> Of course we have a more immediate problem that we can't even get Visual to work 
> with Python 2.6.3.

There is a 2.6.4 coming out, you could check release candidate 2.6.4rc2

--Scott David Daniels
Sco...@Ac...

Re: [Visualpython-users] Visual on Ubuntu *Karmic*? Has *anyone* managed?

[Lenore H. <lh...@si...>]

On Oct 20, 2009, at 22:10 , Guy K. Kloss wrote:

> On Wed, 21 Oct 2009 16:03:56 Bruce Sherwood wrote:
>> For what it's worth, I successfully compiled and ran Visual on  
>> Windows for
>> Python 2.6.2 using the same Boost 1.39 libraries that didn't work  
>> with
>> Python  2.6.3. So now I wonder whether the problem isn't with Python
>> 2.6.3, not with Boost.
>
> That is very good. We're starting to get somewhere with this. So,
> unfortunately it's mostly out of our hands to fix this problem. It  
> seems like
> we need to have a fix for Boost to play along with Python >= 2.6.3  
> or some
> adaptation to Python. However, I've got the feeling that maybe the  
> "virtue" of
> the older Python version was potentially a shortcoming that was fixed

Python 2.6.3 apparently needs svn version of boost to work.  See http://groups.google.com/group/python-ogre-developers/browse_thread/thread/130617f8c69f1bcb#

Apparently there exist versions of boost that support python 3.  https://svn.boost.org/svn/boost/website/public_html/beta/feed/history/boost_1_41_0.qbk

Lenore

Re: [Visualpython-users] Visual on Ubuntu *Karmic*? Has *anyone* managed?

[Guy K. K. <g....@ma...>]

On Wed, 21 Oct 2009 16:03:56 Bruce Sherwood wrote:
> For what it's worth, I successfully compiled and ran Visual on Windows for 
> Python 2.6.2 using the same Boost 1.39 libraries that didn't work with
>  Python  2.6.3. So now I wonder whether the problem isn't with Python
>  2.6.3, not with Boost.

That is very good. We're starting to get somewhere with this. So, 
unfortunately it's mostly out of our hands to fix this problem. It seems like 
we need to have a fix for Boost to play along with Python >= 2.6.3 or some 
adaptation to Python. However, I've got the feeling that maybe the "virtue" of 
the older Python version was potentially a shortcoming that was fixed 
recently, and now we (Boost or the usage of Boost) have to evolve around that.

Guy

-- 
Guy K. Kloss
Institute of Information and Mathematical Sciences
Te Kura Pūtaiao o Mōhiohio me Pāngarau
Room 2.63, Quad Block A Building
Massey University, Auckland, Albany
Private Bag 102 904, North Shore Mail Centre
voice: +64 9 414-0800 ext. 9585   fax: +64 9 441-8181
eMail: G....@ma...  http://iims.massey.ac.nz

Re: [Visualpython-users] Visual on Ubuntu *Karmic*? Has *anyone* managed?

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

For what it's worth, I successfully compiled and ran Visual on Windows for 
Python 2.6.2 using the same Boost 1.39 libraries that didn't work with Python 
2.6.3. So now I wonder whether the problem isn't with Python 2.6.3, not with Boost.

Bruce Sherwood

Bruce Sherwood wrote:
> In a bug report at
> 
> https://bugs.launchpad.net/ubuntu/+source/python-visual/+bug/408663
> 
> Guy reported various aspects of the problem with the new Ubuntu Karmic 
> and gave a pointer to http://ur1.ca/du0m which documents the same 
> problem with PySide that we're experiencing with Visual, and which was 
> just reported on Windows using Python 2.6.3 instead of 2.6.2.
> 
> I've just reproduced this same problem after compiling from source on 
> Windows using Boost 1.39 and Python 2.6.3, as soon as you execute from 
> visual import *:
> 
> Traceback (most recent call last):
>    File "<pyshell#0>", line 1
>      from visual import *
>    File "C:\Python26\lib\site-packages\visual\__init__.py", line 59
>      import cvisual
> AttributeError: 'Boost.Python.StaticProperty' object attribute '__doc__' 
> is read-only
> 
> It would appear that something is seriously wrong with the Boost Python 
> libraries and/or Python 2.6.3 or both. It seems unlikely that both 
> PySide and Visual are doing something wrong, especially since things 
> worked in the past.
> 
> As far as Visual is concerned, until and unless someone figures out the 
> underlying problem I guess we have to stay with Python 2.6.2...? Guy, 
> when you've tried building on Karmic, is that Python 2.6.3?
> 
> Bruce Sherwood
> 
> Guy K. Kloss wrote:
>> The subject pretty much says it all ...
>>
>> Has anybody managed to get Visual to work on Ubuntu Karmic?
>>
>> For me neither the pre-compiled repository binaries in versino 5.11 worked, 
>> neither the back ported ones from Debian Sid (5.12), nor a direct compile.
>>
>> I'm always hitting a road block, and I'm supposing the reason could be the 
>> boost 1.38 libraries as present in Karmic.
>>
>> Any stories to share where it worked?
>>
>> Guy
>>
> 
> ------------------------------------------------------------------------------
> Come build with us! The BlackBerry(R) Developer Conference in SF, CA
> is the only developer event you need to attend this year. Jumpstart your
> developing skills, take BlackBerry mobile applications to market and stay 
> ahead of the curve. Join us from November 9 - 12, 2009. Register now!
> http://p.sf.net/sfu/devconference
> _______________________________________________
> Visualpython-users mailing list
> Vis...@li...
> https://lists.sourceforge.net/lists/listinfo/visualpython-users

Re: [Visualpython-users] Visual on Ubuntu *Karmic*? Has *anyone* managed?

[Guy K. K. <g....@ma...>]

On Wed, 21 Oct 2009 15:05:01 Bruce Sherwood wrote:
> As far as Visual is concerned, until and unless someone figures out the 
> underlying problem I guess we have to stay with Python 2.6.2...? Guy, 
> when you've tried building on Karmic, is that Python 2.6.3?

On Karmic that was 2.6.4. And it did work on Jaunty, which is Python 2.6.2.

So it could be either Python or Boost. However if I'd make a guess, I'd rather 
guess it's the Boost libs. The Python C API is known to be quite robust within 
major releases.

Guy

-- 
Guy K. Kloss
Institute of Information and Mathematical Sciences
Te Kura Pūtaiao o Mōhiohio me Pāngarau
Room 2.63, Quad Block A Building
Massey University, Auckland, Albany
Private Bag 102 904, North Shore Mail Centre
voice: +64 9 414-0800 ext. 9585   fax: +64 9 441-8181
eMail: G....@ma...  http://iims.massey.ac.nz

[Visualpython-users] Visual and Python 3.x

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

I occasionally do searches to see what the status of numpy is for Python 3.x, 
since without numpy it's not possible to build Visual for Python 3.x. What I 
read today is that it is unlikely that numpy will be ported to Python 3.x until 
late 2010.

Of course we have a more immediate problem that we can't even get Visual to work 
with Python 2.6.3.

Bruce Sherwood

Re: [Visualpython-users] Visual on Ubuntu *Karmic*? Has *anyone* managed?

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

In a bug report at

https://bugs.launchpad.net/ubuntu/+source/python-visual/+bug/408663

Guy reported various aspects of the problem with the new Ubuntu Karmic 
and gave a pointer to http://ur1.ca/du0m which documents the same 
problem with PySide that we're experiencing with Visual, and which was 
just reported on Windows using Python 2.6.3 instead of 2.6.2.

I've just reproduced this same problem after compiling from source on 
Windows using Boost 1.39 and Python 2.6.3, as soon as you execute from 
visual import *:

Traceback (most recent call last):
   File "<pyshell#0>", line 1
     from visual import *
   File "C:\Python26\lib\site-packages\visual\__init__.py", line 59
     import cvisual
AttributeError: 'Boost.Python.StaticProperty' object attribute '__doc__' 
is read-only

It would appear that something is seriously wrong with the Boost Python 
libraries and/or Python 2.6.3 or both. It seems unlikely that both 
PySide and Visual are doing something wrong, especially since things 
worked in the past.

As far as Visual is concerned, until and unless someone figures out the 
underlying problem I guess we have to stay with Python 2.6.2...? Guy, 
when you've tried building on Karmic, is that Python 2.6.3?

Bruce Sherwood

Guy K. Kloss wrote:
> The subject pretty much says it all ...
> 
> Has anybody managed to get Visual to work on Ubuntu Karmic?
> 
> For me neither the pre-compiled repository binaries in versino 5.11 worked, 
> neither the back ported ones from Debian Sid (5.12), nor a direct compile.
> 
> I'm always hitting a road block, and I'm supposing the reason could be the 
> boost 1.38 libraries as present in Karmic.
> 
> Any stories to share where it worked?
> 
> Guy
> 

[Visualpython-users] import error with python 2.6.3

[Victor C. <vm...@ya...>]

Hi, everybody!

I am new for this forum, even though I use VPython for a couple of years (not too much, but I found it a great piece of Software).

I run VPython 5.13 on Windows (Python 2.6.2). 

Recently I tried to switch to 2.6.3, but could not import the visual module.

*** Python 2.6.3 (r263rc1:75186, Oct  2 2009, 20:40:30) [MSC v.1500 32 bit (Intel)] on win32. ***
>>> from visual import *
Traceback (most recent call last):
  File "<interactive input>", line 1, in <module>
  File "C:\Python26\lib\site-packages\visual\__init__.py", line 59, in <module>
    import cvisual
AttributeError: 'Boost.Python.StaticProperty' object attribute '__doc__' is read-only
>>> 

Did anyone have this problem? Can it be related to VPython 5.13 or Python 2.6.3?

Thank you for any idea how to fix it!

Vic



      

[Visualpython-users] Visual on Ubuntu *Karmic*? Has *anyone* managed?

[Guy K. K. <g....@ma...>]

The subject pretty much says it all ...

Has anybody managed to get Visual to work on Ubuntu Karmic?

For me neither the pre-compiled repository binaries in versino 5.11 worked, 
neither the back ported ones from Debian Sid (5.12), nor a direct compile.

I'm always hitting a road block, and I'm supposing the reason could be the 
boost 1.38 libraries as present in Karmic.

Any stories to share where it worked?

Guy

-- 
Guy K. Kloss
Institute of Information and Mathematical Sciences
Te Kura Pūtaiao o Mōhiohio me Pāngarau
Massey University, Albany (North Shore City, Auckland)
473 State Highway 17, Gate 1, Mailroom, Quad B Building
voice: +64 9 414-0800 ext. 9585   fax: +64 9 441-8181
G....@ma... http://www.massey.ac.nz/~gkloss

Re: [Visualpython-users] Why no screenshot function?

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

Now that the issues of images and movies have been talked through, it would be 
good if someone (or someones) would develop and test code for programmatically 
saving a screen image on Windows/Mac/Linux and offer it for inclusion in Visual. 
The design criteria would seem to be these:

1) Should work on all three platforms (which may mean three different mechanisms).

2) One output format is presumably sufficient. The simplest is probably targa 
(which is also used by Visual materials), and there are plenty of applications 
for converting targa to other formats. Targa files are basically just sequences 
of 3-byte RGB triples per pixel. But feel free to propose and support a 
different choice.

3) Include a proposed syntax for invoking the save.

I can't work on this myself because I'm deeply involved in the launch of a 3rd 
edition of a physics textbook.

Bruce Sherwood

P.S. On a related issue, there was a proposed patch to deal with some issues in 
packaging a stand-alone Visual application, but David Scherer and I concluded 
that the solution (to hunt around in various folders for material files) was not 
a maintainable scheme, so someone might like to look into this issue and propose 
a different solution.

Re: [Visualpython-users] Why no screenshot function?

[John Z. <joh...@wa...>]

Hi All,

Bruce is certainly right that for publication quality images, something like 
the POVray export is the way to go. However, for animations I have to agree 
with those who suggest the benefit of some window grab utility. A resolution 
of 1024x768 or even 640x480 (ala Youtube) is plenty good for demonstration 
movies. We frequently use these in our stereo visualizations. 

I use the PIL image grab method on Windows and xwd on linux for grabbing 
frames. It would be nice to have a platform independent way of doing this for 
any Vpython visualization. So I give a +1 to consideration this feature in a 
future version.

--John

On Monday 19 October 2009 20:44:01 Bruce Sherwood wrote:
> But that's the fundamental problem. If I understand your note correctly,
> you want "decent image resolution". The only way I know to get that is to
> use ray-tracing, through POV-ray, which is time-consuming, but no bitmap
> image can give comparable results. Moreover, the real-time shading in
> Visual is no match for ray-tracing, no matter what the resolution in dots
> per inch.
>
> However, as was pointed out you might be able to get by with using a very
> large Visual window on a very large monitor, if the dots per inch is
> adequate for your purposes and you can tolerate the shading and lighting
> done by Visual.
>
> Bruce Sherwood
>
> Samuel Pelaez wrote:
> > Hi guys,
> >
> > I definetly agree with Anders. In my work I use Visual to make an
> > animation of the time evolution of a system of particles. In order to
> > save such an animation I must save each frame using the povexport module
> > and then compile every frame with povray. Finally I create the proper
> > gif animation.
> >
> > However the povray compilation is very time-consuming if you want to get
> > a decent image resolution. If there is a "native" way (as native as
> > C/C++ can be in vpython) to take screenshots within vpython, it would
> > make things easier and most probably less time-expensive.
> >
> > Samuel.
> >
> >> El 19 de oct de 2009, 10:31 p.m., "Guy K. Kloss" <g....@ma...
> >> <mailto:g....@ma...>> escribió:
> >>
> >> On Tue, 20 Oct 2009 09:23:41 Anders Wallin wrote: > Could this perhaps
> >> be wrapped as a separate pyth...
> >>
> >> Visual is mainly C/C++. So I assume that could (easily) be added and
> >> exposed
> >> through the Boost.Python wrappers to make it all work.
> >>
> >> Guy -- Guy K. Kloss Institute of Information and Mathematical Sciences
> >> Te Kura Pūtaiao o Mōhiohio...
> >>
> >> Come build with us! The BlackBerry(R) Developer Conference in SF, CA
> >> is the only developer event you...
>
> ---------------------------------------------------------------------------
>--- Come build with us! The BlackBerry(R) Developer Conference in SF, CA is
> the only developer event you need to attend this year. Jumpstart your
> developing skills, take BlackBerry mobile applications to market and stay
> ahead of the curve. Join us from November 9 - 12, 2009. Register now!
> http://p.sf.net/sfu/devconference
> _______________________________________________
> Visualpython-users mailing list
> Vis...@li...
> https://lists.sourceforge.net/lists/listinfo/visualpython-users

-- 
John M. Zelle, Ph.D.             Wartburg College
Professor of Computer Science    Waverly, IA
joh...@wa...          (319) 352-8360

Re: [Visualpython-users] Why no screenshot function?

[Kadir H. <kha...@ya...>]

In Windowns environment, I use screenshots captured by PIL utilities, to create simple movies by using available Windows tools (like MovieMaker).

Screenshot PNG quality for a "normal" size screen is much much higher than the video output I get from them. You loose on quality more while converting to video than you loose on capture.

Blender (I believe also available on Linux) also has a movie editor where you can create movies from screen captures with better quality. You can get larger screen size videos with better quality.

However, video outputs are usually very large in file size, and I prefer to create executables rather than movies these days. That gives the best image quality (as we see on Visual screen, as usual).

POV quality is something else. It is the additional realism provided by ray tracing, not resolution (not pixel resolution at least, but "color" resolution, i.e., "true" shading). But it has a very high cost (yet).

Depending on your need, plain screen capture should do, or if possible or practical I suggest to use executables.

Kadir




________________________________
From: Samuel Pelaez <sam...@gm...>
To: Guy K. Kloss <g....@ma...>
Cc: vis...@li...
Sent: Tue, October 20, 2009 3:44:31 AM
Subject: Re: [Visualpython-users] Why no screenshot function?


Hi guys,
I definetly agree with Anders. In my work I use Visual to make an animation of the time evolution of a system of particles. In order to save such an animation I must save each frame using the povexport module and then compile every frame with povray. Finally I create the proper gif animation.
However the povray compilation is very time-consuming if you want to get a decent image resolution. If there is a "native" way (as native as C/C++ can be in vpython) to take screenshots within vpython, it would make things easier and most probably less time-expensive.
Samuel. 
El 19 de oct de 2009, 10:31 p.m., "Guy K. Kloss" <g....@ma...> escribió:
>
>
>On Tue, 20 Oct 2009 09:23:41 Anders Wallin wrote:
>> Could this perhaps be wrapped as a separate pyth...Visual is mainly C/C++. So I assume that could (easily) be added and exposed
>>through the Boost.Python wrappers to make it all work.
>
>>Guy
>
>-- 
>Guy K. Kloss
>Institute of Information and Mathematical Sciences
>Te Kura Pūtaiao o Mōhiohio...
>Come build with us! The BlackBerry(R) Developer Conference in SF, CA
>is the only developer event you...


      

Re: [Visualpython-users] Why no screenshot function?

[Guy K. K. <g....@ma...>]

On Tue, 20 Oct 2009 14:44:01 Bruce Sherwood wrote:
> you want "decent image resolution"

Well, I guess if someone is using a scriptable screenshot utility to save 
frames that are rendered from Visual, and it's done through the Visual API, 
then they *know* that all they're getting is what's rendered in the window. If 
one wants *more* then they can always go with the POV output.

As long as the API documents that the save_frame(file_name, format='PNG') 
method (or however it may finally look like) will save what one sees on screen 
I don't see a big problem with that. It's about delivering good enough quality 
for the purpose that one can expect from reading the API documentation. And I 
believe e. g. PNG is good enough, JPEG is not (compression artefacts).

Just my thoughts on this topic.

Guy

-- 
Guy K. Kloss
Institute of Information and Mathematical Sciences
Te Kura Pūtaiao o Mōhiohio me Pāngarau
Massey University, Albany (North Shore City, Auckland)
473 State Highway 17, Gate 1, Mailroom, Quad B Building
voice: +64 9 414-0800 ext. 9585   fax: +64 9 441-8181
G....@ma... http://www.massey.ac.nz/~gkloss

Re: [Visualpython-users] Why no screenshot function?

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

But that's the fundamental problem. If I understand your note correctly, you 
want "decent image resolution". The only way I know to get that is to use 
ray-tracing, through POV-ray, which is time-consuming, but no bitmap image can 
give comparable results. Moreover, the real-time shading in Visual is no match 
for ray-tracing, no matter what the resolution in dots per inch.

However, as was pointed out you might be able to get by with using a very large 
Visual window on a very large monitor, if the dots per inch is adequate for your 
purposes and you can tolerate the shading and lighting done by Visual.

Bruce Sherwood

Samuel Pelaez wrote:
> Hi guys,
> 
> I definetly agree with Anders. In my work I use Visual to make an 
> animation of the time evolution of a system of particles. In order to 
> save such an animation I must save each frame using the povexport module 
> and then compile every frame with povray. Finally I create the proper 
> gif animation.
> 
> However the povray compilation is very time-consuming if you want to get 
> a decent image resolution. If there is a "native" way (as native as 
> C/C++ can be in vpython) to take screenshots within vpython, it would 
> make things easier and most probably less time-expensive.
> 
> Samuel.
> 
>> El 19 de oct de 2009, 10:31 p.m., "Guy K. Kloss" <g....@ma... 
>> <mailto:g....@ma...>> escribió:
>>
>> On Tue, 20 Oct 2009 09:23:41 Anders Wallin wrote: > Could this perhaps 
>> be wrapped as a separate pyth...
>>
>> Visual is mainly C/C++. So I assume that could (easily) be added and 
>> exposed
>> through the Boost.Python wrappers to make it all work.
>>
>> Guy -- Guy K. Kloss Institute of Information and Mathematical Sciences 
>> Te Kura Pūtaiao o Mōhiohio...
>>
>> Come build with us! The BlackBerry(R) Developer Conference in SF, CA 
>> is the only developer event you...
>>

Re: [Visualpython-users] Why no screenshot function?

[Samuel P. <sam...@gm...>]

Hi guys,

I definetly agree with Anders. In my work I use Visual to make an animation
of the time evolution of a system of particles. In order to save such an
animation I must save each frame using the povexport module and then compile
every frame with povray. Finally I create the proper gif animation.

However the povray compilation is very time-consuming if you want to get a
decent image resolution. If there is a "native" way (as native as C/C++ can
be in vpython) to take screenshots within vpython, it would make things
easier and most probably less time-expensive.

Samuel.

El 19 de oct de 2009, 10:31 p.m., "Guy K. Kloss" <g....@ma...>
escribió:

On Tue, 20 Oct 2009 09:23:41 Anders Wallin wrote: > Could this perhaps be
wrapped as a separate pyth...
Visual is mainly C/C++. So I assume that could (easily) be added and exposed
through the Boost.Python wrappers to make it all work.

Guy -- Guy K. Kloss Institute of Information and Mathematical Sciences Te
Kura Pūtaiao o Mōhiohio...

Come build with us! The BlackBerry(R) Developer Conference in SF, CA is the
only developer event you...

Re: [Visualpython-users] Visualpython-users Digest, Vol 41, Issue 4

[Cross, J. <JC...@CS...>]

Jorge,

If you are on a 64bit system, I have an up-to-date RPM for Fedora 11 here:
http://www.users.csbsju.edu/~jcross/vpython/
I also put instructions up of what I did as well as a SRC RPM.

Jason

------------------------------

Message: 2
Date: Mon, 19 Oct 2009 12:40:14 +0000
From: jorge a secas <ta...@ho...>
Subject: [Visualpython-users] help in fedora 11
To: <vis...@li...>
Message-ID: <SNT...@ph...l>
Content-Type: text/plain; charset="iso-8859-1"


hi Everybody

i've tried to install vpython in fedora 11 but it doesn't work

this is the mistake

make[1]: *** [cvisualmodule.lo] Error 1

thanks to all

Re: [Visualpython-users] Why no screenshot function?

[Guy K. K. <g....@ma...>]

On Tue, 20 Oct 2009 09:23:41 Anders Wallin wrote:
> Could this perhaps be wrapped as a separate python extension and would
> it then work with visual??

Visual is mainly C/C++. So I assume that could (easily) be added and exposed 
through the Boost.Python wrappers to make it all work.

Guy

-- 
Guy K. Kloss
Institute of Information and Mathematical Sciences
Te Kura Pūtaiao o Mōhiohio me Pāngarau
Massey University, Albany (North Shore City, Auckland)
473 State Highway 17, Gate 1, Mailroom, Quad B Building
voice: +64 9 414-0800 ext. 9585   fax: +64 9 441-8181
G....@ma... http://www.massey.ac.nz/~gkloss

Re: [Visualpython-users] Why no screenshot function?

[Anders W. <and...@gm...>]

It turns out the command 'scrot' works better on my system, it takes a
screenshot of the whole screen. I could then batch-process each file
to crop the visual window and then assemble into a movie.
But all of this seems like a lot of trouble when the c-code to do a
screenshot using SDL/OpenGL is something like:

void Screendump(char *destFile, short W, short H) {
 FILE   *out = fopen(destFile, "w");
 char   pixel_data[3*W*H];
 short  TGAhead[] = {0, 2, 0, 0, 0, 0, W, H, 24};
 glReadBuffer(GL_FRONT);
 glReadPixels(0, 0, W, H, GL_BGR, GL_UNSIGNED_BYTE, pixel_data);
 fwrite(&TGAhead, sizeof(TGAhead), 1, out);
 fwrite(pixel_data, 3*W*H, 1, out);
 fclose(out);
}

Could this perhaps be wrapped as a separate python extension and would
it then work with visual??

Re: [Visualpython-users] Why no screenshot function?

[Anders W. <and...@gm...>]

> Here is a pretty workable solution under Linux.

this seems to work otherwise, but relies on the ImageMagick  'import'
function which does not work well on my system. With non-openGL
windows it sometimes produces OK results, but with an OpenGL window (I
tried visual-scenes and glxgears) it always produces a rubbish
miff-file.
This may be because I'm on Ubuntu 9.04 and I'm using onboard intel-graphics
glxinfo says:
OpenGL version string: 1.4 Mesa 7.4

any suggestions what to do?

Re: [Visualpython-users] help in fedora 11

[Vineyard, M. F. <vin...@un...>]

Hi Jorge,

I built Visual on my fedora 11 machine several months ago. I remember I had to fix a couple of things to get it to compile. One was to change /usr/include/boost/python/detail/translate_exception.hpp as described in the INSTALL.txt file. Another was to create the soft link libboost_thread.so -> libboost_thread-mt.so.1.37.0 in /usr/lib. There may have been others. I can't remember. Try these and if you still have problems, you can send me your log file and perhaps I can help.

Cheers,

Mike

--
Michael F. Vineyard
Frank & Marie Louise Bailey Professor of Physics
Department of Physics & Astronomy
Union College
Schenectady, NY 12308
Email: vin...@un...
Web: http://www1.union.edu/vineyarm
Phone: (518) 388-8353
Fax: (518) 388-6947



-----Original Message-----
From: jorge a secas [mailto:ta...@ho...]
Sent: Mon 10/19/2009 8:40 AM
To: vis...@li...
Subject: [Visualpython-users] help in fedora 11
 

hi Everybody

i've tried to install vpython in fedora 11 but it doesn't work 

this is the mistake

make[1]: *** [cvisualmodule.lo] Error 1

thanks to all
 		 	   		  
_________________________________________________________________
Infórmate, mantente en contacto y encuéntralo todo, a la vez. Con la nueva Toolbar de MSN  nunca has tenido tantas ventajas en tan poco espacio. 
http://toolbar.es.msn.com/

Re: [Visualpython-users] help in fedora 11

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

This is basically a question as to whether anyone on this list has successfully 
built Visual on Fedora and can say what they had to do. I had some off-list 
email with Jorge and was unable to help him even after looking at his logs 
because I have no access to or experience with Fedora. It looked like a possible 
Boost problem, but that's only a guess.

Bruce Sherwood

jorge a secas wrote:
> hi Everybody
> 
> i've tried to install vpython in fedora 11 but it doesn't work
> 
> this is the mistake
> 
> make[1]: *** [cvisualmodule.lo] Error 1
> 
> thanks to all
> 

Re: [Visualpython-users] Why no screenshot function?

[Jim T. <jt...@mi...>]

Here is a pretty workable solution under Linux.  I had found some 
information that made me think I could get this to work on Windows (see 
the comments below) but I never got around to finishing and releasing 
this. It is supposed to be part of the next release of this:  
http://www.missioncognition.net/wp-content/uploads/2009/06/VisualPyODE_v0.1_src.tar.gz 
and you can get copies of the license files there.

In short you can either let the user select the window, or you can pass 
the name of the window.  I could not find any saved code that worked 
using the window name but I'm sure that I was able to get this to work 
using whatever name the visual window defaults to.  I think I was also 
able to change the title of the visual window and use that as well.

Right now my system is broken for running Visual so I can't verify 
anything at the moment.  But if you are interested, go to 
http://www.missioncognition.net/visualpyode/ to see a video I made using 
the code below (or something very closely resembling it).  You will 
notice the scene is panning, I also have code which captures the user 
view actions and then can restore them later as a playback.  That way I 
could create a real-time pan despite the fact that the screen capturing 
is a much slower process.  If anybody is interested in that, holler and 
I'll post it here as well.  One of these days I might get around to 
doing an update of visualPyODE and these will be included.  
Unfortunately I have very little feedback on that project so I don't 
know if there is anybody using it and it is on my back burner as well -- 
so obviously it is not my top priority.

JT

Python code below:

"""
videograb.py
A module to perform video capture of an x-windows window.

Written By:
    James Thomas
    Email: jt...@mi...
    Web: http://missioncognition.net/

Copyright 2009

This library is free software; you can redistribute it and/or
modify it under the terms of EITHER:
  (1) The GNU Lesser General Public License as published by the Free
      Software Foundation; either version 2.1 of the License, or (at
      your option) any later version. The text of the GNU Lesser
      General Public License is included with this library in the
      file LICENSE.
  (2) The BSD-style license that is included with this library in
      the file LICENSE-BSD.

This library is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the files
LICENSE and LICENSE-BSD for more details.
"""

import subprocess
import re
import tempfile
import shutil

# Note for possible MS Windows implementation.  Apparently the window 
can be obtained as follows:
#  import win32gui
#  window = win32gui.FindWindow ( None, WindowTitle )

class VideoCaptureXWindows:
    """Capture a mpeg video from an X-Windows display.  Requires 
xwininfo, import, and convert
    utilities and the mpeg generation library to be installed."""
    def __init__(self, WindowTitle=None):
        """Setup for recording an mpeg video.  The window title is not 
specified then
        the user will be prompted to click the desired window."""
       
        # Initialize variables
        self._windowID = None
        self._frameCounter = 0

        # Generate unique temporary directory to store the captures
        self._fileLocation = tempfile.mkdtemp()

        # Setup window ID determination based on whether a window title 
was specified or not
        if WindowTitle is None:
            # Get the x-windows information without parameters so the 
user will have to select a window.
            command = ['xwininfo']

            # Print a message telling the user to click a window
            # It would be cool if we could come up with a better way to 
notify the user
            # Perhaps some text overlay over the entire desktop.
            print 'Click the window you want to record'
           
            # Set the regular expression string for extracting the 
window ID from the results
            reString = 'xwininfo: Window id: ([0-9x]*)'

        else:
            # Get the x-windows information about the window that has 
the specified title
            command = ['xwininfo',
                       '-tree', '-root',
                       '| grep', WindowTitle]

            # Set the regular expression string for extracting the 
window ID from the results
            reString = '([0-9x]*) "%s"' % title
       
        # Run the command
        proc = subprocess.Popen(command, stdout=subprocess.PIPE)
           
        # Get the results.  Blocks until the results are ready.
        (resultsText, errText) = proc.communicate()

        # Check for error status
        if proc.returncode:
            raise subprocess.CalledProcessError(proc.returncode, cmdText)

        # Extract the window id from the results string.
        result = re.findall('xwininfo: Window id: ([0-9x]*)', resultsText)
        if result:
            self._windowID = result[0]

    def CaptureFrame(self):
        """Capture a still image of the current display in the temp 
directory."""
        if self._windowID:
            command = ['import',
                       '-window', self._windowID,
                       '%s/capture-%06d.miff' % (self._fileLocation, 
self._frameCounter)]
            subprocess.check_call(command)
           
            # Increment the frame counter
            self._frameCounter += 1

    def EncodeMovie(self, FrameTiming, Filename):
        """Generate an mpeg movie from the sequence of frames captured.  
FrameTiming is the
        time spacing between frames specified in seconds as a floating 
point number.  Due to
        the limitation of the convert utility the timing resolution is 
limited to 10 milliseconds.
        Unfortunately that means typical video frame rates are limited 
to 25, 33 1/3, or 50 frames
        per second.  If you want real-time to be preserved make sure you 
animation frame spacing matches."""
        if self._frameCounter > 0:
            delay = int(100 * FrameTiming)
           
            # Add the .mpg extension if it was not provided.
            if Filename.find('.mpg') < 0:
                Filename += '.mpg'

            # Use convert to make the movie.
            command = ['convert',
                       '-delay', '%d' % delay,
                       '%s/*.miff' % self._fileLocation,
                       Filename]
            subprocess.check_call(command)
   
    def DeleteCapturedFrames(self):
        """Delete the temp directory holding the sequence of captured 
frames."""
        shutil.rmtree(self._fileLocation)

[Visualpython-users] help in fedora 11

[jorge a s. <ta...@ho...>]

hi Everybody

i've tried to install vpython in fedora 11 but it doesn't work 

this is the mistake

make[1]: *** [cvisualmodule.lo] Error 1

thanks to all
 		 	   		  
_________________________________________________________________
Infórmate, mantente en contacto y encuéntralo todo, a la vez. Con la nueva Toolbar de MSN  nunca has tenido tantas ventajas en tan poco espacio. 
http://toolbar.es.msn.com/

Re: [Visualpython-users] Why no screenshot function?

[Guy K. K. <g....@ma...>]

On Mon, 19 Oct 2009 06:58:44 Bruce Sherwood wrote:
> And there is PIL, the Python Imaging Library, which can do lots of image 
> conversions. But the fundamental problem is that you'll have a
>  low-resolution  bitmap image, which isn't good for many applications.

I've been thinking about this issue in the past as well. Here are some 
thoughts that crossed me:

 * Screenshot utilities exist for the different operating systems

 * They are different in usage, and they require manual interaction

 * A *scriptable* tool that would work from within Visual could be desirable:
   - which is platform independent, and works the same way on all platforms
   - which just dumps the current frame to a file
   - potentially just as a PIL image, which could be saved easily to any
     image file format

 * Benefits of such a tool:
   - easy, independent grabbing of frame
   - later easy joining of frames to a movie sequence possible
   - especially suitable for non-interactive simulation/rendering
   - I'm using ffmpeg2theora all the time to aggregate PNG sequences
     into an Ogg Theora movie file that can be played with any modern
     browser

So, even though the quality of a grabbed frame like that would be inferior to 
a POVray rendered scene, for certain applications (e. g. creating movies, 
simple documentation of a current state) it would be by far good enough.

If someone has an easy solution, you'd have my vote to embed it into Visual!

Guy


-- 
Guy K. Kloss
Institute of Information and Mathematical Sciences
Te Kura Pūtaiao o Mōhiohio me Pāngarau
Massey University, Albany (North Shore City, Auckland)
473 State Highway 17, Gate 1, Mailroom, Quad B Building
voice: +64 9 414-0800 ext. 9585   fax: +64 9 441-8181
G....@ma... http://www.massey.ac.nz/~gkloss

Re: [Visualpython-users] Why no screenshot function?

[Robert X. <nne...@gm...>]

Turning on full-screen view (scene.fullscreen = True) usually gives a good
enough bitmap resolution for most purposes.

Robert

2009/10/18 Bruce Sherwood <Bru...@nc...>

> And there is PIL, the Python Imaging Library, which can do lots of image
> conversions. But the fundamental problem is that you'll have a
> low-resolution
> bitmap image, which isn't good for many applications.
>
> Bruce Sherwood
>
> Stanley Sokolow wrote:
> > Bruce is correct that there are screenshot functions in the operating
> > systems, including LInux.   In Ubuntu Linux and probably most
> > distributions, there's a widget that can be added to the desktop panel
> > for taking a screenshot.   On Windows, I use the Shift-PrintScrn key
> > combination, then open the accessory program called Paint and paste
> > (Ctrl-V) the clipboard onto the canvas.   Using Paint you can crop and
> > make other changes, then save as jpeg or other formats.  I've used this
> > to make step-by-step instruction manuals for using software, with a
> > screenshot of each important step.
> >
> > Another useful program is the ImageMagick library, which is available
> > for many programming languages, including Python:
> > http://www.imagemagick.org/script/index.php.   It can do lots of image
> > conversions.
> >
> > Stan
> >
> > ------------------------------------
> > On Sun, Oct 18, 2009 at 8:00 AM, Bruce Sherwood <Bru...@nc...
> > <mailto:Bru...@nc...>> wrote:
> >
> >     That could be useful; perhaps the reason it's not there is that
> >     there hasn't
> >     been much need expressed for such a facility. Note that on Windows
> >     and Mac (and
> >     maybe on Linux?) there are facilities built-in to get screen shots.
> >     For example,
> >     Alt-Print Screen on Windows captures a bitmap image of the active
> >     window, which
> >     can then be pasted into various applications, including Word.
> >
> >     Converting to jpeg is NOT a small job. A jpeg is extremely different
> >     from a
> >     bitmap image. Any easily-implemented screen capture will necessarily
> >     be a bitmap
> >     image, and therefore necessarily low-resolution for print purposes.
> >
> >     For many purposes one wants something much better than a bitmap
> >     image. In the
> >     contributed section of vpython.org <http://vpython.org> the
> >     povexport utility generates a file that
> >     can be read by the freeware POV-ray application to generate a
> >     high-resolution
> >     ray-traced image of a VPython scene.
> >
> >     A separate issue is that it would be very useful, and probably not
> very
> >     difficult, to generate a PostScript file from a (two-dimensional)
> >     graph in order
> >     to produce high-resolution prints of graphs.
> >
> >     Bruce Sherwood
> >
> >     Anders Wallin wrote:
> >      > Hi all,
> >      >
> >      > Why is there no screenshot function in visual?
> >      >
> >      > I've looked a bit at SDL/OpenGL in C++ under Ubuntu, and it's
> about a
> >      > 5 min job to google for the correct screenshot code and get it
> >      > working. It saves a 24-bit TGA-file, but I imagine changing that
> to
> >      > bmp or jpeg or similar is not a big job either. I must be missing
> >      > something?
> >      >
> >      > thanks,
> >      >
> >      > Anders
> >
> >
> ------------------------------------------------------------------------------
> >     Come build with us! The BlackBerry(R) Developer Conference in SF, CA
> >     is the only developer event you need to attend this year. Jumpstart
> your
> >     developing skills, take BlackBerry mobile applications to market and
> >     stay
> >     ahead of the curve. Join us from November 9 - 12, 2009. Register now!
> >     http://p.sf.net/sfu/devconference
> >     _______________________________________________
> >     Visualpython-users mailing list
> >     Vis...@li...
> >     <mailto:Vis...@li...>
> >     https://lists.sourceforge.net/lists/listinfo/visualpython-users
> >
> >
>
>
> ------------------------------------------------------------------------------
> Come build with us! The BlackBerry(R) Developer Conference in SF, CA
> is the only developer event you need to attend this year. Jumpstart your
> developing skills, take BlackBerry mobile applications to market and stay
> ahead of the curve. Join us from November 9 - 12, 2009. Register now!
> http://p.sf.net/sfu/devconference
> _______________________________________________
> Visualpython-users mailing list
> Vis...@li...
> https://lists.sourceforge.net/lists/listinfo/visualpython-users
>