visualpython-users

Re: [Visualpython-users] Extra VPython window with gdisplay in 2.9.1

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

On Thu, 2004-04-08 at 14:40, Bruce Sherwood wrote:
> Is this with the experimental Boost-based VPython that Jonathan 
> Brandmeyer recently released? This is a known bug, and he's working on it.
> 
> Bruce Sherwood
> 
> Andrew Dougherty wrote:
> 
> > When I run the graphtest.py demo (or actually any program using gdisplay),
> > I get the regular 'Graph' window with the demo, but I also
> > seem to always get a second, unused window labeled 'VPython'.
> > 
> > Does anyone else see this?
> > 

I've checked in a patch for this bug to CVS, it will be released soon.

-Jonathan Brandmeyer

Re: [Visualpython-users] Extra VPython window with gdisplay in 2.9.1

[Andrew D. <dou...@la...>]

On Thu, 8 Apr 2004, Bruce Sherwood wrote:

> Is this with the experimental Boost-based VPython that Jonathan
> Brandmeyer recently released? This is a known bug, and he's working on it.

Yup, the 2.9.1 experimental release.  I just wanted to make sure it wasn't
a local problem on my end.

So far, all my other programs have worked just fine with the new
Boost-based version, so it's looking pretty nice!

-- 
    Andy Dougherty		dou...@la...
    Dept. of Physics
    Lafayette College, Easton PA 18042

Re: [Visualpython-users] Release: VPython 2.9.1

[Andrew D. <dou...@la...>]

On Wed, 7 Apr 2004, Jonathan Brandmeyer wrote:

> This release corrects bugs found in the 2.9.0 release:

Nice.  Thanks.  I ran into a few glitches while trying to build this on a
Debian Linux/SPARC64 system.  I've enclosed suggested patches (apply with
patch -p1 -N).

The changes are:

INSTALL.txt:
    1.  Debian's "Testing", unfortunately, has python2.3, but only
    Boost 1.30.2.  I put in some notes to that effect.  If it's too much
    or too verbose, go ahead and adjust it to taste.

    2.  Using PYTHONPATH and --prefix to install somewhere safe and out of
    the way only works if PYTHONPATH already exists.  I didn't think
    configure should do that (because you might need to be root to
    make the directory) so I just put in an explicit instruction.

    3.  I fixed a few trivial typos.

configure.ac:
    1.  If compilation of a program with #include <Python.h> fails, the
most likely reason is that Python.h doesn't exist, but there could also be
other local configuration errors.  (In my case, there was a
bit of 32-bit/64-bit confusion in python's pyport.h header.)  I've
suggested a somewhat more general message.

cvisual/Makefile.in:
    1.  The 'make install' step will fail if $prefix/bin doesn't exist
    already.  I added in a conditional mkdir -p command.  (I
    vaguely recall that mkdir -p is not completely portable, but probably
    should be ok everywhere that VPython is likely to be installed.)

diff -r -u visual-2.9.1/INSTALL.txt visual-2.9.1-andy/INSTALL.txt
--- visual-2.9.1/INSTALL.txt	2004-04-07 16:28:55.000000000 -0400
+++ visual-2.9.1-andy/INSTALL.txt	2004-04-08 13:22:48.000000000 -0400
@@ -91,6 +91,11 @@
 			libgtk1.2-dev
 			g++-3.2 (or g++-3.3).
 			libboost-python-dev
+
+		(As of April 8, 2004, Debian's "testing" distribution only contains
+		Boost libraries version 1.30.2.  You can either try installing
+		1.31.0 yourself or use the python2.2 packages.  You'll need
+		python2.2-dev, python2.2-numeric, and python2.2-tk.)

 		Users of Debian Woody and derivatives will need to obtain backports
 		of the above from (eg.) www.backports.org.
@@ -156,7 +161,8 @@
 	simultaneously.  After choosing an approprate prefix, such as /home/jonathan, run
 	configure as normally but specify the PYTHONPATH environment variable to be
 	[prefix]/lib/python2.3/site-packages, replacing python2.3 with python2.2 if that
-	is your interpreter.
+	is your interpreter.  You must create this PYTHONPATH directory if it doesn't
+	already exist.


 III.  Build
@@ -174,9 +180,9 @@
 	the CXXFLAGS variable.

 	The Makefiles support the following targets:
-	all: (the default) compiles are required software
+	all: (the default) compiles all the required software
 	clean: deletes files created by the compiler
-	distclean: deltes all files created by the configure and make programs.  This
+	distclean: deletes all files created by the configure and make programs.  This
 		option is useful if you want to start over from a clean source tree.
 	install: copies all required files in the locations specified by configure
 	install-strip: Performs the 'install' target and strips the final object files.
diff -r -u visual-2.9.1/configure.ac visual-2.9.1-andy/configure.ac
--- visual-2.9.1/configure.ac	2004-04-07 16:18:45.000000000 -0400
+++ visual-2.9.1-andy/configure.ac	2004-04-08 10:28:10.000000000 -0400
@@ -29,7 +29,7 @@


 AM_CHECK_PYTHON_HEADERS([], [
-	AC_MSG_ERROR([A suitable python interpreter was found, but you do not have the header files required for building C/C++ extensions to python.])])
+	AC_MSG_ERROR([A suitable python interpreter was found, but you do not have the header files required for building C/C++ extensions to python, or another problem was encountered compiling a program that includes <Python.h>.])])

 dnl Copied from the official gtk+-2 configure.in
 AC_MSG_CHECKING([for some Win32 platform])
diff -r -u visual-2.9.1/cvisual/Makefile.in visual-2.9.1-andy/cvisual/Makefile.in
--- visual-2.9.1/cvisual/Makefile.in	2004-04-07 16:18:46.000000000 -0400
+++ visual-2.9.1-andy/cvisual/Makefile.in	2004-04-08 11:22:35.000000000 -0400
@@ -22,6 +22,9 @@
 CXXDEPMODE = @CXXDEPMODE@
 STRIP = @STRIP@

+# Program to be determined by configure
+MKDIR_P = mkdir -p
+
 # System configurable variables determined by configure
 PYTHON_INCLUDES = @PYTHON_INCLUDES@
 PYTHON_PREFIX = @PYTHON_PREFIX@
@@ -247,6 +250,8 @@
 	@echo $(LIBTOOL) --mode=install $(INSTALL) $< $(DESTDIR)$(pythondir)/$< \
 		>> $(LOGFILE)
 	@$(LIBTOOL) --mode=install $(INSTALL) $< $(DESTDIR)$(pythondir)/$<
+	@echo test -d $(DESTDIR)$(bindir) || $(MKDIR_P) $(DESTDIR)$(bindir) >> $(LOGFILE)
+	@test -d $(DESTDIR)$(bindir) || $(MKDIR_P) $(DESTDIR)$(bindir)
 	@echo $(INSTALL) -m 755 vpython $(DESTDIR)$(bindir)/vpython >> $(LOGFILE)
 	@$(INSTALL) -m 755 vpython $(DESTDIR)$(bindir)/vpython
 else

-- 
    Andy Dougherty		dou...@la...
    Dept. of Physics
    Lafayette College, Easton PA 18042

Re: [Visualpython-users] Extra VPython window with gdisplay in 2.9.1

[Bruce S. <bas...@un...>]

Is this with the experimental Boost-based VPython that Jonathan 
Brandmeyer recently released? This is a known bug, and he's working on it.

Bruce Sherwood

Andrew Dougherty wrote:

> When I run the graphtest.py demo (or actually any program using gdisplay),
> I get the regular 'Graph' window with the demo, but I also
> seem to always get a second, unused window labeled 'VPython'.
> 
> Does anyone else see this?
> 

[Visualpython-users] Small vpython script suggestion

[Andrew D. <dou...@la...>]

I'm usually not working on demos, so I find the vpython's script default
working directory inconvenient.  Here's a little patch I apply to my local
copy.  If vpython gets no arguments, it starts up in the demo directory as
before.  If vpython *does* get an argument, however, it opens up the
indicated file.

--- visual-2.9.1/cvisual/vpython.in	2004-04-07 16:18:46.000000000 -0400
+++ visual-2.9.1-andy/cvisual/vpython.in	2004-04-08 13:26:04.000000000 -0400
@@ -2,5 +2,15 @@
 PYTHON=@PYTHON@
 visualdemodir=@visualdemodir@
 visualidledir=@visual_idledir@
-cd $visualdemodir
-$PYTHON $visualidledir/@IDLE@
\ No newline at end of file
+case "$1" in
+'-h') echo "USAGE:  $0 [file]"
+    echo "Default is to start in the demo directory: $visualdemodir"
+    exit 1
+    ;;
+'')
+    cd $visualdemodir
+    $PYTHON $visualidledir/@IDLE@
+    ;;
+*)  $PYTHON $visualidledir/@IDLE@ "$*"
+    ;;
+esac

-- 
    Andy Dougherty		dou...@la...
    Dept. of Physics
    Lafayette College, Easton PA 18042

[Visualpython-users] Extra VPython window with gdisplay in 2.9.1

[Andrew D. <dou...@la...>]

When I run the graphtest.py demo (or actually any program using gdisplay),
I get the regular 'Graph' window with the demo, but I also
seem to always get a second, unused window labeled 'VPython'.

Does anyone else see this?

-- 
    Andy Dougherty		dou...@la...
    Dept. of Physics
    Lafayette College, Easton PA 18042

[Visualpython-users] Release: VPython 2.9.1

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

This release corrects bugs found in the 2.9.0 release:

display.kb.getkey() was inadvertently renamed display.kb.getkeys()
(corrected).  Thanks to James Roper for finding this bug.

Implicit and explicit conversion from two-tuples to vectors left the z
component of the resulting vector undefined.  On some builds this would
be unnoticeable (setting z to some very small number close to zero) and
on others it would set z to be some very large number.

The instructions for building from source have been corrected in a few
places.  See INSTALL.txt in the source tarball for details.

The Windows download does not include the leading drive letter in the
filenames, which cause problems for some people.  Actually, I "fixed"
this some time ago, but the resulting upload was found later to be
corrupt.

The new downloads may be found at http://www4.ncsu.edu/~jdbrandm/.

-Jonathan Brandmeyer

Re: [Visualpython-users] Vpython For Low-Tech Artists

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

On Tue, 2004-04-06 at 15:42, Joel Kahn wrote:
> Thanks for the replies so far. I should make it clear
> that I'm running python in Windows--XP on my machine,
> and presumably in various versions from 95 through XP
> on other machines that may be involved. Any CD-ROM
> setups of vpython that I use will need to be in this
> context. If I encounter Mac/Linux/&c situations, I'll
> deal with them as they occur.
> 
> Joel

I would use Bruce's suggestion with one small enhancement.
Create a file named autorun.bat in the Python23 directory with this
line:
pythonw.exe .\Lib\idlelib\idle.pyw

Also create a file autorun.inf (same place) with these contents:
[autorun]
open=autorun.bat
icon=py.ico

Then burn the disk with the complete contents of the Python23 directory
and you will get a Python "Live" CD that will automatically open up IDLE
when inserted into the host computer.  Running from CD isn't exactly
fast, though.  The biggest slowdown will be noticed when executing
import statements.

HTH,
Jonathan Brandmeyer

[Visualpython-users] Vpython For Low-Tech Artists

[Joel K. <jj...@ya...>]

Thanks for the replies so far. I should make it clear
that I'm running python in Windows--XP on my machine,
and presumably in various versions from 95 through XP
on other machines that may be involved. Any CD-ROM
setups of vpython that I use will need to be in this
context. If I encounter Mac/Linux/&c situations, I'll
deal with them as they occur.

Joel


__________________________________
Do you Yahoo!?
Yahoo! Small Business $15K Web Design Giveaway 
http://promotions.yahoo.com/design_giveaway/

Re: [Visualpython-users] Vpython For Low-Tech Artists

[Martin G. <ge...@la...>]

Regarding item #2, while I have not done this, it shouldn't be
too hard to do, starting with Morphix (which is intended for
building custom live-CD linux distributions).  This might not
be what you had in mind, though---or was it?

Martin Gelfand
Dept of Physics, Colorado State

On Mon, 5 Apr 2004, Joel Kahn wrote:

> I have two related questions.
>
> 1. I am looking at ways to use vpython as a set of
> tools to help artists who have relatively limited
> computer expertise and who mainly work in traditional
> physical media--textiles, sculpture, jewelry, &c. Does
> anyone have info on work that has already been done
> along these general lines?
>
> 2. In connection with the above, I want to have an
> installation of vpython that can effectively and
> reliably run entirely off of a CD-ROM on a computer
> that has no python stuff installed at all. I'm not
> sure whether my attempts so far to set this up are
> going to function right or not. Has anyone out there
> made this work? If so, can I get some details?
>
> Any help will be greatly appreciated. I'll be glad to
> pass progress reports on to interested parties.
>
> Joel Kahn
> Springfield, Missouri
>
>
> __________________________________
> Do you Yahoo!?
> Yahoo! Small Business $15K Web Design Giveaway
> http://promotions.yahoo.com/design_giveaway/
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by: IBM Linux Tutorials
> Free Linux tutorial presented by Daniel Robbins, President and CEO of
> GenToo technologies. Learn everything from fundamentals to system
> administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click
> _______________________________________________
> Visualpython-users mailing list
> Vis...@li...
> https://lists.sourceforge.net/lists/listinfo/visualpython-users
>

Re: [Visualpython-users] Vpython For Low-Tech Artists

[Bruce S. <bas...@un...>]

I don't happen to know about people using VPython for artistic purposes, 
but on question 2 I'll comment that there may be a solution for Windows. 
It is possible to install Python without touching the Windows registry, 
so you don't need administrative access. I have for example run this 
version (including VPython) out of a remote file on a public-cluster 
Windows machine on which I couldn't install software myself.

If I remember correctly, what I did was install Python in the special 
nonadministrative mode (which if I remember correctly was an option in 
the install process), then run the VPython installer, telling it where 
Python was. Then the entire Python23 folder could simply be written to 
CD and used from there, I would think. I think the only (mild) 
limitation is that *.py files won't automatically be linked to Python. 
But as long as you start from IDLE, rather than double-clicking a *.py 
file, this won't matter.

Bruce Sherwood

Joel Kahn wrote:

>I have two related questions.
>
>1. I am looking at ways to use vpython as a set of
>tools to help artists who have relatively limited
>computer expertise and who mainly work in traditional
>physical media--textiles, sculpture, jewelry, &c. Does
>anyone have info on work that has already been done
>along these general lines?
>
>2. In connection with the above, I want to have an
>installation of vpython that can effectively and
>reliably run entirely off of a CD-ROM on a computer
>that has no python stuff installed at all. I'm not
>sure whether my attempts so far to set this up are
>going to function right or not. Has anyone out there
>made this work? If so, can I get some details?
>
>Any help will be greatly appreciated. I'll be glad to
>pass progress reports on to interested parties.
>
>Joel Kahn
>Springfield, Missouri
>
>
>__________________________________
>Do you Yahoo!?
>Yahoo! Small Business $15K Web Design Giveaway 
>http://promotions.yahoo.com/design_giveaway/
>
>
>-------------------------------------------------------
>This SF.Net email is sponsored by: IBM Linux Tutorials
>Free Linux tutorial presented by Daniel Robbins, President and CEO of
>GenToo technologies. Learn everything from fundamentals to system
>administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click
>_______________________________________________
>Visualpython-users mailing list
>Vis...@li...
>https://lists.sourceforge.net/lists/listinfo/visualpython-users
>  
>

[Visualpython-users] Vpython For Low-Tech Artists

[Joel K. <jj...@ya...>]

I have two related questions.

1. I am looking at ways to use vpython as a set of
tools to help artists who have relatively limited
computer expertise and who mainly work in traditional
physical media--textiles, sculpture, jewelry, &c. Does
anyone have info on work that has already been done
along these general lines?

2. In connection with the above, I want to have an
installation of vpython that can effectively and
reliably run entirely off of a CD-ROM on a computer
that has no python stuff installed at all. I'm not
sure whether my attempts so far to set this up are
going to function right or not. Has anyone out there
made this work? If so, can I get some details?

Any help will be greatly appreciated. I'll be glad to
pass progress reports on to interested parties.

Joel Kahn
Springfield, Missouri


__________________________________
Do you Yahoo!?
Yahoo! Small Business $15K Web Design Giveaway 
http://promotions.yahoo.com/design_giveaway/

[Visualpython-users] EMWave.py

[Bruce S. <bas...@un...>]

In the contributed section at http://vpython.org there is a new program 
EMWave.py by Rob Salgado that will be of high interest to those who 
teach about electromagnetic waves. Rob has arranged to show not only the 
propagating electric and magnetic fields but also their time 
derivatives. The time derivative of the magnetic field produces the 
electric field, and the time derivative of the electric field produces 
the magnetic field, and.....

Thanks, Rob!

Bruce Sherwood

Re: [Visualpython-users] Announcement: VPython 2.9 release

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

On Sun, 2004-03-21 at 22:08, Arthur wrote:
> > Windows users using Python 2.3 should first uninstall VPython, and using
> > Winzip or some other utility, unpack the file vpython-msw-2.9.tar.gz to
> > the root drive.  This download assumes that you are using Python 2.3 in
> > the stock configuration on drive C:\.
> 
> Jonathan - 
> 
> I've been looking forward to the new VPython.
> 
> I am having an issue untarring the Windows binary.  I believe the issue is
> having the "c:\" drive hardcoded.  The file unpacks fine if I turn off "use
> folder names" - but then it unpacks all files in a flat directory.  If I
> don't turn odd "use folder names" it seems to choke.  Looking at other
> distros, it seems the norm is to include the structure, but without the
> drive letter hardcoded.
> 
> Art 

Point taken, I'll update the package tomorrow afternoon.

Also note that this is not how I expect to release the final Windows
binary; it's only an interim measure until I can get Dr. Sherwood's
existing installer ported to include the few extra .dll's VPython
requires.

Thanks,
Jonathan Brandmeyer

FW: [Visualpython-users] Announcement: VPython 2.9 release

[Arthur <ajs...@op...>]

> Windows users using Python 2.3 should first uninstall VPython, and using
> Winzip or some other utility, unpack the file vpython-msw-2.9.tar.gz to
> the root drive.  This download assumes that you are using Python 2.3 in
> the stock configuration on drive C:\.

Jonathan - 

I've been looking forward to the new VPython.

I am having an issue untarring the Windows binary.  I believe the issue is
having the "c:\" drive hardcoded.  The file unpacks fine if I turn off "use
folder names" - but then it unpacks all files in a flat directory.  If I
don't turn odd "use folder names" it seems to choke.  Looking at other
distros, it seems the norm is to include the structure, but without the
drive letter hardcoded.

Art 

[Visualpython-users] Announcement: VPython 2.9 release

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

I am pleased to announce the release of VPython 2.9, a public beta
release of what will become version 3.0.  This release contains changes
that are mostly "under the hood" but allow us to extend Visual in
several new ways by replacing the Python <-> C++ language interface with
Boost.Python.  The biggest immediately available user-visible changes
include:

-Docstrings for all free functions and class member functions,
-The ability to subclass existing classes in Visual for extending
existing objects in Python itself,
-A few more vector operations (see help(vector) for details),
-New classes vector_array and scalar_array: these have an interface
similar to that provided by Numeric, but it provides R3-vector specific
operations that are significantly faster than Numeric and are more
readable than the equivalent Numeric code,
-Exception reporting for calling a function with the wrong arguments is
automated by Boost,
-And an increased time to compile (sorry, but at least you only do this
once).

There is a big change in the way the C++ parts of Visual are compiled. 
Rather than display the individual commands verbatim on the screen, they
are logged to a separate file cvisual/build.log.

The purpose of producing this release series is to enable existing users
of VPython to test their existing programs with the new version to
ensure that the new interface is as close to 100% backwards-compatable
as possible.  This also gives existing users and educators a chance to
comment on the new interfaces, exceptions, and built-in documentation
strings.  I expect to make a handful of 2.9.x releases based on your
input before naming the result 3.0.

The release packages may be found at http://www4.ncsu.edu/~jdbrandm/. 
There are four packages.
Windows users using Python 2.3 should first uninstall VPython, and using
Winzip or some other utility, unpack the file vpython-msw-2.9.tar.gz to
the root drive.  This download assumes that you are using Python 2.3 in
the stock configuration on drive C:\.

Existing VPython users on OSX 10.2 or 10.3 will only need to untar the
package.  While in the "/" directory, issue `tar -xzf
path/to/vpython-osx10.2-2.9.0.tar.gz` (or
vpython-osx10.3-2.9.0.tar.gz).  This will overwrite your existing
installation.  You can list the contents of the tarball with the command
`tar -tzf path/to/vpython...`

Linux users should download the source package and follow the
instructions in the enclosed INSTALL.txt file.

I look forward to your feedback,
-Jonathan Brandmeyer

[Visualpython-users] Dear admin

[<boo...@ya...>]

Dear program admin,

you seem to have a pretty interesting opportunity going on.

Do you have the amount of members you really wanted to have ?

What if I would tell you that you can reach 12,000 safe-lists
opportunity seekers, known gold investors ?

Interested ?

What if I told you that you ad can be seen by those 12,000
opportunity seekers for only 5$ ?

Simply compose an ad for your program,

and send it to me via e-mail at : boo...@ya... with the batch number
of your 5$ spend into my e-gold account # 1229368 (Simon Papineau)

I will shortly respond you and your ad will be on it's way.

Wishing you the best,
Simon Papineau

Re: [Visualpython-users] obtaining pos in world co-ordinates of a frame obj

[Jonathan B. <jdb...@un...>]

Bruce Sherwood wrote:

> I've added Jonathan's routine to the on-line documentation for "frame" 
> and also updated the Doc zip file that can be downloaded. Thanks, 
> Jonathan.
>
> This does leave open the question however as to whether there ought 
> not be a method in Visual for resolving pos and axis information. In 
> particular, I believe that even Jonathan's routine isn't really 
> complete if this frame is inside another frame: one should really 
> recurse up the enclosing frames.
>
> Bruce Sherwood
>
> Jonathan Brandmeyer wrote:
>
>> Let me amend that routine.  You also have to normalize the y_axis vector
>> (smacks self on head).
>>
>> def world_space_pos( frame, local):
>>     """Returns the position of local in world space."""
>>     x_axis = norm(frame.axis)
>>     z_axis = norm(cross(frame.axis, frame.up))
>>     y_axis = norm(cross(z_axis, x_axis))
>>     return frame.pos + local.x*x_axis + local.y*y_axis + local.z*z_axis
>
Bruce is right: the above function should be named "parent_space_pos" 
instead.
For completeness, here is what the recursive version would look like, in 
the event anyone needs it right now:

def world_space_pos(frame, local):
    x_axis = norm(frame.axis)
    z_axis = norm(cross(frame.axis, frame.up))
    y_axis = norm(cross(z_axis, x_axis))
    parent_space = frame.pos + local.x*x_axis + local.y*y_axis + 
local.z*z_axis
    if (frame.frame == None):
       return parent_space
    else:
       return world_space_pos(frame.frame, parent_space)

Re: [Visualpython-users] obtaining pos in world co-ordinates of a frame obj

[Jonathan B. <jdb...@un...>]

Bruce Sherwood wrote:

> I've added Jonathan's routine to the on-line documentation for "frame" 
> and also updated the Doc zip file that can be downloaded. Thanks, 
> Jonathan.
>
> This does leave open the question however as to whether there ought 
> not be a method in Visual for resolving pos and axis information. In 
> particular, I believe that even Jonathan's routine isn't really 
> complete if this frame is inside another frame: one should really 
> recurse up the enclosing frames.
>
> Bruce Sherwood
>
How about a pair of  member functions of the frame object with the 
following Python signatures:

def world_space_project(self, local):
    """Returns the vector local in the global coordinate system, 
following transforms recursively within nested frames."""

def parent_space_project(self, local):
    """Returns the vector local in the coordinate system of the parent.  
Iff self.frame==None, this is the same as world_space_project()."""

Comments?  Questions?
-Jonathan Brandmeyer

Re: [Visualpython-users] point source?

[Bruce S. <bas...@un...>]

It would indeed be incredibly useful to have a point source of light, 
for many reasons, including illustrating the phases of the Moon. But the 
current lighting model in Visual is very basic, limited to sources at 
infinity. It will require much work to get more sophisticated lighting. 
Jonathan Brandmeyer is hoping to work on this soon, after he finishes a 
very major project to use Boost to connect between Visual and Python.

Here's a routine which, after a mouse click, changes your point of view 
to be looking at the cyan sphere from near the yellow sphere. The key 
notion is that the center of the scene is under the control of the 
program, but the camera direction and distance are normally under the 
control of the user (using right and left mouse-button drags). You can 
disable user control (userzoom = 0 and/or userspin = 0). Note that 
scene.fov ("field of view") is an angle in radians such that the tangent 
of half that angle is the magnitude of the range divided by the distance 
from camera to center.

from visual import *
from __future__ import division

scene.range = 3.5
s1 = sphere(radius=0.5, color=color.yellow)
s2 = sphere(pos=(3,0,0), radius=0.4, color=color.cyan)
scene.mouse.getclick() # wait for mouse click
scene.center = s2.pos # about which user rotates with right mouse button
scene.forward = (1,0,0) # pointing from camera toward center
# And adjust distance from center to camera:
scene.fov = 2*atan(mag(scene.range)/(mag(s2.pos-s1.pos)-s1.radius))
# Now you are near the yellow sphere, looking toward the cyan sphere.

Bruce Sherwood

Lew Riley wrote:

> Hi Folks,
>
> I have two feature requests (unless I've overlooked things that 
> already exist).
>
> 1) It would be incredibly useful to have a point source of light (with 
> 1/r^2 intensity dependence) in addition to the "off-stage" lighting.
>
> 2) I would find it helpful to be able to set the camera position 
> directly in addition to the indirect influences of display.forward and 
> display.center.
>
> I am working on animations of the Ptolemaic and Copernican models of 
> the inner planets. I would like to show the phases of the inner 
> planets. Feature 1 would allow me to use the Sun to illuminate the 
> planets without needing to place the Sun off-camera. Feature 2 would 
> provide me an easy way to place the camera on the Earth.
>
> Feature 2 would also be helpful to anyone illustrating reference 
> frames. That is, one could easily place the observer in various frames 
> of reference to show how observations differ.
>
> Finally, would someone please explain how display.mouse.camera is 
> calculated from display.forward and display.center (and what else)?
>
> Thank you for maintaining an incredibly useful tool.
>
> Cheers,
>
> Lew

[Visualpython-users] point source?

[Lew R. <lr...@ur...>]

Hi Folks,

I have two feature requests (unless I've overlooked things that already 
exist).

1) It would be incredibly useful to have a point source of light (with 
1/r^2 intensity dependence) in addition to the "off-stage" lighting.

2) I would find it helpful to be able to set the camera position 
directly in addition to the indirect influences of display.forward and 
display.center.

I am working on animations of the Ptolemaic and Copernican models of the 
inner planets. I would like to show the phases of the inner planets. 
Feature 1 would allow me to use the Sun to illuminate the planets 
without needing to place the Sun off-camera. Feature 2 would provide me 
an easy way to place the camera on the Earth.

Feature 2 would also be helpful to anyone illustrating reference frames. 
That is, one could easily place the observer in various frames of 
reference to show how observations differ.

Finally, would someone please explain how display.mouse.camera is 
calculated from display.forward and display.center (and what else)?

Thank you for maintaining an incredibly useful tool.

Cheers,

Lew
-- 
___________________________________________________
Lew Riley        http://webpages.ursinus.edu/lriley
Ursinus College Department of Physics and Astronomy
(610) 409-3000 ext. 2293       (610) 409-3660 (FAX)

Re: [Visualpython-users] obtaining pos in world co-ordinates of a frame obj

[Bruce S. <bas...@un...>]

I've added Jonathan's routine to the on-line documentation for "frame" 
and also updated the Doc zip file that can be downloaded. Thanks, Jonathan.

This does leave open the question however as to whether there ought not 
be a method in Visual for resolving pos and axis information. In 
particular, I believe that even Jonathan's routine isn't really complete 
if this frame is inside another frame: one should really recurse up the 
enclosing frames.

Bruce Sherwood

Jonathan Brandmeyer wrote:

> Let me amend that routine.  You also have to normalize the y_axis vector
> (smacks self on head).
> 
> def world_space_pos( frame, local):
>     """Returns the position of local in world space."""
>     x_axis = norm(frame.axis)
>     z_axis = norm(cross(frame.axis, frame.up))
>     y_axis = norm(cross(z_axis, x_axis))
>     return frame.pos + local.x*x_axis + local.y*y_axis + local.z*z_axis

Re: [Visualpython-users] obtaining pos in world co-ordinates of a frame obj

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

On Sun, 2004-03-14 at 17:13, Jonathan Brandmeyer wrote:
> On Sun, 2004-03-14 at 15:52, Axawire wrote:
> > hi
> > 
> > i have created a few frame objects and moved them about in the scene and
> > rotated them. if i select an object in a frame how do i get its pos in world
> > co-ordinates (co-ordinates of the whole scene)?
> > 
> > thanks
> > 
> 
> Simple case: the orientation of the frame has not changed (ie, up points
> along +y and axis points along +x)  simply add the object's pos to the
> frame's pos.
> 
> Harder case: the orientation of the frame is different.  In this case
> you will need to calculate the unit vectors representing the axes of the
> frames local coordinate system, and calculate the object's displacement
> along those unit vectors.
> 
> # Requirements:
> # frame has the following attributes: vector pos, vector axis, vector up
> #    (met by all primitive objects in Visual).
> # local has the following attributes: double x, double y, double z.
> #    (met by vectors and primitive objects in Visual).
> # frame.axis and frame.up are not colinear. (undefined coordinate 
> #    system)
> def world_space_pos( frame, local):
>     """Returns the position of local in world space."""
>     x_axis = norm(frame.axis)
>     z_axis = norm(cross(frame.axis, frame.up))
>     y_axis = cross(z_axis, x_axis))
>     return frame.pos + local.x*x_axis + local.y*y_axis + local.z*z_axis
> 

Let me amend that routine.  You also have to normalize the y_axis vector
(smacks self on head).

def world_space_pos( frame, local):
    """Returns the position of local in world space."""
    x_axis = norm(frame.axis)
    z_axis = norm(cross(frame.axis, frame.up))
    y_axis = norm(cross(z_axis, x_axis))
    return frame.pos + local.x*x_axis + local.y*y_axis + local.z*z_axis

-Jonathan Brandmeyer

Re: [Visualpython-users] obtaining pos in world co-ordinates of a frame obj

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

On Sun, 2004-03-14 at 15:52, Axawire wrote:
> hi
> 
> i have created a few frame objects and moved them about in the scene and
> rotated them. if i select an object in a frame how do i get its pos in world
> co-ordinates (co-ordinates of the whole scene)?
> 
> thanks
> 

Simple case: the orientation of the frame has not changed (ie, up points
along +y and axis points along +x)  simply add the object's pos to the
frame's pos.

Harder case: the orientation of the frame is different.  In this case
you will need to calculate the unit vectors representing the axes of the
frames local coordinate system, and calculate the object's displacement
along those unit vectors.

# Requirements:
# frame has the following attributes: vector pos, vector axis, vector up
#    (met by all primitive objects in Visual).
# local has the following attributes: double x, double y, double z.
#    (met by vectors and primitive objects in Visual).
# frame.axis and frame.up are not colinear. (undefined coordinate 
#    system)
def world_space_pos( frame, local):
    """Returns the position of local in world space."""
    x_axis = norm(frame.axis)
    z_axis = norm(cross(frame.axis, frame.up))
    y_axis = cross(z_axis, x_axis))
    return frame.pos + local.x*x_axis + local.y*y_axis + local.z*z_axis


HTH,
-Jonathan Brandmeyer
	

[Visualpython-users] obtaining pos in world co-ordinates of a frame obj

[Axawire <ax...@ch...>]

hi

i have created a few frame objects and moved them about in the scene and
rotated them. if i select an object in a frame how do i get its pos in world
co-ordinates (co-ordinates of the whole scene)?

thanks

rich
---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.610 / Virus Database: 390 - Release Date: 3/3/2004