pyode-user

[Pyode-user] Status of the wrappers?

[Matthias B. <ba...@ir...>]

Hi,

I would like to ask about the status of both ODE wrappers, Timothy's
original wrappers and Brett's rewrite. Are there any plans to continue
any of them? I suppose you know that I also have a wrapper
(http://i31www.ira.uka.de/~baas/pyode/), but it's also quite out of date
and, probably as everybody else here, I don't have much time to maintain
it. I've recently got a couple of mails from people asking about when
there will be a new version including trimesh support.
That's why I was reminded that it would really be much better if there
was only *one* binding with several people behind it.
So is there still any interest in creating a "unified" ODE binding?

Cheers,

- Matthias -

Re: [Pyode-user] Status of the wrappers?

[Bernie R. <br...@ec...>]

For what it's worth, I'm using the version at ira.uka.de and I find it 
works well.  I've modified it slightly to expose a couple of additional 
methods that I needed access to (setCategoryBits() and setCollideBits()), 
and found that it was quite easy to do even though I'd never used Pyrex 
before.  I'm also building PyODE with the most recent version of ODE (0.5).

The main reasons I chose the ira.uka.de implementation were the fact that 
it had some nice tutorials that made it very easy to get up to speed on 
using it, and it provided a way to implement the near_callback() in Python 
(which I needed for my application).  I'm not clear as to whether the other 
implementations provide this capability or not.

Certainly the fact that ODE is written in C (as opposed to C++) makes it a 
very natural fit for the Pyrex-based approach.  The C++ interface to ODE is 
essentially just a wrapper, and the idea of wrapping a wrapper is 
unappealing to me.  With the Pyrex approach, you get an object-oriented 
interface (which I consider essential) while still using a single layer of 
wrapper.

In any case...  perhaps a good way to kick-start this process is to 
identify what specific functionality it is that we need that is not 
currently available (e.g. trimesh support, which is high on my list) and 
discuss how to add it.  If turns out that converging on a single 
implementation is impractical for whatever reason, it would at least be 
nice to provide some interchangeability between the three.

And yes, I'd be willing to put some time into actually writing code (as 
opposed to just commenting on work that other people have already done or 
will be doing).

Looking forward to hearing everyone's thoughts, and the status off all 
three implementations...




At 11:46 PM 6/20/2004 +0200, Matthias Baas wrote:
>Hi,
>
>I would like to ask about the status of both ODE wrappers, Timothy's
>original wrappers and Brett's rewrite. Are there any plans to continue
>any of them? I suppose you know that I also have a wrapper
>(http://i31www.ira.uka.de/~baas/pyode/), but it's also quite out of date
>and, probably as everybody else here, I don't have much time to maintain
>it. I've recently got a couple of mails from people asking about when
>there will be a new version including trimesh support.
>That's why I was reminded that it would really be much better if there
>was only *one* binding with several people behind it.
>So is there still any interest in creating a "unified" ODE binding?
>
>Cheers,
>
>- Matthias -
>
>
>
>-------------------------------------------------------
>This SF.Net email is sponsored by The 2004 JavaOne(SM) Conference
>Learn from the experts at JavaOne(SM), Sun's Worldwide Java Developer
>Conference, June 28 - July 1 at the Moscone Center in San Francisco, CA
>REGISTER AND SAVE! http://java.sun.com/javaone/sf Priority Code NWMGYKND
>_______________________________________________
>Pyode-user mailing list
>Pyo...@li...
>https://lists.sourceforge.net/lists/listinfo/pyode-user

--
   Bernie Roehl
   University of Waterloo Dept of Electrical and Computer Engineering
   Mail: br...@ec...   Voice: (519) 888-4567 x 2607 [work]
   URL: http://ece.uwaterloo.ca/~broehl 

Re: [Pyode-user] Status of the wrappers?

[brett h. <bha...@ya...>]

Hello Matthias and Bernie,

Unfortunately, i have been very busy with other work in the past couple months, and i have not yet
had time to finish my version of the ODE wrapper.  I still plan to come back to it, but this may
not be for a coule months.  In the mean time, if someone beats me to the punch using another
method, i'll gladly be using their binding when it comes out, and fix bugs in it when i have time.

If not, then i plan to move forward using Boost Python to wrap the c++ wrappers.  Wrapping a
wrapper is not ideal, but from the perspective where several python wrappers must be created and
maintained, it makes sense that they all use the same wrapping language.  I am also maintaining
the PyOSG wrappers, and soon i'll be doing the Cassowary wrappers.  I have looked at the Pyrex
documentation, and it is impressive and has some real elegance.  So it is a blancing act, do we
shoehorn ODE to fit into the Boost model, or do it the clean and quick way with Pyrex?  

Thoughts?
Btw, is anybody in the San Francisco area?

-brett

--- Bernie Roehl <br...@ec...> wrote:
> For what it's worth, I'm using the version at ira.uka.de and I find it 
> works well.  I've modified it slightly to expose a couple of additional 
> methods that I needed access to (setCategoryBits() and setCollideBits()), 
> and found that it was quite easy to do even though I'd never used Pyrex 
> before.  I'm also building PyODE with the most recent version of ODE (0.5).
> 
> The main reasons I chose the ira.uka.de implementation were the fact that 
> it had some nice tutorials that made it very easy to get up to speed on 
> using it, and it provided a way to implement the near_callback() in Python 
> (which I needed for my application).  I'm not clear as to whether the other 
> implementations provide this capability or not.
> 
> Certainly the fact that ODE is written in C (as opposed to C++) makes it a 
> very natural fit for the Pyrex-based approach.  The C++ interface to ODE is 
> essentially just a wrapper, and the idea of wrapping a wrapper is 
> unappealing to me.  With the Pyrex approach, you get an object-oriented 
> interface (which I consider essential) while still using a single layer of 
> wrapper.
> 
> In any case...  perhaps a good way to kick-start this process is to 
> identify what specific functionality it is that we need that is not 
> currently available (e.g. trimesh support, which is high on my list) and 
> discuss how to add it.  If turns out that converging on a single 
> implementation is impractical for whatever reason, it would at least be 
> nice to provide some interchangeability between the three.
> 
> And yes, I'd be willing to put some time into actually writing code (as 
> opposed to just commenting on work that other people have already done or 
> will be doing).
> 
> Looking forward to hearing everyone's thoughts, and the status off all 
> three implementations...
> 
> 
> 
> 
> At 11:46 PM 6/20/2004 +0200, Matthias Baas wrote:
> >Hi,
> >
> >I would like to ask about the status of both ODE wrappers, Timothy's
> >original wrappers and Brett's rewrite. Are there any plans to continue
> >any of them? I suppose you know that I also have a wrapper
> >(http://i31www.ira.uka.de/~baas/pyode/), but it's also quite out of date
> >and, probably as everybody else here, I don't have much time to maintain
> >it. I've recently got a couple of mails from people asking about when
> >there will be a new version including trimesh support.
> >That's why I was reminded that it would really be much better if there
> >was only *one* binding with several people behind it.
> >So is there still any interest in creating a "unified" ODE binding?
> >
> >Cheers,
> >
> >- Matthias -
> >
> >
> >
> >-------------------------------------------------------
> >This SF.Net email is sponsored by The 2004 JavaOne(SM) Conference
> >Learn from the experts at JavaOne(SM), Sun's Worldwide Java Developer
> >Conference, June 28 - July 1 at the Moscone Center in San Francisco, CA
> >REGISTER AND SAVE! http://java.sun.com/javaone/sf Priority Code NWMGYKND
> >_______________________________________________
> >Pyode-user mailing list
> >Pyo...@li...
> >https://lists.sourceforge.net/lists/listinfo/pyode-user
> 
> --
>    Bernie Roehl
>    University of Waterloo Dept of Electrical and Computer Engineering
>    Mail: br...@ec...   Voice: (519) 888-4567 x 2607 [work]
>    URL: http://ece.uwaterloo.ca/~broehl 
> 
> 
> 
> -------------------------------------------------------
> This SF.Net email is sponsored by The 2004 JavaOne(SM) Conference
> Learn from the experts at JavaOne(SM), Sun's Worldwide Java Developer
> Conference, June 28 - July 1 at the Moscone Center in San Francisco, CA
> REGISTER AND SAVE! http://java.sun.com/javaone/sf Priority Code NWMGYKND
> _______________________________________________
> Pyode-user mailing list
> Pyo...@li...
> https://lists.sourceforge.net/lists/listinfo/pyode-user
> 



		
__________________________________
Do you Yahoo!?
Yahoo! Mail - 50x more storage than other providers!
http://promotions.yahoo.com/new_mail

Re: [Pyode-user] Status of the wrappers?

[Bernie R. <br...@ec...>]

At 11:05 AM 6/21/2004 -0700, brett hartshorn wrote:
>Wrapping a
>wrapper is not ideal, but from the perspective where several python 
>wrappers must be created and
>maintained, it makes sense that they all use the same wrapping language.

Perhaps, though I think it depends on the circumstances.

For example, in the project I'm working on, I'm wrapping a C++ graphics 
library using SWIG and
wrapping ODE using Pyrex.  My reasoning is that the C++ graphics API is one 
project, the
ODE wrapper is another, and my main project simply imports modules from the 
other two.

In fact, I initially just used Mattias' PyODE without even knowing that it 
was written
using Pyrex -- it was just a .pyd that I imported.  The fact that I am now 
modifying some of the
code in that wrapper doesn't really change anything --  the main project 
doesn't care where the .pyd
files came from or what tools were used to create them.

That means we can use the right tool for the job in each case.  Since ODE 
is written in C, it makes
sense to use Pyrex.  Since the graphics library is written in C++, it makes 
sense to use SWIG.
For other things, it makes sense to use Boost.  I could easily see a Python 
application that
imports modules that were created using all three tools, and the 
application developer doesn't need
to know.

>So it is a blancing act, do we
>shoehorn ODE to fit into the Boost model, or do it the clean and quick way 
>with Pyrex?

I vote for clean and quick.  :-)

>Btw, is anybody in the San Francisco area?

I'm in Waterloo, Ontario Canada (about an hour west of Toronto).


--
   Bernie Roehl
   University of Waterloo Dept of Electrical and Computer Engineering
   Mail: br...@ec...   Voice: (519) 888-4567 x 2607 [work]
   URL: http://ece.uwaterloo.ca/~broehl 

Re: [Pyode-user] Status of the wrappers?

[brett h. <bha...@ya...>]

Hello Bernie and Matthias,

Clean and quick with Pyrex sounds like a good plan, i wish you luck.  I won't be able to help with
any of the code for the next 2-3 months.  But in two months i am quitting my day job to make more
time for working on free software projects, and then i may have time to help.

I want to see the ODE wrappers done and done well.  So i am putting a $300 bounty on them.  This
bounty is not meant to make people fight, everyone should still work together as much as possible.
 I just know from my own experiences it can be hard to work on free software and pay the bills, so
this bounty is just meant to help out the developer a little bit.

Here's the conditions for the bounty:
  1. All the features of ODE are exposed in Python (within reason, not every little thing, but all
of the important stuff)
  2. The code is clean and easy to build (clean: i should have no problems in reading it) (easy to
build: can be built on all platforms)
  3. It is licensed under the GNU GPL or LGPL


Thoughts?
-brett


--- Bernie Roehl <br...@ec...> wrote:
> At 11:05 AM 6/21/2004 -0700, brett hartshorn wrote:
> >Wrapping a
> >wrapper is not ideal, but from the perspective where several python 
> >wrappers must be created and
> >maintained, it makes sense that they all use the same wrapping language.
> 
> Perhaps, though I think it depends on the circumstances.
> 
> For example, in the project I'm working on, I'm wrapping a C++ graphics 
> library using SWIG and
> wrapping ODE using Pyrex.  My reasoning is that the C++ graphics API is one 
> project, the
> ODE wrapper is another, and my main project simply imports modules from the 
> other two.
> 
> In fact, I initially just used Mattias' PyODE without even knowing that it 
> was written
> using Pyrex -- it was just a .pyd that I imported.  The fact that I am now 
> modifying some of the
> code in that wrapper doesn't really change anything --  the main project 
> doesn't care where the .pyd
> files came from or what tools were used to create them.
> 
> That means we can use the right tool for the job in each case.  Since ODE 
> is written in C, it makes
> sense to use Pyrex.  Since the graphics library is written in C++, it makes 
> sense to use SWIG.
> For other things, it makes sense to use Boost.  I could easily see a Python 
> application that
> imports modules that were created using all three tools, and the 
> application developer doesn't need
> to know.
> 
> >So it is a blancing act, do we
> >shoehorn ODE to fit into the Boost model, or do it the clean and quick way 
> >with Pyrex?
> 
> I vote for clean and quick.  :-)
> 
> >Btw, is anybody in the San Francisco area?
> 
> I'm in Waterloo, Ontario Canada (about an hour west of Toronto).
> 
> 
> --
>    Bernie Roehl
>    University of Waterloo Dept of Electrical and Computer Engineering
>    Mail: br...@ec...   Voice: (519) 888-4567 x 2607 [work]
>    URL: http://ece.uwaterloo.ca/~broehl 
> 
> 
> 
> -------------------------------------------------------
> This SF.Net email sponsored by Black Hat Briefings & Training.
> Attend Black Hat Briefings & Training, Las Vegas July 24-29 - 
> digital self defense, top technical experts, no vendor pitches, 
> unmatched networking opportunities. Visit www.blackhat.com
> _______________________________________________
> Pyode-user mailing list
> Pyo...@li...
> https://lists.sourceforge.net/lists/listinfo/pyode-user
> 



		
__________________________________
Do you Yahoo!?
Yahoo! Mail is new and improved - Check it out!
http://promotions.yahoo.com/new_mail

Re: [Pyode-user] Status of the wrappers?

[Bernie R. <br...@ec...>]

Hmm... that's very nice of you to offer.  I would suggest the money goes to 
Matthias, for these reasons...

1) The code is his, I'm just making small changes to it

2) When I do commercial projects, $300 isn't much money

3) When I do non-commercial projects, $300 is way too much!  *smile*



At 02:38 PM 6/22/2004, brett hartshorn wrote:
>Hello Bernie and Matthias,
>
>Clean and quick with Pyrex sounds like a good plan, i wish you luck.  I 
>won't be able to help with
>any of the code for the next 2-3 months.  But in two months i am quitting 
>my day job to make more
>time for working on free software projects, and then i may have time to help.
>
>I want to see the ODE wrappers done and done well.  So i am putting a $300 
>bounty on them.  This
>bounty is not meant to make people fight, everyone should still work 
>together as much as possible.
>  I just know from my own experiences it can be hard to work on free 
> software and pay the bills, so
>this bounty is just meant to help out the developer a little bit.
>
>Here's the conditions for the bounty:
>   1. All the features of ODE are exposed in Python (within reason, not 
> every little thing, but all
>of the important stuff)
>   2. The code is clean and easy to build (clean: i should have no 
> problems in reading it) (easy to
>build: can be built on all platforms)
>   3. It is licensed under the GNU GPL or LGPL
>
>
>Thoughts?
>-brett
>
>
>--- Bernie Roehl <br...@ec...> wrote:
> > At 11:05 AM 6/21/2004 -0700, brett hartshorn wrote:
> > >Wrapping a
> > >wrapper is not ideal, but from the perspective where several python
> > >wrappers must be created and
> > >maintained, it makes sense that they all use the same wrapping language.
> >
> > Perhaps, though I think it depends on the circumstances.
> >
> > For example, in the project I'm working on, I'm wrapping a C++ graphics
> > library using SWIG and
> > wrapping ODE using Pyrex.  My reasoning is that the C++ graphics API is 
> one
> > project, the
> > ODE wrapper is another, and my main project simply imports modules from 
> the
> > other two.
> >
> > In fact, I initially just used Mattias' PyODE without even knowing that it
> > was written
> > using Pyrex -- it was just a .pyd that I imported.  The fact that I am now
> > modifying some of the
> > code in that wrapper doesn't really change anything --  the main project
> > doesn't care where the .pyd
> > files came from or what tools were used to create them.
> >
> > That means we can use the right tool for the job in each case.  Since ODE
> > is written in C, it makes
> > sense to use Pyrex.  Since the graphics library is written in C++, it 
> makes
> > sense to use SWIG.
> > For other things, it makes sense to use Boost.  I could easily see a 
> Python
> > application that
> > imports modules that were created using all three tools, and the
> > application developer doesn't need
> > to know.
> >
> > >So it is a blancing act, do we
> > >shoehorn ODE to fit into the Boost model, or do it the clean and quick 
> way
> > >with Pyrex?
> >
> > I vote for clean and quick.  :-)
> >
> > >Btw, is anybody in the San Francisco area?
> >
> > I'm in Waterloo, Ontario Canada (about an hour west of Toronto).
> >
> >
> > --
> >    Bernie Roehl
> >    University of Waterloo Dept of Electrical and Computer Engineering
> >    Mail: br...@ec...   Voice: (519) 888-4567 x 2607 [work]
> >    URL: http://ece.uwaterloo.ca/~broehl
> >
> >
> >
> > -------------------------------------------------------
> > This SF.Net email sponsored by Black Hat Briefings & Training.
> > Attend Black Hat Briefings & Training, Las Vegas July 24-29 -
> > digital self defense, top technical experts, no vendor pitches,
> > unmatched networking opportunities. Visit www.blackhat.com
> > _______________________________________________
> > Pyode-user mailing list
> > Pyo...@li...
> > https://lists.sourceforge.net/lists/listinfo/pyode-user
> >
>
>
>
>
>__________________________________
>Do you Yahoo!?
>Yahoo! Mail is new and improved - Check it out!
>http://promotions.yahoo.com/new_mail

--
    Bernie Roehl
    University of Waterloo Dept of Electrical and Computer Engineering
    Mail: br...@ec...    Voice:  (519) 888-4567 x 2607 [work]
    URL: http://ece.uwaterloo.ca/~broehl 

Re: [Pyode-user] Status of the wrappers?

[brett h. <bha...@ya...>]

The money has been kindly provided by AwareMedia, so you can thank them for it.  They are funding
artists, and programming is one of the art forms they are supporting.

I think we should vote on who gets the bounty when we have a release done.  Democracy i think is
the only fair way to decide these things.

-brett

--- Bernie Roehl <br...@ec...> wrote:
> Hmm... that's very nice of you to offer.  I would suggest the money goes to 
> Matthias, for these reasons...
> 
> 1) The code is his, I'm just making small changes to it
> 
> 2) When I do commercial projects, $300 isn't much money
> 
> 3) When I do non-commercial projects, $300 is way too much!  *smile*
> 
> 
> 



		
__________________________________
Do you Yahoo!?
New and Improved Yahoo! Mail - Send 10MB messages!
http://promotions.yahoo.com/new_mail 

Re: [Pyode-user] Status of the wrappers?

[Timothy S. <ti...@st...>]

On Tue, 2004-06-22 at 13:25, Bernie Roehl wrote:
> >So it is a blancing act, do we
> >shoehorn ODE to fit into the Boost model, or do it the clean and quick w=
ay=20
> >with Pyrex?
>=20
> I vote for clean and quick.  :-)

As do I.

Please tell me your sourceforge.net usernames so I can add you to the
PyODE project. Shall I'll remove the current code from CVS and import
Matthias's code?

> >Btw, is anybody in the San Francisco area?
>=20
> I'm in Waterloo, Ontario Canada (about an hour west of Toronto).

I'm in Durban, South Africa.

--=20
Timothy Stranex <ti...@st...>

Re: [Pyode-user] Status of the wrappers?

[Matthias B. <ba...@ir...>]

Timothy Stranex wrote:
> Please tell me your sourceforge.net usernames so I can add you to the
> PyODE project. 

My sf user name is "mbaas".

> Shall I'll remove the current code from CVS and import
> Matthias's code?

Maybe before getting our hands dirty in the code we should clarify those 
points I've posted in my other message (which will also give me the 
opportunity to translate some of the comments that are still in German 
in my files... ;-) ).

- Matthias -

Re: [Pyode-user] Status of the wrappers?

[Bernie R. <br...@ec...>]

At 02:28 PM 6/22/2004, Timothy Stranex wrote:
>Please tell me your sourceforge.net usernames so I can add you to the
>PyODE project. Shall I'll remove the current code from CVS and import
>Matthias's code?

I'm broehl on sourceforge.

I would suggest keeping what's on CVS now (or moving it into a separate
branch) and adding a branch for Matthias'  code.  I'll email him the changes
I made to add setCategoryBits() and setCollideBits().


--
    Bernie Roehl
    University of Waterloo Dept of Electrical and Computer Engineering
    Mail: br...@ec...    Voice:  (519) 888-4567 x 2607 [work]
    URL: http://ece.uwaterloo.ca/~broehl