openvrml-develop — OpenVRML developers mailing list

Re: [Openvrml-develop] OpenVRML and SGI

[Xavier D. <Xav...@im...>]

> That's downright goofy. It looks like something libtool's doing, though.
> Are you using CVS or a distribution tarball? If you're using CVS, what
> version of libtool is on your machine?

I am not uisng CVS.  I just downloaded the tar.gz.
However here is the version of libtool used:

./libtool --version
ltmain.sh (GNU libtool) 1.3.4 (1.385.2.196 1999/12/07 21:47:57)

> -- 
> Braden McDaniel                 It is hard to know if nothing is /
> http://endoframe.com            actually nothing
> e-mail: br...@en...    And thus difficult to know if a policy /
> Jabber: br...@ja...       of doing nothing is successful
>                                                 -- Radiohead

-- 
+------------------+----------------------------------------------------+
|Xavier DECORET| iMAGIS, Laboratoire GRAVIR/IMAG  (CNRS,INRIA,INPG,UJF) |
|              | INRIA Rhone-Alpes, 655 Av de l'Europe, 38330 Montbonnot|
|Doctorant     | France. Tel: +33 4 76 61 54 58 - Fax: +33 4 76 61 54 40|
+------------------+--------+-------------------------------------------+
|Xav...@im...  |  http://www-imagis.imag.fr/~Xavier.Decoret   |
+---------------------------+-------------------------------------------+

Re: [Openvrml-develop] OpenVRML and SGI

[Braden M. <br...@en...>]

On 13 Feb 2001, Xavier Decoret wrote:

>
> I can't compile openVRML on my SGI O2.
>
> I have the latest IRIX distributions and freeware (including the
> gcc-2.95.2).
> I use
> > ./configure
>   --prefix=/disc/decoret/OPENVRML
>   --with-glut-prefix=/usr/local
>   --with-jpeg-prefix=/usr/freeware
>   --with-zlib-prefix=/usr/freeware
>   --with-png-prefix=/usr/freeware
>   --with-x
> and
> >  gmake
>
> But it dies later with
>
> /bin/sh ../../../libtool --mode=link c++ -I/usr/freeware/include -I/usr/freeware/include -I/usr/freeware/include -g -O2  -o antlr-prelinked.lo Vrml97Scanner.lo Vrml97Parser.lo ../../../lib/antlr/libantlr.la
> rm -fr /antlr-prelinked.ox
> mkdir /antlr-prelinked.ox
> Cannot create directory "/antlr-prelinked.ox": Permission denied
>
> Why does it try to work in /?

That's downright goofy. It looks like something libtool's doing, though.
Are you using CVS or a distribution tarball? If you're using CVS, what
version of libtool is on your machine?

-- 
Braden McDaniel                 It is hard to know if nothing is /
http://endoframe.com            actually nothing
e-mail: br...@en...    And thus difficult to know if a policy /
Jabber: br...@ja...       of doing nothing is successful
                                                -- Radiohead

[Openvrml-develop] OpenVRML and SGI

[Xavier D. <Xav...@im...>]

I can't compile openVRML on my SGI O2.

I have the latest IRIX distributions and freeware (including the
gcc-2.95.2).
I use
> ./configure
  --prefix=/disc/decoret/OPENVRML
  --with-glut-prefix=/usr/local
  --with-jpeg-prefix=/usr/freeware
  --with-zlib-prefix=/usr/freeware
  --with-png-prefix=/usr/freeware
  --with-x
and
>  gmake

But it dies later with

/bin/sh ../../../libtool --mode=link c++ -I/usr/freeware/include -I/usr/freeware/include -I/usr/freeware/include -g -O2  -o antlr-prelinked.lo Vrml97Scanner.lo Vrml97Parser.lo ../../../lib/antlr/libantlr.la
rm -fr /antlr-prelinked.ox
mkdir /antlr-prelinked.ox
Cannot create directory "/antlr-prelinked.ox": Permission denied

Why does it try to work in /?

Does any one have a version of openVRML on SGI?

-- 
+------------------+----------------------------------------------------+
|Xavier DECORET| iMAGIS, Laboratoire GRAVIR/IMAG  (CNRS,INRIA,INPG,UJF) |
|              | INRIA Rhone-Alpes, 655 Av de l'Europe, 38330 Montbonnot|
|Doctorant     | France. Tel: +33 4 76 61 54 58 - Fax: +33 4 76 61 54 40|
+------------------+--------+-------------------------------------------+
|Xav...@im...  |  http://www-imagis.imag.fr/~Xavier.Decoret   |
+---------------------------+-------------------------------------------+

[Openvrml-develop] VrmlNodePtr patch

[Braden M. <br...@en...>]

I've uploaded a patch to move the library from using raw VrmlNode pointers
to VrmlNodePtr, a refcounted smart pointer:

<http://sourceforge.net/patch/index.php?func=detailpatch&patch_id=103653&group_id=7151>

Some work has yet to be done before this will be ready to check in, but
this patch will give folks a concrete idea of where this change will put
us. Comments welcome.

-- 
Braden McDaniel                 It is hard to know if nothing is /
http://endoframe.com            actually nothing
e-mail: br...@en...    And thus difficult to know if a policy /
Jabber: br...@ja...       of doing nothing is successful
                                                -- Radiohead

Re: [Openvrml-develop] Simple Object to Object Collision Detection

[Sylvain C. <syl...@to...>]

I donr know if you know about this one but seem quite pertinent at this stage
I think:
 SOLID - Software Library for Interference Detection
LPGL, designed with vrml in mind, use stl, can be used with or without qhull.
Maybe could be retrofitted into openvrml?
You'll find more links in this pages to others collision library too.

Sylvain Carette
VRML designer-composer

Braden McDaniel wrote:

> On Fri, 2 Feb 2001, Patrick AS Sinclair wrote:
>
> > I'm using the libVRML97 in my work but I need very simple object to
> > object collision detection. Basically, I want to check whether a point
> > is inside certain objects in the scene, such as Anchor nodes. Is there
> > an easy way to do this or do I have to calculate the bounding boxes
> > etc. and check whether the point is inside them myself?
>
> cks has done some work adding bounding boxes/spheres with the goal of
> adding object-avatar collision detection. I'm not sure how adaptable this
> work is to interobject collision detection; hopefully he'll chime in here.
>
> In any event, wiring OpenVRML to do interobject collisions is going to
> take some work. I'd definitely consider a patch to add such functionality.
> I think Contact and/or Cortona have implemented  extensions for
> interobject collisions; it may be a good idea to model the interface after
> what they've done.
>
> --
> Braden McDaniel                 It is hard to know if nothing is /
> http://endoframe.com            actually nothing
> e-mail: br...@en...    And thus difficult to know if a policy /
> Jabber: br...@ja...       of doing nothing is successful
>                                                 -- Radiohead
>
> _______________________________________________
> Openvrml-develop mailing list
> Ope...@li...
> http://lists.sourceforge.net/lists/listinfo/openvrml-develop

Re: [Openvrml-develop] Simple Object to Object Collision Detection

[Christopher K. S. J. <cs...@qu...>]

Patrick AS Sinclair wrote:
> 
> I'm using the libVRML97 in my work but I need very simple object to
> object collision detection. Basically, I want to check whether a point
> is inside certain objects in the scene, such as Anchor nodes. Is there
> an easy way to do this or do I have to calculate the bounding boxes
> etc. and check whether the point is inside them myself?
> 

 Check out VrmlNode::getBVolume(), and the comments in
VrmlBVolume.h (and friends). I'm switching jobs right now,
so I'm a little overloaded, but if you've got specific
questions, I can probably answer. Note that if you want
exact results, instead of just bvolumes, then you're on
your own, and possibly in for a lot of work. ray-based
picking and avatar/object collision are on the requirements
list, but aren't implemented yet. Can you
elaborate on your problem/requirements? (If I had to
guess, sounds like a 3d pointing device?)

-cks

Re: [Openvrml-develop] Simple Object to Object Collision Detection

[Braden M. <br...@en...>]

On Fri, 2 Feb 2001, Patrick AS Sinclair wrote:

> I'm using the libVRML97 in my work but I need very simple object to
> object collision detection. Basically, I want to check whether a point
> is inside certain objects in the scene, such as Anchor nodes. Is there
> an easy way to do this or do I have to calculate the bounding boxes
> etc. and check whether the point is inside them myself?

cks has done some work adding bounding boxes/spheres with the goal of
adding object-avatar collision detection. I'm not sure how adaptable this
work is to interobject collision detection; hopefully he'll chime in here.

In any event, wiring OpenVRML to do interobject collisions is going to
take some work. I'd definitely consider a patch to add such functionality.
I think Contact and/or Cortona have implemented  extensions for
interobject collisions; it may be a good idea to model the interface after
what they've done.

-- 
Braden McDaniel                 It is hard to know if nothing is /
http://endoframe.com            actually nothing
e-mail: br...@en...    And thus difficult to know if a policy /
Jabber: br...@ja...       of doing nothing is successful
                                                -- Radiohead

[Openvrml-develop] Simple Object to Object Collision Detection

[Patrick AS S. <pa...@ec...>]

Hi,

I'm using the libVRML97 in my work but I need very simple object to
object collision detection. Basically, I want to check whether a point
is inside certain objects in the scene, such as Anchor nodes. Is there
an easy way to do this or do I have to calculate the bounding boxes
etc. and check whether the point is inside them myself?

Thanks,

Patrick Sinclair

[Openvrml-develop] Using a refcounted smart pointer for nodes

[Braden M. <br...@en...>]

I've discussed using a refcounted smart pointer for nodes (instead of the
intrusive refcount currently in place), and I'm currently implementing
this. I'm tracking my activity in this task:

<http://sourceforge.net/pm/task.php?func=detailtask&project_task_id=23714&group_id=7151&group_project_id=2579

... to which I've recently posted a fairly extensive comment. This work
has pretty broad ramifications in the library, and I thought I'd post a
note about it in case anyone would like to discuss the issue.

-- 
Braden McDaniel                 It is hard to know if nothing is /
http://endoframe.com            actually nothing
e-mail: br...@en...    And thus difficult to know if a policy /
Jabber: br...@ja...       of doing nothing is successful
                                                -- Radiohead

Re: [www-vrml] Re: [Openvrml-develop] ANN: OpenVRML 0.10.0

[clayton c. <dr...@sm...>]

clayton cottingham wrote:
> 
> Jan-Olof Janson wrote:
> >
> > Can I download that Mandrake binary? Does it support the full VRML97
> > spec? Does it support MesaGL or similar?
> >
> 


ok since on e person wanted them every one get s them 

lemme know if they work for you

they can be located here:
http://www.hyperbomb.com/openvrml/

Re: [Openvrml-develop] ANN: OpenVRML 0.10.0

[Clayton C. a. d. <dr...@sm...>]

On Sun, 21 Jan 2001 20:28:55 -0500 (EST), Braden McDaniel said:

> OpenVRML 0.10.0 has been released.
>  


im happy to report
on mandrake 7.2 
happily compiled both
openvrml and gtklookat

kudos goin out again!


-- 
back in the day
we didn't have no
old school
-dr. frog
http://www.hyperbomb.com
it sells itself

[Openvrml-develop] ANN: OpenVRML 0.10.0

[Braden M. <br...@en...>]

OpenVRML 0.10.0 has been released.

This release of the library includes primarily organizational changes that
are of more interest to developers than end-users. However, since it is
probably true in general that developers compose our meager interest
base, if you are one of our existing users, you probably ought to upgrade.
The changes, summarized:

- Changed package directory structure.

- Minor bugfixes.

- Changed library names from libvrml97 and libvrml97gl to libopenvrml
  and libopenvrml-gl, respectively.

- Changed header file installation locations.

- API cleanup: improved const-correctness many places and changed many
  of the VrmlField subclass methods to match the semantic used by the
  corresponding objects in the ECMAScript binding.

- JavaScript support restored to the default build.

- Repaired Win32 and Mac builds.

***

OpenVRML can be downloaded from

  <http://sourceforge.net/project/showfiles.php?group_id=7151>

The OpenVRML home page is

  <http://openvrml.sourceforge.net>

Re: [Openvrml-develop] Our SFImage story

[Braden M. <br...@en...>]

On Wed, 20 Dec 2000, Christopher K. St. John wrote:

> Braden McDaniel wrote:
> >
> > In either of those cases, our current design would require dynamically
> > converting to the 4-byte-element format when one of these bindings
> > attempts to access the data.
>
>  The java bindings (EAI and Script node) already use an array
> of bytes.

Both the IDL and Java bindings to the EAI use int. (Perhaps you were
thinking of the obsolete Cosmo EAI?) This makes the Java Script node
binding the oddball (which is also the case in respects other than this
one).

> The netscape javascript api, if I'm remembering right,
> lets you fiddle with the array access operators in the code
> that binds the script object to the underlying C/C++ object.
> You'd case out on numcomponents, but it shouldn't be very much
> code. (That assumes I'm remembering correctly...)
>
>  Besides, doing image processing in javascript is not likely to be
> fast for a whole raft of other reasons, so I don't see it as a likely
> bottleneck.
>
> > I think we need to store the integer format in VrmlSFImage.
>
>  Then you'd have to convert back and forth to an array of bytes
> when you were using a java interface.

Yup. To make this a little less painful, the Java binding could cache the
byte data. And as I mentioned before, we can go ahead and cache this data
in VrmlSFImage for the time being.

(As far as caching the byte data in the renderer, we really need to do
that too: the data used by the renderer may be rescaled from its original
dimensions, and that should be opaque to the scene graph interface.)

>  Unless it turns out that the javascript implementation is
> just totally undoable, I'd vote for leaving things the way
> they are.

It's not impossible, and if the ECMAScript binding were the oddball here,
I'd agree with you. But the Java binding is the oddball.

-- 
Braden N. McDaniel                              e-mail: br...@en...
<http://endoframe.com>                          Jabber: br...@ja...

[Openvrml-develop] MovieTexture: I need some confirmation

[clayton c. <dr...@sm...>]

heya
ive been exp with the movie texture

ive come across some things i just wanna make
sure that they are bugs/features not me

i've found system mpeg video {those cont vid & aud }
dont work

ive also found that the texture repeats wierd
if the width and hieght are not equal
if they are equal {i have one thats 32X32}


as well the 'loop true' stops if the scene manipulated/navigated but not
everytime


let me know if you need examples/ have any pointers or other
observations!

thanks

Re: [Openvrml-develop] lookAt

[S.K.Bose <bo...@pa...>]

On Wed, 20 Dec 2000, Jason Goh wrote:

> When I compile the lookAt project, the following errors appear:
> --------------------Configuration: lookat - Win32 Debug--------------------
> >>Compiling...
> >>lookat.cpp
> >>g:\openvrml-0.9.0\libvrml97gl\src\vrml97gl\vieweropengl.h(346) : warning C4099: 'GLUtesselator' : type name first seen using 'class' now seen using 'struct'
>         c:\program files\microsoft visual studio\vc98\include\gl\glu.h(155) : see declaration of 'GLUtesselator'
> >>Compiling...
> >>ViewerGlut.cpp
> >>g:\openvrml-0.9.0\libvrml97gl\src\vrml97gl\vieweropengl.h(346) : warning C4099: 'GLUtesselator' : type name first seen using 'class' now seen using 'struct'
>         c:\program files\microsoft visual studio\vc98\include\gl\glu.h(155) : see declaration of 'GLUtesselator'
> >>Linking...
> >>LINK : fatal error LNK1104: cannot open file "../../libvrml97core/win32/debug/libvrml97core.lib"
> >>Error executing link.exe.
> 
> >>lookat.exe - 1 error(s), 2 warning(s)
> 
> How do i create the libvrml97core.lib from the given codes?
> 
> 
> 
In new release, it supposed to create all dependent library first and then
it will do linking to make lookat.exe. Probably there is some problem for
creating libvrml97core.lib. Did you install "java" and "ANTLR" ? Otherwise
you use Makefile. See more details in "Install" within win32 directory.

Bose

Re: [Openvrml-develop] Our SFImage story

[Christopher K. S. J. <cs...@qu...>]

Braden McDaniel wrote:
> 
> In either of those cases, our current design would require dynamically
> converting to the 4-byte-element format when one of these bindings
> attempts to access the data.
>

 The java bindings (EAI and Script node) already use an array
of bytes. The netscape javascript api, if I'm remembering right,
lets you fiddle with the array access operators in the code
that binds the script object to the underlying C/C++ object.
You'd case out on numcomponents, but it shouldn't be very much
code. (That assumes I'm remembering correctly...)

 Besides, doing image processing in javascript is not likely to be
fast for a whole raft of other reasons, so I don't see it as a likely
bottleneck.


> I think we need to store the integer format in VrmlSFImage.
>

 Then you'd have to convert back and forth to an array of bytes
when you were using a java interface.

 Unless it turns out that the javascript implementation is
just totally undoable, I'd vote for leaving things the way
they are.


-cks

[Openvrml-develop] Our SFImage story

[Braden M. <br...@en...>]

Right now, we store SFImages as an array of bytes, rather than as an array
of 4-byte ints as they are represented in VRML syntax. This is
problematic:

  * The ECMAScript binding includes access to the pixel data array--as
    4-byte ints. It looks like we currently punt on this.

  * PixelTexture's (the only pre-defined node type with an SFImage) image
    field is exposed, so an hypothetical EAI should be able to access it.

In either of those cases, our current design would require dynamically
converting to the 4-byte-element format when one of these bindings
attempts to access the data. That seems extremely likely to have
unacceptable performance.

I think we need to store the integer format in VrmlSFImage. The byte
format can remain there in parallel, though I suspect we might eventually
want to move it to a cache in the renderer.

-- 
Braden N. McDaniel                              e-mail: br...@en...
<http://endoframe.com>                          Jabber: br...@ja...

[Openvrml-develop] lookAt

[Jason G. <jas...@re...>]

When I compile the lookAt project, the following errors appear:
--------------------Configuration: lookat - Win32 =
Debug--------------------
>>Compiling...
>>lookat.cpp
>>g:\openvrml-0.9.0\libvrml97gl\src\vrml97gl\vieweropengl.h(346) : =
warning C4099: 'GLUtesselator' : type name first seen using 'class' now =
seen using 'struct'
        c:\program files\microsoft visual =
studio\vc98\include\gl\glu.h(155) : see declaration of 'GLUtesselator'
>>Compiling...
>>ViewerGlut.cpp
>>g:\openvrml-0.9.0\libvrml97gl\src\vrml97gl\vieweropengl.h(346) : =
warning C4099: 'GLUtesselator' : type name first seen using 'class' now =
seen using 'struct'
        c:\program files\microsoft visual =
studio\vc98\include\gl\glu.h(155) : see declaration of 'GLUtesselator'
>>Linking...
>>LINK : fatal error LNK1104: cannot open file =
"../../libvrml97core/win32/debug/libvrml97core.lib"
>>Error executing link.exe.

>>lookat.exe - 1 error(s), 2 warning(s)

How do i create the libvrml97core.lib from the given codes?

[Openvrml-develop] probs with email domains

[clayton c. <dr...@sm...>]

i checked in on this 	
if you do a whois on endofframe it says the domain is available

Re: [Openvrml-develop] Re: openvrml.spec file

[Dr F. <dr...@sm...>]

braden,
did you get that last message of mine?

i dont thinks so

anytime i send you anything directly it errors
out telling me my domain {smartt.com} doesnt exist!



here is the recursive delete patch for the install section of the spec file!

Re: [Openvrml-develop] Re: openvrml.spec file

[Braden M. <br...@en...>]

On 16 Dec 2000, Dr Frog wrote:

> On Fri, 15 Dec 2000 22:19:22 -0500 (EST), Braden McDaniel said:
>
> >  No. Make your modifications to 'openvrml.spec.in'. This file is processed
> >  by configure, which simply replaces variables that look like @THIS@ when
> >  outputting 'openvrml.spec'.
>
> ok cool i think i getcha
> then run cvs diff -u
> which produces the patch files,um
> right?

Right. Read the info files.

-- 
Braden N. McDaniel                              e-mail: br...@en...
<http://endoframe.com>                          Jabber: br...@ja...

Re: [Openvrml-develop] Re: openvrml.spec file

[Dr F. <dr...@sm...>]

On Fri, 15 Dec 2000 22:19:22 -0500 (EST), Braden McDaniel said:

> On Fri, 15 Dec 2000, clayton cottingham wrote:
>  
>  > so should i just rename the
>  > openvrml.spec file to
>  > openvrml.spec.in and
>  > then do a cvs diff -u ?
>  
>  No. Make your modifications to 'openvrml.spec.in'. This file is processed
>  by configure, which simply replaces variables that look like @THIS@ when
>  outputting 'openvrml.spec'.
>  





ok cool i think i getcha 
then run cvs diff -u 
which produces the patch files,um 
right?



>  -- 
>  Braden N. McDaniel                              e-mail: br...@en...
>  <http://endoframe.com>                          Jabber: br...@ja...
>  
>  
>  _______________________________________________
>  Openvrml-develop mailing list
>  Ope...@li...
>  http://lists.sourceforge.net/mailman/listinfo/openvrml-develop
>  
>  

Re: [Openvrml-develop] Re: openvrml.spec file

[Braden M. <br...@en...>]

On Fri, 15 Dec 2000, clayton cottingham wrote:

> so should i just rename the
> openvrml.spec file to
> openvrml.spec.in and
> then do a cvs diff -u ?

No. Make your modifications to 'openvrml.spec.in'. This file is processed
by configure, which simply replaces variables that look like @THIS@ when
outputting 'openvrml.spec'.

-- 
Braden N. McDaniel                              e-mail: br...@en...
<http://endoframe.com>                          Jabber: br...@ja...

[Openvrml-develop] Re: openvrml.spec file

[clayton c. <dr...@sm...>]

so should i just rename the
openvrml.spec file to 
openvrml.spec.in and 
then do a cvs diff -u ?

Re: [Openvrml-develop] oopenvrml.spec file

[Braden M. <br...@en...>]

On 14 Dec 2000, Dr Frog wrote:

> um i cvs co libvrml97 and i could
> not see openvrml.spec
>
> i did see a openvrmlspec.in

That is the file you want to edit. openvrml.spec is generated from that
file by configure.

> please advise
> on how to
> cvs diff -u

That command will recursively search directories for files that have been
modified since they were checked out, and write a machine-readable change
summary to stdout. Refer to the cvs and diff info files for detailed
information on their use.

-- 
Braden N. McDaniel                              e-mail: br...@en...
<http://endoframe.com>                          Jabber: br...@ja...