Hi Marcus and everybody else,
On Wed, 2005-06-22 at 11:25 +0200, Marcus Lindblom wrote:
> We've written a few _very_ simple manips ourselves, but the main thing
> is that OpenSG does not abstract window events (key/mouse/etc), so it's
> hard to do generally at the moment. (Our manipulators use a very simple
> window-event system that we've attached on top of our own Win32Window
> class, which calls the Win32 API CreateWindow() and attaches a
> OSGWin32Window to it.)
>
> Generic manips require OSG to handle GUI stuff, which it does not do.
^^^
nitpick: please don't call it that, that's the abbreviation for the
'other' system. ;) OpenSG is too long for a namespace prefix, but
hopefully short enough to be used in full for mails.
> (As of now, there needs to be different manips for QT/Win32/GLUT & X.)
>
> Of course, moving the SSM's input functions to an abstract interface and
> writing some basic event producers for the various supported GUI API's
> would perhaps solve this in a simple and acceptable way.
>
> What do Dirk & Co think?
the SimpleSceneManager is really a Simple(SceneManager), not a
(SimpleScene)Manager as some people think. ;) I wrote it to shorten the
amount of code needed to make the Tutorials work, therefore it's
interface is designed to allow easy passthrough of GLUT events. It was
never really intended it to be a full-blown interaction framework. And
it shows. I've openend it up a little bit over time, but it's really not
all that flexible.
Nonetheless (and, in hindsight, not surprisingly) it has found a lot
more use than I expected, so things like object manipulation and more
extensibility are becoming important. But I don't see how I can do it
justice by taking it on, I'm having a hard enough time to do all the
things I want to do in the core without worrying about another big
subsystem, especially in an area that I don't have much experience with.
If somebody else wants to take on the NSSSM (NotSoSimpleSceneManager ;),
I'd be delighted. Having standard viewers and manipulators was and still
is one of Inventor's greatest strengths, and getting some of that over
to OpenSG would be a big plus.
I know of at least three manipulator concepts that exist (two out of IGD
and OpenManip). It would be nice if the people that wrote these concepts
could send a short mail about their basic design, how they integrated it
into the existing system, and what does and does not work yet. For
OpenManip that might be the paper
( http://www.eg.org/EG/DL/PE/OPENSG03/08braitmaier.pdf ), unless Michael
wants to add something. From the other two I know one of them is on
vacation, but maybe the other can give us a quick overview?
I know the Open Scenegraph folks have a more elaborate event handling
system, without manipulators as far as I know, so that might be a source
either of inspiration or maybe even of code. I haven't looked at it, so
I don't know how good it is, but it's a start. And then of course there
the original Open Inventor itself, which has been made open some time
ago. So there's no shortage of examples, but no real solution either.
My conclusion: I'd love to see it, I'm happy to help coordinating things
as far as I can, but I don't see myself taking the lead at this point.
I'd like to do it, but I don't have enough time to commit to it to make
it go anywhere, and that would be a pity.
My .02$
Dirk
-------------------------------------------------------
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_id=7477&alloc_id=16492&op=click
_______________________________________________
Opensg-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/opensg-users
