[Bf-python] questions about python sub-interpreters, merge of bpy and gamepy

[Willian Padovani Germano]

From: "maci_ray" <maci_ray at yahoo.de>
(...)
> I think it would cause a lot of problems when game-engine scripts will
> stay running after end, or did you ment static scripts running while
> gameengine starts?

Yes, when we integrate normal scripts properly, they should be able to stay
"up" just like other Blender windows.

> Remind that the game objects only do exist from start of the engine till
> it's end.

Yes, one of the things mentioned in that thread was that it would be nice if
the game engine could be stopped and the current state of objects exposed to
the bpython API.

> I took some experiments with a slighly modified bpy. In
> BL_KetsjiEmbedStart.cpp I modified the call to BPY_start_python() with
> a flag indicating bpy runs in a game mode. With this flags only the
> modules I needed (that should be tested and be game engine proven) were
> initialized. Some special functionality is enabled, also, real mesh
> replacement that I really needed. Maybe we should investigate further
> on this...

Sure.

> Don't forget the _completely_ different approches of implementation
> of the APIs.

They much probably "encode" the data in different ways, but it should still
be possible to use both together, taking data from one and passing to the
other.  But I haven't played with this, it's all a pure guess.

> BTW: I also found I some information about gameengine/gameengine python
> and stuff. So if you have questions, maybe I can help.
>
> Some references:
> http://www.blender.org/modules/bc2002/Erwin/architecture/ketsjiintro.htm
> http://www.python.org/doc/PyCPP.html

Ah, thanks, I didn't know about them.

--
Willian, wgermano at ig.com.br