[Bf-python] Remote IPython & PyCrust Shell for Blender

Dietrich Bollmann diresu at web.de
Thu Mar 27 11:11:21 CET 2008


Hi Stany,

Thanks for your friendly statement :)

On Wed, 2008-03-26 at 17:13 +0100, Stani wrote:
> Willian Padovani Germano schreef:
> > Hi,
> >
> > Didn't read Dietrich's patch, but I checked the discussion on the patch 
> > tracker and the official page. Indeed looks like an interesting 
> > candidate, but I agree with others about the need to have a clean patch, 
> > specially if a branch is used for further developments.
> >
> > http://formgames.org/blender/command-port/
> > http://projects.blender.org/tracker/index.php?func=detail&aid=6887&group_id=9&atid=127
> >   
> Dietrich, the functionality you achieved looks very nice. Of course the 
> best way would be to work together with the Blender developers. For the 
> SPE audience it only makes sense to interoperate with the standard 
> Blender. Unfortunately I only know Python and not C, so I don' t know if 
> I can be of much help to clean your patch.

No problem.

I personally would be much happier to have the command port in the
official blender trunk.  But this seems to be much more difficult than I
initially thought.  There are not so much people using Blender from
other programs and the official developers are quite busy and do not
have the time to look at every patch.

The second problem is that a command port is something complex and
relies on lots of things which are complicated and difficult to
understand: threads, sockets, mutexes, condition variables,
protocols, ... Nothing which makes a computer graphics specialist happy
to deal with.

The next thing is, that I developed the thing myself and added
everything I found necessary or at least helpful.  Still I know
perfectly that others might feel different about this.  I personally
think the patch is "clean" as it is and probably would remove things
others would prefer to have when trying to make it smaller.  I therefor
depend on others telling me what I should change and what not.  But this
means, that others have to apply the patch as a patch itself is
unreadable if it makes changes to more than 3 files.  They than would
have to read the sources which takes a huge amount of time also... 

All these reasons add and make the integration difficult.
Unfortunately :(

> > Anyway, the main issue now (as usual) is the upcoming events refactor in 
> > Blender. It would be wiser to wait a little more before coding anything 
> > non trivial, because things will change considerably and we have chance 
> > to come up with better / cleaner ways to interface with Blender data.
> >
> > For now I suggest you guys directly interested in scripting Blender via 
> > external apps to create a basic plan, detailing what you need to have, 
> > preferred methods (sockets, simple signals, etc.). A page in our bpython 
> > dev section in the wiki would be ideal for this, but as a start one of 
> > you could write something here and the others could add to it, then we 
> > put the results in the wiki.
> >   
> I've asked editing rights for the wiki. Willian, can you provide me the 
> exact url?

Looking forward to your statements :)

Best wishes, Dietrich

> Thanks,
> Stani
> 
> _______________________________________________
> Bf-python mailing list
> Bf-python at blender.org
> http://lists.blender.org/mailman/listinfo/bf-python
> 




More information about the Bf-python mailing list