[Bf-committers] Blender 2.5 api proposal
joeedh at gmail.com
Mon Dec 1 09:19:24 CET 2008
I did a little work on solving the instability issues with the current API
(e.g. getting the Draw module to not crash) so I know just how unstable it
is. The API's original goal was import/export scripts (which we now have
many of, so I'd say it succeeded there) but for many other purposes it has a
lot of issues.
It's entirely possible that the current way the API gets into the guts of
blender (especially the way it calls internal functions) won't be tolerable
in 2.5, in which case it'd take quite a lot of rework to get it to maintain
functionality, and like Cam has said, many or most scripts will probably
break anyway, no matter what we do.
That being said, if there's a way to have *some* sort of compatibility API,
I'm all for it. I just don't really think it's possible to do, without
writing a lot of new code (though one idea would be to write a set of
emulation modules in python, that emulate the old API; just don't know if
it'd be doable).
On Sun, Nov 30, 2008 at 11:32 PM, Geoffrey Bantle <hairbat at yahoo.com> wrote:
> Just my 2cents on this one,
> As someone who relies on a lot of code written for the current API in a
> production environment I really think it's time to put it out to pasture.
> Too many things just don't work as expected and often-times script writers
> are put in the position of having to read the API source itself to figure
> out what exactly is going on since the docs cannot hope to give a clear view
> of things.
> Worse yet the current API is fraught with problems that make it prone to
> crashing Blender and causing corruption of the internal database. It's time
> for a change and Campbell's proposal has my full support.
> Geoffrey Bantle
> Bf-committers mailing list
> Bf-committers at blender.org
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Bf-committers