[Bf-committers] Call for help?
zdys at zdys.de
Mon Mar 21 14:12:34 CET 2005
> A couple days ago there was discussion about getting some new developers.
> Has it already died? I think bf-blender would be helped greatly with some
> new dev experience. I'm not part of the official bf-blender, but I could
> join in and help on the interface code if needs be. Anyway, just a thought.
I'll have to finish some pending jobs before I can dare to start coding in c++.
I did some c++ some 10 years ago at a beginners level. Since then I did a lot
of java and switched to python about 2 years ago. I am interested in joining
the blender dev, but still feel a bit of in an inadequate knowledge level, so I
keep quite and read. I owe a lot to the blender community. I actually make a
living with this software and just really do not want to disappoint anyone, so
I explain my actual passiveness and at the end would like give an idea for the
[--passiveness due to:
I just started a project-management tool for rendering in an environment with
multiple computers. It will cover the topic of calculating the costs of a
rendering process, allow upload and distribution of blend files to be then
taken and rendered by attached clients. The result ought to be a summary which
can be presented to the customer, in a detailed form. A lot of additional -
production relevant - goodies will be included, it will be open source and have
the same license as blender :-) For now it's PyQT [Eric3 as IDE] and plain
Python enhancing development speed and allowing to do a 'fast rewrite' in c++,
once a production state has been reached.
Lately I had the need of doing s.th. I named an 'AnimatedExtrusion'. This is
the extrusion of a mesh along a path. This with changing of the path's position
and the originating mesh. It's alpha, has a bug with the 'curve follow', but
did the job.
Blend file including scripts: AnimatedExtruder.py, ExternalEditor.py at:
Result file [sorry for the WMF, would have to reencode from customers requested
[script working at sec 57-64 and 73-78]
First the script calculates a mesh which has as many vertex points as the
extrusion ist be, e.g.: extrusion 10 frames, mesh 5 vertices= 50 vertices. Then
it surfaces the vertices, like an extrusion would do. After this it 'simulates'
the movement of the base object, writing it's coors as values of a python list
in a text-object. When filled, the text-object is completed with some code to
place the mesh vertices corresponding to the listvalues where every list
position is understood as a frame. Finally the created object is associated
with this text object, making it a script called on the frame-change event for
the originaly created mesh.
I know it's far from being perfect, but it was my first python/blender
experience and saved me some animation time. btw: I noticed s.th. that could be
a bug, the 'undo' buffer is announced to be 'overfilled' while doing the
calculation, you can cause this error to appear with an animation of e.g. ~100
and a base mesh of 5 vertices - see the given blend file. The calculating is
slow, presumably for the allocation of memory for the Text-Object while writing
the massive ammount of coor data? Writing to a file and loading this didn't
work. Trying to use some coherent functionality with socket communication
stopped when I realised that threading.Thread didn't work. Anyway, this
functionality could be integrated somewhere near to the particle system... ah
and absolute vertex saving has no call from within python.
I hope it's okay if I take my time to decide when, how and if I can join the
blender dev without giving any particular promise but showing sincere interest.
> -- Austin
> Bf-committers mailing list
> Bf-committers at projects.blender.org
tel: +49(0)60214590690 | mobil: +49(0)1713113868
skype: luigi.monaco | public.key at ocanom.com
BOFH excuse #180:
More information about the Bf-committers