[Bf-python] Attribute casing

[Joseph Gilbert]

It seemed like a nice system for the Render module because most buttons 
in the UI were toggles for bits. However it does leave a few things to 
be desired - such as the abilty to 'get' the value of a bit rather than 
just setting the bit. Combined with attributes it might work nice for 
individual bits. i.e. you could have a 'blah' attribute who's setter 
calls the bitToggle function in genutils and who's getter gets the value 
and returns a bool. This works nice on single toggles.
e.g. object.attribute = blender.TRUE
(this is a call to the toggleBit function)
or
if object.attribute: (calls the getter)
...do x
For attributes that take a group of bits i.e. something that takes more 
than TRUE/FALSE -  it might be better to get/set a list of 
BPy_Constants. The setter for such could parse the list of constants and 
then call the bitToggle(true) for each constant passed, while the getter 
would return a list of constants maybe.
Just an idea :)

Ken Hughes wrote:

> Yehoshua Sapir wrote:
>
>> Gilbert, Joseph wrote:
>>
>>> Yes correctNorMap sounds like it should be correct for the texture
>>> button. Col, Nor, etc are part of a bit toggle regarding 
>>> mtex->mapto. These should be module constants. Col, Nor, Var are 
>>> attributes that relate to
>>> mtex->colFac, mtex->norFac, and mtex->varFac attributes of the Mtex
>>> struct.
>>>
>> My patch avoids exposing the bitfield to the Python scripter and 
>> instead provides Python Booleans. (For example, there's a 
>> correctNorMap attribute that gives you True or False rather than 
>> having to do math with some bitfield and a constant.) The interface 
>> is a lot nicer IMO. Anyway, because of this, I have separate 
>> variables for the map to buttons.
>
>
> I think this is a good approach to making the BPython API resemble the 
> UI as much as possible.  In fact, I was prepping for "tp_getset-ing" 
> sceneRender.c (Blender.Scene.RenderData) and noticed that Joseph had 
> done this for the myriad of button in the Scene UI.  (Plus he had a 
> nice generic bitfield manipulation procedure which I modified slightly 
> and put into gen_utils.c)  So while I'm adding new atributes to 
> modules, can I get a consensus that I also implement enable/disable 
> attribures for bitfield access?
>
> Ken
>
>
> _______________________________________________
> Bf-python mailing list
> Bf-python at projects.blender.org
> http://projects.blender.org/mailman/listinfo/bf-python
>
>
>