[Bf-committers] Change Representation and Initialisation of Matrices to Conform with Standard Notation

[Campbell Barton]

Changes to the __repr__ function should evaluate, since this is common practice,
see: http://docs.python.org/py3k/library/functions.html#repr

currently mathutils and bpy do this where possible, eg.

>>> Euler()
Euler((0.0, 0.0, 0.0), 'XYZ')

>>> bpy.context.object
bpy.data.objects['Armature']

# note how str does _not_ always do this, though for mathutils we only
have repr at the moment.
>>> print(bpy.context.object)
'<bpy_struct, Object("Armature")>'


If we want a better way to represent mathutils data, I'd prefer to use
__str__, or add a method that returns a pretty formatted matrix.

Another reason this would be nice is repr on floats gives hard to read
results sometimes, a custom printing function can give aligned columns
without too many decimal places, same goes for vectors and other
types.
eg:

Matrix(((1.0, 3.142857074737549, 0.0, 0.0),
        (0.0, 1.0, 0.0, 1.0000000195414814e-24),
        (0.0, 0.0, 1.0, 0.0),
        (3.1415927410125732, 0.0, 0.0, 1.0)))

a str could display this in a more humanly readable way - something like..

http://pastie.org/2993250
---
<Matrix [1.0,    0.0,      0.0, 3.1415],
        [3.1428, 1.0, 100001.0, 0.0   ],
        [0.0,    0.0,      1.0, 0.0   ],
        [0.0,    1.0,      0.0, 1.0   ]
        >
---

The main annoyance of this is that script authors will need to know
that str(matrix) gives a different result to repr(matrix), but we can
document it so don't think its too bad.
print(matrix), uses str(), so its mainly the console where devs will
see the repr output by default.

I don't think we'll get much resistance to having humanly readable
__str__ output, the functionality is fairly easy to add and this kind
of change can be made at any time, so I rather keep this discussion
focused on changing behavior of the api.

On Sat, Dec 10, 2011 at 2:01 AM, Andrew Hale <trumanblending at gmail.com> wrote:
> Perhaps we could consider this as two separate proposals;
>
>   1. Change the repr function for matrices so that they print columns as
>   columns.
>   2. Change how Python gets/sets data from C so that matrices in Python
>   maintain the standard math notation of matrix[row][column]
>
> The first should not effect any scripts as it only relates to printing in
> the console but should hopefully decrease confusion with users. The second
> will break compatibility for some scripts and so will require more
> discussion.
>
> Thanks,
> Andrew
>
> On Sat, Dec 10, 2011 at 12:27 AM, Domino Marama
> <domino at dominodesigns.info>wrote:
>
>> On Fri, 2011-12-09 at 23:57 +1100, Andrew Hale wrote:
>> > Also, note that MatLab stores arrays in column major format but still
>> uses
>> > matrix[row, column] for indexing entries.
>>
>> numpy also uses row, column indexing
>>
>> http://docs.scipy.org/doc/numpy/reference/generated/numpy.matrix.html
>>
>> >>> import numpy
>> >>> m = numpy.matrix("1 2; 3 4; 5 6")
>> >>> m
>> matrix([[1, 2],
>>        [3, 4],
>>        [5, 6]])
>> >>> m[0]
>> matrix([[1, 2]])
>> >>> m[0,1]
>> 2
>>
>> This is the order I'd expect for any matrix implementation without
>> reading any docs too..
>>
>>
>> _______________________________________________
>> Bf-committers mailing list
>> Bf-committers at blender.org
>> http://lists.blender.org/mailman/listinfo/bf-committers
>>
> _______________________________________________
> Bf-committers mailing list
> Bf-committers at blender.org
> http://lists.blender.org/mailman/listinfo/bf-committers



-- 
- Campbell