[Numpy-discussion] Array resize question

[Neil Martinsen-Burrell]

On 2009-06-16 09:42 , Cristi Constantin wrote:
> Good day.
> I have this array:
>
> a = array([[u'0', u'0', u'0', u'0', u'0', u' '],
> [u'1', u'1', u'1', u'1', u'1', u' '],
> [u'2', u'2', u'2', u'2', u'2', u' '],
> [u'3', u'3', u'3', u'3', u'3', u' '],
> [u'4', u'4', u'4', u'4', u'4', u'']],
> dtype='<U1')
>
> I want to resize it, but i don't want to alter the order of elements.
>
> a.resize((5,10)) # Will result in
> array([[u'0', u'0', u'0', u'0', u'0', u' ', u'1', u'1', u'1', u'1'],
> [u'1', u' ', u'2', u'2', u'2', u'2', u'2', u' ', u'3', u'3'],
> [u'3', u'3', u'3', u' ', u'4', u'4', u'4', u'4', u'4', u''],
> [u'', u'', u'', u'', u'', u'', u'', u'', u'', u''],
> [u'', u'', u'', u'', u'', u'', u'', u'', u'', u'']],
> dtype='<U1')
>
> That means all my values are mutilated. What i want is the order to be
> kept and only the last elements to become empty. Like this:
> array([[u'0', u'0', u'0', u'0', u'0', u' ', u'', u'', u'', u''],
> [u'1', u'1', u'1', u'1', u'1', u' ', u'', u'', u'', u''],
> [u'2', u'2', u'2', u'2', u'2', u' ', u'', u'', u'', u''],
> [u'3', u'3', u'3', u'3', u'3', u' ', u'', u'', u'', u''],
> [u'4', u'4', u'4', u'4', u'4', u' ', u'', u'', u'', u'']],
> dtype='<U1')
>
> I tried to play with resize like this:
> a.resize((5,10), refcheck=True, order=False)
> # SystemError: NULL result without error in PyObject_Call
>
> vCont1.resize((5,10),True,False)
> # TypeError: an integer is required
>
> Can anyone tell me how this "resize" function works ?
> I already checked the help file :
> http://docs.scipy.org/doc/numpy/reference/generated/numpy.ndarray.resize.html
> Thank you in advance.

Resize takes the existing elements and puts them in a new order.  This 
is its purpose.  What you want to do is to stack two arrays together to 
make a new array:

In [4]: x = np.random.random((5,5))

In [6]: x.shape
Out[6]: (5, 5)

In [7]: y = np.zeros((5,5))

In [8]: y.shape
Out[8]: (5, 5)

In [10]: z = np.hstack((x,y))

In [11]: z.shape
Out[11]: (5, 10)

In [12]: z
Out[12]:
array([[ 0.72215359,  0.32388934,  0.24858866,  0.40907379,  0.26072476,
          0.        ,  0.        ,  0.        ,  0.        ,  0.        ],
        [ 0.59085241,  0.88075534,  0.2288914 ,  0.49258006,  0.28175061,
          0.        ,  0.        ,  0.        ,  0.        ,  0.        ],
        [ 0.50355137,  0.30180634,  0.09177751,  0.08608373,  0.04114688,
          0.        ,  0.        ,  0.        ,  0.        ,  0.        ],
        [ 0.06053053,  0.80426792,  0.21038812,  0.28098004,  0.88956146,
          0.        ,  0.        ,  0.        ,  0.        ,  0.        ],
        [ 0.17359959,  0.4629072 ,  0.30100704,  0.45434713,  0.86597028,
          0.        ,  0.        ,  0.        ,  0.        ,  0.        ]])

This should work the same with your unicode arrays in place of the 
floating point arrays here.  (To make an array with all empty strings, 
you can do y = np.empty((5,5), dtype='U'); y[:] = '')

-Neil