[Numpy-discussion] Database with Nulls to Numpy Structure

Christopher Barker Chris.Barker@noaa....
Fri Oct 2 22:38:55 CDT 2009

Mark Janikas wrote:
> Thanks for the input!  I wonder if I can resize my own record array?  I.e. one call to truncate... Ill give it a go.

you should be able too, yes. Be careful though, you can't call resize() 
if there are any other references to the array.

>  But the resize works great as it doesn't make a copy:

Actually, it's not that simple. With numpy arrays, there is the array 
object itself, and there is the data block that the array points to. Whn 
you call resize() it may make a copy of the data block (which is why it 
won't work if there are other references to it), while keeping the same 
python object.

> In [12]: a = NUM.arange(10)
> In [13]: id(a)
> Out[13]: 190182896
> In [14]: a.resize(5,)
> In [15]: a
> Out[15]: array([0, 1, 2, 3, 4])
> In [16]: id(a)
> Out[16]: 190182896

So this shows you have the same python object. I think there is a way to 
get the value of the pointer to the data block, but I dont' know off the 
top of my head how.

> Whereas the slice seems to make a copy/reassign:
> In [18]: a = a[0:2]
> In [19]: id(a)
> Out[19]: 189981184

slicing creates a new python object, but it doesn't copy the actual data:

In [4]: b = a[2:5]

In [5]: a is b
Out[5]: False

In [6]: a[2:5] = 10

In [7]: a
Out[7]: array([ 0,  1, 10, 10, 10,  5,  6,  7,  8,  9])

In [8]: b
Out[8]: array([10, 10, 10])

so you can see a and b are different python objects, but they share the 
same data block.


Christopher Barker, Ph.D.

Emergency Response Division
NOAA/NOS/OR&R            (206) 526-6959   voice
7600 Sand Point Way NE   (206) 526-6329   fax
Seattle, WA  98115       (206) 526-6317   main reception


More information about the NumPy-Discussion mailing list