| Array with float indices. [message #85000] |
Fri, 21 June 2013 05:01  |
Nikola
Messages: 53
Registered: November 2009
|
Member |
|
|
I just discovered a bug in the code that I'm writing. It was caused by a nasty feature of IDL that I wasn't aware before. This behaviour - maybe naively - I would call erratic.
Briefly, I use an array of indeces (xind) to extract a subarray (y) from an array (x). Something like:
x = FINDGEN(5)/5
xind =[2, 3, 4]
y = x[xind]
The indeces xind are computed by a subroutine that gives a FLOAT output. It may happen that the indeces are out of the range of the original array (x), e.g. xind = [3, 4, 5]. I was naively assuming that my code would crash on that. Nevertheless, it turned out that the code can live with that as long as the indeces are submitted as a float array with arbitrary values?!? Using the same example as above:
print, x[5]
% Attempt to subscript X with <INT ( 5)> is out of range.
% Execution halted at: $MAIN$
print, x[5.]
% Attempt to subscript X with <FLOAT ( 5.00000)> is out of range.
% Execution halted at: $MAIN$
print, x[[5.]]
0.800000
print, x[[15.]]
0.800000
Is this actually a feature of IDL? Is it documented somewhere? I found it to be quite dangerous as a potential source of bugs. I expected that
(1) x[any scalar] = x[[any scalar]]
(2) x[float scalar] = x[FLOOR(float scalar) or ROUND(float scalar)]
but not
x[[float]] = x[float < N_ELEMENTS(x)-1]
Note: It's common for 7.1 and 8.2.3.
|
|
|
|
|
|
|
|
| Re: Array with float indices. [message #85003 is a reply to message #85001] |
Fri, 21 June 2013 05:30  |
Lajos Foldy
Messages: 176
Registered: December 2011
|
Senior Member |
|
|
On Friday, June 21, 2013 2:21:20 PM UTC+2, David Fanning wrote:
> Nikola Vitas writes:
>
>
>
>> Is this actually a feature of IDL? Is it documented somewhere? I found it to be quite dangerous as a potential source of bugs.
>
>
>
> Yes, it is one of the lesser known features, but it has been a feature
>
> of IDL from forever. It is certainly documented *somewhere*. I just re-
>
> read the "Array - subscripts" section of the documentation, though, and
>
> I didn't find it there. So, who knows where it is. I can't seem to find
>
> anything in the IDL Help these days. :-(
>
Language > Arrays > Using Arrays as Array Subscripts
Array Subscripts and Clipping
If an element of the subscript array is less than or equal to zero, the first element of the subscripted array is selected. If an element of the subscript array is greater than or equal to the last subscript in the subscripted array, the last element is selected.
regards,
Lajos
|
|
|
|
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
