ind_resolve
Resolves a single list of indices to their multi-dimensional representation.
Prototype
function ind_resolve ( indices [*] : byte, short, integer or long, dsizes [*] : byte, short, integer or long ) return_val : integer or long
Arguments
indicesA list of indices into a one-dimensional array that was converted to a one-dimensional array from a multi-dimensional array.
dsizesA list of dimension sizes of the multi-dimensioned array mentioned above.
Description
ind_resolve takes a list of indices into a one-dimensional array (that was converted from a multi-dimensional array) and returns the list of indices that they represent in the multi-dimensional array. The array returned will be dimensioned N x M, where N is the number of indices and M is the length of dsizes (the number of dimensions of the multi-dimensional array).
As of version 6.0.0, this function allows "long" input for the indices and/or dimension sizes, which potentially allows you to have values greater than or equal to 2 gigabytes (GB) on a 64-bit system. This function will return longs if on a 64-bit system and any of the values are >= 2 GB.
See Also
Examples
Example 1
The following example demonstrates the usefulness of the ind_resolve function in conjunction with the ind function:
; ; Create sample multi-dimensioned array. ; The following creates a 3D array of size 2 x 2 x 4. ; a = (/(/(/1,2,3,4/), (/5,6,7,8/)/), (/(/9,10,9,8/),(/7,6,5,4/)/)/) ; ; Returns the index locations where "a" is greater than 5. ; NCL's ndtooned is used to temporarily ; reshape "a" as is required by the ind function. ; ; a1D = ndtooned(a) dsizes_a = dimsizes(a) indices = ind_resolve(ind(a1D.gt.5),dsizes_a) print(indices) ; Set all elements of "a" which are greater than 5 to -999. dim_ida = dimsizes(indices) npts = dim_ida(0) ; number of elements > 5 (here 9) ndim = dim_ida(1) ; rank of "a" (here 3) a@_FillValue = -999 do n=0,npts-1 a(indices(n,0),indices(n,1),indices(n,2)) = a@_FillValue end do print(a)
There are nine index locations where a is greater than 5: (0,1,1), (0,1,2), (0,1,3), (1,0,0), (1,0,1), (1,0,2), (1,0,3), (1,1,0), and (1,1,1). The above script, then, will return an array dimensioned 9 x 3 with the following values:
Variable: indices
Type: integer
Total Size: 108 bytes
27 values
Number of Dimensions: 2
Dimensions and sizes: [9] x [3]
Coordinates:
Number Of Attributes: 1
_FillValue : -999
(0,0) 0
(0,1) 1
(0,2) 1
(1,0) 0
(1,1) 1
(1,2) 2
(2,0) 0
(2,1) 1
(2,2) 3
(3,0) 1
(3,1) 0
(3,2) 0
(4,0) 1
(4,1) 0
(4,2) 1
(5,0) 1
(5,1) 0
(5,2) 2
(6,0) 1
(6,1) 0
(6,2) 3
(7,0) 1
(7,1) 1
(7,2) 0
(8,0) 1
(8,1) 1
(8,2) 1
After the do loop, the array a contains the following:
Variable: a
Type: integer
Total Size: 64 bytes
16 values
Number of Dimensions: 3
Dimensions and sizes: [2] x [2] x [4]
Coordinates:
(0,0,0) 1
(0,0,1) 2
(0,0,2) 3
(0,0,3) 4
(0,1,0) 5
(0,1,1) -999
(0,1,2) -999
(0,1,3) -999
(1,0,0) -999
(1,0,1) -999
(1,0,2) -999
(1,0,3) -999
(1,1,0) -999
(1,1,1) -999
(1,1,2) 5
(1,1,3) 4
Example 2
Determine all locations of a multi-dimensional array (x) where the elements match the minimum or maximum values. This is a more general approach than offered by the maxind and minind functions:
xMax = max(x) xMin = min(x) x1D = ndtooned(x) ; only do this once indMax = ind_resolve(ind(x1D.eq.xMax),dimsizes(x)) ; locations of max indMin = ind_resolve(ind(x1D.eq.xMin),dimsizes(x)) ; locations of min delete (x1D)
Example 3
Assume x is a three-dimensional array with coordinate variables (time,lat,lon). Print the coordinates where -3 < x < 9. Before executing various statements, check to make sure that at least some values of x satisfy the criteria. Use the num function to test if there are values between -3 and 9, then print the times, latitudes, and longitudes:
if (num(x.gt.-3. .and. x.lt.9.) .gt. 0) then x1D = ndtooned(x) i1D = ind(x1D.gt.-3. .and. x1D.lt.9.) i = ind_resolve(i1D, dimsizes(x) ) ; 2D [npts,3] (rank(x)=3) print(time(i(:,0)) +" "+ lat(i(:,1)) +" "+lon(i(:,2)) ) delete(x1d) ; no longer needed delete(i1d) end if