FILE ARRAY is a variable that assigns an array type and
structure to a disk file, so that you can access and modify the disk
file as if it were an
array (see Arrays). Since
interaction with disk files takes longer than interaction with internal
memory, working with file arrays is slower than working with ordinary
arrays. However, ordinary arrays are limited in size by the size of
your work memory, whereas file arrays are limited in size by the size of
your hard disk, which is usually much bigger.
You create a file array by using one of the functions
cdblfarr. (See bytfarr, intfarr,
lonfarr, fltfarr, dblfarr, cfltfarr, and
cdblfarr.) For instance:
float file array with dimensions 200 by 100, associated
with file file.dat. If the file does not exist yet, then it is
created the first time you write data to it. You use a file array just
like an ordinary array, except that the
+ summation flag is not
(yet) implemented for file arrays.
File arrays assume that the data values in the disk file have the byte order that is appropriate for the current computer system. FZ files are always stored in little-endian order, so file arrays reading from FZ files currently only works on little-endian machines.