Array Syntax

Using the earlier declarations, references can be made to:

• whole arrays (conformable)
  • A = 0 -- sets whole array A to zero; scalars always conform with arrays.
  • B = C + 1 -- adds one to all elements, of C and then assigns each element to the corresponding elements of B.

    For this to be legal Fortran 90 both arrays in the RHS expression must conform (B and C must be same shape and size). The assignment could have been written B(:) = C(:) + 1 demonstrating how a whole array can be referenced by subscripting it with a colon. (This is shorthand for lower_bound:upper_bound and is exactly equivalent to using only its name with no subscripts or parentheses.)

• elements
  • A(1) = 0.0 sets one element to zero,
  • B = A(3) + C(5,1) sets whole array B to the sum of two elements.
  If present, array subscripts must be integer valued expressions.

  A particular element of an array is accessed by subscripting the array name with an integer which is within the bounds of the declared extent. Subscripting directly with a REAL, COMPLEX, CHARACTER, DOUBLE PRECISION or LOGICAL is an error. This, and indeed the previous example, demonstrates how scalars (literals and variables) conform to arrays; scalars can be used in many contexts in place of an array.

• array sections
  • A(2:6) = 0 sets section of A to zero,
  • B(-1:0,1:2)=C(1:2,2:3)+1 adds one to the subsection of C and assigns to the subsection of B.
  The above examples demonstrate how parts (or subsections) of arrays can be referenced. An array section can be specified using the colon range notation first encountered in the SELECT CASE construct. In addition, the sequence of elements can also have a stride (like DO loops) meaning that the section specification is denoted by a subscript-triplet which is a linear function.

Care must be taken when referring to different sections of the same array on both sides of an assignment statement, for example,

    DO i = 2,15
     A(i) = A(i) + A(i-1)
    END DO

is not the same as

    A(2:15) = A(2:15) + A(1:14)

in the first case, a general element i of A has the value,

    A(i) = A(i) + A(i-1) + ... + A(2) + A(1)

but in the vectorised statement it has the value,

    A(i) = A(i) + A(i-1)

In summary both scalars and arrays can be thought of as objects. (More or less) the same operations can be performed on each with the array operations being performed in parallel. It is not possible to have a scalar on the LHS of an assignment and a non scalar array reference on the RHS unless that section is an argument to a reduction function.