Intel® Fortran Compiler Classic and Intel® Fortran Compiler Developer Guide and Reference

ID 767251
Date 3/22/2024
Public
Document Table of Contents

Implied-DO Lists in I/O Lists

In a data transfer statement, an implied-DO list acts as though it were a part of an I/O statement within a DO loop. It takes the following form:

( list, do-var = expr 1, expr 2 [, expr 3])

list

Is a list of variables, expressions, or constants (see Simple List Items in I/O Lists ).

do-var

Is the name of a scalar integer or real variable. The variable must not be one of the input items in list.

expr

Are scalar numeric expressions of type integer or real. They do not all have to be the same type, or the same type as the DO variable.

The implied-DO loop is initiated, executed, and terminated in the same way as a DO construct.

The list is the range of the implied-DO loop. Items in that list can refer to do-var, but they must not change the value of do-var.

Two nested implied-DO lists must not have the same (or an associated) DO variable.

Use an implied-DO list to do the following:

  • Specify iteration of part of an I/O list

  • Transfer part of an array

  • Transfer array items in a sequence different from the order of subscript progression

If the I/O statement containing an implied-DO list terminates abnormally (with an END, EOR, or ERR branch or with an IOSTAT value other than zero), the DO variable becomes undefined.

Examples

The following two output statements are equivalent:

  WRITE (3,200) (A,B,C, I=1,3)            ! An implied-DO list

  WRITE (3,200) A,B,C,A,B,C,A,B,C         ! A simple item list

The following example shows nested implied-DO lists. Execution of the innermost list is repeated most often:

  WRITE (6,150) ((FORM(K,L), L=1,10), K=1,10,2)

The inner DO loop is executed 10 times for each iteration of the outer loop; the second subscript (L) advances from 1 through 10 for each increment of the first subscript (K). This is the reverse of the normal array element order. Note that K is incremented by 2, so only the odd-numbered rows of the array are output.

In the following example, the entire list of the implied-DO list (P(1), Q(1,1), Q(1,2)...,Q(1,10)) are read before I is incremented to 2:

  READ (5,999) (P(I), (Q(I,J), J=1,10), I=1,5)

The following example uses fixed subscripts and subscripts that vary according to the implied-DO list:

  READ (3,5555) (BOX(1,J), J=1,10)

Input values are assigned to BOX(1,1) through BOX(1,10), but other elements of the array are not affected.

The following example shows how a DO variable can be output directly:

  WRITE (6,1111) (I, I=1,20)

Integers 1 through 20 are written.

Consider the following:

        INTEGER mydata(25)
        READ (10, 9000) (mydata(I), I=6,10,1)
  9000  FORMAT (5I3)

In this example, the iolist specifies to put the input data into elements 6 through 10 of the array called mydata. The third value in the implied-DO loop, the increment, is optional. If you leave it out, the increment value defaults to 1.

See Also