Digital PDFs

AA-R953A-TK

March 1983

161 pages

Original

7.4MB

view download

OCR Version

7.2MB

view download

Document:	F4LangRef Mar83
Order Number:	AA-R953A-TK
Revision:	0
Pages:	161
Original Filename:	AA-R953A-TK_F4LangRef_Mar83.pdf

OCR Text

PDP-11 FORTRAN IV
Language Reference Manual
Order No. AA-R953A-TK

March 1983

This document describes the fundamentals of the FORTRAN IV language
elements, as implemented for the PDP-11 systems.

SUPERSESSION/UPDATE INFORMATION:

This manual contains
information concerning

FORTRAN

March

1983

SOFTWARE VERSION TO
OPERATING SYSTEM AND VERSION:

FORTRAN |V V2.6
e RSTS/E V8.0
* RT-11 V5B
e RSX-11M V4.1

¢ RSX-11M-PLUS V2.1
VAX/VMS V3.0

digital equipment corporation - maynard, massachusetts

First Printing, June
Revised: December

1974
1974

December

1975

June

1977

September

1979

March 1983

The
and

information
should

not

this
be

document

construed

subject

change

commitment

Corporation.
Digital Equipment Corporation assumes
for any errors that may appear in this document.
The

software

and
may
license.

described

wused

this

copied

document

only

furnished

accordance

without
Digital

responsibility

under

with

notice

Equipment

the

license

terms

such

No responsibility is assumed for the use or reliability of software on
equipment that is not supplied by Digital Equipment Corporation or its
affiliated companies.

1974,

Printed

The

postpaid

READER'S

requests
the user's
documentation.

The

following

are

1975,

1977,

1979,

1983

Digital Equipment Corporation
All Rights Reserved.

COMMENTS
critical

trademarks

form

U.S.A.

the

evaluation

Digital

last

page

assist

Equipment

this

document

preparing

Corporation:

DEC

DIBOL

DEC/CMS

EduSystem

UNIBUS

DECnet

IAS

VAX

DECsystem-10

MASSBUS

VMS

DECSYSTEM-20

PDP

DECUS

PDT

Logo

DECwriter

RSTS

RSX

mflannan

ZK2266

HOW TO ORDER ADDITIONAL DOCUMENTATION
in Continental USA and Puerto Rico call

800-258-1710

In New Hampshire, Alaska, and Hawaii call 603-884-6660
In Canada call

613-234-7726 (Ottawa-Hull)

800-267-6146 (all other Canadian)
DIRECT MAIL ORDERS (USA & PUERTO RICO)*

DIRECT MAIL ORDERS (CANADA)

Digitat Equipment of Canada Ltd.

940 Belfast Road
Ottawa, Ontario K1G 4C2

Attn: A&SG Business Manager
DIRECT MAIL ORDERS (INTERNATIONAL)

Digital Equipment Corporation

P.O. Box CS52008

A&SG Business Manager

Nashua, New Hampshire 03061

c/o Digital's local subsidiary or
approved distributor

*Any prepaid order from Puerto Rico must be placed
with the local Digital subsidiary (809-754-7575)

Internal orders should be placed through the Software Distribution Center (SDC), Digital Equipment

Corporation, Northboro, Massachusetts 01532

future

CONTENTS

Page

PREFACE

e
s

e
e

+«

.«

¢«

&«

.«

¢«

&«

Logical Constants
.
Hollerith Constants

«
.

¢«
.

«
«

o
.

&
&

=
N

Constants

Type
.

Literals

Rules
.

for
¢

Specification

Implication

.«

+«

¢«

.«

.+«

.«

Data

Array

Array References Without
Adjustable Arrays
.
«
¢«

Declarators

¢«

Storage

Type

EXPRESSIONS
=

¢«
.

|
I
I

2-13

.
.

2-14
2-15

Logical

2-12

2-12
2-13

type

Subscripts
o
«
o
&

2-12

Data

Array

Arithmetic Expressions
Use of Parentheses .
Relational

Constants
o

Type

Subscripts

Type

Array

AU
B WN

Data

Hollerlth

Data
ARRAYS

Constants

Data

COMPONENTS

Precision

VARIABLES

.« o

+«

¢«

Alphanumeric

el e I e

¢«

Field

Constants

Complex

Indlcator

¢«

Constants

Double

Real

STRUCTURE

NAMES

STATEMENT

FORTRAN

DN
NN

UNIT

PROGRAM

Number

Sequence

O COWVWOOOIJAOAATUH
BN

Field

Integer

Field

Statement

CONSTANTS

Field

Statement

Continuation

TYPES

Formatting

Indicator

Debugging

LINE

Set

FORTRAN

Label

Comment

e
s

Statement

¢
e

e
o

NN
BB W WN N

¢
+

Character

Tab-Character

OV D WWWN
-

.«
+

Character-per-Column

DATA
AU
D W)

.«
«

PROGRAMS

[

FORTRAN

= (WP

—

OVERVIEW

FORTRAN

NN
N

LANGUAGE

SYMBOLIC

PDP-11

ELEMENTS

FORMATTING

WK
=

NN NN NNDNNNNDNNNBNNNNONNDNNNDNNNND
MDD
DNNODN

CHAPTER

Statements
Comments .

o B e

S py W

DWWWWwWwwwwhDoND
NP

INTRODUCTION

SOOI
OTd B B RWWWWWwWwwwN -

CHAPTER

Arithmetic

Expressions

iii

.
.

.
«

Expression

2-16

2-17
2-18

.
&

CONTENTS

Statement

STATEMENTS
.«
4
¢
¢
o
o
Arithmetic IF Statement

o
.

o
o
¢«
¢

o
o

s
o

o
o

o
&

Logical

&«

o«

Control

o«

DO LOOPS
+
¢
o
¢
¢

=+
¢

&«
o

«
o

o
o

o«
o

o
o

LOOPS

Control Transfers in
Extended Range .
. ¢
STATEMENT

CALL STATEMENT .
RETURN STATEMENT

o
¢

o
o

o
¢

o
o

s
o

o
o

¢«

STA,TEMENT

L]
L

SPECIFICATION

IMPLICIT STATEMENT .
¢
&
o
4
TYPE DECLARATION STATEMENTS

4
.

¢«

+«

¢«

Use

Virtual

Arrays

Equivalent

¢«

o«

¢«

Blocks

STATEMENT

Subprograms

STATEMENT

ARGUMENTS

¢«

STATEMENT

the

&«

!
WWNHHFOLOIJAADd
WNH

References

Common

STATEMENT

DATA

STATEMENT

Extending

Array

Arrays

EXTERNAL

TR
N

Making

BLOCK

STATEMENT

EQUIVALENCE

STATEMENTS

Restrictions

PROGRAM

STATEMENT

Virtual

DATA

STATEMENT

VIRTUAL

DIMENSION
COMMON

[

STATEMENT

[

¢«

o
o
¢

END

.
4
o

PAUSE

STOP

o
.

CONTINUE

Statement

Iteration

Nested

STATEMENT

|
LY >
ST~
S
=
e |
U|
Hi= O OWW O]

LGN

Assigned

S
S A
T L N LSO S
Y
Y.
|I
[I
TR I
B
IOULH WWND DN -

Statement

S IGNGRS

Statement

=
N

¢«

N
w

s
.

L]
o

W
-

STATEMENTS

Computed

WCOJOAUMd

WWWWWNNON R

A
&
&

&
&

S
9

S
o

T
s

Y
L

Y
o
.

B
O
S

ooy
o
e
o
o
o
o
o
o
e
o

¢«

USER-WRITTEN SUBPROGRAMS
Statement Functions
.

.
.

¢
¢«

&«
¢«

¢
&«

4
¢

o
¢

¢
o

o
o

o
&

¢«

Subprograms

AT
YA
YO
{
I
OO dWwwNnr

L]
.

®
L)

L4
[}

L
.

*
.

L]
L]

L
[}

L
.

L
[

L4
L]

.
.

7
7

SUBPROGRAM

w N

NN
L]
N
WN

SUBPROGRAMS

"Rules

[]
.

W N

Governing

Adjustable

Function

Subprogram

Arrays

Subprograms

Subroutine
FORTRAN

N NN

CHAPTER

STATEMENTS

Unconditional

STATEMENT

W
W

[1~Y

STATEMENT

ASSIGNMENT
STATEMENT

ASSIGN

CONTROL

AT
A

CHAPTER

ASSIGNMENT

LOGICAL

CHAPTER

STATEMENTS

ARITHMETIC

CHAPTER

ASSIGNMENT

CHAPTER

W ww
|
ww

Page

LIBRARY

FUNCTIONS

INPUT/OUTPUT

STATEMENTS

I/O OVERVIEW
Records
.

.
.

L]
[

*
[}

®
.

*
.

L4
[]

Ll
[

o«

Arguments

Files

o«

ACCeSS

MOAES

o«

CONTENTS

Page

1-2

Direct Access
.
© e s
I/0 STATEMENT COMPONENTS e o o
Logical Unit Numbers . « « «
Format Specifiers
=« « o« o+ o
Direct Access Record Numbers

AcCCeESS

e
o
«
o
.

s
o
«
o
«

e
o
¢
o
«

o
o
o
o
«

o
o
o
o
o«

o
s
o
o
o

e
e
o
o
o

o
o
o
o
o

o
o
o«
o
o

o
o
o
»
o+

1-3
1-3
1-3
1-3
1-3

o
o
o
o
e
.

o
o
o
»
o
.

o
o
o
o
e
o

1-3
1-4
1-5
[1-6
1-7
7-7

.
e
«
«
o«
o
.
.
o

.
o«
o
o
o
@
.
.
o

7-9
1-10
7-12

FIELD DESCRIPTORS
'« & &« o o o s o o o o s o o
I Field DesSCriptor « « o o o o o o o s o o o
O Field Descriptor « « « o o o o o o s o s o
F Field Descriptor « « « o o ¢ s o o o o« o o
E Field DescCriptor « « « o o o o o s o o o s«
D Field DesScCriptor « « o o o o o o o o o o o
G Field DescCriptor +« o + o o« o o o o s o s o
L Field Descriptor +« « o + o o s s o s o o s
A Field Descriptor « o « o o « o o o ¢« o o +
H Field Descriptor « « o« o o o o o o o o o «
X Field DesSCriptor « o o o o o o o o o o o o
T Field Descriptor « « « o« o« o o o o o o o o
Q Field Descriptor « « « o ¢ o o o o o o o «
Dollar Sign Descriptor + « « « o« o o o o o o«
Colon DeSCripPtor « « « « o o s o o o o« o o «
Complex Data Editing « « « &+« o o« o ¢ o & « «
Scale Factor . . .
.
.
e o o o o s+ o«
Repeat Counts and Group Repeat Counts
.+ « .
Default Field Descriptors
. « « ¢« o« o o o o+
CARRIAGE CONTROL CHARACTERS
. « ¢ « ¢« o« o « o+
FORMAT SPECIFICATION SEPARATORS
. ¢« ¢« o o o o
EXTERNAL FIELD SEPARATORS
.+ ¢ ¢+ ¢ o o o o « «
RUN-TIME FORMATS . . . « .+ &
.
.
FORMAT CONTROL INTERACTION WITH INPUT/OUTPUT
LISTS
¢« « o s o o o o o o o o o o o o o o o o«
SUMMARY OF RULES FOR FORMAT STATEMENTS . . . .
General RULlES
o &+ o + o o o o o o o o o o o
Input RULIES
o o o« ¢ o o o o o o o o o o oo
Output RULES &« « o o ¢ s o o o o o o o o o o

o
o
o«
o
o+
o
s+
o«
«
o
o
«
o«
o«
»
o+
o
.
o«
.
&
o
.

»
o«
»
o«
o+
o«
«
o«

End-of-File Condition and Error Condition
Parameters . . .
¢t e o e o o o s o e o e

7.2.5
Input/Output LlStS e e o s s s e o e e e o
7.2.5.1
Simple LiStS « ¢« ¢ o o o o o o o o o o o
7e2.5.2
Implied DO LiStS o « o o o o o o o o o o
7.3
SEQUENTIAL INPUT/OUTPUT
.
e s e e s e e e
7.3.1
Formatted Sequential Input Statements
. .
7.3.2
Formatted Sequential Output Statements . .
7.3.3
List-Directed Input Statements . . . . .
7.3.4
List-Directed Output Statements
. « « « «
7435
Unformatted Sequential Input Statement
«
7.3.6
Unformatted Sequential Output Statements «
7.4
DIRECT ACCESS INPUT/OUTPUT . « ¢ o o o o o s
7.4.1
Unformatted Direct Access Input Statement
7.4.2
Unformatted Direct Access Output Statement
7.5
ENCODE AND DECODE STATEMENTS .+ o o « o o o o

7-15
7-15
7-16
7=17

[)

[]

[)
o

Keyword

8-23
8-23

BUFFERCOUNT

Keyword

8-21
8-22
8=22

BLOCKSIZE

ASSOCIATEVARIABLE

o+
o«

8-10
8-11
8-12
8-12
8-13
8-13
8-14
8-14
8-15
8-17
8-17
8-18
8-18
8-109
8-20

o
&
o

8=2
8=2
8-4
8-5
8-6
8=7
8-8
8-9

¢«

[}

Keyword

ACCESS

STATEMENT

71-14

STATEMENTS

OPEN

INPUT/OUTPUT

AUXILIARY

[

CHAPTER

8.7
8.7.1
8.7.2
8.7.3

8.1
8.1.1
8.1.2
8.1.3
8.1.4
8.1.5
8.1.6
8.1.7
8.1.8
8.1.9
8.1.10
8.1.11
8.1.12
8.1.13
8.1.14
8.1.15
8.1.16
8.1.17
8.1.18
8.2
8.3
- 8.4
8.5
8.6

1-14

STATEMENTS

FORMAT

CHAPTER

7.2.4

Sequential

7.1.3.2
7.2
7.2.1
7¢2.2
7.2.3

7.1.3.1

APPENDIX

+¢

NOSPANBLOCKS

o
o

&
@

Keyword

+«

o«

Keyword

@
o

Keyword

.«

¢«

TYPE

Keyword

UNIT

Keyword

.
.

.
¢

¢
¢

&
¢

¢
o

o
¢

&«
o

o
o

o
«

o
o

&
&

&
.

¢
¢

4
¢

¢
¢

o
¢

o
o

o
¢«

o
o«

o
o

¢«

o«

.
SET

CLOSE STATEMENT
REWIND STATEMENT
BACKSPACE

FIND STATEMENT +
&+
ENDFILE STATEMENT
DEFINE

STATEMENT

FILE

CHARACTER

A.l

FORTRAN

A,2

ASCII

A.3

RADIX-50

FORTRAN

STATEMENT

.«

Keyword

RECORDSIZE
SHARED

Keyword

READONLY

O HWNEHO
CO

o b
b

e b

b
.

= e

NAME

WN)

[]

MAXREC

Keyword

INITIALSIZE

]

FORM

2 b b

=
&

o
o

R OCWOOUWYWOOOODIIIJdO
DN

o
&

NN H

o
o

— o

o«

«
o

o«

¢
+

¢
«

*®

&
«

W WOWWw
W OWw
i
WO OWWOWWIWOWWOUWYOOY
WY

¢«

o
.

Keyword

¢«

ERR Keyword
.
.
¢
EXTENDSIZE Keyword

DISPOSE

N AU

WCOWYW O OWOYVWLOYWWOWWOVYWWOWWOUWYYOWYY
o

CONTENTS

SETS

CHARACTER

SET

CONSTANTS

LANGUAGE

.
AND

CHAR ACTER

SUMMARY

B.l

EXPRESSION

OPERATORS

B.2

STATEMENTS

B.3

LIBRARY

¢«

«
.

.
.

FUNCTIONS

INDEX

1-1

FORTRAN

1-2

Line

1-3
2-1

Array

4-1

Nested

4-2
5-1

Control Transfers and Extended
Equivalence of Array Storage .

Form

Required

Order

Storage

Example

and

¢
Lines

©¢

Range

o«

.
.

.
.
.

of
.«

LoOPS

Statements

Coding

Formatting

FIGURE

O 00 ~JWw-~J U

FIGURES

Statements

Effect

Data

Magnitude

Default Field Widths .
.
.
«
Carriage Control Characters

¢«
.

Summary

OPEN

FORMAT

Statement

CodesS

Keyword

Values

.«

Formats
¢«
¢
¢
«
. «
o«
o«

.
«
o«

.
&
&

o«

.«

*
L

Wb NDUNWN

Assignment

00000 O

for

Rules

Conversion

Types of User-Written Subprograms
List-Directed Output Formats .
. .

Names

Symbolic

o]
LU
D
B
|
N w
NN
[ Ol
Ll|
= |

Identified

Data Type Storage Requirements .
.
«
Result Data Type for Exponentiation

Entities

W W

TABLE

WOWWOW-JNWNN
N
|

TABLES

CONTENTS

ASCII

Character

Set

Expression Operators .
.
FORTRAN Library Functions

vii

PREFACE

MANUAL

OBJECTIVES

This manual
designed as

describes
the
elements
a reference, rather than

of
PDP-11
FORTRAN
tutorial, document.

and

1is

This document serves as the
FORTRAN
language
reference
manual
for
several
operating systems that run on the PDP-1l1 family of computers.
Therefore, no information specific to an operating system is presented
here.
For
that
information,
refer
to
the
user's guide for each
system,

INTENDED

AUDIENCE

Because this is
understanding of

STRUCTURE

This manual

a
reference
document,
readers
who
FORTRAN will derive maximum benefit.

THIS

basic

DOCUMENT

contains

nine

chapters

and

Chapter 1 consists of general
and

have

introduces

basic

two

appendixes.

information

facts

needed

concerning

for

FORTRAN

writing

FORTRAN

programs.

Chapter 2 describes the components of FORTRAN
as symbolic names, constants, and variables.

Chapter
used in

3 describes assignment
the program.

Chapter
control

4
deals
with
control
statements,
which
from one point in the program to another.

Chapter 5 describes specification
characteristics
of
symbols used
type and array dimensions.

Chapter 6 discusses subprograms,
supplied with PDP-11 FORTRAN 1IV.

Chapter

covers

Chapter

describes

with

formatted

FORTRAN

I/0

the

statements,

which

define

such

values

transfer

statements, which define the
in the program, such as data

both

user-written

and

those

input/output.

FORMAT

statements.

statements

used

conjunction

Chapter
such as

contains information on
auxiliary
CLOSE, and DEFINE FILE.

I/0

statements,

OPEN,

Appendix

FORTRAN

1V.

Appendix

summarizes

the

character

summarizes

the

language

sets

supported

elements

PDP-11

FORTRAN

1v.

ASSOCIATE

The

DOCUMENTS

following

documents

are

interest

PDP-11

FORTRAN

programmers:

RT-11/RSTS/E

FORTRAN IV User's

Guide

RSX,

FORTRAN IV User's

Guide

CONVENTIONS

The

VAX/VMS

USED

THIS

DOCUMENT

following conventions
®

Uppercase
should

words

type

the

are used

and

Lowercase words and
are to substitute a

Brackets

Braces

([])

({})

letters

words

and

used

indicate

that

you

used in examples indicate
value of your choice.

that

you

can

optional

lists

this manual:

letters

letters
word or

enclose

from

examples

shown.

elements.

which

one

element

1is

chosen.

Ellipses
repeated

addition,

(...)
indicate that
or more times.

the

following

characters:
Tab

character

Space

the

preceding

item(s)

one

character

(TAB)

characters

denote

special

nonprinting

CHAPTER
INTRODUCTION

l.1

LANGUAGE

PDP-11

FORTRAN

OVERVIEW

The PDP-11 FORTRAN IV language is based on American National
Standard
(ANS) FORTRAN X3.9-1966 but includes the following enhancements to ANS
FORTRAN:

You can
If
the

use any arithmetic
expression is not

integer

type.

Mixed-mode

The

expressions

following

data

expression
an integer

can

type

contain

has

been

as an
type,

array
subscript.
it is converted to

elements

any data

type.

added:

LOGICAL*1

The IMPLICIT statement
symbolic names.

The

following

redefines

input/output

the

(I/0)

implied

statements

have

data

type

been

added:

TYPE
PRINT
READ
WRITE
FIND

Device-oriented
'

I/0

(u'r)

(u'r
(u'r)

Unformatted

direct-access

I/0

OPEN

CLOSE
DEFINE

File control and
specification

FILE

ENCODE
DECODE

Formatted

data

The specifications END=s and/or ERR=s
or
WRITE
statements
in
order
to
statement
specified
by
s
when
an
condition occurs.

Alphanumeric
apostrophes)

List-directed I/0 can be
a format specification.

Constants

and

1literals
(strings
can be used in place

expressions

REWRITE,

TYPE,

and

conversion

memory

WRITE,

attribute

used

are

of
of

can be included in
READ
transfer
control to the
end-of-file
or
error

characters
bounded
Hollerith constants.

perform

permitted

formatted

statements.

the

I/0

I/O without

1lists

INTRODUCTION

The

statement

PDP-11

increment

FORTRAN

parameter

can

have

negative

value.
®

For

readability,

label
®¢

PROGRAM

You

can

You

statement

include

FORTRAN

point

you can
statement.

can

optionally

used

explanatory

statement.

These

include
D

debugging

statements

column

These

letter

when

you

specify

compiler

they

are

treated

comments.

in
®

statement

can

the

label

use

list

The

arrays

provide

address

space.

programs

sequence

terminated
program

program

and

following

the

program.

the

begin

same

with

line

any

exclamation

simply

command

assigned

consist

FORTRAN

organized

into

statements

that

statement.

END

one

data

referred

PROGRAMS

subprogram.

program
are

placing

compiled

only

otherwise,

the

control

FORTRAN

statements
a

statement

and

units.

outside

parameter

program

unit

program

throughout

normal

the

rest

optional

comments.

unit

computing

executable

optional

areas

program

defines

qualifier;

expression

large

FORTRAN

are

statements

statement.

program

statements

l1.2.1

arithmetic

Virtual

ELEMENTS

FORTRAN

any

computed

PDP-11 FORTRAN IV
of this manual.

1.2

comma

(!).

can

You

main

comment

optional.
®

comments

the

The

use

program
procedure

can

either

consists

and

a
is

main

one

main

subprograms.

Statements

are

grouped

nonexecutable,
program,

into

two

Executable

Nonexecutable

characteristics,

and

general

statements

provide

classes:

specify

describe

editing

and

executable

the

data

actions

and

arrangement

data-conversion

a
and

information.

Statements are divided into physical sections called lines.
A line is
a string of up to 80 characters.
If a statement is too long to fit on
one line, you can continue it on one or more additional lines,
called
continuation
lines.
A
continuation
1line
1is
identified
by
a

continuation character in the sixth column of that line.
(For
information on continuation characters, see Section 1.3.4.)
You

can

identify

statement

transfer

control

statement

label

statement's

with

integer

initial

line.

label

obtain

number

the

placed

that

other

information

the

further

statements
it

first

contains.

five

can
A

columns

INTRODUCTION

l.2.2

PDP-11

FORTRAN

Comments

program

processing

do not affect
documentation
aid

programmer.

describe

the

processes,

actions
and

the

provide

program,

greater

ease

any way.
They are merely
You can use them freely to

identify
in

program

reading

the

sections
source

and

program

listing.
The letter C in the first column of a source line identifies
that
1line
as
a
comment.
In addition, if you place an exclamation
point (!)
in the statement portion of a source line, the rest of
that
line is treated as a comment.
Any

printable

1.2.3
The

character

can

FORTRAN Character

FORTRAN
1.

All

character

comment.

Set

set

uppercase

appear

and

consists

of:

lowercase

letters

through

z)
2.

The

numerals

The

special

through

characters

listed

below:

Character

Name

A OY (AB

Space

Equal

sign

Plus

Minus

Asterisk

Slash

(

Left

tab

sign

parenthesis

)

Right

’

Comma

Period

Apostrophe

parenthesis

Quotation

mark

Dollar

sign

Exclamation

Colon

point

Other printable ASCII characters can appear
in
a
FORTRAN
statement
only
as
a
part of a Hollerith constant or alphanumeric literal
(see
Appendix A for a list of printable characters).
Except in Hollerith constants and alphanumeric literals, the
makes no distinction between uppercase and lowercase letters.

compiler

INTRODUCTION TO

1.3

FORMATTING

Each

FORTRAN

A FORTRAN

line

has

Statement

label

Continuation

Statement

field

Sequence

number

FORTRAN

1IV

LINE

the

PDP-11

following

four

fields:

field

indicator

field

You can format a FORTRAN line in two ways:
1) by typing one character
per
column
(character-per-column)
or
2) by using the tab character
(tab-character)
to
get
from
field
to
field.
You
can
use
character-per-column
formatting
when
punching
cards,
writing on a
coding form, or typing on a terminal keyboard.
However, you
can
use
character-per-column
tab-character
formatting,
in
addition
to
formatting, only when you are typing at a terminal keyboard.

1.3.1
As

Character-per-Column Formatting

shown

in Figure

statement

1-1,

labels,

sequence numbers.
each

field.

Each

column

field

a FORTRAN

line

continuation

Sections

represents

1.3.3

single

1is

divided

indicators,

through

1.3.6

The

character.

1into

fields

for

text,

and

statement

describe

the

use

columns

making

each

follow.
FO RTRAN

CODER

CODING FORM

| mOMEM

DATE

PAGE

€ Connnent i

SY\A"A.A('P:Jl

IDENTIFICATION

FORTRAN STATEMENT

!
<

12345670 91011 12131413161718192021222)M232027282930313233341530317

30394041 424344454647404950513233545536375059606162016405860676868
701
7217374 757627 707964

THIS, PROGRAM CALCULATES PRIME NUMBERS FROM 'L 1O 350, .,

DO 10, A=Vl
+——+-

.\, AN

5,00 2, L bbbttt

J=I
+ +
S 2 b4

S e e et S e S S o b o s S o B e e
+ At
et
S
B iS S S e e ee eS e e o o

+
+

~++—+

=L/A

e T T SIS ARY S

——t

=1/J
B=A-L

—

I F (B)

1F (J,. LJ SQRT,

((FLOAT

TY PE ‘705 !
:AQNTAI NQEL‘ L

‘40
+—+

1,05

i T e

(l)))

e S I S

GQ 1O
4

——

ei e I
Rl ot
S i e o e e T oo e eS e b o
e ee
e
o
e e Ti e e ei o oe S S S

+—+

FORMAT

‘0 5

U4,'IS

e e e

i o o

o o

—~+

+-—+
b+ +

PRIME ") s

DN,

-—

——

Set
S

-+
b+

R A

+—t-

e e e S

+————4—+

o i

eT e

e
S

T S S S

e
SO

e I

s S S e

DIGITAL

IR o

St e

o ee S

HE It S

s o

e oe e

e S e

e e S
U

2o ob S
Y

IS W U

e
U

e S
ST

m
YT E

{

i ol ol 4

0-#+¢—+—#—04—04—+—##0f&#+»0—44&440—##—F4++0—++¢—1‘5%#“1

~+
U

e
st a8

S e
S S S

e e
U

V234 8] l S 101 121314131617101920212223262526
2720930 M IV

PG-3

SRRS

R
e B

e S T

L
b

SRS

oo et S

e e

b S

Ae e

ST SR S

ot a2

S8 SR

dnaderdrabdrabrbhmhash

+——4

o
o S
e

mm am o

obasboadnrudonb

I NIV W40 414740444340 47484950 3132333453503730596061010)04
0300 7 MH 70 N 7] nnnunnnu

EQUIPMENT CORPORATION

MAYNARD,

MASSACHUSETYTS
ZK-613-82

Figure

1-1:

FORTRAN Coding

Form

INTRODUCTION
TO

PDP-11

FORTRAN

Field
Statement

Column(s)

label

Continuation

indicator

Statement
Sequence

through

7 through 72

number

through

To get from one field to another, type each space
individually.
example,
in Figure 1-1, enter the first line, type C, press the
bar five times, and begin typing the comment.

1.3.2

Tab-Character

For
space

Formatting

You

can press the tab character to move to the continuation
indicator
field
or the statement field.
You cannot move to the sequence number
field, however, by pressing the tab.
Figure 1-2
compares
keystrokes
in
lines
using
tab-character
formatting
and
in
those
wusing
character-per-column formatting.

Format Using TAB Character

Character-per-Coiumn Format

112 3 4 sle6|7 8 910111213i41s1617181920
C Ga® FIRST VALUE

FII|R|S|T

VI]A|L

10@B| = J + 5+K
+

1]0

@B IVAL = 1+2

|U{E

5|+

L+|M

I {V[{A]L

Fl+]2

ZK-614-82

Figure

1-2:

Line

Formatting

Example

The statement label field consists of the
characters
that
you
before the first tab character.
The statement label field cannot
more than five characters.
After you type the first
tab
character,
you
can
continuation indicator field or the statement field.
To

enter

the

statement

the

end

enter

type

FORTRAN

then

the

line.

the

continuation

field

statement

indicator

consists

all

field,
the

type

characters

type

any
after

type
have

either

digit.
the

the

The

digit

'
field

without

immediately

starts

with

continuation

after

a'digit.

the

first

indicator

tab.

Note

field,

that

INTRODUCTION

PDP-11

FORTRAN

Many text editors and terminals advance the terminal print carriage to
a
predefined print position when you type the TAB key.
However, this
action is not related to the FORTRAN compiler's interpretation of
the
tab

character

described

above.

You can use the tab character to improve the legibility of
a
FORTRAN
statement.
For compilation, spaces are ignored, except those within a
Hollerith constant or alphanumeric literal.
For
1legibility,
spaces
are
printed
1in
the
source
1listing.
Tabs
also
are
ignored for
compilation purposes in a statement field.
1In the source listing, the
tab
causes
the
character that follows to be printed at the next tab
stop (located at columns 9, 17, 25, 33, and so forth)

1.3.3

Statement Label

Field

A statement label or statement number consists of one to five
decimal
digits
ir
the
statement
label field of a statement's initial line.
Spaces and leading 0s are ignored.
An
all-zero
statement
1label
is
invalid.

Any statement referenced by another statement must hava a
label.
two statements within a program unit can have the same label.

You can use two special indicators in the first column
of
the
label
field:
the
comment indicator and the debugging statement indicator.
These indicators are described in Sections 1.3.3.1 and 1.3.3.2 below.

The

statement

label

field

a continuation

line must be blank.

1.3.3.1
Comment Indicator - The letter C in column
the
line
1is
a comment.
The compiler prints that
program

listing

and

then

ignores

the

1
indicates
that
line in the source

line.

1.3.3.2
Debugging Statement Indicator - The
1letter
D
in
column
1
designates
a
debugging
statement.
The first line of the debugging
statement can have a statement label in the remaining columns
of
the
label
field.
1f
a
debugging
statement
is
continued,
every
continuation line must have
a
D
in
column
1
and
a
continuation
indicator

The

compiler

column

command

specifies

whether

debugging

statements

are

compiled.
If
you
specify
debug-statement
compilation,
debugging
statements are compiled as a part of the source program;
if
you
do
not
specify
debug-statement
compilation,
debugging
statements are
treated as comments.
For a description of compilation commands, refer
to the appropriate user's guide.

1.3.4

Continuation Field

A continuation indicator is any
character,
except
0
or
space,
1in
column
6
of
a
FORTRAN line or any digit, except 0, after the first
tab.
A statement can be divided into continuation lines at any point.
The compiler considers the characters after the continuation character
to follow the last character of the previous
1line,
as
1if
no
Dbreak
occurred
at
that
point.
If a continuation indicator is 0, then the
compiler considers the
1line
to
be
the
first
1line
of
a
FORTRAN
statement.

INTRODUCTION TO

PDP-11 FORTRAN

occur
can
they
but
continued,
Comment lines cannot be
1line(s)
continuation
its
1line and
initial
statement's

a
between
or between

lines.

successive continuation

Statement Field

1.3.5

field.
statement
the
in
placed
The text of a FORTRAN statement is
Because the compiler ignores the tab character and spaces (except in
text
Hollerith constants or alphanumeric literals), you can space the
The use of tabs for
1legibility.
for maximum
any way desired
in
spacing

is discussed

Section

1.3.2

NOTE

1line

extends

beyond

character

position 72, the text following position
is
72 is ignored and no warning message
printed.

1.3.6

Sequence Number Field

in
information can appear
identifying
other
A sequence number or
The compiler
73 through 80 of any line in a FORTRAN program.
columns
cannot
Remember that you
ignores characters in this field.
the sequence number field by tab-character formatting,

1.4

move

PROGRAM UNIT STRUCTURE

program
Figure 1-3 shows the allowed order of statements in a FORTRAN
figure, vertical lines separate statement types that
this
In
unit.
statements,
FORMAT
For example, comment lines,
can be interspersed.
DATA statements, and executable statements are allowed alternatives in
types
statement
indicate
Horizontal lines
the body of the program.
For example, IMPLICIT statements cannot
cannot be interspersed.
that
be interspersed with executable statements, a PROGRAM statement, or an
END statement because each has a definite order in the program.
PROGRAM,FUNCTION,SUBROUTINE, or BLOCK DATA Statements

IMPLICIT

Statements
PARAMETER

Other

Statements

Specification

Comment

FORMAT

Lines

and

StENTRYt

atements

Statements

DATA

Statements

Statement Function
Definitions

Executable

Statements

END Line
ZK-1043-82

Figure 1-3:

Required Order of Statements and Lines
1-7

CHAPTER

FORTRAN

The

basic
e

components

Constants

changed
®

Variables

fixed

Arrays

groups

be
of

names

that
by

values

The

symbolic

computation

using

the

trigonometric

1is

wused

name

name.

performed

followed
unit

that

arguments,

any:

sine

the

argument.

the

function

obtain

used

single

result.

and functions have symbolic names.
A symbolic
characters that identify entities in the program.

variables,

data

Logical

‘Integer

Real

Double

Complex

following

the

exception

Sections

respectively,

arrays,

expressions,

types:

and

functions

can

symbol
such
as

arrays,

string

program

place

and

with

Individual

optionally
a

The

contiguously

symbolic

variables, array elements,
combinations
of
these
components
with
operators.
An
operator
is
a
that
a
certain
kind
of
operation,

specifying

following

stored

constants,

multiplication,

Constants,

are

conjunction

that

cannot

stored values.
statements.

-- names of functions,
arguments.
A
function is

They

represent

individually

specified

references

numbers.

program

collectively by just
called array elements.

Expressions

Variables,

changed

for example, computing
The
resulting
value
reference.
e

are:

references

1lists

performs

such

are

Function
by

statements

values,

referenced

subscript
values

can

COMPONENTS

statements.

symbolic

values

STATEMENT

FORTRAN

program

stored

can

have

name

the

precision

sections

of
2.1

detail

function
and

provide

2.2

the

basic

references,
on

information

components

which

are

symbolic

names

common

all

FORTRAN,

described
and

basic

data

with

Chapter
types,

components.

FORTRAN STATEMENT COMPONENTS

2.1

SYMBOLIC NAMES

Symbolic names identify entities within a FORTRAN program unit. These
The "Typed" column indicates
entities are 1listed in Table 2-1.
whether the entity has a data type, such

Entities Identified by Symbolic Names

Table 2-1:

Typed

Entity

yes

Variables
Arrays

Statement

yes

functions

Processor-defined

functions

Function subprograms
Subroutine subprograms
Common blocks
Main programs
Block data subprograms
arguments

Dummy

yes

yes
no
no
no
no
yes

A symbolic name is a string of letters and digits totaling
of six characters.

and

integer,

real,

Data types are discussed in Section 2.2.

forth.,

The first character must be a letter.

maximum

Examples of valid and invalid symbolic names are:
valid

Invalid/Explanation

NUMBER
K9

5Q
B.4

(begins with a numeral)
(contains a special character)

you
That is,
Symbolic names must be unique within a program unit.
cannot use the same symbolic name to identify two or more entities in
the same program unit.

some
In executable programs consisting of two or more program units,
The
entities must have unique names throughout all the program units.
entities

2.2

are:

Processor-defined functions

Function

Subroutine subprograms

Common

Main

Block

subprograms

blocks

programs

data

subprograms

DATA TYPES

Each basic component
one

several

Integer

data

(constants, variables, and so

types:

-- a whole number.

forth)

represents

FORTRAN

Real

decimal

fraction,
®

Double

maximum
®

The

can

true

basic

type

Whenever

is,

whole

the

two.

number

with

numbers

can

its

specified

construction

can

explicitly

declared.

data
1is

(see

specifies

that

IV,

type

twice

many

complex

part,

the

second

one

three

Section

(as

converted

performed

constants);

without

according

value
to

"numeric
real,

values

occupy

numeric

storage

integer,

two

unit"
or

four

another

rules

for

the

amount

logical

numeric

unit

IMPLICIT

the

3.1).

store

complex

real

decimal

false.

(with

one

than

part.

convention

the

naming
be

number,

representing

represents

component

statements

and

real

value

imaginary

conversion

FORTRAN

that

real

inherent

needed

precision
PDP-11

value

FORTRAN

memory

the

assignment
ANS

can

statement);

COMPONENTS

digits.

first

Logical

implied

type,

pair

the

represents

data

significant

number;

ways:

number,

combination

precision

Complex

STATEMENT

value.

Double

storage

units.

bytes

memory.

PDP-11 FORTRAN IV provides additional data
types for better control of
performance
and
memory requirements.
Table 2-2 lists the data types
available and the amount of memory required (in
bytes).
The
form
*n
appended to a data type name is called a data type
length specifier.

Table

2-2:

Data

Type

Storage

Requirements

Storage

Requirements

(Bytes)

BYTE

LOGICAL

LOGICAL*1

LOGICAL*4

INTEGER

INTEGER*2

REAL

REAL*4

REAL*8

DOUBLE

PRECISION

COMPLEX

COMPLEX*8

BYTE

two

bytes.

The

l-byte

false,

Either two or four
bytes
compiler command qualifier

Only

storage
a

+127.

LOGICAL*1l

two
area

single

Four bytes are
computation.

and

INTEGER*4

1
or

are

allocated

specified.

bytes
can

are

contain

character,

allocated

used

but

synonymous.

2-3

only

the

default

for

computation.

logical

integers

two

depending

The

bytes

the

allocation

values

true

the

range

-128

are

used

for

FORTRAN STATEMENT COMPONENTS

2.3

CONSTANTS

A constant represents a fixed value and can be
value, or a character string.

logical

number,

They
Hollerith constants or alphanumeric literals have no data type.
Section
(see
appear
they
which
in
context
the
of
type
data
the
assume
2.3.6.2).

Integer Constants

2.3.1

It can
An integer constant is a whole number with no decimal point.
An
number.
decimal
a
as
ed
interpret
1is
and
sign
have a leading
integer constant has the following form:
snn

optional

A string

Leading 0s,

sign.

numeric characters.

if any,

are ignored.

A plus
A minus sign must appear before a negative integer constant.
is
constant
unsigned
(an
constant
positive
a
sign is optional before
assumed

to be positive).

Except for the sign, an integer constant cannot
other than the numerals 0 through 9.

constant

integer

The absolute value of an

contain

cannot

character

greater

than

32767.

Examples of valid and

invalid

integer constants are:

Invalid/Explanation

valid
0

-127

+32123

(too

99999999999

large)

(decimal point and

3.14

comma not allowed)

32,767

Integer constants can also be specified in octal form.

of an

The octal form

is:

integer constant

"nn

A string of digits

in the range 0 to 7.

An octal integer constant cannot be negative or greater than "177777.
Examples of valid and invalid octal

integer constants are:

valid

Invalid/Explanation

"107

"108

(contains a digit outside the

"177777

"1377.

(decimal point not allowed)

"177777"

allowed

range)

(trailing quotation mark not allowed)

FORTRAN

2.3.2
A

Real

real

three

STATEMENT

COMPONENTS

Constants

constant

number

with

forms:

basic

real

constant

basic

real

constant

integer

A basic real
formats:

constant

decimal

followed

string

point.

decimal

can

take

one

exponent

digits

one

three

S.nn

snn.nn
snn.

optional

sign.

string

decimal

digits.

The decimal point can appear anywhere in the string.
The
number
of
digits is not limited, but only the leftmost 7 digits are significant
.
Leading 0s (0s to the left of the first nonzero digit) are ignored
in
counting
the leftmost 7 digits.
Thus, in the constant 0.00001234567,
all of the nonzero digits are significant, and
none
of
the
0s
are

significant.

decimal

exponent

has

the

form:

Esnn

optional

integer

sign.

The

exponent

integer
the
A

real

minus

plus

1.0

sign
sign

Except

for

numerals

the

constant

with

used),

represents

constant

value

number

constant.

g),.

occupies

must
is

four

bytes

which

(for

and

seven

the

signs,

the letter E and
positive exponent.

decimal

point,

contain

1.0E6

interpreted

decimal

cannot

preceding

example,

between

and

appears

1in

follow.

The

exponent

real

of a
nonzero
real
0.29E-38 or greater

cannot

nedative

the

character

constant,

real

represents

real

digits.

The magnitude
approximately

typically

for

constant

through

must

power

be multiplied

appear

optional

algebraic

letter

exponent

precision

real
0

constant
cannot
than approximately

1letter

(if

other

than

the

integer

omitted,

exponent.

but

constant
it

be
smaller
1.7E38.

can

than

FORTRAN STATEMENT COMPONENTS

Examples of valid and

invalid real constants are:

valid

Invalid/Explanation

3.14159
621712,
-.00127
+5.0E3

1,234,567
325E-45
-47.E47
100
$25.00

2E-3

(commas not allowed)
(too small)
(too large)
(decimal point missing)
(special character
not

allowed)

Double Precision Constants

2.3.3

integer
A double precision constant is a basic real constant or an
constant followed by a decimal exponent of the following form:
Dsnn

optional

integer

sign.

constant.

A double precision constant occupies eight bytes and is interpreted as
The
a real number with a precision of typically 16 decimal digits.
only
but
limited,
not
is
exponent
the
number of digits that precede
the leftmost 16 digits are significant.

A minus sign must appear before a negativée double precision constant;
Similarly, a
is optional before a positive constant.
sign
a plus
negative exponent
a
for
D
minus sign must appear after the letter
while a plus sign is optional for a positive exponent.

The exponent following the letter D cannot be omitted, but it

can

The magnitude of a nonzero double precision constant cannot be smaller
than approximately 0.29D-38 or greater than approximately 1.7D38.
Examples of valid and

invalid double precision constants are:

valid

Invalid/Explanation

1234567890D+5
+2.71828182846182D00
-72.5D-15

1234567890D45
1234567890.0D-89
+2.7182812846182

1DO0

2.3.4

(too large)
(too small)
(no Dsnn present;

this is a valid
real

constant)

Complex Constants

comma
by a
A complex constant is a pair of real constants separated
The first real constant represents the
in parentheses.
and enclosed
constant
real
second
the
and
number
complex
the
of
part
real
represents

the

imaginary part.

FORTRAN

complex

constant

has

the

STATEMENT

following

COMPONENTS

form:

(rc,rc)
rc

The

real

parentheses

required.
A

constant.

See

complex

and

comma

Section

constant

are

2.3.2

part

for

occupies

eight

complex

rules

for

bytes

number.

Examples

the

constant

forming

and

1is

real

and

are

constants.

interpreted

valid

and

invalid

valid

complex

constants

are:

Invalid/Explanation

(1.70391,-1.70391)

(1,2)

(+12739E3,0.)

(integers

(1.23,)

(second

are

real

not

allowed)

constant

missing)

(1.0,1.0D0

(double
are

2.3.5
A

Logical

logical

two

constants

allowed)

Constants

constant

logical

precision

not

specifies

constants

are

true

false.

Thus,

only

the

possible:

following

. TRUE.

.FALSE.

The

delimiting

2.3.6
A

are

required

part

each

constant.

Hollerith Constants

Hollerith

character
A

periods

constant

count

Hollerith

and

a string of
letter H.

the

constant

has

the

printable

following

characters

preceded

form:

nHclc2c3...cn

unsigned,

characters

The

printable

maximum

Examples

nonzero

the

integer
constant
stating
the
string (including spaces and tabs).

character.

number

valid

and

characters
invalid

valid

16HTODAY'S

1HB

number

255.

Hollerith

constants

are:

Invalid/Explanation

DATE

1IS:

3HABCD

(wrong

number

characters)

FORTRAN STATEMENT COMPONENTS

2.3.6.1

Alphanumeric Literals - An alphanumeric literal is

string

printable ASCII characters enclosed by apostrophes that represents
of
The form is:
an alternate form of a Hollerith constant.
'‘clc2c3...cn!

character.

A printable

Both delimiting apostrophes must be present.

alphanumeric

Within

represented

between

character

1literal,

consecutive

two

character

apostrophe

the

apostrophes

(with no space or other

them).

characters
of
number
the
The length of the alphanumeric literal is
including spaces and tabs, except that two
between the apostrophes,
is
A tab
consecutive apostrophes represent a single apostrophe.
stored as a single character but is displayed as spaces up to the next
The length must be

tab stop.

Examples of valid and

invalid

in the range from 1 to 255.

literals are:

valid

Invalid/Explanation

'WHAT?'

'HEADINGS

(must contain trailing

'TODAY''S DATE 1IS:

(must contain at least

"NOW OR NEVER"

(quotation

HE SAID,

2.3.6.2

constants

according

"HELLO"'

Data Type
are

the

used

Rules
1in

for

apostrophe)

one

character)

Hollerith

Constants - When

Hollerith

numeric expressions,

following rules:

marks

in
wused
be
cannot
place of apostrophes)

they assume a data type

including
When the constant is used with a binary operator,
assignment operator, the data type of the constant is the
the
data type of

the other

operand.

Statement

For example:

Data Type
of Constant

Length of
Constant

REAL*4
INTEGER*2
REAL*8

4
2
8

INTEGER*2 ICOUNT
DOUBLE

REAL*8

RALPHA = 4HABCD
JCOUNT = ICOUNT + 'XY'
DOUBLE = 8HABCDEFGH

When a specific data type is required,
for

the constant.

For

example:

Data

type

that

1is

type

Length

Statement

of Constant

Constant

X=Y (1HA)

INTEGER*2

assumed

FORTRAN STATEMENT COMPONENTS
®

When

the

constant

assumed.

For

used

APAC

When

the

type

constant

in any
example:

When

(2HAB)
1HC-1HA

,NOT,

length

the

constant

2-2

lists

Table

the

character

2.4

VARIABLES

variable

Section

2,1

variable

change

Type

INTEGER*2

data

Length of
Constant

Constant

2
2
2

assigned
type

Two

the

and
a

variables

the

and
partial
EQUIVALENCE

subprogram

memory

new

type,

requirements
can

for

are

storage

the

establish

statements,

data

the

data

considered
of

If variables
associated)

the

execution

input
of

the

is defined
(for
variable
becomes

data

same

variable

you

variable.

The

type

by
undefined;

data

represents,

its

type
the

is
data

variable

statements,

are

and

dummy

variable

during

associated

assignment),
is,

Association
statements,
arguments

associated

location,

that

refers.

COMMON

storage

nam €.

statem ent

types

use

arguments

the

storage

example,

you

each other when
they
refer
partially associated when
part
refers is the same
as
part
or

actual

by a DATA
statement.

different

with

type

the

however,

with

when

defined

same

(see

are

one variable
which the Other
and

references.,

They

occur

value

constants.

are

type of

type.

location

The

location;

value

data

When data
of
any
converted, if necessary,
to

associated

location.

each

name).

that

the

with

symbolic

stored

data

You

required

storage.

indicates

association

program

assignment

value,

right.

associated

assigning

to which
location to

A variable is
contains data

other

currently

variable,

location

variable

name

the

characters

byte

form

storage

variable.

same

before

its

type
declaration
statements or IMPLICIT
names that begin with certa
in letters.

the

variable

one

classified

wusing

truncated

symbolic

value

the

choosing

is less than the length impl
ied by the
to the constant on the right
.
When the
greater than the length impli
ed by the
data

number

value

Precision,

the

type

context,

INTEGER*?2

occupies

that

are

for

Variables

all

Constant

INTEGER*2
INTEGER*2

'B'

constant

Each

1,2,3

type

length of the constant
type, spaces are appended

type,

can

data

Data

the

data

Length

Type

other

Statement

none

used

For

argument,

Constant

(9HABCDEFGHTI)

assumed.

actual

Data

Statement
CALL

example:

and

you

can

with
be

(or

definedqd

execution

the

partially

value

one

value
of
cannot predict

the

its

FORTRAN STATEMENT COMPONENTS

Data Type Specification

2.4.1

that given
ion 5.2) specify le:
statements (see dSect
Type declaration repr
examp
For
.
esent specifie data types

variables are to

COMPLEX VARl

DOUBLE PRECISION VAR2

Dbe associated
variable VARl is tocompl
indicate that the which
These statements stora
ex data, and
ain
cont
will
ge location
with an 8-byte
e double
8-byt
an
with
ed
ciat
asso
be
is to
that the variable VAR2 ion.
precision storage locat
a more general scope: n, ita
(see Section 5.1) has
The IMPLICIT statement the
icit type declaratio in
expl
an
of
nce
signifies that, in g withabse
or any letter with
r,
lette
fied
speci
a
nnin
variable name begi
type.
a specified range, is to represent a specified data

only once.
the data type of a variable
You can explicitly declare prece
statement.
CIT
IMPLI
an
over
dence
takes
n
explicit declaratio
2.4.2

Data Type by Implication

or explicit type
r IMPLICIT statements begi
In the absence of eithe
nning with I, J,
names
with
s
variable
declaration statements,ed all
bles with
integer variables. edVaria
K, L, M, or N are assum anyto be
be real
to
assum
are
r
lette
other
names beginning withle:
For examp

variables.

Integer Variables

Real Variables
ALPHA

2.5

JCOUNT

BETA

ITEM

TOTAL

NTOTAL

ARRAYS

associated with a
guous storage locations indiv
An array is a group of conti
idual storage
The
name.
the array
single symbolic name, eleme
a subscript
by
to
red
refer
are
nts,
locations, called arrayname. Section 2.5.2 discusses subscripts
.
appended to the array
n
. For example, a colum
one to seven dimensions
An array can have from imens
must
you
,
value
a
to
refer
To
.
array
ional
of figures is a one-d A table of more than one column of figures is
r.
specify the row numbearray
must specify both
. To refer to a value you
a two-dimensional
covers several
that
es
figur
of
r. A table
row number and column numbe
a value in this
to
refer
To
.
array
al
nsion
pages 1is a three-dime
page number,
and
r,
numbe
n
array, you must specify row number, colum
The following FORTRAN statements establish arrays:
e Type declaration statements (see Section 5.2)
e

The DIMENSION statement (see Section 5.3)
The COMMON statement (see Section 5.4)

The VIRTUAL statement (see Section 5.5)

FORTRAN

These

define

array,

and

the

element

with

may contain array
declarators
(see
Section
the" name
of the array, the number
of dimensions
number

program

execution

defined

2.5.1

Array

array

DATA

can

declarator
a

specifies

program

declarator

(df,d]

The

defined

has

the

unit

symbolic

number

the

For

2.1

the

The

name

the

form

The

dimension.

number
of

elements.
An

IUNIT

value

number

array

unit.

subprogram

main

2.5.2

array

during

element

input

program

Section
before
can

statement,

execution

name

that
the

identifies

properties

an
that

array,
a

that

1is,

symbolic

integer

upper

bound

can

array

constant

indicates

dimensions

the

name.,)

the

integer

array.

number

range

from

one

name,

dimensions

seven.

(10,10,10)

dimension

elements

can

the

array

specifying

arguments.

to
the

See

declarators
program.

that

one

are

adjustable

subprogram

equal

not

process
Section

are

not

use

arrays

with

well

6.1.2

for

constant

the
1000

are

the

arrays

dimension

array

not

program

used

adjustable

different
as

IUNIT

contains

within

can

elements

product

above

declarator

You

dimension

the

constant

declarators

number

each

IUNIT

array

arrays.

the

above,

dimension.

only

that

specifies

example

each

appear

define

declarators

array.

declarator

elements.

single

declarators

Dimension

name

Dimension
within

three-dimensional

associated

form:

declarators

number

that

consists

storage

DIMENSION

IUNIT

indicates

following

gives

dimension

array.

example,

the

assignment

symbolic
ang

A dimension declarator
that
variable that specifies
the

the

...)

(Section

The

defined

Declarators

within

array

2.5.1)

demension,

statement.

execution

array.

each

the same type as the arra
y name (see
or an entire array
can
be
defined

array

statement.

is considered

program

entire

elements

data of
element

array

during

input

contains

array

it
2.5.4).

and

COMPONENTS

statements

that

STATEMENT

name as
information.

permitted

Subscripts

A subscript qualifies
expressions,
called

array

subscript

name.,
A
subscript
is
expressions,
that
are

2-11

1list

enclosed

FORTRAN STATEMENT COMPONENTS

array is
ent in theit qual
e which elem
that determin
parentheses and
ifies.
name
y
arra
the
referenced. The subscript is appended to
A subscript has the following form:
(sl,s)...)

A subscript expression.

ic
, oOr arithmet
tant, variable
ession can be a asubscons
A subscript expr
is
it
ger,
inte
type
of
t is not al part.
the value of cation crip
expression., If
tion
frac
any
of
converted to integer by trun
cript expressionn
contain one subsfor
y reference must
A subscripted arra
each dimensio
(one
for each dimension defined for that array
declarator).

2.5.3

Array Storage

ns
k of the dimensioRAN
ion, you can thines.
ier in this sect
As discussed earl
FORT
ver,
Howe
ls or plan
, columns, and leve
of an array as LoOws
es. A
ar sequence of inwvalu
line
a
as
ry
memo
in
y
arra
an
es
stor
always
first
the
ent
with its first elem
y 1is stored ent
one-dimensional arra
of
tion
loca
age
stor
last
the
in
elem
and its last onal
storage location
ost
lefm
the
that
so
array is stored
multidimensi
the sequence. Amost
t
"order of subscrip
is called they stor
This
dly.
rapi
vary
subscripts
two,
one,
in
age
ple, Figure 2-1 shows arra
progression." For exam
.
and three dimensions

Data Type of an Array

2.5.4

the data
the same way as er
y is specified in by
The data type of an ,arra
of the
lett
ial
the init
ly
that is, implicit
type of a variable
t.
emen
stat
name or explicitly by a type declaration
e
data type. Any ofvalu
array have theconvsame
All the values anin an
the
type
data
d to the
array element a is DOUBLEertePREC
assigned to
for
ISION statement, each
in
d
name
is
y
arra
an
array. If
for
tion
loca
te storage
example, the compiler allocates e anof 8-by
any type is assigned to any
a valu
element of the array.y, When
ision.
element of that arra it is converted to double prec
Array References Without Subscripts

2.5.5

ify an array name
statements, you can spec
In the following types of
array is to be used
re
enti
the
to indicate that

without a subscript
(or defined):

Type declaration statements

COMMON statement

DATA statement

EQUIVALENCE statement

FORTRAN

FUNCTION

SUBROUTINE

Input/output

STATEMENT

COMPONENTS

statement

statement
statements

You can also use unsu
bscripted array
names
as
actual
references
to external procedures,
Unsubscripted array
permitted in any othe
r type of statement.

arguments

names

are

not

One-Dimensional Array BRC (6)

[ 1 [BRO(M)[ 2 [BRC(2)] 3 |BRC(3)| 4 |BRC4)| 5 |BRC(5

)| 6 [BRc(e)]

Memory Positions

Two-Dimensional Array BAN (3,4)

[ BAN(1,1)]

4 | BAN(1,2)|

BAN(1,3) { 10 | BAN(1,4)

2 | BAN(2,1)|

5 | BAN(2,2)|

BAN(2,3) | 11 | BAN(2,4)

3 | BAN(3,1)|

6 | BAN(3,2){

BAN(3,3) | 12 | BAN(3,4)

Memory Positions

Three-Dimensional Array BOS (3,3,3)

191

BOS(1,1,3) | 22 | BOS(1,2,3) | 25

20 | BOS(2,1,3) | 23 | BOS(2,2,3) |

101

BOS(1,1,2)

—

11]BOS(2,1,2)
1

[BOS(1.1.1)|

[BOS(2,1,1) | 5

3 | BOS(3,1,1)

B0OS(1,3,3)

B0OS(2,3,3)

|13 | BOS(1,2,2) [ 16 | BOS(1,3,2) § | 27 | BOS(3,3.3)
|14 [ BOS(2,2,2) | 17| BOS(2,3.2)

1 BOS(1,2,1) | 7 | BOS(1,3,1) | | 18|

BOS(3,3.2)

[ BOS(2,2,1) | 8 | BOS(2,3,1)

6 | BOS(3,2,1) | 9 | BOS(3,3,1)

Memory Positions
ZK-616-82

Figure

2.5.6

Adjustable

array

Chapter

2.6
An

Array

Storage

Adjustable Arrays

dimensions.
the

2-1:

arrays

allow

use

bounds
for

subprograms

adjustable

and the array
more information.

name

manipulate

array
as

arrays

subprogram,

variable

you

specify

subprogram

arguments.

See

EXPRESSIONS
expression

component,

represents

such

single

constant

value.

variable,

can
or

be
a

single

combination

basic
basic

FORTRAN STATEMENT COMPONENTS

computations
s. Operators specc ify
one or more operator
components with d,
onent(s) to
comp
basi
the
using the value(s) of
to be performe
obtain a single value.
cal.
ic, relational, al orand logi
sified as arithmetvalu
Expressions are clasons
cal
logi
tion
rela
es;
produce numeric
Arithmetic expressi
es.
valu
expressions produce logical

Arithmetic Expressions

2.6.1

and
arithmetic elements
ons are formed with
Arithmetic expressis.
a
ds
yiel
on
essi
expr
an
The evaluation of such
arithmetic operator .
value
single numeric

An arithmetic element can be any of the followin
e

A numeric constant

A numeric variable

A numeric array element

An arithmetic expression enclosed in parentheses

An arithmetic function reference

include
be interpreted to when
used above, can also
The term "numeric," e aslogi
used
data
ger
inte
cal data is treated as
logical data, sinc
in an arithmetic context.
g the
be performed usinresu
ify a computation a tonume
Arithmetic operatorsic spec
lt.
a
as
e
valu
ric
produce
values of arithmet elements to are:
The operators and their meanings

Operator

Function

Exponentiation

Multiplication

Division

Addition and unary plus

Subtraction and unary minus

with
s because each is used
called binary operator
These operators are
y
unar
also
are
ls
The plus (+) and minus (-) symbo

two elements.
immediately preceding an arithmetic element to
operators when writtennega
tive value.
denote a positive or
hmetic operator with any valid arithmetic element,
You can use any aritTabl
e 2-3.
except as noted in
nt must have a defined value before it can be
A variable or array eleme
expression.
used in an arithmetic

FORTRAN

Table

2-3

shows

exponent,

and

the

Table

allowed

data

Result

the

Data

Type

Real

Double

Complex

A
by

negative
an

value

valid

element

integer

cannot

except

Arithmetic

expressions

result

The

the

precedence

are

the

are

result

case

has

performed

first.

an
in

element

another

the

same

base

and

double

ranked

and

Second

and

Third

they

algebraically
example,

type

double

the

precision

precision,

order

determined
and

precedence

operators

are

evaluated

equivalent

equal
by

precedence
the

the

operations

is:

left-to-right

power.

2.6.1.1
Use of Parentheses - You
particular
order
of
evaluation.
in

parentheses,

value

1is

used

that

(such

compiler

order

in
3+4-1, the addition is performed
Exponentiation, however, is evalua
ted right to
A**B**C, B**C is evaluated first
and then A is

resulting

First

enclosed

data

precedence

The

with

0-value

Precedence

* %

two

and

exponentiated

real

are

Operator

appear,

evaluated

and

exponentiated

this

operators

can

element;

exponentiation,

The

operators.

base

Complex

exponent.

When

Double

element.

higher

Real

element,

Exponentiation

Integer

Note:

any

for

types

exponentiation.

Integer

only

data

Exponent

Integer

base

result

Base

COMPONENTS

combinations

type

2-3:

STATEMENT

and

evaluation.

order

that

For

before

the

subtraction.

left.

For

raised

example,

use
parentheses
When
part
of
an

part

any

can

the

is
evaluated
evaluation of the

the

in
resulting

to
force
expression

first
and
remainder of

a
is
the

the

FORTRAN STATEMENT COMPONENTS

examples,

following

the

expression.

operators indicate the order of evaluation:

the

numbers

below

the

4 +3%2-6/2=7T

(443)

2 -6 / 2 =11

+3*2-6)

((4+43)

2 - 6)

/2=2

/ 2 = 4

expressions within
As shown in the third and fourth examples, order
of precedence,
normal
the
to
ing
accord
ed
evaluat
parentheses are
parentheses.
within
eses
unless you override the order by using parenth

nt
Using parentheses to specify the evaluation order is often importa
are
that
orders
ion
evaluat
since
tions
computa
high-accuracy
in
algebraically equivalent might not be computationally equivalent due
to rounding and normalization.

order 1is important,
Using parentheses to specify the evaluation doubt
exists as to the
any
furthermore, in difficult expressions. If
parentheses
Extra
eses.
parenth
use
ion,
express
resulting value of an
do not affect the result, but lack of sufficient parentheses does.

- If every element in
2.6.1.2 Data type of an Arithmetic Expression type,
the value produced
data
same
the
of
is
an arithmetic expression
s of different

by the expression is also of that data type. 1f element
type of the result
data types are combined in an expression, the data
of each operation is determined by a rank associated with each data
type, on the following basis:

Rank

Data Type
Logical

Integer

Real

Double precision

Complex

(Low)

(High)

on two arithmetic
The data type of the value produced by an operation data
type of the
types is the
data
different
of
elements
the value
,
example
For
on.
operati
the
in
highest-ranked element
element is real.
resulting £from an operation on an integer and a real
The data type of an expression is the data type of the result of the
last operation in that expression.

FORTRAN

Operations

are

classified

Integer

operations

integer

elements.

context

are

STATEMENT

COMPONENTS

data

treated

Integer

operations

are

entities

wused

that

can

rounded.

For

example:

The

value

Real

1/3
of

integers.)

result

not

operations

are

performed

real,

integer,

expression

Real

combinations

elements.

Any

integer

giving

then

elements

each

present

using

part

real

only

and
converted

are

fractional

evaluated

however,

precision

double

operations

precision

making

precision

it
the
element.

expression

then

operation

most

The

real

element

not

its

accuracy.

Any
is

equal

real

integer
to

real

arithmetic.

converted

real

The
Note,

integer
a

real

element

double

precision

significant

portion
of
a
double
significant portion is 0.
The
double precision arithmetic.

least

evaluated

Converting

a
increase

not

1logical

that
in
the
statement
Y
=
(I/J)*X,
an
operation
is
performed
on
I
and
J
and
multiplication is performed on that result and X.
Double

on
any

truncated,

division

only

arithmetic

arithmetic,

division

type
expression

integer

from

elements

data

performed

1/3

this

operations

follows:

(Logical

fraction

1/3

type

double

precision
element
example, the real number:

For

does

0.3333333

converted

to:

0.3333333000000000D0
not

to:

0.3333333333333333D90
Complex

operations -containing
a
complex
to

real

data

elements

are

type,

imaginary

2.6.2

relational

separated

true

the

converted

stated

real

elements

described.

data

type

expression

are

Double

converted
precision

rounding

the

least

and

the

resulting

value

complex.

Expressions

expression
a

integer

portion.
The
real
element
thus
obtained
is
as
the
real
part
of
a
complex
number;
the
part is 0.
The expression is then
evaluated
using

arithmetic

Relational

operation

previously

significant
designated
complex

element,

consists

relational

operator.

relationship

exists

two
The

and

arithmetic
value

false

the

expressions
expression

does

not.

FORTRAN

STATEMENT COMPONENTS

A relational operator tests for a relationship between two
expressions.

These

operators

arithmetic

are:

Relationship

Operator
.LT.

Less

than

.LE.

Less

than or equal

.EQ.

Equal

.NE.

Not

.GT.

Greater

than

.GE.

Greater

than or equal

The delimiting

periods are a

equal

required part

of each

operator.

L.NE,.
and
L[EQ.
the
by
only
related
be
can
Complex expressions
Complex entities are equal if their corresponding real and
operators.
are

imaginary parts

both

equal.

In an arithmetic relational expression, the arithmetic expressions are
are then
These values
values.
their
to obtain
evaluated
first
operator
the
by
stated
relationship
the
whether
determine
to
compared
exists.

For

example:

APPLE+PEACH

PEAR+ORANGE

.GT.

real
the
sum of
"The
relationship,
the
states
expression
This
real
the
of
sum
the
than
greater
1is
PEACH
and
APPLE
variables
of
variables PEAR and ORANGE." If that relationship exists, the value
false.
is
expression
the
of
value
the
not,
1if
true;
is
expression
the

All

operators

relational

have

the

same

precedence.

you

can

Arithmetic

operators have a higher precedence than relational operators.

As in any other arithmetic expression,

the

alter

relational

order

expression.

evaluation

However,

use

parentheses

operators

are

the arithmetic expressions in a

since

arithmetic

evaluated before relational operators, you need not enclose the entire
arithmetic expression

parentheses.

You can compare two numeric expressions of different in a relational
the expression with the
the value of
this case,
In
expression.
type
data
the higher-ranked
lower-ranked data type is converted to
before

the

comparison

2.6.3

Logical

made.

Expressions

Logical expressions are formed with logical elements and 1logical
A logical expression yields a single logical value, either
operators.
true

false.

A logical element can be any of the following:
e

integer

logical

constant

integer

logical

variable

integer

logical

array element

FORTRAN

The

®¢

relational

logical

operators

Operator

COMPONENTS

expression

integer

STATEMENT

enclosed

function

parentheses

reference

are:

Example

.AND.

Meaning

_AND.

Logical

true

conjunction:

if,

and

only

The

if,

true,

.OR.

,OR.

Logical

disjunction

expression

both,
.XOR,

,XOR.

exclusive

.EQV.

Logical
true

The
A

.NOT.

delimiting

logical

assigned
which

all

periods

expression

its

the

evaluated:

either

The

true

and

only

logical

The

if,

are

OR):
A

The
B,

expression

false,

negation:

true

and

only

logical

operators

are

evaluated

according

and

whether

The

if,

expression

both

value,

Logical

if,

the expression is false
have the same value.

equivalence:

if,

the same
false.
.NOT.

OR:

and

(inclusive

true

vice versa;
but
if both elements
.EQV.

true.

Logical

true

expression

both

is
have

true

expression

false.

required.
an

order

precedence

operators.
The
following list gives the order in
operators that can appear in a
logical
expression
are

Operator

Precedence

* %

First

Second

+, -

(Highest)

Third

Relational
Operators

Fourth

.NOT.

Fifth

.AND.

Sixth

.OR.

Seventh

. XOR., .EQV,
Operators

Eighth

equal

rank

example:

A*B+C*ABC

The

sequence

.EQ.

which

are

X*Y+DM/ZZ

evaluated

.AND.

evaluation

from

.NOT.

occurs

left

K*B

.GT.,

right.

is:

(((A*B)+(C*ABC)).EQ.((X*Y)+(DM/ZZ))).AND.(

.NOT.((K*B).GT.TT))

For

FORTRAN STATEMENT COMPONENTS

As in arithmetic expressions, you can use
normal sequence of evaluation.

parentheses

alter

the

Two consecutive logical operators are not allowed unless the second is
.NOT. .

essions
Some logical expressions are evaluated before all their subexpr
A .AND.
on
expressi
the
.FALSE.,
is
A
if
For example,
are evaluated.
ion
(F(X,Y) .GT. 2.0) .AND. B is .FALSE.. The value of the express
the
Thus,
F(X,Y).
ng
evaluati
without
A
can be determined by testing
function subprogram F may not be called, and side-effects resulting
from the call, such as changing variables in COMMON, cannot occur.

g
When a logical operator operates on logical elements, the onresultin
integer
operates
operator
logical
a
When
logical.
1is
type
data
the
elements, the logical operation 1is <carried out bit-by-bit onof the
corresponding bits of the internal (binary) representation
logical
integer elements. The resulting data type is integer. When a value
is
logical
the
values,
logical
and
integer
combines
operator
carried
is
n
operatio
the
then
and
first converted to an integer value,
out as for two integer elements. The resulting data type is integer.
Example:

INTEGER
I = "65

J
K

=
=

I.0R."100
I.AND."23

In this example, I has the value "165 and K has the value "21.

CHAPTER
ASSIGNMENT

Assignment

statements

element.

The

three

3.1
An

kinds

assign

STATEMENTS

single

value

assignment

statements

Arithmetic

assignment

statement

Logical

ASSIGN

assignment

arithmetic

ASSIGNMENT

numeric

array

are:

STATEMENT

statement
element.

assigns

The arithmetic assignment statement has
=

variable

statement

assignment

array

statement

ARITHMETIC

variable

arithmetic

value

the following form:

variable

array

element.

The

arithmetic

equal

rather,

KOUNT

This

sign

does

means
=

expression.

KOUNT

statement

not

"is

mean

replaced

means,

KOUNT

integer

with

constant

1."

"teplace

the

sum

Although

the symbolic name on
undefined,
values
must have

The
the
If

the

expression

real

entity
v

and

have

converted

the

yield

that

the

directly

summarizes

expression

must

expression

the

the
data

the
been
on

same

data

the

value

produces

left

equal

For

to,"

mathematics;

example:

variable

references

"is

by."

the
data

the

type

conversion

current
the

value

the

KOUNT

1left
of
the
equal
sign
previously assigned to all
right

the

value

equal

Proper

the

are

before

for

equal

size.

than

For

32767

the

assigned.

assignment

the

can
be
symbolic

example,

invalid

variable.

assigns

different,

integer
and

sign.

INTEGER*2

statement

types

rules

the

greater

sign

types,

data

value

current

the

value

Table

statements.

value

3-1

ASSIGNMENT STATEMENTS

Table

Conversion Rules for Assignment Statements

3-1:

Expression

(E)

Variable
or

Array

Assign

Integer

Logical

Append fraction
(.0) to E and

Real

Double

Append

fraction

E and

sign to Msl por-

tion of V; LSl

portion of V is 0

Truncate real part
of E to integer
and assign to Vj;
imaginary part
of E is not used

Truncate E to
integer and
assign to V

Assign E to V

Assign Msl por-

Assign real part

Lsl portion of

inary part of
is not used

tion

as-

Complex

Truncate E to
integer and
assign to V

assign

(.0)

Precision

Double

Precision

Real

Integer or Logical

(V)

Element

Assign E to MsI

rounded

tion

part

portion of V is
zero, imaginary

of E to Msl pir-

sl portion of

imag-

E to V;

Assign real part

Assign E to V

portion

not

used

Append

Complex

fraction

(.) to E and assign to real part
of V; imaginary
part of vV is 0.0

Assign

Assign Msl por-

tion of

real part of V;
imaginary part
of

nary part

MS = most significant (high order);

portion of E is
rounded; imagiis

Assign

real

part of V; LSl

0.0

E to

LS = least

of V

0.0

significant

(low

order)

Examples of valid and invalid assignment statements are:
valid

(4.*
BETA = -1./(2.*X)+A*A/(X*X))
PI
SUM

3,14159

=
=

SUM+l.

Invalid/Explanation

3.14 = A-B

(entity on the left must be

variable

-J = 1I**4

(entity on the left must be

variable

invalid

array element)

ALPHA = ((X+6)*B*B/(X-Y) (entity

expression
not

the

right

because the parentheses are

balanced)

ASSIGNMENT

3.2

LOGICAL ASSIGNMENT

The logical
false) to a

variable

The

assignment

logical

Note

statement

array

logical

variable

logical

expression.

all

logical

value

(true

has

the

following

form:

that

must

Otherwise,

the

assigns

element.

statement

Values,
to

STATEMENT

value,

but

assignment

STATEMENTS

will

either

numeric

Examples

array

logical

element.

data

conversions

values

variables

not

logical

have

type

will

any

logical

logical,

must

array

elements

assignment

and

must

made

yield

according

logical

Table

3-1,

meaning.
have

been

previously

assigned

statements

are:

valid
LOGICAL
PAGEND

PAGEND,
=

ABIG

,FALSE.

PRNTOK

ABIG

PRNTOK,

LINE

.LE.

.,GT.

132

.AND.

,AND.

.NOT.

.GT.

PAGEND

.AND.

.GT.

Invalid/Explanation
X=,TRUE.

(entity

the

left

must

logical)

ASSIGN
3.3
The

ASSIGN
ASSIGN

variable.
destination

STATEMENT
statement

The
in a

assigns

statement

variable
can
then
subsequent assigned

label

be
GO

integer

wused
to
specify a
TO
statement
(see

value

transfer
Section

4.1.3).

The

ASSIGN
ASSIGN

statement
s

has

the

label

the

ASSIGN

statement.

integer

variable.

ASSIGN

similar

statement

form:

The

following

executable

assigns

the

arithmetic

statement

assignment

the

label

same

program

the

statement,

unit

variable.

except

that

the

ASSIGNMENT

STATEMENTS

variable becomes defined for use as a statement
1label
reference
and
becomes
undefined
as an integer variable;
that is, the value cannot
be used for purposes of output or arithmetic.

The

statement

program

label must

unit.

It may

not

refer

to an executable statement

refer

FORMAT

in the

same

statement.

The ASSIGN statement must
be
executed
before
the
assigned
GO
TO
statement(s) in which the assigned variable is to be used.
The ASSIGN
statement and the assigned GO TO statement(s) must occur in
the
same
program
For

unit.

example:
ASSIGN

This

100

statement

100.

then

NUMBER

undefined

NUMBER

associates

Arithmetic

statement,

the

variable

operations

become

invalid.

NUMBER with

the

variable,

the
For

statement

the

label

following

example:

NUMBER+1

and

does

not

result

value

101

being

stored

NUMBER.

Assigning

the

assignment

statement:

variable

value

with

the

following

arithmetic

NUMBER=10

dissociates the
longer
be used
value

variable from statement
100.
in an assigned GO TO statement,

The
variable
can
no
but has the arithmetic

10.

Examples:

valid
ASSIGN

ASSIGN

99999

NSTART
TO

KSTOP

Invalid/Explanation

ASSIGN

250

ERROR

(variable must

integer)

CHAPTER

CONTROL

Statements

are

written.

However,

control

You

use

can

another

the

same

also

govern

program

The

normally

section

control

program

iterative

termination.

control
®

statement

conditionally

executes
--

The

following
examples

sections
the

they

are

transfer

point

within

Control

program

statements
execution, and

iterative

transfers

---

returns

the

from

suspends

program

end

they

control

specified

the

executable

subprogram
a

subprogram

program

the

execution

execution
program

statements,
are

unit

times

control

these

which

terminates

describe

ways

program

processing

control

temporarily

marks

transfers

control

statement

flow

statement

transfers

statement

within

and

control

statement

unit

END

number

program

which

subprogram.

unit.

specified

statement

control

statement

STOP

specifies

RETURN

statements

calling
PAUSE

transfer

program

are:

statement

transfers

CONTINUE

CALL

program

statement

the

order

normal

statement

group

the

another program
processing, suspension

statements

conditionally
®¢

interrupt

statements

unit

STATEMENTS

executed

you may

unit

giving

their

forms

used.

GO TO
4.1
GO

GO
TO

STATEMENTS

statements

types

transfer

statements

control

Unconditional

Computed

statement

Assigned

statement

within

are:

statement

program

unit.

The

three

CONTROL STATEMENTS

Unconditional GO TO Statement

4.1.1

The unconditional GO

statement

statement every time it is executed.

transfers

control

same

the

The unconditional GO TO statement has the form:
GO

A statement

label.

1in

The statement identified by s must be an executable statement
same program unit as the GO TO statement.

the

Examples:

GO TO

99999

Computed GO TO Statement

4.1.2

The

7734

computed

statement

transfers

control

specified by the value of an arithmetic expression.
GO TO
slist

(slist)([,]

statement

one

labels

An arithmetic expression whose value is in

the

range

A list, called the transfer

1list,

executable statements separated by commas.

where n is the number of statement labels in the transfer list.
The computed GO TO statement evaluates e and, if

the

result

integer

data

type.

Control

necessary,

converts

1is transferred to the

statement label in position e in the transfer list.

If the value of e is less than 1 or greater than the number of labels
in the transfer list, control is transferred to the first executable

statement after the computed GO TO.
Examples:

GO TO

(12,24,36),INDEX

GO TO

(320,330,340,350,360), SITU(J,K)+1

In the first example, if INDEX has a value of 2, execution will be
transferred to statement 24, In the second example, if SITU(J,K) has
a value of 2, execution will be transferred to statement 340.

4.1.3

Assigned GO TO Statement

The assigned GO TO statement transfers control to a statement label
placed in a variable by an ASSIGN statement. Thus, the transfer
destination can be changed, depending on the most recently executed
ASSIGN

statement.

CONTROL

The

assigned

integer

v([[,]

statement

STATEMENTS

has

the

following

labels

executable

form:

(slist)]

variable.

slist
A

list

one

commas.

statements

The

assigned GO TO statement transfers control to the
label
was
most recently assigned to the variable v.

the

The

ASSIGN

statements
value

Examples

of assigned

ASSIGN

See

associated

control

200

1is

ASSIGN

statement(s),

transferred

unit.
slist.

statements

whose
Section 3.3

slist

must

used,

and

the

executable

the

assigned

are:

IGO

(This

example

ASSIGN

the

which

in the same program
must be a member of

statement

statement,

statements

separated

450

IBEG,

(This

equivalent

200.)

450.)

IBEG

(300,450,1000,25)

example

equivalent

IF
4.2
An

1IF
IF

STATEMENTS

statement

specified

For

Logical

statement

each

type,

the

decision

based

The

two

executes

types

statement

Arithmetic
arithmetic

based

transfer

evaluation

arithmetic

(e)

arithmetic

sl,

the

value

statement

s2,

control
an

expression

execute

contained

has

transfers
of

the

control

arithmetic

following

one

expression.

form:

expression.

sl,s2,s3
Labels

the

Statement
statement

are:

statement

on the

only

statements

statement.

statements,

The

control

met.

Arithmetic

4.2.1
The

statement
IF

transfers

condition

executable

statements

the

same

program

unit.

three

CONTROL

(sl,s2,s3)

All three labels
refer

three

different

are

STATEMENTS

The arithmetic IF statement first
If

evaluates

Then,

parentheses.

is:

Less

than

Equal

Greater

however,

required;

Control

0
O

than

need

not

the

(e)

1in

passes

Label

expression
to:

examples

Some

(THETA-CHI)

50,50,100

This statement transfers control to statement 50

if the real

THETA
1is less than or equal to the real variable CHI.
to statement 100 only if THETA is greater than CHI.
IF

(NUMBER/2*2-NUMBER)

the

4.2.2

value

Logical

to statement 40 if the value
of
the
it transfers control to statement 20

odd.

Statement

The logical IF
statement
conditionally
executes
a
single
statement based on the evaluation of a logical expression.
The

logical
IF

(e)

variable

Control passes

20,40,20

This statement transfers control
integer variable NUMBER is even;
if

they

statements.

statement has

the

following

FORTRAN

form:

logical

expression.

A
complete
executable
another

FORTRAN
statement

logical

statement.
except a DO

The
statement
can
be
any
statement, an END statement, or

statement.

The logical IF statement first evaluates the logical
expression
(e).
I1If
the
wvalue
of
the
expression
1is
true,
the
statement (st) is
executed.,
If the value of the expression is false, control
transfers
to
the
next
executable
statement
after
the
1logical
IF
and the
statement (st) is not executed.
Note that
e
must
yield
a
1logical
value.

Examples

logical

.GT.

(REF(J,K)

LOGICAL

statements

.,OR.

.NE.

HOLD)

ENDRUN

(ENDRUN)

CALL

.,LT.

EXIT

are:
1)

REF(J,K)

250

REF(J,K)*(-1.5DO0)

CONTROL

STATEMENTS

DO
4.3
The

DO
DO

STATEMENT
statement

statements.
statement,

and

"DO

loop."

The

specifies

The
the

statement

s[,]

iterative

sequence
DO

has

statement

the

processing

statements

together

following

the

range

1is

called

with

its

sequence
of

the

called

form:

v=el,e2[,e3]

The
label
physically

of
an
executable
statement.
follow in the same program unit.

variable.

The

statement

must

integer

el,e2,e3
Integer

The

variable

initial,
omit the
s

expressions,

identifies

statement

The

the

must

called

arithmetic

®¢

END

RETURN

statement

range

the

and

e2,

and

first
for

loop

then

the

executed

Conversely,

be
if

increment

the

initial,
of

the

loop.

The

terminal

(e3)

negative,

parameter

(e3)

equal

exact

must

cannot

to
be

the

and

the
less

e2,

increment

parameter

positive,

el,

and

statements

The

that

statement.

expressions

initial

repeatedly.

than

statements

terminal,

executable

parameter
greater

the

terminal

4.3.1,

increment
must

all

the

evaluates
the

The

including

value

Section

the

If you
used.

statement

The

statement

are

statement

wvariable.

The

el,

statement

values

respectively.

control

(e2)

variable;

statement

determine

are

control

be:

the

terminal

not

statement,

The

terminal,
and
increment
parameters, respectively.
increment parameter, a default increment value of 1 is

assigned

the

mechanism

the

range

to
of

the

explained

terminal

parameter

initial
than

parameters,

parameter
equal

(el).
el.

CONTROL STATEMENTS

The number of executions of the DO range, called the iteration
is

count,

given by:
-

the greatest
is,
that
1in X,
where [X] means the greatest integer
integer less than the absolute value of X and with the same sign.
1f the iteration count is 0 or negative, the DO loop is executed once.

DO Iteration Control

4.3.1

After each execution of the DO range, the following actions are taken:
algebraically

added

The value of the increment parameter is

The iteration count is decremented by 1.

transfers
control
1If the iteration count is greater than 0,
to the first executable statement after the DO statement for

the control

another

variable.

iteration

the

range.

If the iteration count is 0, execution of

the

statement

terminates.

You can also terminate execution of a DO statement by wusing a
statement within the range that transfers control outside the loop.
The control variable of the DO statement remains defined with its
current

value.

When execution of a DO loop terminates, if other DO

1loops

its

transfers outward to the next outer DO
control
statement,
terminal
If no other DO
loop in the DO nesting structure (see Section 4.3.2).
loop shares the terminal statement, or if this DO loop is outermost,

control transfers to the first executable statement after the terminal
statement,

You cannot alter the value of the control variable within the range of
However, you can reference the control variable for
the DO 1loop.
other purposes

in statements within the range.

long
as
The range of a DO statement can contain other DO statements,
as these nested DO loops meet certain requirements. See Section
4.3.2.

You cannot transfer control into the range of a DO

1loop.

Exceptions

during

the

to this rule are described in Sections 4.3.3 and 4.3.4.
Examples

DO statements

follow.

valid
DO

This

100

K=1,50,2

statement

iteration.

specifies

iterations;

K=49

final

CONTROL

This

statement
iteration.
DO

specifies

STATEMENTS

iterations;

J=-2

during

the

final

IVAR=5

during

the

final

IVAR=1l,5

This statement
iteration.

specifies

iterations;

Invalid/Explanation
DO

NUMBER=5,40,4

(the

M=2.,10

statement

decimal

label

point

has

missing)

been

typed

for

comma)

Note

that

the

DO40M

2.10

last

invalid

unintentionally

4,3.2

Nested

DO loop can
inner-nested

outer

loop.

the

arithmetic

include

one

loop must

nested

assignment

loops

lie
can

Loops

K=1,10

. 35

CONTINUE

M=1,20

outer

loop

nested
not

Transfers

loop;

the

The

range

statement.

the

Figure

4-1

Nested
Loops

K=1,10

CONTINUE

M=1,15

CONTINUE

Figure

loops.

DO 25 L=1,20

Within

statement.

Incorrectly

DO 35 L=2,50,2

Control

within

terminal

Nested

|_ L. 45

complete

completely

loops.

Correctly

4.,3.3

statement

Loops

Nested

illustrates

valid

example,

loop,

you

however,

permitted.

4-1:

CONTINUE

Loops

can
a

Nested

transfer

control

from

outer

an
loop

inner
to

loop
inner

CONTROL

STATEMENTS

If two or more nested DO loops share the same terminal statement,
you
can
transfer
control to that statement only from within the range of
the innermost loop.
Since the shared terminal statement
1is
part
of
the
innermost
1loop,
any
transfer to the terminal statement from an
outer

loop

Extended

4.3.4

invalid

transfer.

Range

A DO loop has an extended range if it
contains
a
control
statement
that
transfers control out of the loop and if, after execution of one
or more statements, another control
statement
returns
control
back
into the loop.
Thus, the range of the loop is extended to include all
executable statements between the destination statement of
the
first
transfer

The

and

the

following

statement

rules

A transfer
from

its

govern

into
in

the

returns

use

control
DO

the

statement

loop.

extended

the range of a DO statement

extended

Statements

that

range:

is permitted only

range.

the extended

range must

not

change

the

control

range

control

variable.

Figure 4-2

illustrates

valid

and

1invalid

extended

transfers.

valiad
Control Transfers

K=1,10

DO 15 L=2,20
GO

15
DO

Invalid
Control Transfers

Loop

20
B

|_ L 35

50
Extended

CONTINUE

A=B+C

L=2,20

A=B+C
DO

M=1,15

D=E/F

CONTINUE

X=A*D

M=1,15

CONTINUE

X=A*D

.
D=E/F
.

CONTINUE
.
CONTINUE

Range

Figure

DO 50 K=1,10
20

4-2:

Control

Transfers

and

Extended

Range

CONTROL

STATEMENTS

CONTINUE
4,4

CONTINUE

The

STATEMENT

CONTINUE

statement

statement.
loop

transfers

used

control

primarily

when

that

loop

would

otherwise

statement

such

The

CONTINUE

statement

has

the

end

arithmetic

the

terminal

with

executable

statement

prohibited

control

IF.

following

form:

CONTINUE

CALL
4.5
The

CALL
CALL

STATEMENT
statement

procedure.

(See

Chapter

executes

can
for

also

the

SUBROUTINE

specify

greater

CALL

the

list

other

for

the

definition

has

the

following

external

subroutine.

and

statement

CALL

subprogram

argument

detail

subroutine) .
The

use

form:

s[([a)l[,[al])...)]

The

name

SUBROUTINE

dummy

other

argument

external

subprogram

associated

with

other
a

procedure.

external

SUBROUTINE

procedure;

subprogram

you

actual

specify

values
then

argument.

(Section

argument

the
1list
transfers control

list,

the

with

the

dummy

the

subroutine.

The

arguments

data

type

the

with

variables,

arrays,

constants,

CALL

the

array

describes

CALL

statement

arguments

first

executable

CALL

CURVE

ICALL

PNTOUT

CALL

EXIT

the

arguments.)

associates

subroutine.

statement

the

It
the

statement

must agree in number,
order,
and
arguments in the subroutine.
They can be
elements, constants,
expressions,
Hollerith

array
name

literals,

the

argument

list

subprogram

refers

array.

Examples

actual

dummy

alphanumeric

unsubscripted

6.1

statements

are:

(BASE,3.14159+X,Y,LIMIT,R(LT+2))
(A,N,'ABCD')

names.

the

entire

CONTROL STATEMENTS

RETURN
4.6

RETURN STATEMENT

It has the following form:

the calling program.

subprogram

The RETURN statement is used to return control from

RETURN

When a RETURN statement is executed in a function subprogram,

control

is returned to the statement that contains the function reference (see
Chapter 6). When a RETURN statement is executed 1in a subroutine
subprogram, control is returned to the first executable statement
following

statement.

the CALL

example:

RETURN statement

SUBROUTINE SIZCHK
IF (N) 10,20,30
=-1

K=0

K=+1

(N,K)

RETURN
RETURN
RETURN
END

PAUSE
4,7

PAUSE

and
The PAUSE statement temporarily suspends program execution
displays a message on the terminal to permit you to take some action.
The PAUSE statement has the following form:
PAUSE
disp

[disp]

An alphanumeric literal, a decimal digit string of

digits,

octal

The disp argument is

one

five

constant.

optional.

The

effect

PAUSE

statement

If it is running as a
depends on how your program is being executed.
batch job, the contents of disp are written to the system output file
but the program is not suspended.

1f the program is running in interactive mode, the

contents

disp

followed by the prompt sequence
are displayed at your terminal,
you should then enter a
suspended;
is
program
the
that
indicating

command, after which execution
control
executable statement following the PAUSE.
specific to the operating system,
Some examples

PAUSE

statements

PAUSE

999

PAUSE

'"MOUNT NEXT TAPE'

resumes
Because

with the first
the command 1is

it is not given here.

are:

CONTROL

STATEMENTS

STOP
4.8

STOP

STATEMENT

The

STOP

the

operating

The

STOP

statement

terminates

statement

STOP

program

system.

has

the

following

execution

and

returns

control

form:

[disp]

disp
An

alphanumeric

digits,
The

literal,

octal

disp

when

argument, if
execution stops.

Example

STOP

'END

present,

statements

decimal

constant.

digit

specifies

string

message

one

five

displayed

are:

RUN'

STOP

END
4.9

END

The

END

last
The

STATEMENT
statement

source
END

line

marks

the end of a program
every program unit.

statement

has

the

following

unit,

must

Dbe

the

form:

END

The END statement
continued,
In

main

reaching

must

not

occur

program,

the

END

statement,

RETURN

statement

subprogram,

STOP

statement

program
is

continuation

prevents

execution

implicitly

1line

execution

terminates.

executed.

from

CHAPTER

SPECIFICATION STATEMENTS

Specification
allocate

statements

and

characteristics

The

specfication
e

are

initialize
the

symbolic

statements

IMPLICIT

statement

symbolic

names

Type

declaration

type

statement

the

number

statement

arrays,

used

specifies

symbolic

array

statements

and

the

VIRTUAL

the

one

implied

DATA

and
®

statement

array

PROGRAM

procedure
--

elements

statement

number

contiguous

BLOCK

DATA

which

initial

common

blocks

The

following

and

examples

sections
of

their

for

one

two

the

specifed

initial

values

program

execution

assigns

values

describe

data

symbolic

establishes

may

these

areas

arrays

entities

symbolic

names

before

statement

the

storage

assigns

type

dimensions

space

declares

dimension

associates

unit
¢

declares

each

program

data

location

statement

external

reserves

normal

statement

storage

EXTERNAL

be
®

same

vyou
other

program.

explicitly

elements

reserves

outside

EQUIVALENCE

the
¢

statement

located

let

define

names

declares

the

storage
e

that

and

are:
--

DIMENSION

COMMON

names

statement

specified

and

nonexecutable

variables

BLOCK

assigned

statements,

usage.

names

variables,

name

DATA

main

giving

arrays,

program

entities

with

unit

contained

their

forms

IMPLICIT
5.1
By

IMPLICIT

STATEMENT

default,

all

interpreted

names

beginning

integer

data,

with

and

the

all

1letters

names

beginning

through

with

any

are

other

SPECIFICATION

STATEMENTS

The IMPLICIT statement permits you to

letter are interpreted as real.

change these default data typing rules.

The

IMPLICIT statement has the following form:
IMPLICIT typ(al,al...)[,typ(al,al...)])...

typ

(See Table 2-2.)

One of the data type specifiers.

¢ or cl-c2,
An alphabetic specification in one of two forms:
The latter form specifies a
where c¢ is an alphabetic character.
in
which must occur
through c¢2,
from ¢l
range of letters,
order.

alphabetical

The IMPLICIT statement assigns the specified data type to all symbolic
or any letter in a
names that begin with any specified 1letter,
specified range, and which have no explicit data type declaration.
For

example:

IMPLICIT
IMPLICIT

INTEGER (I,J,K,L,M,N)
REAL (A-H, 0-2Z)

These statements specify the default in the absence

explicit

any

statement.

IMPLICIT statements must precede all other

and

all

executable

You cannot

IMPLICIT

label

Any data type can be
following

examples

specification

statements,

statements.

specified

show:

IMPLICIT DOUBLE

IMPLICIT COMPLEX

PRECISION

(S,Y),

(D)

LOGICAL*l

statement,

IMPLICIT

the

type

(L,A-C)

TYPE DECLARATION
5.2

TYPE DECLARATION STATEMENTS

The type

specified

statement

declaration
symbolic

explicitly

gives

data

names.

The type declaration statement has the following form:
typ

typ

vi,v]...

One of the data

type specifiers

The symbolic name

function subprogram,

(see Table 2-2).

variable,

array,

or an array declarator.

statement

function,

The following rules apply to a type declaration statement:

A type

declaration

statement

must

precede

all

executable

statements.

You can declare the data type of a symbolic name only once.
5-2

SPECIFICATION

You

cannot

You

can

appending

array

label

symbolic

use

name

can

being

length

attribute

*s,

specified
specifier and an
first.
Examples
INTEGER

type

declaration

array

statement.

statement

declarator

(see

declare

Section

array

2.5.1)

followed

data

type

length

specifier

one of the acceptable lengths for the data
Table 2-2).
Such a
specification
overrides

that

the

REAL

type

name.,

where
s
declared
(see

form

STATEMENTS

the

statement

implies,

and

item.

assigns

new

the

type
the

length

If
you
specify
both
a
data
type
length
declarator, the data type length specifier goes
type declaration statements are:

array

COUNT,

MATRIX (4,4),

SUM

MAN, IABS

LOGICAL

SWITCH

INTEGER*2

REAL*

WX1l,

M12*4,

WX3*4,

IVEC*4 (10)

WX5,

WX6*8

DIMENSION
5.3

DIMENSION

The

DIMENSION

and

the

The

DIMENSION

STATEMENT
statement

number

specifies

elements

statement

DIMENSION

has

the

each

the

number

dimensions

dimension.

following

array

form:

a(d)[,a(d)]...

a(d)
An

array

declarator

(see

Section

2,5.1).

The

symbolic

name

array.

d
A

The

dimension

DIMENSION

element

determines
number

product

This

declarator.

statement
each

allocates

dimension

the
its

dimension

statement

defines

and

can

array

defines

You

also

ARRAY

MATRIX

use

array

statements.

name

cannot

element

The

only

label

one

having

125

declarators
However,

array

DIMENSION

For

data

to
type

each
of

Moreover,
array

array

the

array

the

total

equal

example:

the

MATRIX(5,5,5)

each.
You

storage

array.

declarators.

ARRAY(4,4),

VIRTUAL

one

each

length of
a
storage
element.
storage
elements
assigned
to
an

DIMENSION

each,

type
each

declarator.

statements.

real

elements

integer

declaration,
program

elements

unit,

4
of

Dbytes
2

bytes

COMMON,
you

can

use

and
an

SPECIFICATION STATEMENTS

Examples of DIMENSION statements are:
BUD(12,24,10)

DIMENSION

DIMENSION X(5,5,5),Y(4,85),2(100)
DIMENSION MARK(4,4,4,4)

For further information on arrays and on storing array
Section

elements,

see

2.5.

COMMON
5.4

COMMON STATEMENT

A COMMON statement reserves one or more contiguous blocks of

name

symbolic

identifies

each

block;

the

order

statments
common

block.

specify

storage.

you can omit a

COMMON
program unit.
variables and arrays in each

symbolic name for the blank common block in a

also

however,

The COMMON statement has the following form:
COMMON [/[cbl/]
cb

nlist

nlist[[,]/[cb]/ nlist]...

A symbolic name, called a common block name. cb can be Dblank.
If the first cb is blank, you can omit the first pair of slashes.
A list of variable names,
separated by

commas.

A common block name can be the same

However,

and

array

variable

names,

array

declarators

array

name.

cannot be the same as a function name or subroutine name

in the executable program (see Section 2.1).

When you declare common blocks of the same name in different program
units, these names are all associated with the same storage area when
For
the program units are combined into an executable program.
example:

PROGRAM

MAIN

COMMON /BLOCK1/ICOUN, IHOL/BLOCK2/ICHK
CALL

GSUB

END

SUBROUTINE

GSUB

COMMON /BLOCK2 /JCHK (10) /BLOCKL /JCOUN, JHOL

END

In the example, BLOCKl in MAIN and BLOCK1 in GSUB are associated

the same storage area, as are the BLOCK2s,

with

You can have only one blank common block in an executable program, but

you can have any number of named common blocks.

SPECIFICATION

Entities

are

above

the

storage

space

Entities
agree

in
if

the

INTEGER*2

COMMON

with

same

must

results

unit

space

may
to

not

assign
in

for

beginning
Examples

occur

on
of

when

the

way

byte

the

boundary.

(even

program
2-byte

the

list.

program

unit

same

should

correspondence

program

unit.

For

The

byte)

statements

are

combined

into

variable

the

variables

subsequent

however,

units

integer

bytes

(BYTE)

that

blocks;

word

these

data

compiler
all

MONEY

real

arrays
other

block

variable

any

supplies

common

type

filler

are

allocated

/BLK1/LIMA,R/

/ALFA,BET

boundary.

follow.

Program

Subprogram

HEAT,X/BLK1/KILO,Q

SUBROUTINE

FIGURE

.
RETURN

END

statement

block,

and

statement
and

KILO

the

statement.

COMMON

HEAT

basis.

with

statement

COMMON

one-for-one

one-to-one

another

The

first

one
in

the

CALL

associated

occur

high-order

LOGICAL*1

odd

COMMON

Main

common

are

placed

contains

because

the

such

common

COMMON

statement

contains

program

block

allocated

each

block

unit

CENTS.
You

they

entities

common

blocks

JCOUN

MONEY

correspond

common

common
and

MONEY

executable

made

because

program

COMMON

incorrect

ICOUN

CENTS

another

BLOCKl

type

one

COMMON

and

data

them

example,

storage

example,

assigned

with

assigned

STATEMENTS

and

Valid

the

puts

the

main

KILO

and

program
Q

subroutine

blank

common

makes

block,

puts
named

ALFA

and

BLKl.

HEAT

and

common

and

makes

block,

BET
LIMA

the

blank

BLK1l.

correspond
and

The

correspond

Usage

LOGICAL*1

CHARS (9)

COMMON/STRING/ILEN,CHARS

Invalid

Usage

LOGICAL*1 CHARS (9)
COMMON/STRING/CHARS,
ILEN

odd

the

second example,
byte address.

the

integer

variable

ILEN

allocated

SPECIFICATION STATEMENTS

VIRTUAL
5.5

VIRTUAL STATEMENT

allocated in physical
A virtual array is an array whose storage is addres
sable main memory.
ly
main memory outside of the program's direct
ly addressable
direct
frees
m
progra
a
in
arrays
The use of virtual
memory for executable code and other data storage.

specifies the
The VIRTUAL statement specifies a virtual array. s It
dimension.
each
in
element
of
number
the
and
ons,
number of dimensi
The VIRTUAL statement has the following form:
VIRTUAL a(d)

a(d)

[,a(d)]...

An array declarator

(see Section 2.5.1).

The symbolic name of an array.

A dimension declarator.

directly

addressable

space

available

user

total
or 65,536
programs executing on a PDP-11 family computer sis of64K,
1 to 8
from
element
32,767
of
m
maximu
bytes. An array can have a
element
per
byte
1
of
array
L*1
LOGICA
m
maximu
A
.
element
per
bytes
maximum COMPLEX array
The

maximum

would require 32,767 bytes of storage space. A
storage space, a
of 8 bytes per element would require 262,136 bytes of ressab
le memory.
ly-add
direct
on
limit
requirement far beyond the 64K
without
memory
main
l
externa
in
placed
are
arrays
Virtual
memory
le
ressab
ly-add
direct
of
64K
significantly diminishing the
to

available

programs.

NOTE

Virtual arrays are not supported on all
See the
systems.
operating
PDP-11
appropriate PDP-11 FORTRAN IV user's
guide for more information.

Examples:

VIRTUAL A(1000), LARG(180,180), MULT (4,4,4,4,4,4,4)

named A having 1000
The above example defines a one-dimensional arrayhaving
32400 elements,
LARG
named
array
elements, a two-dimensional
16384 elements.
and a seven-dimensional array named MULT having

same way as the
The data type of a virtual array is specified in the
implicitly
either
is,
that
array,
or
e
data type of any other variabl
a type
in
itly
explic
or
name
the
of
letter
first
the
to
ing
accord
declaration statement.

For further information
Section

2.5.

concerning

arrays

and

their

storage,

see

SPECIFICATION

5.5.1

Restrictions

The

names

used

some

virtual

array

and

name

Virtual

virtual

must

not

name

used

virtual

array

EQUIVALENCE

array

initial

array

value

virtual

The

I/0

name of a
specified

The

name

actual

value
7.

virtual

as
of

The

name

wused

(Section

the
of

array

the

array
a

not

COMMON

array

must

virtual

array

the

format

(third

statement

must

not

the

element

must not
argument inside
(Section 7.6).

used

subprogram

argument

format
array or
specifier

virtual

array

element

assigned

run-time
a

virtual

DECODE

not

5.6) .

argument

dummy

specify

statement

element

contain
name

appear

subprogram

corresponding

(Section

to
The

buffer

ENCODE

virtual

argument

must

array element cannot
be
statement (Section 5.8).

DATA

statement.

parentheses)

used

virtual

arrays cannot be
used
specifications
(Section 8.6).
virtual array element must not

elements

statement

Virtual

Arrays

5.4).

Use

arrays
contexts:

The

the

virtual

(Section
2.

STATEMENTS

(Section

assigns

6.1).

virtual

NAME

9.1.13).

array
element
cannot
keyword in an OPEN statement

Examples:

Valid

Usage
VIRTUAL

A(1000),B(2000)

READ(1,*)

10,1I=1,1000

B(I)=-A(I)*2
WRITE(2,*)

CALL

SUB

Invalid

Usage

VIRTUAL

A(10)

DATA

(A(I),I=1,1000)

(A,B)

A(l)/2.5/

COMMON

/X/

EQUIVALENCE
WRITE(1,A)

(A(l),Y)
X,Y

ENCODE(4,100,A(3))

5.5.2
A

Virtual

Array

dummy

argument

associated

with

X,Y

References
declared

actual

(Used

DATA

(Used

COMMON

(Used

EQUIVALENCE

(Used

format

specifier)

(Used

ENCODE

output

statement)
statement)

actual

variable

argument

associated
(see

that

argument

virtual

that

reference

only
with
Section 2.4).
1In

buffer)

Subprograms

array

also

array.

become

statement)

the

virtual

dummy argument
effect, an actual

can

only

name

array

declared
argument

become

virtual

element

a
that

can

simple
is

SPECIFICATION

STATEMENTS

a
Furthermore,
an expression.
virtual array element is treated as
value must have been assigned to the element before it is used as an
The subprogram must not alter the value of the
actual argument.
corresponding dummy

argument.

EXAMPLES:

Valid Usage
A(1000),B(1000)

VIRTUAL
B(3)=0.5

CALL

SCALE(A,1000,B(3))

END

SUBROUTINE

(X,N,W)

SCALE

X (N)

VIRTUAL
S=0

I=1,N

10,

S=S+X(I)*W
TYPE

*,S

END

Invalid

Usage

VIRTUAL

A(1000)

REAL

B(4000)

CALL ABC(A,B,A,(3))
END

SUBROUTINE

ABC(X,Y,Z)

(Actual argument is virtual)
(Actual argument is nonvirtual)
(Actual argument is virtual array

REAL X (1000)
VIRTUAL Y (4000)
Z=2.3

element)

END

EQUIVALENCE
5.6

EQUIVALENCE STATEMENT

The EQUIVALENCE statement partially or totally associates two or

entities in the same program unit with the same storage location.

The EQUIVALENCE statement has the following form:
EQUIVALENCE
nlist

(nlist)

[,(nlist)]...

A list of variables, array elements, and arrays separated
commas. You must specify at least two entities in each list.

1list

The EQUIVALENCE statement allocates all of the entities in

each

In an EQUIVALENCE statement, each expression in a subscript

reference

beginning at the same storage location.

must be an integer constant.

Dummy arguments, virtual arrays, and virtual array elements may not be
used

in an EQUIVALLENCE statement.

SPECIFICATION

You

must

way

byte
An

not

equivalence

that

subsequent

data

used

EQUIVALENCE

LOGICAL*1l

STATEMENTS

arrays

any

boundary.

array

name

element

You

equivalence

can

the

that

each

store

multiple

variables

begins

with

Examples

Valid

the

one

data

PRECISION

the

complex

statements

(DVAR,IARR(1l))

EQUIVALENCE

statement

IARR

occupy

Invalid

Usage

the

makes

same

above

(ILEN,

example,

Making
you

Arrays

make

an integer
variable shares

variable.

the

four
as

elements

the

double

the

integer

precision

variable

integer

variable

ILEN

allocated

the
elements

arrays

same

element

another

the

first

You

must

the

not

to
to

normal
the

use

the

linear

array

and
of

the

statement

locations

storage

element

for

made

equivalent,

example,

the
in

same
way

assign

array.
that

the

same

You

also

array

elements.

For

array

equivalent

equivalence

attempt

first

array

and

set
the

sixth

example,
the

examples
DIMENSION

TABLE

EQUIVALENCE
a

these

result

storage

the

use

(2,2),

the

TRIPLE

(TABLE(2,2),

these

EQUIVALENCE

both

element
overlap

storage

must

element

statement

you

first

array.

the

odd

third

inconsistent

one

then

the

made equivalent to the first
elements of the
first
array
second array.

elements

memory

are

I1f,

five

EQUIVALENCE

arrays

space.

array

the

assign

element

equal-sized

storage

last
elements of

two

first

two

7-element

five

make

the

array,

the

Some

of one array
equivalent
to
an
element
of
EQUIVALENCE
statement
also
sets
equivalences
corresponding elements of the two arrays.
Thus,
if
the

first

second

you
integer

element

between

another

Equivalent

array,

attempt

data
types,
such
Furthermore, you can
single component
of

BYTS(2))

the

another

location

first

are:

storage

address.

When

the

odd

BYTES (10)

EQUIVALENCE

5.6.1

with

such

DVAR

EQUIVALENCE

the

type

IARR(4)

LOGICAL*1

numeric

address.

example,
variable, the

DVAR.

byte

refers

For

complex

part

same

elements

allocated

Usage

INTEGER*2

array

statement

different

the

type.

EQUIVALENCE

DOUBLE

This

real

components

higher-ranked
data
variable equivalent to

storage

other

type

array.

entity

with

other

not
with

cannot

element

between

the

other

follow.

(2,2,2)

TRIPLE(1,2,2))

statements,

the

space
allocated
to
statements align the arrays.

entire

array

TRIPLE.

TABLE

Figure

5-1

part

shows

how

SPECIFICATION

STATEMENTS

Array TABLE

Element
Number

TRIPLE(1,1,1)
TRIPLE(2,1,1)
TRIPLE(l,2,1)
TRIPLE(2,2,1)
TRIPLE(1,1,2)
TRIPLE(2,1,2)
TRIPLE(1,2,2)
TRIPLE(2,2,2)

Element
Number

Array
Element

WU
WN -

Array TRIPLE

TABLE(1,1)
TABLE(2,1)
TABLE (1,2)
TABLE (2,2)

Equivalence of Array Storage

Figure 5-1:

The following statements also align the two arrays as shown in

Figure

5-~1:

EQUIVALENCE
EQUIVALENCE

(TABLE,TRIPLE(2,2,1))
(TRIPLE(l,1,2), TABLE(2,1))

In the EQUIVALENCE statement only, you can identify an array element
(that is, the linear element number), even
with a single subscript
For
though the array was defined as a multidimensional array.

example,
Figure

the

EQUIVALENCE

5.6.2

aligns the two arrays as shown in

statement

following

5-1:

(TABLE(4),

TRIPLE(7))

Extending Common Blocks

When you make entities equivalent to other entities stored in a common

block, the common block can be extended beyond its original boundaries
But
to include the entities specified in the EQUIVALENCE statement.
you can extend the common block only beyond the last element of the
previously established common block.

extend

cannot

You

the

common

B(5)

B(6)

block in such a way as to place the extended portion before the first
The following examples show
element of the existing common block.
valid and invalid extensions of the common block:

valid

COMMON

A(2)

A(l)

DIMENSION A(4),B(6)
A

EQUIVALENCE (A(2),B(1l))

B(l)

A(3)

A(4)

B(2)

B(3)

B(4)

~—

Extended
Portion

Existing
Common
Invalid

DIMENSION A(4),B(6)
COMMON

B(1l)

EQUIVALENCE (A(2),B(3))

A(l)

A(2)

A(3)

A(4)

B(2)

B(3)

B(4)

B(5)

\V_/\

Ex tended
Portion

If you assign two entites to

equivalent

each

other.

common

— -

—~

Existing Common

blocks,

you

cannot

B(6)
7

Extended
Portion

make

them

SPECIFICATION

STATEMENTS

EXTERNAL
5.7

EXTERNAL

The

EXTERNAL

arguments

external

The

statement

subprogram

STATEMENT

other

procedure

(Section

EXTERNAL

symbolic

EXTERNAL

external

name

Note,

however,

(for

example,

not

use

subprogram

declares

function

actual

subroutine

(Section

6.3).

form:

the

name.

The

CALL

the

each

name

subprogram

reference

complete

SQRT(B)

name

name.

that

Such

function

name
can

1in

then

can

use

CALL

dummy

the
appear

the

argument

list

actual

associated

reference used as an
SUBR(A,SQR
(B) ,C))
T
represents
name

dummy

statement.

function

need

not

argument

value,

defined

the

EXTERNAL

Main

Program

EXTERNAL

names

function

library

following

subprogram

name.

that

procedure

user-supplied

statement.

example

external

FORTRAN

the

subprogram;

subprogram

EXTERNAL

statement

argument

has

name

with

procedure

argument

you

{,v}...

associated

The

can

6.2)

statement

EXTERNAL

The

lets

subprograms.

statement

is:

Subprograms

SIN,COS,SINDEG

SUBROUTINE
EXTERNAL

TRIG

(X,F,Y)

F(X)

RETURN

CALL

TRIG

(ANGLE,SIN,SINE)

CALL

TRIG

(ANGLE,COS,COSINE)

CALL

TRIG

(ANGLE,SINDEG,SINE)

END

.
.

FUNCTION

SINDEG

SIN

(X)

(X*3.14159/180)

RETURN

END

In
the

the

example,

FORTRAN

statements

function
second
TRIG,

pass

the

reference

statement
the

SIN

library,

second

and

COS

and
name

F (X)
TRIG.

statement

SIN (X)

COS (X)

SINDEG (X)

are

SINDEG
a

trigonometric
is

functions
supplied
1in
user-supplied function.
The CALL

function

subsequently
Depending
is

equivalent

the

subroutine

invokes
which
to

the

CALL

one

TRIG.

function

statement
the

The

the

invoked

following:

SPECIFICATION STATEMENTS

DATA
5.8

DATA STATEMENT

The DATA statement assigns initial values to
array elements before program execution,

variables,

arrays,

and

array

The DATA statement has the following form:
DATA nlist/clist/[[,)nlist/clist/]...
nlist

A list of one or more

variable

names,

array

names, separated by commas, to which the values in clist
basis.
one-for-one
be assigned sequentially on a

element
are to

Subscript expressions must be integer constants.

clist

A list of constants, separated

commas,

assigned

Clist constants have one of the following forms:

nlist.

val

Used when you specify clist as n * val. Specifies the number of
times the same value is to be assigned to successive entities in
the associated nlist. The value of n is a nonzero, unsigned
constant.

integer

The DATA statement assigns the constant values in each <clist to the
entites in the preceding nlist. Values are assigned one for one in
the order in which they appear, from left to right.

The number of constants must

correspond

entities in the preceding nlist.

exactly

the

number

When an unsubscripted array name appears in nlist, values are assigned
to every element. The associated constant list must therefore contain
enough values to fill the array. Array elements are filled 1in the
order of subscript progression.

Dummy arguments, virtual arrays, and virtual array elements may not be
initialized

in DATA statements.

When a Hollerith constant or alphanumeric literal is assigned to a
variable or array element, the number of characters that can be
(see Table 2-2).
assigned depends on the data type of the component
capacity of the
the
than
s
character
fewer
contains
If the constant
variable or array element, the constant is extended on the right with
1f the constant contains more characters than can be stored,
spaces.
the constant is truncated on the right.
Examples of
INTEGER

the DATA statement are:
A(10)

BYTE BELL,TAB,LF,FF,ACHR,ZCHR

DATA A, BELL,TAB,LF,FF,ACHR,ZCHR/10*0,7,9,10,12'A"',1HZ/

SPECIFICATION

the

array

example,

the

ASCII

BELL,

TAB,

and

byte

variables

Some

other

REAL

examples

statements
control

LF,

assign

character

all

codes

are

FF.

elements

assigned

of
the

are:

X (5)

COMPLEX
DATA

DATA

STATEMENTS

X/2*—3.,4.,2*0.37/,Z/(l.0,-3.0)/

PROGRAM
5.9

PROGRAM

STATEMENT

The

PROGRAM

statement

assigns

The

PROGRAM

statement

has

following

PROGRAM

the

symbolic

name

main

it,

program

unit.,

form:

nam

The

symbolic

PROGRAM

first
the

statement

name

any

the

same

executable

The

name,

PROGRAM

optional.

the main

entity

within

the

name

program

statement

any
(see

must

you

program.

use

The

the

symbolic

main program.
subprogram, entry,
Section

not

have

It
or

must

name must

also

must

common

statement

not

the

block

2.1),

the

not

label.

BLOCK DATA
5.10
The

BLOCK
BLOCK

only

DATA

purpose

blocks.

statement
is

BLOCK

the

special

type

blocks

and

common

BLOCK

DATA

statements.

DATA

BLOCK

begins

declare

Therefore,

nonexecutable

The

STATEMENT

statement

DATA

[nam]

symbolic

name.

can

use

has

the

program

following

of
to

program
define

wunit

unit

data

can

contain

whose
common

only

form:

nam

You

EQUIVALENCE,

The

last

only
and

statement

statement.,

type
DATA

declaration,
statements

BLOCK

DATA

program

‘

BLOCK

DATA

program

BLOCK

DATA

statement

unit

must

not

IMPLICIT,

following

contain

have

any

DIMENSION,
BLOCK

unit

DATA

must

executable

statement

label.

COMMON,

statement.

END

statements.

SPECIFICATION STATEMENTS

If you use a BLOCK DATA program unit to initialize any entity in a
you must provide a complete set of data type
block,
common
specification statements for all the entities 1in the block, even
though some of the entities are not assigned an initial value in a
DATA statement. You can use the same BLOCK DATA program unit to
define initial values for more than one common block.
An example of a BLOCK DATA program unit is:
BLOCK

DATA BLKDAT

INTEGER
LOGICAL

S,X
T,W

DOUBLE PRECISION
DIMENSION R(3)

COMMON /AREAl/R,S,T,U,/AREA2/W,X,Y

DATA R/1.0,2*2.0/,T/.FALSE./,U/0.214537D-7/,W/.TRUE./,Y/3.5/
END

In the example, enough information
implicitly declare the data type
blocks AREA1l and AREA2. Not all the

statement.

1is provided to explicitly or
of every variable in the common
variables appear in the DATA

CHAPTER

SUBPROGRAMS

subprogram

procedure.

one

subprogram

the

subprogram

some

cases,

different

program

Subprograms

the

are

FORTRAN

User-written

same

two

kinds:

subprograms

the

called

actual

arguments

dummy

discussion

subprogram

referencing
a

argument

the

arguments

basis

argument
and

any

also

cases,

the

library

references

the

supplied

in
in

part

back

functions:

Section

The

subprogram

the

actual

referencing

6.1

which

subprogram

receive

the

system-supplied

the

subprogram,

results.

describes

the

specifies

arguments.

program.
actual

user-written

passes

uses

and

dummy

subprograms;

and

subprograms.

1is

ends.

reference

For

example,

when

assigned

returns,
1is

the

(I,J(3),4)

of
and

the

list

the

dummy

subprogram.

corresponding
argument

corresponding

actual

argument

actual
in

actual

value

is
4.

the

argument.

and

dummy

association

next.

arguments,
a

the

association
the

M has

the

retention

dummy

transferred

value
to

subprogram

an associated list
associated with J(3),

the

assigned

There

control

takes

value

subprogram

one

located,

an entity which passes a value to or
from
a
arguments
are
specified
in
the
statement
the subprogram.
Dummy
arguments
are
specified
in
the
of the subprogram and are associated with actual arguments

dummy

subprogram

When

other

ARGUMENTS

one-to-one

argument;

statement
and,

Actual

definition
Each

computing
references

references

arguments,

6.2

describes

SUBPROGRAM

that

kinds:

FORTRAN

that

passed

below,

Section

subprogram.

turn

function

compute

called

6.1

define

statement

user-written

three

arguments,

may

6.3

program

actual

entities,

Section

wunit

either

reference

Values
the

are

Processor-defined
cases,

that

functions

kind

values,

when

referencing

system.

Subroutines

arguments;

statements

unit.

many

invoked

The

Functions

program

one

group

the

Statement

executed.

There

is
in

from

arguments

and

(K,L,M)

associated

with

SUBPROGRAMS

Rules Governing Subprogram Arguments

6.1.1

constants,

Actual arguments can be

elements, or subprogram names.

array

Actual

appear

the subprogram definition

variables,

arrays,

expressions,

Dummy arguments as specified in
variable

unsubscripted

names.

must agree in order, number, and data type with the

arguments

dummy arguments with which they are associated.

are

Dummy arguments

symbolic

names

associated

become

which

with

variables, arrays, or subprograms defined or declared in other program

units.

associated

dummy

with

argument
actual

wundefined

Although dummy arguments are not variables,

each dummy argument
array, or subprogram.

not

currently

argument.

arrays,

subprograms,

may be declared as though it were a variable,
the
Each dummy argument name is declared with

the associated actual argument.

attributes of

Of
If the actual argument is a constant, expression, subprogram name,
may
argument
dummy
g
correspondin
the
reference,
array element
virtual
not

be modified.

A dummy argument declared as an array can be associated only with an
1is an array or array element of the same data
actual argument that
dummy

argument

array

If the actual argument is an array element, the dummy

argument

array

type.

If the actual argument is an array, the

must not be larger

than the actual argument array.

will be associated with elements of the actual argument array starting

In this case, the dummy argument array must
from the actual argument.
not be larger than the number of elements remaining in the actual
argument

array.

Valid

Usage

PROGRAM

MAIN

DIMENSION

CALL

X(A,

A(10),

B(5.5)

B(1l,2))

END

X(Y,Z)
DIMENSION Y (10), Z(5,2)
SUBROUTINE
END

Invalid

Usage

PROGRAM

MAIN

DIMENSION

CALL

A(10),

B(5,5)

X(A,B(1,21)

END

SUBROUTINE

X (C,D)

DIMENSION C(12)

DIMENSION D(5,5)
END

(dummy array must not be larger than actual
array)

(dummy array must not be larger than number

of elements remaining in actual array)

SUBPROGRAMS

6.1.2

Adjustable

Arrays

An adjustable array is a dummy argument array
with
dimensions
that
can
be
changed
or
dimensions of the associated actual argument
program.

well

The

adjustable

constants,

following
®

The

rules

array

the

govern

adjustable

declarator

dimension

the

use

array

The

adjustable

argument

array
is

must

size

the

adjustable

size

the

actual

become

the

match the
referencing

integer

variables,

arrays:

dummy

argument

associated

with

array

must

less

array.

in the adjustable array
adjustable dimensions must be

Variables

the

contains

subprogram

declarators.

Variables

array

adjustable

array

arguments

the

subprogram,

arguments

must

have

defined

than

declarator

the

actual

equal

that

integer

declarator
and

array.

The

the

For

that

in
"adjusted"

adjustable

must

subprogram.
e

declared

represent

data

must

type.
be

dummy

corresponding

actual

value.

example,

PROGRAM

MAIN

DIMENSION

SUM1

Al1(10,35),

A2(3,56)

SUM(Al1,10,35)

SUM2

SUM(A2,3,56)

SUM3

SUM(A1l,10,10)

END

FUNCTION

SUM

DIMENSION

(A,M,N)

A (M,N)

SUM

1,N

1,M

SUM

0.0

SUM

A (I,J)

RETURN

END

the

example,

array.

The

control

6.2
A

the

You

can

places
of

that

statement

information

user-written
a

actual

Note

USER-WRITTEN

are

A2.

statements

can

subprogram

For

and

function

series

perform

either

subprograms

your

program,
or

well

declarators,

sum

and

arguments

iteration

use

operations

and

the

see

the

adjustable

specified

are

specify

Section

sections

used

the

size

to
of

2.5.1.

SUBPROGRAMS

subprogram

that

dummy

array

arrays

computes

FORTRAN

statement

computing

procedure.

arithmetic

operations

and

statements

perform

thus
in

avoid

each

computing

having

place.

group

computing

FORTRAN
procedure

duplicate

FORTRAN
procedure

statements.
in

several

the

series

SUBPROGRAMS

Table 6-1 1lists
There are three types of user-written subprograms.
it, and the
define
to
needed
ts
statemen
the
am,
each type of subprogr

method of transferring control to the subprogram.

Table 6-1:

Types of User-Written Subprograms

Subprogram

Method

definition

Function reference

function

Statement function

Function subprogram

FUNCTION

Function reference

Subroutine subprogram

SUBROUTINE

Statement

Control Transfer

Defining Statements

RETURN

CALL statement

RETURN

A function reference consists of the function name and the function
arguments and is used in an expression. The function returns a value
that is used in place of the reference in the expression in which it
appears.

Function and subroutine subprograms can change

their

cannot,

either

values

the

arguments and the calling program can use the changed values.

A subprogram can refer to other subprograms,

directly or indirectly, refer to itself.

but

Statement Functions

6.2.1

A statement function is a single-statement computation specified by a
When you reference the statement function name, with
symbolic name.
its arguments, in an expression, the computation defined by the
statement Ffunction name is performed and the resulting value replaces
Statement
the statement function name in evaluating the expression.
unit.
program
single
a
within
d
reference
and
defined
are
functions
The statement function definition statement has the following form:
£
f

([pl,pl...]1)=e

The name of the statement function.
A dummy

argument.

expression.

The expression (e) is an arithmetic or logical expression that defines
the computation to be performed.

SUBPROGRAMS

statement

function

reference

has

the

following

form:

(lal[,al...])

The

name

actual

the

function.

argument.

When a statement function reference
appears
1in
an
expression,
the
values of the actual arguments are associated with the dummy arguments
in
the
statement
function
definition.
The
expression
in
the
definition is then evaluated.
The resulting value is used to complete
the evaluation of the expression containing the function reference.
The

following
e

rules

govern

the

Statement function
unit.,

use

names

statement

must

functions:

. unique

within

the

same

program
e

statement
unit as its

A
statement
function
definition
statement
reference to another statement function, which
earlier in the same program unit.

Statement function
definitions
must
be
executable statements
(see Figure 1-3).

The data type of the resulting value assigned to the
determined
either implicitly by the first letter of
or explicitly by a type declaration statement.

Statement
order,

function reference
definition.

function

number,

dummy

and

must

appear

arguments

data

type

the

same

can
must

placed

program

include
a
be defined

before

all

name
1is
the name,

serve

only

indicate

arguments

for

the

statement

function.

Statement function
usual
rules
for
2.1).

Statement

within

each

arrays

declared
e

dummy

argument
uniqueness
of

function

statement

the

and

same

used

dummy

names

do
not
symbolic names

arguments

function

names

within

the

definition.

the

same

must

dummy

program

follow
the
(see Section
unique

Variables

arguments

can

or
be

unit.

The
data
type
of
statement
function
dummy
arguments
1is
determined
either implicitly by the first letter of the name,
or explicitly by a type declaration statement.

Examples

statement

function

definitions

are:

valid
VOLUME (RADIUS)

AVG

(A+B+C)/3

SINH

(A,B,C)
(X)

4.189*RADIUS**3

(EXP(X)

EXP

(-X))*0.5

Invalid/Explanation
AXG

only

(A,B,C,3.)

(A+B+C)/3

constant

argument).

cannot

dummy

SUBPROGRAMS

Examples

statement

refer

the

second

function

statement

references
function

follow.

definition

The

examples

below

above.

valid
GRADE
IF

AVG

(AVG

(TEST1,TEST2,XLAB)

(P,D,Q).LT.AVG(X,Y,Z))GO

300

Invalid/Explanation
FINAL

(An
in

actual
data

6.2.2

AVG

(TEST3,TEST4,LAB2)

argument

type;

Function

A function
When
you

and

this

its

case,

corresponding
LAB2

dummy

integer

argument

but

above

must
is

agree

real.)

Subprograms

subprogram
reference

is a
the

program unit referenced by
function
name,
with
its

a symbolic
arguments,

name.
in an

expression, the program unit defined by the function name is executed;
and
the resulting value of the function replaces the function name in
evaluating the
expression.
A
function
subprogram
consists
of
a
FUNCTION

statement

computing

procedure.

The

FUNCTION
[typ]

typ

followed

statement

FUNCTION

One

the

The

name

has

the

series

following

statements

that

define

form:

nam[*m] [([p[,pP]...])]

data

type

specifiers

(see

Table

2-2).

nam

the

function.

data

dummy

function

has

the

type

length

specifier

(see

Table

2-2).

argument.

reference

following

that

transfers

control

function

subprogram

form:

nam

([a[,al...])

The

symbolic

nam

When

actual

the

name

the

function.

argument.

the

function

subprogram

used

expression,

control
1is
transferred
to
the
subprogram;
and the values of the
actual arguments
(if any)
in the
function
reference
are
associated
with
the
dummy
arguments
(if
any)
in the FUNCTION statement.
The
statements in the subprogram are
then
executed.
A
value
must
be
assigned to the name of the function as though it were a variable.

Finally,

RETURN

statement

control
to
the
<calling
implied RETURN.
The value
to complete the evaluation

executed

the

function

and

returns

program
unit.
An END statement acts as an
assigned to the function's name is now used
of the expression containing the name.

SUBPROGRAMS

The

following
®

The

rules

govern

use

FUNCTION statement
function subprogram.
FUNCTION

function

SUBROUTINE,

function

cannot

The

data

type

The

example

type

function

can

itself

subprograms:

first

the

have

not

statement

contain

these

FUNCTION

statement.

reference

another

either

directly

function

name

can

the

function

must

the

have

the

referencing

subprogram

but

indirectly.

specified

FUNCTION

same

statements:

statement

statement.

name

label.

another

explicitly

and

FUNCTION

must

DATA,

function

not

subprogram

declaration

must

BLOCK

subprogram

must

statement

reference

implicitly

the

data

type

either

the

program.

is:

ROOT (A)

=1.0

EXP(X)

EMINX
ROOT
IF
X

=
=

1./EX
((EX+EMINX)*.5+COS(X)-A)/((EX

(ABS (X-ROOT).LT.lE-6)
=

EMINX)*.5-SIN
(X))

RETURN

ROOT

this

END

The
to

function
obtain

the

example

root

cosh(X)

F(X)

The

value

this

root

is:

Xi+l

uses the Newton-Raphson
following function:

the

cos(X)

passed

iteration

method

argument.

The

iteration

formula

for

cosh (Xi)+cos (Xi)-A

sinh (Xi)-sin(X1i)
The

caluclation

less

than

The

function

uses

(see

Section

6.3).

6.2.3
A

Subroutine

subroutine

name.,

When

program
to

the

returned
a

the

FORTRAN

defined
the

SUBROUTINE

the

difference

between

and

library

functions

EXP,

SIN,

COS,

Xi+l

and

ABS

Subprograms
is

reference

statement

computing

until

subprogram
you

unit
to

repeated

1.0E-6.

the

statement

procedure.

unit

subroutine

function

subroutine

program

the

and

referenced
name

name

function

name.

followed

series

CALL

symbolic

statement,

the

executed.

contrast

subprogram,

value

subroutine
a

subprogram
statements

consists
that

is
of

define

SUBPROGRAMS

The SUBROUTINE statement has
SUBROUTINE

nam

the

following

form:

[([pl,p]...1)]

nam

The

name

dummy

the

subroutine.

argument.

You must use a CALL statement to
transfer
control
to
a
subroutine
subprogram,
and
a
RETURN statement to return control to the calling

program

unit.

Section

4.5

describes

the CALL

statement.

When control is transferred to
the
subroutine,
the
values
of
the
actual
arguments
(if
any) in the CALL statement are associated with
the
corresponding
dummy
arguments
(if
any)
in
the
SUBROUTINE
statement.
The
statements
in
the
subprogram
are
then executed.
Finally, a RETURN statement is
executed
in
the
subroutine
and
it
returns

control

implied

The

the

calling

program.

END

statement

acts

RETURN.

following
®

rules govern

The SUBROUTINE

the

use

subroutine

statement must

the

statement must

not have

subprograms:

first

statement

subroutine.

The

SUBROUTINE

A subroutine
DATA,

subprogram must

another

SUBROUTINE

not

contain

statement

label.

FUNCTION,

BLOCK

statement.

A subroutine subprogram can reference another
it cannot reference itself either directly or

subprogram,
indirectly.

but

Example:

The subroutine in the following
example
computes
the
volume
of
a
regular
polyhedron,
given
the number of faces and the length of one
edge.
It uses the computed GO TO statement to determine
whether
the
polyhedron
is
a
tetrahedron,
cube,
octahedron,
dodecahedron,
or
icosahedron.
The GO TO statement also transfers control to the proper
procedure
for
calculating the volume.
If the number of faces is not
4, 6, 8, 12, or 20, the subroutine displays an error
message
on
the
user's

terminal.

Main

Program

COMMON

NFACES, EDGE, VOLUME

NFACES,EDGE

CALL

PLYVOL

TYPE

STOP
END

'VOLUME=',6VOLUME

SUBPROGRAMS

Subroutine
SUBROUTINE
COMMON

CUBED

PLYVOL

NFACES, EDGE,VOLUME

Goro

= EDGE**3
(6,6,6,1,6,2,6,3,6,6,6,4,6,6,6,6,6,6,6,5)
,NFACES

GOTO

VOLUME

CUBED

0,11785

CUBED

0.47140

CUBED

7.66312

CUBED

2.18170

RETURN

VOLUME
RETURN

TYPE

100

100,

FORMAT

NFACES

REGULAR

POLYHEDRON

HAS

',I3,

FACES.'/)

VOLUME=0.0
RETURN

END

6.3

FORTRAN

LIBRARY

FORTRAN library
in the same way
For

FUNCTIONS

functions are system-supplied
subprograms
as user-written function subprograms.

referenced

example:

3.14159

ABS is a
absolute

FORTRAN
value

library
of
X-1

3.14159;

the

The FORTRAN
gives
the

ABS (X-1)

result

function.
As a result of this reference, the
1is calculated and multiplied by the constant

assigned

library functions are
data
type
of
each

the

variable

“"R.

listed in
Appendix
B,
which
also
1library
function
and of the actual

arguments.

The

FORTRAN

1library

functions.

Note

that

Intrinsic

Functions

functions

the

and

also

are

processor-defined

the

Basic

External

called

functions
Functions

processor-defined

include

both

the

described

ANS

FORTRAN.

Normally, a name in the table of processor-defined functions refers to
the
FORTRAN
library
function with that name.
However, the name can
refer
to
a
user-defined
function
wunder
any
of
the
following
conditions:

The name appears in
different data type

a type declaration
from that shown in

statement specifying
the table.

The name is used in
different data type

a function reference with arguments
from that shown in the table.

Processor-defined function names apply only to
the
program
unit
in
which
they are referenced.
Thus, they can be used for other purposes
in
other
program
units.
In
addition,
the
data
type
of
a
processor-defined
function
does
not
change
because of an IMPLICIT
statement,

CHAPTER

INPUT/OUTPUT

FORTRAN

programs

use

READ

REWRITE,

TYPE,

and

statements

are

translation

wused

and

with

editing

statements

format

data

(readable

character)

form.

Each

statement

refers

WRITE

STATEMENTS

statements

external

READ

for

output.

specifiers

between

the

input,
Some

that

internal

logical

ACCEPT,

rather,

unit.

TYPE,

they

The

user's

terminal

system

line

(binary)

PRINT
data

and

TYPE

and

the

(I/0)

Direct

Internal

and

I/0

data

program.

See

statements

between

are

unit

refer
and

to
an

normally
is

that

from

1I/0

records,

access

ENCODE

and

between

variables

units;
logical

connected

form

DECODE

format

specifiers

statements

form

See

arrays

the

and

from

terminal.

Section

statements

and

the

categories:

selected

files.

See

record
7.4.

translate

within

the

and

FORTRAN

7.6.

1I/0

(binary)

three

from

logical

sequentially
I/0 device such as a

transfers

Section

into

and

implicit

connected

normally

the

connected

records

direct

contain

character)

grouped

transfers

Formatted

internal

(readable

not

program

statement

are

and

from

transfer

statements.

Access

number,
®

the

statements
PRINT

statements

Sequential I/0
files,
or
to
Section 7.3.

statements
between

these

form

printer.

Input/output

I/0

and

transfer

WRITE,
of

control

which data is to be transferred.
A logical unit can be
device or file by the OPEN statement
(see Section 9.1).
The

and

forms

the

are

within

called

used

the

formatted
translate

program

and

1/0
data

external

records.

I/0 statements that
do
not
contain
format
specifiers
are
called
unformatted
1I/0 statements.
Unformatted I/0 statements transfer data
without translation.
Unformatted I/0
1is
generally
used
when
data
output

similar)
the

data

external

I/0

and

how

unformatted

record

the

associated

format

transferred.

data

I/0,

transferred.

saves

all

data

contain,

file

terms
and

input

execution

preserves

conserves

can

subsequently

I/O

process,

usually

transfer

one

will

Unformatted

translation
data,

that

depend
In

program

statements

data

program.

greater

storage

the

same

(or

eliminating

precision

the

amount

space.

records.
way

the

time

The

records

are

separated,

transferred.

the

I/O

formatted

specifier

statement

jointly

specifies

1/0,

the

determine

the

1I/0

the

amount

statement

amount

data

and

its

data

INPUT/OUTPUT

STATEMENTS

Executing an input or output statement initiates
transfer
of
a
new
record.
Normally,
the
data
transferred
by
an
I/0
statement
constitutes
one
record.
However,
formatted
I/0
statements
can
transfer

than

one

record.

Section 7.1 describes general FORTRAN input/output concepts.
Section
7.2
describes the components of FORTRAN I/O statements.
Sections 7.3
through

7.6

describe

each

category

I/0

statement.

I/0 OVERVIEW

7.1

The following sections describe in general terms
the
characteristics
of
FORTRAN I/0 processing:
records, files and access modes.
See the
appropriate PDP-11 FORTRAN IV
wuser's
guide
for
more
detail
about
FORTRAN

7.1.1

I/0

processing.

Records

A record is a collection
of
data
items,
called
fields,
that
are
logically
related
and
are
processed
as
a unit.
Each FORTRAN I/O
statement transfers one record.
Formatted I/0 statements may transfer
additional

records.

If an input statement does not use all of the data fields in a record,
the remaining fields are ignored.
If an input statement requires more
data fields than the record contains, an error occurs.
If an output statement
record can contain, an

7.1.2

attempts to write more
error condition occurs.

data

fields

than

PFiles

A file is a collection of logically related
records,
arranged
1in
sequential order, and treated as a unit.
The arrangement of a file
determined

when

the

file

created.

Files can be stored on disk or on
magnetic
devices such as terminals, card readers, and
as

the

sequential

tape.
Other
line printers

a
is

peripheral
are treated

files.

In the
sequential
file
organization,
records
appear
in
physical
sequence.
Each record, except the first, has another record preceding
it, and each record, except the last, has another record following it.
The
physical order in which records appear is always identical to the
order in which the records were originally written to the file.

7.1.3

Access Modes

Access mode is the method your program
uses
to
retrieve
and
store
records
in
a file.
The access mode is specified as part of each 1/0
statement.
PDP-11 FORTRAN IV supports
two
kinds
of
access
modes:
sequential and direct.

7.1.3.1
Sequential Access - Sequential access means that records
processed
1n
sequence.
For
a
sequential
organization
file,
sequence

the

physical

sequence

7-2

the

records.

are
the

INPUT/OUTPUT

7.1.3.2

Direct

specifies
number

records

7.2

the
each

must

1/0

Access
order
I/0

Direct

processing

statement.

STATEMENT

access
by

For

fixed-length,

STATEMENTS

means

including

sequential

that

the

direct

access

organization

program
record

files,

the

COMPONENTS

The following sections
logical unit numbers,

key

expressions,

control

7.2.1
A

I/0

error

Logical

logical

file

Unit

unit

number

logical

with

unit

end-of-file

value

number

Explicitly

through

TImplicitly

guide

numbers

Format

specifiers

the

are

The

statement

The

name

7.3.3

7.2.3

A direct
position
The
to

value
the

7.2.4
If

WRITE,

1/0

7.3.4

Direct

99,

file

device

Statement
The

the

(see

specific

constant

one

Section

appropriate

use

integer

through

two

9.1).

PDP-11

FORTRAN

implicit

1logical

unit

Record

record

number
in

error

REWRITE,

greater

number

End-of-File
or

you

run-time

format

Section

8.7

statements.

can

use

than

and

end-of-file

ENCODE

can

any

(see

Section

describes

asterisk

formatting.

is an integer value
direct access file,

records

Condition

and

instead

See

the

Sections

Numbers

record
be

1/0

statements

statement.

denote
list-directed
list~directed 1/0.

access

must

FORMAT

I/0

statements,

and

statements,

containing

the

formatted

array

formats

Access

maximum

I/0

detail.

label

FORMAT

between

specifier
and

OPEN

System.

used

describes

sequential

format

refers

must

Specifiers

®¢

interaction

occurs.

that

number

range

describes

greater

8.6).

the

following:

Chapter

the

value

unit

connected

user's

integer

logical

ways
:

7.2.2

condition

Numbers

1/0 device.

variable

describe
the
components
of
I/0
statements:
format specifiers, dire
ct access record numbers,
lists, and parameters
specifying the transfer
of

equal
in

the

Error

Condition

statement

7-3

and

less

file,

condition

DECODE

that

is
can

specifies

than

the

equal

any

READ,

Parameters

encountered,
specify

that

control

INPUT/OUTPUT STATEMENTS

rs have the
ement. The specifie
to a specified stat
to be transferred resp
conditions:
r
erro
and
ile
of-f
endfor
ectively,
following forms,
END=s
ERR=s

to be
The label of an executable statement to which control is
transferred.
de either
ENCODE or DECODE statementin cananyinclu
A READ, WRITE, REWRITE, above
. The
order
specifications,
the
or both of must
and/or
r,
numbe
d
recor
r,
numbe
unit
follow the
specification(s)
format.
to
ification must refer the
in the END=s or ERR=s spec
The statement label emen
of
that
as
unit
ram
prog
same
the
in
t with
an executable stat
1/0

statement.

records exist in a
occurs when no more produ
An end-of-file conditionwhen
ced by the ENDFILE
d
recor
ile
end-f
an
sequential file, oOr
a READ statement
1f
is encountered.
statement (see Section 9.8)e condi
1/0 operation, it
an
g
durin
tion
f-fil
encounters an end-o
specification.
END=s
named in the
transfers control to the statementnt,
an error condition occurs.
If no END=s specification is prese
1f a READ, WRITE, REWRITE, ENCODE or DECODE

statement encounters

sfers control to the
an I/0 operation, it tran
error condition during
ion., If no ERR=5S
ficat
speci
ERR=S
the
statement whose label appears in
is present, the 1/0 error terminates program execution.

s
or REWRITE statement or directa acces
An END= specification in a WRITE
d
recor
write
or
read
to
pt
attem
you
If

READ statement 1is ignored.er than the maximum specified for the
using a record number great
logical unit, an error condition occurs.
describes system
FORTRAN IV user's guide proce
The appropriate PDP-11 can
ssing. These
to control error
subroutines that you used use
I1/0 system
the
from
n
matio
infor
n
to obtai
subroutines can also be occur
red.
that
on the type of error

Examples of I/O statements follow.

READ (8,END=550)

(MATRIX(K),K=1,100)

control to statement 550 if an end-of-file
This statement transfers al
unit 8.
condition occurs on logic
WRITE

(6,50,ERR=390)

to statement 390 if an
This statement transfers controlment.
in execution of the WRITE state

error

occurs

READ (l,FORM,ERR=150,END=200) ARRAY

OcCCuUurs
ol to statement 150 if an error
This statement transfers contr
if the
200
ment
state
to
and
ment
state
READ

in the execution of the

end-of-file condition occurs.

7.2.5

Input/Output Lists

ment contains
t, ENCODE, or DECODE state
The I/0 list in an input, outpu
which or to
from
nts
eleme
array
and
s,
the names of variables, array
7-4

INPUT/OUTPUT

which
can

data

also

I/O

will

contain

list

has

transferred.

constants

the

and

STATEMENTS

The

I/0

list

expressions

following

output.

output

statement

form:

s[,s]...

A
The

simple

list

implied

1list.

I/0

statement assigns values to, or
elements in the order in which they

list

transfers
values
appear, from left

7.2.5.1
Simple Lists - A simple I/0 list
variable, array, array element, constant,

list

consists

simple

I/0

parentheses.

For

WRITE

When

you

ACCEPT
a

use

TYPE,

K(3),

reads
PRINT

order

varying most
two-dimensional array:

the

name

ARRAY,

statement

ARRAY

(2,1),

array

READ

list

fill

writes

rapidly.

commas

I/0

every

all

initial

subscript

1list,

element

the

values

element

progression,

For

and

enclosed

name
to

the

with

assigns

example,

subscripts,

values

ARRAY

(3,1),

statement,

subscripts
READ

The

can
be
a
single
expression.
A simple I/0
element or a group of two or

of
in

READ
the

array;

the

with

the

following

array.

array

and

leftmost

defines

ARRAY(3,3)

that

array
data

statement

element

separated

(L+4)/2,

with

subscript

enough

begins

the

I/0

example:

DIMENSION

simple
elements

unsubscripted

transfer

proceeds

1list

(5,10)

statement

WRITE,

Data

either

from,
the
to right.

ARRAY

later

(1,1250)

from

appears

the

(1,2),

input
and

variables

the

list.

For

READ

record(s)
on

the

through
I1/0

list

statement,
ARRAY (1,1),

ARRAY
can

(3,3).
be

used

example:

J,K,ARRAY (J,K)

1250

FORMAT

(I1,X,I1,X,F6.2)

input

record

contains

the

following

values:

1,3,721.73
When
to

the
J

READ

and

values

statement

the

for

the

output

1/0

a
An
a

before

this

example,

function
input

executed,
K,

(to

thereby

Then
that

the

are

left

output

subprogram

statement

subscript

I/0

expression

I/0

expression

1list

must

I/0

input

value
be

721.73

used

use

list

the

any

actual

expression

array

contain

reference.

to
way

expression.

I/0

must

this

subscripts

valid

further

assigned
subscript

assigned

subscripts

contain

value

the

operations.
not

1/0.

not

array

first

attempt

performs

list
in

the

establishing

their

may

not

statement

that

of)

list.
statement

However,
For

Variables

appear

is
to

ARRAY(J,K).

ARRAY (1,3).
must

second

expression,

refer

except

INPUT/OUTPUT STATEMENTS

element
7.2.5.2 Implied DO Lists - An implied DO list is an I/0 list
a
within
t
statemen
I/O
an
of
part
a
were
it
that functions as though
Implied DO lists can be used to:

DO loop.

Specify iteration of part of an 1/0 list

Transfer

Transfer array elements in a sequence different from the order

part

an array

of subscript progression

implied DO list has

the form:

(list,i=el,e2[,e3])
list

list.

1/0

integer

variable.

el,e2,e3

Integer

expressions.

same forms
and the same functions that they have in the DO statement (see Section
4.3). The list immediately preceding the DO loop parameter is the
range of the implied DO loop. Elements in that list can reference i,

The variable i and the parameters el, e2, and e3 have the
but they must not alter the value of i.
WRITE

(A,B,C,

(3,200)

Some examples are:

I=1,3)

The statement in this example behaves as though you had written:
WRITE
Another

is:

example

WRITE

A,B,C,A,B,C,A,B,C,

(3,200)

(I,(J,P(I),Q(I,J),J=1,L),I=1,M)

(6)

The I/0 list in this example consists of an implied DO list containing
The implied DO lists
another implied DO 1list nested within it,
the Js for
varying
fields,
*M
(1+3*L)
of
together will write a total
each

value

In a series of nested implied DO lists, the parentheses indicate the
Execution of the innermost lists is
nesting (see Section 4.3.2).
repeated most often.
150

WRITE

FORMAT

For example:

(6,150)

(F10.2)

((FORM(K,L), L=1,10), K=1,10,2)

Because 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. This is the reverse of the
order of subscript progression. In addition, K is incremented by 2,
so only the odd-numbered rows of the array are output.

The

entire

control

list

READ

this

This
then
The

P(l),

1list

For

(Q(I,J),

Q(1,1),

(3,5555)

Q(1l,2)

that

the

control

WRITE

(6,1111)

This

statement

7.3

SEQUENTIAL

(I,

simply

variable

can

you

vary

can

read

use

according

external

prints

the

integers

statements

I/0

statements
specifier,

control

1In

wusing

statements

translation
formatted

Both

that

contain

that

each

translation

list-directed

simple

can

both

through

implied

B0OX(1,10)
the

and

array.

directly.

For

20,

the

formatted
is

read

sequential

and

I/0

translation.

data

I/0

logical

unit.

list-directed

the

correct

statements

using

format

and

records.

the

I1/0

internal

and

external

statements

are

without

wusing

and

list

method
FORMAT

list-directed

When

you

records,

records

1I/0

read

files

you

must

ensure

binary

format.

transfer

from

internal

formatted
types

between
output

between

transfer
the

data
or

records

data

sequential

same

with

sequentially

sequential

input

formatted

refer

record

Unformatted

the

effect,

statements.

combination

output

transfer

format

elements

without

I/0

sequential

obtaining

records

the

Instead

form,

also

«control

form.

List-directed

INPUT/OUTPUT

sequential

specifiers

before

I=1,20)

Sequential I/O statements transfer
either files or I/0 devices.
Formatted

the

I=1,5)

statément assigns input values to BOX(1,1)
through
terminates without affecting any other element of
value

before

J=1,10)

example:

transmitted

eeesQ(1,10)

arrays,

subscripts

(BOX(1,J),

1is

example:

J=1,10),

multidimensional

subscripts
and
For example:
READ

implied

incremented.

processing

fixed
list,

STATEMENTS

(p(1),

example,

(5,999)

incremented
When

variable

INPUT/OUTPUT

data

READ
ACCEPT

7.3.1

Formatted

Sequential

Input

Statements

The

formatted
sequential
READ
statement
transfers
data
from
specified logical unit.
If a formatted sequential READ statement
not

have

logical

unit

number,

uses

implicit

logical

unit.

the
does

INPUT/OUTPUT

STATEMENTS

The formatted sequential ACCEPT statement is similar to a formatted
sequential READ statement except that an implicit logical unit number
is

always

used.

Formatted sequential input statements have the following forms:
READ

s])
[1list]
(u,f[,END=s][,ERR=

READ

f{,list]

f[,list]

ACCEPT
u

logical

format

number.

unit

specifier.

label

The

executable

statement.

list

I/0

list.

A statement of the following form
system-defined
200,

READ

logical

data

causes

read

from

ALPHA,BETA,GAMMA

transferred

Characters

unit:

formatted

statements

sequential

are

form specified by the format specifier.
internal
translated to the
The resulting values are assigned to the elements of the I/0 list.

If the number of list elements is less than the number of input record

fields,

the excess portion of the record is

ignored.

Usually a single formatted record is transferred by the execution of a
However, the format specifier
input statement.
formatted sequential
can specify that more than one record is to be read during execution
of

single

statement.

input

If the FORMAT statement associated with a

formatted

statement

input

contains a Hollerith constant, input data is read and stored directly
See Section
into the storage location of the format specification.
8.]—.9.

only
occurs
transfer
data
If no I/0 list is present,
record and the storage location of the format specifier.
READ

(5,100)

FORMAT

100

(15H DATA GOES

HERE)

These statements read 15 characters from the next
unit

REVIEW

The FORMAT
100

the

characters

the
between
For example:

are:

record

SECTIONS

statement becomes:
(1l5HREVIEW

FORMAT

SECTIONS)

Other examples of formatted sequential input statements follow:
READ

300

(1,300)

FORMAT

ARRAY

(20F8.2)

logical

INPUT/OUTPUT

These
to

statements

read

record

ARRAY.

READ

100

100,

from

logical

unit

and

assign

fields

ICOUNT,ALPHA,BETA

FORMAT

(I5,

F8.2,

These

statements

read

assign

fields

STATEMENTS

F5.,1)

record

integer

from

variable

BETA.

implicit

ICOUNT

and

1logical

real

wunit,

variables

ALPHA

and

WRITE
TYPE

PRINT
7.3.2

Formatted

Sequential

The
formatted
sequential
specified logical unit.

The

formatted

formatted
to

The

sequential

sequential
implicit logical

formatted
WRITE

TYPE

Output

Statements

WRITE

statement

TYPE

WRITE

and

statement,

unit.

sequential

output

transfers

statements
except

statements

are

that

have

the

data

similar

output

following

the

directed

forms:

(u,f[,ERR=s])[list]

f[,list]

PRINT f[,list]
u

logical

format

unit

number.

£
specifier.

The

label

executable

statement.

list
An

I/O0

list,

The

I/0 list specifies a sequence of
values
that
are
converted
to
characters and positioned as specified by the
format specifier.
1If no
I/0 list is present, data transfer occurs
only
between
the
storage
location of the format specifier and the record.
The

data

transferred

normally

constitutes

one

specify

single

Numeric

that

additional

output
data

conversion

records

under

external

calculations,

1loss

unformatted

are

format

format.

precision,

use

formatted
formatted

statement.

output
to

precision

output.

sequential

record.

written

control
such
may

data

output

However,

is
is

result.

statement

the

during

format

can

execution

during

the

rounded
input

for

avoid

additional
loss

INPUT/OUTPUT STATEMENTS

not exceed
The records transmitted by a formatted WRITE statement must
, a line
example
For
accept.
can
device
ed
the length that the specifi
characters.

printer typically cannot print a record longer than 132
Examples of formatted sequential output statements follow.
WRITE

FORMAT

650

(6,

650)

HELLO THERE')

the

These statements write

alphanumeric

FORMAT statement to logical unit 6.

literal

the

fields,

contained

WRITE (1,95) AYE,BEE,CEE
FORMAT (3F8.5)

These statements write one record,
unit

logical

WRITE (1,950)
FORMAT (F8.5)

950

consisting

three

AYE,BEE,CEE

These statements write three separate records, consisting of one field
each,

to logical unit

In the last example, the rightmost parenthesis of the FORMAT statement

is reached before all elements of the I/O list are output. Each time
1is
this occurs, the current record is terminated and a new record
more
a
For
written.
are
records
separate
Thus, three
initiated.

complete explanation, see Section 8.7.

READ

ACCEPT
List-Directed Input Statements

7.3.3

d
The list-directed READ statement transfers records from the specifie
and
form,
internal
to
external
from
data
the
logical unit, translates
assigns the resulting values to the elements of the 1/0 1list in the
order in which those elements appear, from left to right. The 1/0
list is required. 1I1f a list-directed READ statement does not 1include
a logical wunit number, an implicit logical unit number is used. The

list-directed ACCEPT statement is similar to a 1list-directed READ
statement except that an implicit logical unit number is always used.
The list-directed input statements have the following forms:
READ

(u,*[,END=s][,ERR=s])

READ

*,list

list

*,list

logical

unit

number.

Indicates list-directed formatting.

INPUT/OUTPUT

The

label

executable

STATEMENTS

Sstatement.

list
An

The

I/0

list,

external

record

separators.

The

constant

null

repetition

constants

repetition

null

input

constants

before
and

has

constant

after
F

not

permitted.
constant

for
A

arithmetic
null

value

represent

1is

The

form

r*c

permitted

except

value
®¢

One

comma,

®¢

slash

The

slash

value

and

record;

Each

input

exhausted

I/0
before
is

The

end

Spaces

(see

FORTRAN

the

value

(see

real

Section

integer

parentheses.
the

between

the

constant
octal

element

Spaces

first

constant,

second

constant

either

constants

and

its

two

between

the

list

element

part

commas

the

not

corresponding

according

consecutive

occur

are

rules

3-1).

either

complex

pair
and

performed

can

for

constant

a
in

and

list

with

commas.

remains

complex

null

unchanged.

constant,

but

constant.

where
r
constant.

specified
null

nonzero,

Spaces

are

not

above.

value

where

be:

spaces

with

will

all

1If

the
ignored.

beginning

tabs
surrounding

terminates

remaining

list.

without

separator

record

and

corresponding

used

logical

Table

Spaces

the

can

statement

record
of

r occurrences
<constant
and
¢
within the constant

all

the

and

values.

form

comma,

r occurrences
constant.

separator

the

values

r*c

form

enclosed

indicates

integer

satisfy

types

specified

indicates

integer

parenthesis

conversion

entire

unsigned

The form
unsigned

the

form

the

Hollerith

data

that

has

opening

assignment

cannot

can

form

separating

constant.

these

comma-and

(false).

the

values

parenthesis.

the

specifies

null

match,

value

intervening
A

the

closing

describe

constant

between

and

(true)

paragraphs

separated

the

sequence

value

complex

occur

be:

2.3).
can

contain

can

following

Each

must

value

I/0

list

read

one

slash
values in

equivalent
of

processing
elements

separator
a

record

spaces

record

a
are

space

are

tabs

the

input

records

as
required
or the I/0 list
used, the remainder

occurs
are

character.

ignored.

statement

unchanged.

to
is
of

INPUT/OUTPUT STATEMENTS
follows.

An example

includes:

The program unit

PRECISION T

DOUBLE
COMPLEX
LOGICAL

READ

D,E
L,M

(1,*)

I,R,D,E,L,M,J,K,S,T,A,B

The external record contains:

(3, 2 )

(3.4E0,4.2¢0),

4 6.3

, t,F,,3*14.6 /

The following values are assigned to the I/O list elements:
1/0 List

Value

Element
HnRIOEHO=H

6.3

(3.4,4.2)
(3.0,2.0)
. TRUE.

.FALSE.
14
14.6

14.6DO

and J will be unchanged.

WRITE
TYPE
PRINT

7.3.4

List-Directed Output Statements

The list-directed WRITE statement transfers the elements in the 1/0
list to the specified logical unit, translating and editing each value
according to the data type of the value.

The list-directed TYPE and PRINT statements are similar to the
list-directed WRITE statement, except that output is directed to an
implicit

logical

unit.

List-directed output statements have the following forms:
WRITE
TYPE
PRINT

(u,*[,ERR=s])
*,list
*,list

list

INPUT/OUTPUT

STATEMENTS

logical

unit

number.

Indicates

list-directed

The

formatting.

label

executable

statement.

list
An

I/0

list.

The

values in the I/0 list are converted to character
form and written
a fixed format according to the data type of the value.
Table 7-1
lists the output
formats
for
each
data
type.
The
I/0
1list
is
required.

Table

Data

7-1:

List-Directed

Type

Output

LOGICAL*1

Format

INTEGER*?2

INTEGER*4

I12

REAL*4

1PG1l5.7

REAL*8

1PG25.16

COMPLEX*8

IX,'(',1PGl4.7,

Hollerith

1X,An

',"',

the

Hollerith

List-directed

output

values,

slash

literals

separators,

the

Hollerith
Each

statements

1I/0

record

statement

1list

Note

may

not

produce
forms

are

not

input

output

using

begins

with

space

writes

one

output

1list-directed

8.7).

for

Some

examples
PRINT

TYPE

octal

values,

values.

without

delimiting
that

single

record.

unit

consists

(see

record,

of:

A(5)

A/5*3.4/

WRITE

(1,*)

'ARRAYAVALUESAFOLLOW'

WRITE

(1,*)

A,5

the

following

records

will

contain

formatting.

control.

records

null

Alphanumeric

records

carriage

complete

the

follow:

DIMENSION

then

constant)

*,
'THEAARRAYAZAIS',Z
*,'THEAANSWERAIS', (I,XX(I),I=1,10)

program

DATA

1pGl4.7,")"

length

list-directed

Each output value is contained within
Hollerith constants that are longer than

for

repeated

itself

that

constants

output

’

LOGICAL*4

apostrophes.

Formats

output:

ARRAYAVALUESAFOLLOW

AAA3 . 400000AAAAAAAZ . 400000AAAAAAA3 . 400000AAAAAAAS . 400000
ANAASZ s 40000 0AAAAAAAAAND

Each
Section

except

INPUT/OUTPUT STATEMENTS

READ
Unformatted Sequential Input Statement

7.3.5

The unformatted sequential READ statement transfers one unformatted
record from the specified logical unit, and assigns the untranslated
fields of the record to the 1/0 list elements in the order in which
The data type of each element
from left to right.
they appear,
the amount of data input to the element.

determines

The unformatted sequential READ statement has the following form:
[list]
(u[,END=s][,ERR=s])

READ
u

logical

number.

unit

label

The

executable

statement.

list
an

I/0

list.

An unformatted sequential READ statement reads exactly one record,
If the I/0 list does not use all the values in the record (that is,
the
there are more values in the record than elements in the list),
list
of
number
the
If
discarded.
1is
record
the
of
remainder
values

elements is greater than the number of
error

1in ,the

record,

occurs.

I/0

list,

The unformatted sequential READ statement must only be used to

read

If an unformatted sequential READ statement contains

one

full

record

skipped.

records created by unformatted sequential WRITE statements.
Examples of unformatted sequential
READ

(1)

FIELDl,

input statements follow.

FIELD2

This statement reads one record from
values

to variables FIELD1l and

READ

FIELD2.

logical

wunit

and

assigns

(8)

This statement advances

logical unit 8 by one record.

WRITE
7.3.6

Unformatted Sequential Output Statements

the
transfers
statement
WRITE
sequential
unformatted
The
untranslated values of the elements in the I/0 list to the specified
logical unit as

one unformatted

record.

The unformatted sequential WRITE statement has the following form:
WRITE

(ul,ERR=s])[list]

INPUT/OUTPUT

STATEMENTS

logical

unit

number.

]

The

label

executable

statement.

list
An
If

list.

unformatted

record
Some

I/0

output

examples

WRITE

This

statement

LIST

contains

unit.

I/0

1list,

one

null

outputs

the

unit

writes

contents

elements

through

array

logical

unit

(4)

statement

7.4

DIRECT

ACCESS

access

I/0

record

statement

contains

INPUT/OUTPUT

and

establishes

null

statements

numbers,

9.1)

statement

specified

(LIST(K),K=1,5)

logical

This

Direct

the

follow.

(1)

WRITE

from
a

record

the

transfer

direct

records,

access

files.

number.

attributes

The

the

specified
Each

OPEN

direct

statement

access

record

access

I/0

(Section

file.

Unformatted direct access I/0 statements transfer records of
binary
data
without
translation.
The DEFINE FILE statement (Section 9.9)
may be used to specify the attributes of the direct access file.

READ
7.4.1
The

Unformatted

unformatted

record

from

assigns

the

Direct

Access

direct

access

file

currently

the

untranslated

Input

READ

fields

Statement

statement

connected

the

transfers
the

record

the

specified

the

specified
unit,

and

I/0

list

following

form:

elements.

The

unformatted
READ

direct

access

(u'r[,ERR=s])

[list]

READ

statement

logical

unit

number.

The

record

The

label

number.

list
An

I/0

list.

executable

statement.

has

the

INPUT/OUTPUT

STATEMENTS

is,
If the I/0 list does not use all the fields in the record (that
there are more fields in the record than elements in the list), the
1list
of
number
the
If
1is discarded.
record
remainder of the
the

than

greater

elements

number

of record fields, an error

occurs.

Examples of unformatted direct access
(1'10)

READ

input statements

LIST(1l),LIST(8)

This statement reads record 10 of a file
2

assigns

follow.

logical

wunit

integer values to specified elements of array LIST.

READ

(4'IREC)

(RHO(N) ,N=1,5)

This statement reads the record specified by the value of
file on

logical

and

unit

and assigns

real values

IREC of

to array RHO.

WRITE
Unformatted Direct Access Output Statement

7.4.2

the
transfers
statement
WRITE
access
direct
unformatted
The
untranslated values of the elements in the I/0 list to the specified
record of the file currently connected to the specified unit.
WRITE

The unformatted direct access

statement

the

has

following

Form:

(u'r[,ERR=s])

WRITE

[list]

logical

unit

number.

The

record

number.

executable

I/0

statement

label.

list

list.

the record,

the

into the record,

new

If the values specified by the I/0 list do not fill
unused portion of the record is filled with zeroes.
If

the

error

I/0 list specifies more data than can

fit

occurs.

You can use a WRITE statement either
update

an existing

record.

to write a

record

Example:

WRITE

(2'35)

(NUM(K),K=1,10)

This statement outputs 10
connected

logical

unit

integer values to record 35

the

file

INPUT/OUTPUT

STATEMENTS

ENCODE
DECODE
7.5

ENCODE

AND

DECODE

STATEMENTS

The ENCODE and DECODE statements transfer data according
to
format
specifiers,
translating
the
data from internal to character form,

and

vice

versa.

however,

these

arrays

The

ENCODE

Unlike
data

the

and

conventional

transfers

FORTRAN

DECODE

take

formatted

place

1I/0

statements,

entirely

between

following

forms:

variables

program.

statements

have

the

ENCODE

ENCODE (c,f,b[,ERR=s])
[list]

DECODE (¢, f,b[,ERR=s])
[list]

integer

number
form.
to

1In

format

The

DECODE

(bytes)

statement,
to

formatted

specifier.

internal

statement,

translated

the

number

form.

¢
to

1is

the

character
characters

corresponds

the

specified,

record.

TIf

than

one

record

occurs.

name

the

characters

translated

length

error

expression.

formatted

characters

array,

array

record.

after

translation

statement,
it
internal form,

contains

The

executable

(see

Section

label

element

the
to

the

ENCODE

variable.

statement,

external

form.

characters

corresponds

receives

the

DECODE

translated

statement.

list
An

I/O0

I/0
In

list

1list
the

contains

DECODE

translation

"and

to a WRITE
elements
to

stores

the

transmitted,

7.2.6).

data

statement,

Similar

list

the

internal

the

characters

remaining

the

ENCODE

translated

1list

statement,

receives

character
the

the

form.

data

after

form.

statement, the
character form

the

1In

ENCODE
statement
translates
the
according to the format specifier,

fewer

character

than

positions

characters
are

filled

are
with

spaces.

Similar

character

format
If

and

array,

stores

its

array

number

The

virtual

characters
the

that

are

processed

buffer

array

element

the

data

ENCODE

depends

example,
element,

since an INTEGER*2
array
the
maximum
number
of

elements

array.

data

record

process

that

this

elements

prograssion.

virtual

The

statement, the
DECODE
statement
b
to
internal
(binary)
form

specifier,

subscript
a

READ

data

type

the

elements

must

(see
or
b

translates
to

according

the

not

Section

DECODE
in

can
contain
characters is

that

the

1list.

order
the

the
the

name

of
of

5.5).

statement

can

statement.

For

2
characters
per
twice the number of

INPUT/OUTPUT

STATEMENTS

The interaction between the format specifier and the I/0
1list
Section
8.7)
for an ENCODE or DECODE statement is the same as
for

formatted

example

I/0 statement.

the

DOUBLE

ENCODE

PRECISION

INTEGER*2

DATA

DECODE

INBUF,

statements

follows.

OUTBUF

A,B,C,D

INBUF/'12345678'/

DECODE
100

and

(see
that

ENCODE
FORMAT

(8,100,INBUF) A,B,C,D
(8,100,0UTBUF) D,C,B,A
(412)

The DECODE statement translates the 8 characters in INBUF to integer
form
(specified
by
statement 100), and stores them in the integer
variables

A,B,C,D,

=12

follows:

The ENCODE statement translates the values D,C,B,A to character
and stores the characters in the variable OUTBUF, as follows:
OUTBUF

'78563412'

form

CHAPTER
FORMAT

STATEMENTS

FORMAT statements are nonexecutable statements used with formatted I/0
statements,
and
with
ENCODE
and
DECODE statements, to specify the
editing and formatting of the data.
If input is being performed or
a
DECODE statement is being executed, the format statement describes the
format in which the input data is interpreted.
If
output
1is
being
performed
or
an
ENCODE statement
1is
being
executed,
the format
statement describes the format
in
which
the
output
data
will
be
represented.

Throughout this chapter a distinction is made between
"external"
and
"internal"
form.
External
form refers to the ASCII characters in a
data field of a formatted record.
Internal form refers to the
binary
representation of a data value.
FORMAT

statements

have

the

following

FORMAT

(glflslf2s2

...

£fnqgn)

Zero

slash

(/)

record

descriptor

form:

terminators.

field

group of

field

descriptors

enclosed

parentheses.
s

field

separator.

The entire list of field
terminators,
including
specification.
The list

descriptors,
field
separators,
the
parentheses,
1is
called
must be enclosed in parentheses.

The field separators are comma and slash.
terminator.
Section
8.5
describes
in
field

record
format

also
a
functions

record
of the

separators.

field

descriptor

has

the

following

form:

[rlclwl.d]]

field

descriptor

code

descriptor
1is
to
it is assumed to be

be
1,

(I1,0,F,E,D,G,L,A,H,X,T,P,Q,S,

oOF

repeated

(X3

the
field
you omit ¥,

The number of times
(repeat count).
If

—

A slash is
detail the

and
the

FORMAT

STATEMENTS

The

external

The

number

field

width.

characters

The

terms

and

must

less

and

The

term

optional;

descriptors.
The
invalid in others.
The

field

must

descriptors
--

Iw,

Logical

Real,

Character

Editing,

and

8.1

first

control

equal

right

unsigned
to

however,

255;

you

required

the

decimal

integer
and

constants.

and

cannot

use

field

some

point.

must

nonzero.

some

field

descriptors

and

precision,

and

complex

Fw.d,

Ew.d,

nH,

'...',

nX,

character

Dw.d,

Gw.d

Aw
Hollerith
a

number

describes

character

the

are:

Integer

double

all

term

Section
The

than

characters

each

field

output

record

See

Section

8.3

information.

constant

descriptor

nP,

detail.

generally

for

TTMn,

positions)

contains

carriage

information.

During

data transfers, the format specification is scanned
from
1left
to
right.
The
elements
1in the I/0 list are correlated one-for-one
with the corresponding field descriptors.
However,
the
editing
and
Hollerith
constant
field
descriptors
do
not
require
an I/0 list
element.
Section 8.7 describes
in
detail
the
interaction
between
format
You

specifiers

use

data.

You

double

and
or

use

precision,

Section

8.8

You

create

can

format

the

field

and

complex

format

the

list.

descriptor

summarizes

instead

I/0

process

field

integer

descriptor

and

1logical

process

real,

data.

rules

during

FORMAT

for

writing

program

FORMAT

execution

Section

8.6

statement.

statements.
using

describes

run-time

formats.

8.1

FIELD

field

DESCRIPTORS

descriptor

several

data

an external
descriptors
the

field

has

the

The
detail.

field,

Field

The

each

field.

external

8.1.1

describes

items;

but

the

size

and

data

item

following
The field
treat

format
the

external

data

item

medium

called

sections describe each of the
field
descriptors ignore leading spaces in

embedded

and

trailing

spaces

zeroes.

Descriptor

descriptor
following

specifies

form:

transfer

decimal

integer

values.

FORMAT

The

corresponding I/0
logical data type.

Rules

in
@

Effect
The

the

value,
The

the

minus
e

data

the

first

sign,

the

specifies that w
field,
interpreted

assigned

must

first

data

integer

exponent

value

either

integer

the

character
the

I/O

the

element.
it

maximum

element,

the

as
is

list

constant;

exceeds

treated

field,

characters
are
read
as a decimal integer

cannot

field.

list

character

field

nonblank

corresponding

nonblank

the

point

external

appears

Input:

decimal

magnitude

must

descriptor

external

element

external

and

contain
e

Data

field

from

for

list

STATEMENTS

a
a

external

negative

occurs.

field

1is

value.

plus

sign,

1is

treated

field

allowed

error

sign

positive

value.
@
Input

all-blank

field

treated

value

Example:

Format

Rules

Internal

Value

2788

-26

ANANNAN3 L2

2
Effect

The

for

Data

field

external
If

the

2008

Output:

I/0

field

value

312

descriptor

corresponding

Field

in
@

External

specifies

1list

characters

does

not

output

element,

fill

long,
the

the

right

value

Jjustified,

the

decimal integer.

field,

1leading

spaces

are

inserted.
e

the

filled
e

the

have

®
Output

value

exceeds

with

asterisks.

value

minus

the

list

sign

term

w must

therefore

sign,

when

necessary.

Plus

signs

are

field

width,

element

its

entire

negative,

leftmost,

large

the

enough

nonblank
to

the

field

character.

provide

for

suppressed.

Example:

Format

Internal

Value

External

Representation

284

-284

174

3244

174
*x

-473

29,812

* k%

Not

permitted:

error

will

The
minus

FORMAT

8.1.2

O Field

The

has

the

field

STATEMENTS

Descriptor

descriptor

following

specifies

transfer

octal

integer

values.

either

integer

form:

The corresponding I/0
logical data type.

Rules

in
@

Effect
The

for

list

Data

field

element

must

Input:

descriptor

specifies

that

characters

are

read

value,

and

through

from
the
external
field, interpreted as an octal
assigned to the corresponding I/0 list element,
®¢

The

external

it
cannot
field.
@
¢

all-blank

field

the

the
Input

field

can

contain

value

the

corresponding

contain

only

sign,

treated

external

list

numerals

decimal

point,

data

element,

the

value
exceeds

error

exponent

the

allowed

size

occurs.

Example:

Format

Rules

77777

31274

1623

15AAAAN

53248

974

Effect
The

the

Internal
Decimal Value

Field

in
¢

External

for

Data

field

Not

descriptor

external

field

are

octal

(2's

value

the

permitted:

error

Output:

corresponding

signs

32767

specifies

1I/0

1list

characters

output;

complement)
does

not

output

element,

long,

negative

value

the

octal

value

right

justified,

octal

integer.

transmitted

of
an

its

spaces

are

form.
fill

the

field,

leading

inserted.
e

the

filled
Output

value

exceeds

with

asterisks.

the

field

width,

the

entire

field

Example:

Format

Internal

(Decimal)

32767

-32767

02
04

14261

Value

External (octal)
Representation
ATTT777

100001
* *

13.52

AA33

Not

permitted:

error

1is

FORMAT

8.1.3

F Field

STATEMENTS

Descriptor

The F field descriptor
specifies
the
precision values.
It has the following

transfer
form:

real

double

Fw.d

The corresponding I/0 list
precision
data
type;
or
part

Rules

complex

Effect

The

data

for

field

from
the
precision

element
it must
type.

Data

must be of either
be either the real

real
or
double
or the imaginary

Input:

descriptor

external
value, and

specifies

that

characters

are

field,
interpreted
as
a real
assigned to the
corresponding

read

or double
I/0
1list

element.
Any decimal point, signs, or exponent field present
in the external field are included in the w count
and
d
is
part

If
the

the

characters

point

1is

include

used.

decimal
point,
the
rightmost d digits of

decimal

the

decimal
w.

point,

the

characters

point

1is

placed

position
not

include

before

the

If the w characters include an
exponent
field
(see
Section
2.3.2
for
real
constants
and
Section
2.3.3
for
double
precision constants), the exponent is
used
to
evaluate
the
number's
magnitude
before
the
decimal
point
position
is
determined.

If the first nonblank
minus sign, the field

If the first
appears
in

character of the
external
field
is treated as a negative value.

nonblank character is a
the
field,
the
field

plus sign, or if
1is treated as a

1is

no
sign
positive

value.

w must

Input

all-blank
be

field

greater

treated

than

equal

value

d+1l.

Examples:

Format

Rules

External

Field

Internal

F8.5
F8.5

-1234.567

~-1234,56

F8.5

24,77E+2

2477.0

F5.2

1234567.89

Effect

123456789

Value

for

Data

123.45678

123.45

Output:

The
F
field
descriptor
transfers
the
value
of
the
corresponding I/0 list element, rounded to d decimal positions
and right justified, to an external field w characters 1long.

If the value
inserted.

If the
filled

does

not

fill

value exceeds the
with asterisks.

the

field

field,

width,

leading

the

entire

spaces

field

are

FORMAT

Plus

signs

w must
width

are

suppressed.

be greater
should
be

after the point,
digits
to
the
negative
Output

STATEMENTS

than or equal to d+l;
however,
the
field
large enough to contain the number of digits
plus
1left

for the
of
the

point,
point,

plus
plus

the
for

number
of
a possible

sign.

Examples::

Format

Internal Value

F8.5

External

2.3547188

F9.3

8789.7361

F2.1

51.44

Representation

N2.35472
N8789.736
* %

F10.4

-23.24352

AN-23.2435

F5.2

325.013

*hk kK

F5.2

-.2

-0.20

8.1.4

E Field

Descriptor

The E field descriptor specifies transfer of real
values in exponential form.
It has the following

or double
form:

precision

Ew.d

The

corresponding

I/0

list

data

type;

complex

data

type.

precision
part

Rules

Effect

On input, the
descriptor.
Input

for

element

must

either

either
the

real

the

double

imaginary

Input:

field

descriptor

does

not

differ

from

the

field

Examples:

Format

External
734.432E3

AN1022,43E-6

0.102243E-2

E15.3
E12.5

52.3759663 AAAAA
210.5271D+10

0.523759E+2
0.2105271E+13

precision.

Rules

Effect

for

are
If

right
the

0.734432E+6

that

the

E field
indicator

descriptor

the

I/0

treats
the
list element

D
is

Output:

The
E
field
corresponding

and

Value

E12.4

single

Internal

E9.3

In the last example, note
exponent
field indicator

Field

descriptor

I/0

justified,

value

does

transfers
the
element, rounded to

1list
to

not

fill

external
the

field

value
of
the
d decimal digits

characters

characters,

leading

1long.
spaces

inserted.
the

filled

value

exceeds

with

asterisks.

the

characters,

the

entire

field

1is

FORMAT

Output

standard

STATEMENTS

form:

that

is,

value is negative, an optional 0, a
the right of the decimal point, and
one

the

following

two

minus

sign

the

decimal point, 4 digits
a 4-character exponent

to
in

forms:

E+nn
E-nn
nn

A
®

Plus

2-digit
signs

must

width

integer

are

suppressed.

greater

must

constant.

not

than

stated

after the point, plus
the
point,
plus one
the
Output

Internal

for
a

for

the

d+7;

that

1is,

the

field

the

number

digits

than

point, plus 1
negative

for

possible

the 0 before
plus 4 for

sign,

Value

External

Representation

E9.2

475867.222

N0 .48E+06

E12.5

475867.222

A0,47587E+06

0.00069

E10.3
E5.3

less

Examples:

E12.3

The

exponent.

Format

8.1.5

equal

Field

field

AAND
L 690E-03

-0.5555
56.12

-0.556E+00
kkkkk

Descriptor

descriptor

specifies

transfer

form

values
in
exponential
following form:

with

real

double

instead

precision

real

has

the

Dw.d

The

corresponding

precision
part of a
Rules

Effect

On input, the
descriptors.
Input

I/0

data
type;
complex data

for

list

element

or it
type.

must

must
be

either

either
the

real

the

differ

from

the

double

imaginary

Input:

field

descriptor

does

not

field

Examples:
Format

External

Field

Internal

Value

D10.2

12345AA00A

0.1234500000D+8

D10.2

AA123.45AN

0.1234500000D+3

D15.3

367.4981763D-04

0.3674981763D-1

Rules

Effect

for

Input:

There

only

one

difference

output:
If
of the letter

you
E.

use

the

between
descriptor,

the
the

and

letter

E
D

descriptors
is

output

instead

FORMAT

Output

STATEMENTS

Examples:

Format

Internal Value

D14.3
D23.12

5413.87625793

0.0363

D9.6

8.1.6

External

AAANAND «363D-01
AANAAND .541387625793D+04

1.2

G Field

Value

khkkkhkkxkhkkkk

Descriptor

The G field descriptor specifies transfer of real or double
precision
values,
combining
Eor F-type formats according to the size of the
number being output.
It has the following form:
Gw.d

The corresponding I/0 list
precision
data
type;
or
part

Rules

complex

Effect

On input, the
descriptors.
Rules

Effect

data

for

element
it must

must be of either
be either the real

real
oxr
double
or the imaginary

type.

Input:

field

for

descriptor

does

not

differ

from

the

Output:

The
G
field
descriptor
transfers
the
value
of
the
corresponding I/0 list element, rounded to d decimal positions
and right justified, to an external field w characters long.
The

form

magnitude

Table

which

the

value,

8-1:

Data

Effect

Magnitude

value

written

described

Data Magnitude

Effective

function

Table

the

8-1.

Formats

Conversion

0.1

Ew.d

0.1

<m¢<«<K

1,0

1.0

10.0

F(w-4).(d-1),

10d-2

104d-1

F(w=4).1,

'aAnnn!

10d-1

10d

F(w=4).0,

'aAnnA!'

10d

F(w=4).d,

NOTE:
The 'aAaaA'
in the second column of
four spaces are to follow the numeric data

'AAAA!
'anaa’

Ew.d

Table
8-1
specifies
representation.

that

FORMAT

Plus

w must be greater than or equal
width
must not be stated to be
after the point, plus 1 for the
before
the point, plus 1 for a
for a possible exponent.

Output

signs

are

suppressed.
to d+7;
that
1is,
the
field
less than the number of digits
point, plus 1 for a possible 0
possible negative sign, plus 4

Examples:

Format

Internal

G1l3.6
G1l3.6
G1l3.6
G13.6

Value

External

Representation

0.01234567
-0.12345678
1.23456789
12,.34567890

A0.123457E-01
-0.123457A00N
AAY . 23457A0AN
AAL2,3457 ALAAA

Gl3.6

123.45678901

AN123,457 000N

Gl3.6

-1234.56789012

AN=1234,.57A00N

Gl3.6

12345.67890123

ANL12345.7AMAN

Gl3.6

123456.78901234

AAL23457 .AAAAN

Gl3.6

-1234567.89012345

-0.123457E+07

Compare the
same values

above examples
output with an

Format

with the following
equivalent F field

Internal

Value

examples, which
descriptor.

External

0.01234567

F13.6

-0.12345678

ADAN-0.123457

F13.6

1.23456789

AANANL
. 234568

12,34567890

ANANL2 345679

F13.6

123.45678901

AAN123.456789

F13.6
F13.6
F13.6

-1234.,56789012
12345,.67890123
123456.,78901234

A=-1234,567890
A12345.678901
123456.789012

F13.6

-1234567.89012345

field

following

the

ANAAND 012346

F13.6

L Field

show

Representation

F13.6

8.1.7
The

STATEMENTS

khkkkkkhhkhhxk

Descriptor

descriptor

specifies

transfer

logical

data.

has

the

either

integer

characters

are

form:

The

corresponding

logical
Rules

data

I/0

list

element

must

Effect

for

Input:

The L field descriptor specifies
from the external field.

If
or

the
t,

list

type.

that

first nonblank character of the field
the value .TRUE.
1is assigned to the

read

is the
letter
T
corresponding I/0

element.,

If the first nonblank character of the field
or
f,
or if the entire field is blank, the
assigned.

Any

other

value

the

external

field

is the
letter
value .FALSE.

produces

error.

F
1is

FORMAT

Rules

in
¢

Effect
The
(1f

for

field descriptor specifies output
the
value
of
the
corresponding

.TRUE.),

field

letter

preceded
Output

Output:

external
The

STATEMENTS

the

letter

w-1

(if

characters
is

the

of either the letter T
I/0
1list
element is

value

1is

.FALSE.)

long.

the

rightmost

position

the

field,

spaces.

Examples:

Format

8.1.8
The

the

Value

External

Representation

. TRUE.

ANANAAT

L1l

.FALSE,

has

Internal

Field

field

Descriptor

descriptor

following

specifies

the

transfer

Hollerith

values.

type,
data.

since

you

characters

from

the

corresponding

I/0

list

form:

The
can

corresponding
use variables

The

value

Rules

in
¢

must

Effect
The

I/0 list element
of any data type
be

for

than

equal

255.

Input:

field

external

less

can be of any data
to store Hollerith

descriptor

record

and

transfers

assigns

them

w
the

element.

The

maximum

number

the

size

the

I/0

characters

list

List

Maximum

Element

LOGICAL*1

LOGICAL*4

INTEGER*?2

INTEGER*4

REAL

REAL*S8

If
W

characters

less

characters

and

stored

depends

Number

the

stored

excess

greater than
1in
the
rightmost characters

can

follows:

8
PRECISION

COMPLEX

Characters

BYTE

DOUBLE

that

element,

trailing

than
are

maximum

number of characters
1I/0 list element,
assigned to that
element.

<corresponding
are

are

ignored.

the

number

assigned

spaces

are

characters

than

the

list

element,

added

fill

the

can
left

element.

that

can
the
Leftmost
only

stored,

justified,

FORMAT

Input

STATEMENTS

Examples:

Format

External Field

A6
A6
A6
A6

PAGEA#
PAGEA#
PAGEA#
PAGEA#

Rules

Effect

for

Internal Value

#
A#
GEA#
PAGEA#AA

(LOGICAL*1)
(INTEGER*2)
(REAL)
(DOUBLE PRECISION)

Output:

The A field descriptor specifies output of
corresponding
I/0
1list
element
to
an
characters

the contents of the
external
field
w

long.

If w is greater than the size of the list
element,
the
data
appears in the field, right justified, with leading spaces.

If w is

less

than the size

leftmost w characters

are

the

1list

transferred.

element,

only

the

Examples:

Output

Format

Internal Value

External Representation

A5
A5
A5

OHMS
VOLTSAAA
AMPERES A

AQHMS
VOLTS
AMPER

H Field Descriptor

8.1.9

The H field descriptor specifies transfer of data between the external
record
and the storage location of the H field descriptor itself.
It
has

the

form

nHclc2c3

The

Hollerith
¢n

...

number

constant:

characters

transferred.

ci
An

ASCII

Rules

Effect

for

Input:

The H field descriptor specifies acceptance
of
n
characters
from
the
external
field
and
their
assignment to the same
storage location as the characters of the H descriptor,
which
are overlaid by the input data, character for character.
in

Rules

character.

Effect

the H field descriptor

TYPE

200

Output:

The H field descriptor specifies output of
the
following the letter H to the external field.

An example of
100

for

FORMAT

characters

usage follows.

100

(41HAENTERAPROGRAMATITLE,AUPATOA20ACHARACTERS)

200

FORMAT

(20HATITLEAGOESAHEREAAA)

FORMAT

The

TYPE

statement

descriptor
statement
data
in

spaces
an

words
the

the

TITLE GOES
remainder

pair

For

FORMAT

The

characters

the

terminal.

the keyboard,
statement 200.

HERE.
1If
of
the H
field

from

user's

placing
The new

the user enters
field descriptor

descriptor

may

also

field

The

the
input
characters

1less
than
20
is filled with

specified

literal.

alphanumeric

apostrophes.

the

100

response from
descriptor in

right.

alphanumeric

In an

transfers

statement

accepts the
the H field

replace the
characters,

STATEMENTS

1literal,

the

apostrophe

written

two

example:

('TODAY''SADATEAIS:A',I2,'/',12,'/',12)

apostrophes

wused

1in

this

way

1is

considered

single

character.

8.1.10
The
the

Field

Descriptor

X field descriptor
following form:

specifies

skipping

character

positions.

has

The
The

term n
value

specifies how many character
of
n
must be greater than

equal

255.

Rules

Effect

Rules

The

the

input

in
¢

for

field

Effect
The

field

external
WRITE
90
The

WRITE

where nn
numeral

NUMBER

is
1

are

specifies

The

descriptor
For

field

the

character

specifies

output

spaces

the

example:

NPAGE

‘

(13H1PAGEANUMBERA,I2,16X,23HGRAPHICAANALYSIS,ACONT.)
prints

record

similar

to:

GRAPHIC

the current value of
1in the first H field

Field

that

skipped.

ANALYSIS,

CONT.

the variable NPAGE.
Note
that
descriptor is not printed but is

to advance the printer paper to the top
describes printer carriage control.)

8.1.11

be
skipped.
less than or

Output:

record.

statement

PAGE

descriptor

(6,90)

FORMAT

to
and

Input:

record

for

positions are
or equal to 1

new

the
used

page.

(Section

8.3

the

character

Descriptor

descriptor

be
treated
relative
following form:

specifies
to

the

start

position

the

external

record.

has

to
the

The

term

character

but

record.

indicates
to

not

the

treated.

greater

than

position

The

value

the

number

the
of

external
must

record
greater

characters

of
than

allowed

the

equal

1in

the

FORMAT

Rules

Effect

In an
data
For

for

STATEMENTS

Input:

input statement, the T field descriptor
specifies
that
1is to be input starting with the nth character position.

example:

FORMAT (T7,A3,T1,A3)
READ (5,10) J,K

In the example, a 3-character string starting at character position
7
in the external record is read first, followed by a 3-character string
starting
at
character
position
1;
however,
any
order
can
be
specified.
Rules

Effect

for

Output:

In an output statement, the T field descriptor specifies
that
the
data
1is
to
be
output
starting
at
the nth character
position of the external record.
For example:
PRINT

FORMAT

(T50,'COLUMNA2',T20,'COLUMNAL"')

These statements will print
position 50.
The remainder

8.1.12

"COLUMN 1" at position 20 and COLUMN 2"
of the line contains blank characters.

Q Field Descriptor

The Q field
descriptor
specifies
assignment
to
the
corresponding
variable
in
the
1I/0
1list
of the number of characters in the input
record remaining to be transferred during a READ
operation.
It
has
the

following

form:

The corresponding

I/0 list element must be of

integer or

logical

data

type.

For

example:

1000

READ (4,1000) XRAY,KK,NCHRS, (ICHR(I),I=1,NCHRS)
FORMAT (El15.7,I4,Q,80Al)

These input statements read two fields into the variables XRAY and KK.
The
number
of
characters remaining in the record is stored in NCHRS
and exactly that many characters are read into
the
array
ICHR.
By
placing
the
Q
descriptor first in the format specification, you can
determine the actual length of the input record.

In an output statement, the Q field descriptor has
that the corresponding I/0 list element is skipped.

8.1.13

Dollar

effect

except

Sign Descriptor

The dollar sign character ($) used as a field
descriptor
suppresses,
on
output,
a
carriage
return at the end of the line when the first
character of the line is a space or a plus sign (see
Section
8.3
on

carriage control characters).

In an

input statement,

the $§ descriptor

FORMAT

ignored.

1/0;

The

descriptor

(rather

leaves
the
than returning

will

Thus,

the

200
produce

intended

terminal
it to the

output

the

primarily

for

interactive

print position at the end of
left margin)
so that a typed

same

the text
response

line.

statements
TYPE

100

STATEMENTS

100

FORMAT

('AENTERARADIUSAVALUEA',S)

200,X

FORMAT

(F6.2)

message

the

terminal

the

following

form:

ENTERARADIUSAVALUE

Your

response

(in

this

case,

12.)

can

then

follows:

the

same

1line,

ENTERARADIUSAVALUEALZ2,

Note

that

same

result.

used

carriage

following

two

(11HASIGNAHERE:,S)

200

FORMAT

(l1ll1HS$SSIGNAHERE:)

Colon

colon

list

accomplishes

the

equivalent:

(:)

used

items

are

the

remain.

For

100,3

200,4

field

descriptor

I/0

list.

The

terminates
:

has

format

effect

example:

FORMAT ('AlI=',I2,

'AJ=',I2)
FORMAT ('AK=',I2,:,'AL=',12)

200

These

items

100

character

are

Descriptor

character

control

formats

FORMAT

control

I/0

The

200

8.1.14
The

the

statements

the

following

two

lines:

detail.

I=A3AJ=

K=n4

Section

8.7

8.1.15

Complex

complex

describes

value

output

Data
is

descriptors,

format

Editing

ordered

complex

using

control

any

pair
value

combination

of
1is

real

values.

governed

the

forms

Gw.d.
Rules

in
®

Effect
In

for

input

assigned
imaginary

Therefore,
by

Fw.d,

two

real

Ew.d,

input
field

Dw.d,

Input:
statement,

a
parts,

the

complex

two

I/0
respectively.

successive

1list

fields

element

are

1its

read

real

and

FORMAT

Input

STATEMENTS

Examples:
Format

External

Field

Internal

Value

F8.5,F8.5

1234567812345.67

123.45678,

12345.67

E9.1,F9.3

734,432E8123456789

734.432E8,

123456.789

Rules

Effect

for

Output:

output

statement,

transferred
descriptors.
e

under

The
two
parts
punctuation
or

the

two

control

parts

repeated

complex

value

successive

are

field

are
transferred
consecutively,
without
spacing,
unless
the format specifier states

otherwise.
Output

Examples:

Format

Internal Value

2F8.5
E9.2,'A,A",E5.3

3.456732
56.123

Representation

A2,35472 3.45673
AQ.48E+06AL ,A***x*k

Scale Factor

8.1.16

A scale factor
the
location
complex
The

2.3547188,
47587.,222,

External

is
of

value used in a format
the
decimal
point
in

specifier which
determines
real, double precision, or

values.

scale

factor

has

the

form:

signed

+127.

or
It

that

the

Effect

Rules

unsigned
specifies

decimal
for

point

both

If you do
applies.

not

The

factor

scale

integer
the
is

Input

use

constant

number
to
and

scale

set

the

positions

range

-127

left

the

through
or

right

move.
Output:

factor,

default

the

scale

start

factor

every

1/0

statement.

A scale factor applies to all subsequent
descriptors, until a new scale factor is

The

scale

factor

can

appear

F, E, D, or
specified.

field

descriptor.

For

example:

In the
not to

FORMAT

(X,

example,
the E6.3

14,

E6.3,

the 3P
or the

3P,

2A3,

212,

2F5.3,

applies to the 2F5.3 and
X, I, or A descriptors.

E8.5)

the

E8.5

but

FORMAT

scale

field
10

factor

FORMAT

For

this

(X,

factor

Rules

external

effect.

the

external

toan

2F5.3

E3.2,

2PE4.2,

example,

.01,

moving

has

and

E8.5.

effect

two

F4.2,

records

scale

factor

the

scale

3PE4.2)

are

read,

reinstated

field

does

contain

the

not

corresponding

scale

factor

decimal

multiplies

two

exponent,

multiplication

only

record,

explicit

Input:
contains

the

point

reversion

second

format.

field

it
a

can

for

specifies

assignment

Input

the

Effect

the

decimal

8.7)

suppose

For

both

Section

F3.2,

has

factor

prefix

to
the start of the format.
1In the
scale factor 3P now applies to F3.2.

specification
Additional

E8.5)

applies

(see

example,

scale

example:

FORMAT

occurring
the active
A

reversion

factor.

For

(3P2F5.3,

example,

Format

appear

descriptor.

the

can

STATEMENTS

places

two

input

the

scale

exponent,

the

data

I/0

list

multiplies

point

the

places

value

the

scale

10**-n

and

element.
input

factor

the
100,

value

left.

moving

-2P
the

right.

Examples:

Format

External

Field

Internal

3PE10.5

AAN3T . 614N

3PE10.5

AN3T7 ,614E2

-3PE10.5

3761.4

AAAN3T 614

37614.0

Additional

Rules

Scale

Effect

factors

I/0

list

For

the

apply

variables
F

field

multiplied

Thus,

right;

for

.037614

Output:

only

not

data

output.

descriptor,

the

value

10**n

before

transfer

positive

scale

factor

moves

negative

scale

The

values

factor

the

I/0

the

list

For

the
the

subtracted

value

factor

moves

exponent;

the

left

and

field

(see

the

decimal

point

moves

the

decimal

descriptor,

the

basic

real

Section

2,3.2)

from

the

exponent.

the

decimal

point

multiplied

Thus,
the

negative

scale

factor

increases

the

exponent.

right

moves

the

element

external

left.

change.

Value

record.
to

point

the

constant
by

10**n,

positive
and

part
and

scale

decreases

decimal

the

point

the
to

FORMAT STATEMENTS

®¢

scale

factor

magnitude
range

has

the

effect

data

descriptor,

the

scale

data

value

factor

has

1is

output

field

descriptor

within

the

field

1if

the

effective
descriptor

function.
Moreover, the G field
field descriptor if the magnitude

outside

the

because

supplies
1its
own
scaling
descriptor functions as an E
of

same

its

range.

effect

1In
for

this

case,

the

field

the

descriptor.
Output

Examples:
Format

Internal

Value

External

1PE12.3

-270.139

AAN=-2,.701E+02

1PE12.2

-270.139

AAA=-2,70E+02

-1PE12.2

-270.139

AAN=-0.03E+04

8.1.17

Repeat

You

apply

can

number

Counts
most

and

field

successive

Group

Repeat

descriptors
data

fields

Counts

(except
by

an unsigned nonzero integer constant
repetitions.
This
constant
1is called a
following

two

statements

are

preceding

with
the

Representation

that

the
number
of
count.
For example,

repeat

equivalent:

FORMAT

(El12.4,El12.4,E12.4,15,15,15,15)

FORMAT

(3E12.4,415)

Similarly, you can apply a group of field
descriptors
data
fields
by
enclosing these field descriptors in
them

with

unsigned

nonzero

constant
1is
called a group repeat
two statements are equivalent:

For

repeatedly
parentheses

constant.
example,

FORMAT

(218,3(F8.3,E15.7))

FORMAT

(18,I8,€8.3,E15.7,€8.3,E15.3,€8.3,E15.Z)

field

be
enclosed
specification.
If

integer

count.

you

not

descriptor,

which

in
parentheses
Thus, it could be
specify

group

descriptor

specifying

preceding

field

integer

following

could

not

otherwise

and
treated
as
repeated a desired

repeat

The
the

to
and

count,

default

repeated,

a
group
number of
count

can

repeat
times.

assumed.

8.1.18

Default

Field

Descriptors

If you write the field descriptors I, O, L, F, E, D, G, or
A
specifying
a
w or w.d, default values are supplied based on
type of the I/0 list element.
Note that for F, E, D, and G,
not specify only w or d;
you must specify w.d or nothing.
Table

8-2

lists

the

default

values

for

and

without
the data
you
may

FORMAT

Table

Field

8-2:

STATEMENTS

Default

Descriptor

List

Element

Data

the

8.2

Type

INTEGER*2

I,
L

INTEGER*4
LOGICAL

REAL,

DOUBLE

Notice

Field Widths

COMPLEX

PRECISION

LOGICAL*1

BYTE

INTEGER*?2

LOGICAL*4, INTEGER*4

REAL,

DOUBLE

that

for

the

corresponding

CARRIAGE

field

1/0

CONTROL

list

COMPLEX
PRECISION

descriptor

the

default

the

1length

element.

CHARACTERS

The first character of every record transferred to a
printer
1is
not
printed.
Instead, it is interpreted as a carriage control character.
The FORTRAN I/0
system
recognizes
certain
characters
as
carriage
control
characters.
Table
8-3
1lists
these
characters
and their
effects.

Table

8-3:

Carriage Control

Character

Characters

Effect

(space)

Advances

one

line

(zero)

Advances

two

lines

(one)

Advances

(plus)

Does

advance

(dollar

sign)

not

Advances
suppresses

top

(allows

one
line
carriage

page
overprinting)

before
return
at

printing
the end of

and
the

record

Any character other than those
space
and
1is
deleted
from
accidentally omit the carriage

8.3

the

record

FORMAT

not

printed.

SPECIFICATION

Field descriptors
commas.
You can

listed in Table 8-3
1is
treated
as
a
the
print
line.
Note
that
1if
you
control character, the first
character

in a
also

SEPARATORS

format specification are
use the slash (/) record

generally separated by
terminator to separate

FORMAT

field

descriptors.

record

and

WRITE
40

This

statement

You

(6,40)

(306)

WRITE

(6,50)

multiple

However,
result

parentheses

initiator

The

the

are

the
or

(blank

line)

(blank
(blank

line)
line)

top

the

external

statement
external

This
with

record,

called

the

are
1,

The

above

the

end

(n-1l)

because

format

field

records

record;

record,

the

are

For

blank

current

blank

output

two

and

the

on.

specification

opening

and

closing

record

themselves

example:

following

output:

LINE
LINE

such

Fw.d

specifies

that

from

the

external

record.

contains

read
unless

1less

characters
you

than
from

padded

the

statement

Iis

data

the

characters,

the

short

input

next
field

data
with

field

the
field

leading

input
in

the

zeroes

the field descriptor is numeric, you can avoid having
field by using a comma to terminate the
field.
The
the
field
descriptor's
field width specification.

short

field

termination,

entering

data

(5,100)

FORMAT

the

SEPARATORS

record

would

you

records

page

characters

or spaces.
When
to pad the input
comma
overrides

when

skipped

respectively.

produce
of

FIELD

descriptor

read

input

between

('1l',T51,'HEADINGALINE'//T51,"'SUBHEADINGALINE'//)

50,

field

current

(6,99)

Column

specification

SUBHEADING

records
appear

terminates

beginning
blank

terminator,

input
slashes

skipped

first

format

statements

EXTERNAL

bypass

slash

the

HEADING

8.4

consecutive

first

FORMAT

above

skipped

WRITE

the

to:

slashes

and

N,O,P

terminates

slashes

1in

output

K,L,M

records

The

slash

example:

K,L,M,N,O0,P

(n-1l)

output.

input

For

(16,2F8.4)

records.

second

terminates
record.

equivalent

FORMAT

descriptors,
are

WRITE

use

new

(306/16,2F8.4)

FORMAT

can

blank

slash
a

(6,40)

FORMAT

initiates

STATEMENTS

from

and

terminal

field

descriptors.

I1,J,A,B

(216,2F10.2)

statements

read

the

following

particularly

keyboard.

record:

For

You

can

useful

use

example:

FORMAT

execution,
I

the

following

STATEMENTS

assignments

occur:

1.0

0.35

The physical

end

the

record

Note that the 4 part of a w.d
external
field
separator.
explicit decimal point in the

also

serves

field

terminator.

specification
is
not
affected
by
an
Therefore,
you should always include an
external field for F, E, D and
G
field

descriptions.

You can use a
long.
If
a
is considered

comma to terminate only fields less
than
w
characters
comma follows a field of w characters or more, the comma
part of the next field.

Two successive
commas,
or
a
comma
after
a
field
of
exactly
w
characters,
constitutes a null (zero-length) field.
Depending on the
field descriptor specified, the resulting value assigned is 0, 0.0, or
0.DO,

.FALSE..

You cannot use a comma to terminate a field that is controlled
by
an
A,
H,
or
alphanumeric
1literal
field
descriptor.
However, if the
record reaches its physical end before w characters
are
read,
short
field
termination
occurs;
and
the
characters that were input are
assigned successfully.
Trailing
spaces
are
appended
to
fill
the
corresponding I/0 list element or the field descriptor.

8.5

RUN-TIME

FORMATS

A run-time format 1is a format stored as Hollerith or alphanumeric data
in
an
array.
In the I/0 statement referencing the format, you write
the name of the array instead of a format statement label (see Section
7.2.2).
Virtual arrays must not be used for this purpose.
A

run-time

without
closing

format

the
word
parentheses

array

has

the

FORMAT
and
the
are required.

same

form

statement

label.

FORMAT

statement,

The opening

and

Run-time formats are especially useful when you do
not
know
exactly
which
field
descriptors
will
be required until execution time.
To
solve this problem, you can write a program to create
a
format,
the
choice
of field descriptors being made to depend on the attributes of
the

data.

For

example:

REAL

TABLE (10,5)

REAL

FMT(1l1l),FBIG,FMED,
FSML

DATA FMT(1)/'('/,EMT(1ll)/')'/
DATA FBIG,FMED,FSML/'F8.2','F9.4','F9.6"'/
DATA FMT (3),FMT(5)
,FMT(7) ,FMT(9) /4*',"'/
DO

I=1,10

J=1,5

FMT (2*J) =FMED

IF
IF

(TABLE(I,J).GE.100)
(TABLE(I,J).LE.0.l1)

CONTINUE
WRITE (6,FMT)

CONTINUE
END

FMT(2*J)=FBIG
FMT(2*J)=FSMAL

(TABLE(I,J),J=1,5)"

FORMAT

the

example,

magnitudes
known until
the

real

the

data

array
is

FMT.

stored

odd-numbered

elements

stored

real

the

descriptors
depend

the

left
of

Each

time

the

format

FMT

CONTROL

control

During

format

begins

field

the

format

when

repeat

I/0

statement

else

error

records

counts

are

I/O

and
read

input

initiates
as

indicated

1is

record

read.

On execution, a
specified
wunit

element

the

I/O0

alphanumeric

and

literal

field

descriptors,

external

record

and

format

control,

element.

is
1If

the

reached,

last

executed,
storage

the

H
At

field
the

end

LISTS

list

1I/0

statement.

information

(if

one

exists)

specification.

from

left

The

you

must

specify

descriptor

reads

record

one

when

found,

closing

format

one

control.

the

I/0

right,

the

Thereafter,

format

and
in

I/0

the

field

list

field

data

storage

encountered,

translated
Dbetween
is not found, format

When

output

format.

least

format,

from

the

additional

specification.

list

elements

current

Format

slash occurs in
parenthesis
of

record

if
a
slash
appears
closing parenthesis is
transferred,

list.

literal

descriptor

made

remain

are

descriptors

element

location

¥,

the

I/0

data

list

parenthesis
determines

the

and

the

whether

the

and,
list

I/0

alphanumeric

between

the

specification.
A,

for

format
more

and

format

to
the
also be

correspond

directly

checked

transferred

the external record
control terminates.

control

each

1In

occurs

in
the
format
reached and mote

corresponds

descriptor.

transfer

the
the

discarded

formatted output statement transmits a record
as
format
control
terminates.
Records can

written during format
«control
specification
or
if the last
I/0 list elements remain to be

I/0

depends

be input when
1last
closing

reached

new

list,
field

when

the

is
the

label.

formatted

format

statement

characters

a new record
or when the

specification

the

correlated

format

remaining

The

are

Any

one

of
the

FMT

field

specified.

filled.
the

use

right

assigned

INPUT/OUTPUT

element

contains

control requires that
format specification,
format

through

be
in

other

descriptors

format

and

choice

statement

taken

(1),

Finally,

run-time

FMT

field

The

holding
lost.

action

descriptor

formatted

wunit

can

WITH

execution

the

The

occurs.

execution,

specified

location

specification

except

the

with

TABLE.

stored

elements

format

in the array
data will be

control,

one

that

TABLE.

are

FSML.

recompiled)

INTERACTION

jointly
the

(or

Format

and

referencing

1is not stored
statement, the

list

statement

descriptor
of the I/0

the

data

array

stored

even-~-numbered

the

compiled
into

selection

I/0

read

real

commas

FMED,

1is
Data

and

instead

location.

FORMAT

and

FBIG,

the

the elements of TABLE will not
format specification is stored

parenthesis
FMT.

assigned

magnitude

references

provided

FMT(l11l),

variables

statement

8.6

stored

of
the
data
stored in
just before output.
The

parenthesis

STATEMENTS

field

corresponding

1if

appropriate,

element.

specification
list

elements

one

is
are

be processed.
1If not,
format
control
terminates.
However,
if
additional
list
elements
remain,
part
or
all
of
the
format
specification is reused in a process called format reversion.

FORMAT

Format reversion
starting

from

point.

new

STATEMENTS

the termination of
one.

Format

reversion

are:

the

control

current

reverts

record

and

the group

the

repeat

specification
whose
1left
parenthesis
is
complemented
by
the
next-to-last
right
parenthesis
of the format specification.
If the
format does not contain a group repeat specification,
format
control
returns
to
the
beginning
of the format specification and continues
that

Examples

format

READ

100

(I,l100)

FORMAT

(F8.3,

F8.3)

In this example, three records containing
first
record
assigns values to A and B;
the

third

and

DIMENSION

(6,10)X,(I,B(I),(A(I,J),J=1,5),I=1,5)

FORMAT

SUMMARY

The
and

(E10.3/(I5,E10.3,

In
this
example,
specification that

8.7

are
read.
to C and D;

A(5,5),B(5)

WRITE

two fields
the second

5(F8.5)))

format
reversion
begins with I5.

RULES

FOR

FORMAT

returns

the

group

repeat

STATEMENTS

The following sections summarize the rules for constructing and
wusing
the
format
specifications and their components, and for constructing
external fields and records.
Table 8-4 summarizes the FORMAT codes.

8.7.1

General

Rules

FORMAT

must

omit

statement

field
be

the

must

descriptor
unsigned

repeat

always

such

integer

count

and

rIw

labeled.

constants

field

nX,

the

greater

width

terms
than

¥,
0.

and

You

can

specification.

In a field descriptor such as Fw.d, the term
d
must
Dbe
an
unsigned
integer constant.
If w is specified, then you must
specify 4 in F, E, D, and G field descriptors even if
it
|is
0;
and
the
field
width specification (w) must be greater
than or equal to d.
The decimal point is also required.
You
must either specify both w and d, or omit them both.

In a field descriptor such
as
nHclc2
characters
must
follow
the H format
printing ASCII character in this field

scale

factor

the

form

nP,

...
cn,
exactly
n
code.
You can use any
descriptor.

must

signed

unsigned
integer
<constant
1in
the
range
-127 through 127
inclusive.
The scale factor affects the F, E, D, and G field
descriptors
only.
Once
you
specify
a
scale
factor, it
applies to all subsequent F, E, D, and G field descriptors in
that format specification until another scale factor appears.
You must explicitly specify OP to reinstate a scale factor of
zero.
Format reversion does not affect the scale factor.

FORMAT

STATEMENTS

No repeat count is permitted for H,
X,
T,
literal
field
descriptors
unless
these
enclosed
1in
parentheses
and
treated
as
specification.

the

associated

format

descriptor
A format
same
as

Input

statement

other

than

contains

the

and

I/0

1list,

contain

alphanumeric

H,;,

opening

the

must

specification in an array must
a
format
specification
in

including
FORMAT

8.7.2

I/0

specification

or
alphanumeric
descriptors
are
a
group
repeat

and

statement

c¢losing

label

1least

be
a

one

are

literal.

constructed
the
statement,

FORMAT

parentheses.

only

the
field

The

word

permitted.

Rules

A minus sign must
input
field;
a

precede a negative
value
plus
sign
1is
optional

1in
an
before a

external
positive

value.

input,

must

external

1integer

field

under

<constant.

field

descriptor

cannot

contain

point

or an
exponent.
An
external
field
wunder
descriptor control must contain only the numerals 0
and must not contain a sign, a decimal point, or an
On

input,

descriptor
double
and/or
If

size
that

external
control

field
must

under

precision constant.
It can
an E or D exponent field.
external

field

contains

real

double

contain

the
d

precision

the

The

conversion

field

width

that

G
a

field
real

decimal

point,

the

point

actual

field, as indicated by
specification
of
the
field

descriptor.

If an external field contains an exponent, the
(if
any) of the corresponding field descriptor
on

decimal

O
field
through 7
exponent.

constant

decimal

of
the
fractional
part of
decimal point, overrides
the

corresponding

integer

control
a

scale
has no

factor
effect

field.

specification

must

large

enough

accommodate both the numeric character string of the external
field and any other characters that
are
allowed
(algebraic
sign,

decimal

A comma
field
are

Output

and/or

only

character

the

separator.

shorter

designates

8.7.3

point,

than

null

It
the

exponent).

you

terminates
number

(zero-length)

can
input

use
of

characters

numeric

external

fields

expected.

that
also

fields.

Rules

format

specification

cannot

specify

than
the
external
record can
printer record
cannot
contain
including

the

carriage

control

output

characters

contain.
For example, a line
more
than
133
characters,
character.

FORMAT

STATEMENTS

The field width specification (w) must
be
large
enough
to
accommodate
all
characters
that
the
data
transfer
can
generate, including an algebraic
sign,
decimal
point,
and
exponent.
For example, the field width specification in an E
field descriptor
should
be
1large
enough
to
contain
d+7
characters.

The first character of a record output to a line
printer
or
terminal
is
wused
for carriage control;
it is not printed.
The first character of such a record should be a space, 0, 1,
$,
or
+,
Any other character is treated as a space and is
deleted

from

the

Table

Code

record.

8-4:

Summary

Form

FORMAT

Codes

Effect

Specifies

transfer

decimal

Oow

Specifies

transfer

octal

Fw.d

Specifies

transfer

values

basic

Ew.d

Dw.d

Gw.d

transfer

exponential

form.

transfer

real

Specifies
values

1in

instead

real

double

precision

or double

precision

or double

precision

real

Specifies

transfer

logical

input,
acts
like E code or

exponential

form

transfer

on
acts

values.

double

Specifies

values:
output,

real

integer

values.

form.

values

transfers

real

Specifies

with

integer

f£;

double

precision

1like
F
F code.

code;

data:

input,

output,

transfers

Specifies

transfer

alphanumeric or

Hollerith

transfer

alphanumeric or

Hollerith

values.

nHc...cC

Specifies
values
format

between
an
external
storage location.

Specifies

that

characters

(on
input)
or
that
transmitted (on output).

Specifies
of

the

Specifies
be

the

transferred

Suppresses

position,

character

number
in

carriage

are

spaces

record

the

be
are

external

and

the

skipped
to

Dbe

recotd,

processed.

characters

input

record.

return

during

remaining

interactive

1/0.

Terminates
exhausted.

format

control

the

I/0

1list

CHAPTER

AUXILIARY

INPUT/OUTPUT

The

auxiliary
input/output
functions.,
These statements
®

OPEN

file

and
e

file

terminates

REWIND,

STATEMENTS

statements

perform

file

connection

the

between

the

logical

FIND

writes

condition
the
e

(and

read

between

logical

unit

and

perform

file-positioning

DEFINE

Section

these

special

record

transfer)

that

when

causes

input

end-of-file

statement

record.
FILE

unformatted,

See

END=

7.2

for

associates

direct

access

definition

for

connection

and

required

device.
BACKSPACE,

ENDFILE

unit

attributes

functions.
e

management

are:

device, and specifies
write operations.

establishes

FORTRAN

logical

wunit

reads

with

components

file.

the

I/O

statement

statements.

OPEN
9.1

OPEN

STATEMENT

OPEN

creates

OPEN

statement

can

new

either
file

specify

subsequent

connects

and

file

connects

existing

attributes

that

file

logical

control

unit.

file

logical
1In

creation

processing.

The OPEN statement has the following form:
OPEN (par[,par]...)
par

keyword

specification

one

the

following

key

value

key
A

keyword,

described

the

keyword,

below.

value
Depends

described

9~-1

below.

forms:

unit,

addition,

and

AUXILIARY

INPUT/OUTPUT STATEMENTS

Keywords are divided into several categories based on function:
Keywords that

identify the unit and file.

UNIT

- logical unit number to be used

TYPE
DISPOSE

- file existence status at OPEN
- file existence status after CLOSE

- file name specification for the file

NAME

Keywords that describe the file processing to be performed.
- FORTRAN access method to be used

ACCESS

- write protection

READONLY

Keywords that describe the records

in the file.

BLOCKSIZE
CARRIAGECONTROL

- size of I/0 transfer buffer
- type of printer control

RECORDSIZE

FORM

- type of FORTRAN record formatting
length

record

logical

Keywords that describe file storage allocation when a file

created.

- initial file storage allocation
- file storage allocation increment size

INITIALSIZE
EXTENDSIZE

access

Keywords that provide additional capability for direct
1/0.

access

direct

ASSOCIATEVARIABLE

- variable holding the next

MAXREC

- maximum direct access record number

record

number

Optional keywords that provide improved performance or special
capabilities.

ERR

- statement to which control is transferred
occurs during execution of
error
an
if

BUFFERCOUNT

the

NOSPANBLOCKS

OPEN

number

- records

physical

statement

1/0 buffers

are

not

to use

split

blocks

- other programs can simultaneously

SHARED

the

across
access

file

NOTE

Not all PDP-11 operating systems support
all
keywords
and options.
Consult the
user's
IV
FORTRAN
PDP-11
appropriate
guide for information in system-specific
restrictions.

Table 9-1 lists in alphabetical order the keywords and their
associated values,

including default values.

possible

AUXILIARY

Table

Keyword

ACCESS

9-1:

INPUT/OUTPUT

OPEN

Statement

values?

'SEQUENTIAL"

STATEMENTS

Keyword

Values

Function

Access

Default

method

'SEQUENTIAL"

'DIRECT'
'APPEND'

ASSOCIATEVARIABLE

record

number
direct
BLOCKSIZE

Size

in
access

1/0

associate

variable

System

default

System

default

buffer
BUFFERCOUNT

Number

I/0

buffers
CARRIAGECONTROL

'FORTRAN'

control

‘LIST'
"NONE'

'FORTRAN'

(formatted)
'NONE'
(unformatted)

DISPOSE

'SAVE'

DISP

'"PRINT'

tion

'DELETE'

ERR

'KEEP'

]

File

disposi-

Error

transfer

label
EXTENDSIZE

File

'FORMATTED'

storage

Format

type

error

Volume

system

default

Depends

'UNFORMATTED'

INITIALSIZE

transfer

allocation
increment
FORM

'SAVE'

ACCESS

File

storage

allocation

maximum

allocation

MAXREC

Maximum

number
direct
NAME

File

record

in
access

name

Depends

specification
NOSPANBLOCKS

Records

span

READONLY

Write

tion

not

blocks

unit

and

Records

system
can

span

blocks

protec-

write

protection

(continued

page)

AUXILIARY

Table 9-1

Keyword

INPUT/OUTPUT

(Cont):

STATEMENTS

OPEN Statement Keyword Values

valuesl

Function

RECORDSIZE

Record

SHARED

File sharing

File status

'NEW'

Logical

No default

TYPE

length

Default

allowed

'‘OLD!

'NEW

Depends on TYPE,

not

open

allowed

'SCRATCH'
'UNKNOWN'

UNIT

is an alphanumeric literal,

number

array

element

name.

expression.

integer

statement

label,.

integer

variable

unit

array name, variable name,

name.

Keyword specifications can appear in any order.
Determining
whether
they are optional and which ones are required depends upon the type of
file you are establishing or have established, and upon what you
plan
to

Some

with

it,

examples

OPEN

follow.

(UNIT=1,

ERR=100)

This example creates a new sequential formatted file on
the

default

OPEN

file

wunit

with

name,.

(UNIT=3,

TYPE='SCRATCH',

INITIALSIZE=50,

ACCESS='DIRECT',

RECORDSIZE=64)

This example creates a 50-block sequential file to be used with direct

access.

OPEN

The file

is deleted at program termination.

(UNIT=I, NAME='MTO:MYDATA.DAT', BLOCKSIZE=8192,
TYPE="NEW', ERR=14, RECORDSIZE=1024, RECORDTYPE='FIXED')

This example creates a file on magnetic tape with a large
for

efficient

OPEN

block

size

processing.

(UNIT=I, NAME='MTO:MYDATA.DAT', READONLY, TYPE='OLD',
RECORDSIZE=1024, RECORDTYPE='FIXED', BLOCKSIZE=8192)

This example opens

the file created in the previous example for

Sections 9.1.1 through 9.1.18 describe the keywords

in detail.

input.

AUXILIARY

9.1.1

ACCESS

INPUT/OUTPUT

STATEMENTS

Keyword

The ACCESS keyword specifies
the
method
writing
records.
In
FORTRAN
IV
there
sequential and direct.

This

keyword

has

ACCESS=

acc

the

following

1locating,

are

reading,
or
access
methods:

two

form:

acc

One

the

alphanumeric

ACCESS=

'APPEND'

the

record

last

sequential

the

beginning

ACCESS

implies

the

specified,

access,

the

1literals,

'SEQUENTIAL',

sequential

access

the

you must
file.

In direct access, you specify
I/0 statement, and the system

ASSOCIATEVARIABLE

each

direct

default

read

write

'SEQUENTIAL'.

sequential

sequential
This

1I/0

record

records

that

record number n
selects the nth

the

has

the

integer

(Section
record.

variable

operation,

contains

the

This

file.

following

ASSOCIATEVARIABLE

sequence

records

the

from

the

file

7.2.3)

the

(asv)

that,

record

number

specifier

1is

access.

keyword

Keyword

specifies

access

or
'APPEND'.
positioning after

file.

Direct access to sequential file requires
be fixed-length (see Section 9.1.15.)

9.1.2

and

after

ignored

the
for

form:

asv

9.1.3

integer

variable.

BLOCKSIZE

Keyword

BLOCKSIZE

I/0
the
the

This

specifies

statements

the

appear

size

(in

bytes)

transfer

records

entities specified in the I/0 list.
1In
records to an intermediate I/0 buffer.

keyword

has

BLOCKSIZE

the

following

the

I/0

directly
fact,

transfer

between

the

system

buffer.

file

and

transfers

form:

Dbks

bks
An

integer

For

sequential

transfer

magnetic

tape

expression.
files,

for

files.

device.
See

the

BLOCKSIZE

disk

files

The

determines
or

default

the

1is

the

number

physical

the

system

default

appropriate

information.

PDP-11

FORTRAN

wuser's

disk

blocking

guide

blocks

factor

for

the

for

AUXILIARY

INPUT/OUTPUT STATEMENTS

BUFFERCOUNT Keyword

9,1.4

BUFFERCOUNT specifies the number of buffers to be associated with the
The BLOCKSIZE keyword determines
logical wunit for multibuffered 1/0.
If you do not specify BUFFERCOUNT, or if you
the size of each buffer.
This keyword has the following
specify 0, the system default is used.
form:

BUFFERCOUNT
bc

integer

expression.

CARRIAGECONTROL Keyword

9.1.5

used
to Dbe
CARRIAGECONTROL determines the kind of carriage control
The default for formatted files is 'FORTRAN',
a file.
when printing
'FORTRAN'
'NONE'.
the default for unformatted files, the default is
(see
character
first
the
of
interpretation
FORTRAN
normal
specifies
rather
'LIST' specifies no FORTRAN interpretation, but
Section 8.2);
specifies no implied
'NONE'
and
records;
spacing between
single
carriage

control.

This keyword has

the

CARRIAGECONTROL
cc

following
=

form:

One of the alphanumeric literals

'FORTRAN',

'LIST', or

'NONE'.

DISPOSE Keyword

9.,1.6

DISPOSE determines the disposition of the file connected to the wunit
1If you specify 'SAVE' or 'KEEP', the file is
when the unit is closed.
1If you
this is the default value.
retained after the unit is closed;
submitted to the system line printer
1is
file
the
'PRINT',
specfiy
If you
On some systems, the file is deleted after printing.
spooler.
A read-only file (see Section
'DELETE', the file is deleted.
specfiy
9.1.14) cannot be printed or deleted, and a scratch file (see Section
9.1.17)

cannot be saved or printed.

This keyword has

the

following

forms:

DISPOSE = dis
DISP = dis
dis

One of the alphanumeric

literals

'SAVE',

'KEEP',

'PRINT',

'"DELETE'.

9.1.7

ERR Keyword

ERR transfers control to the executable statement specified by s if an
ERR
The
statement.
OPEN
the
of
execution
during
occurs
error
subsequent
to
not
specification applies only to the OPEN statement,
If an error does occur, no file is opened
I/0 operations on the unit.
or

created.

AUXILIARY

This

keyword
ERR=

The

9.1.8

the

following

executable

label

EXTENDSIZE

when

STATEMENTS

form:

EXTENDSIZE

file

has

INPUT/OUTPUT

Keyword

specifies

the

additional

file

EXTENDSIZE,

statement.

you

number

storage

specify

blocks

the

system

used.

This

keyword

has

the

following

EXTENDSIZE

expression.

which

allocated.

extend

you

default

for

the

not

disk

specify

device

form:

integer

9.1.9

FORM

specifies

Keyword
whether

using formatted or
'"FORMATTED' is the
default.
the

You

same

This

the

file

being

opened

unformatted I/O statements.
default.
For direct access,

must

not

mix

formatted

and

written

access,
is
the

unformatted

unit.

keyword
FORM

has

the

following

read

For sequential
'UNFORMATTED'
I/0

statements

form:

ft
The

9.1.10

alphanumeric

INITIALSIZE

space

This

for

you

new

has

INITIALSIZE

the

file

0,
the
=

'FORMATTED'

'UNFORMATTED'.

Keyword

specifies

specify

keyword

literal

number

blocks

the

1If

you

not

allocation

made.

disk.

initial

following

initial

specify

allocation

INITIALSIZE,

form:

insz

insz
An

9.1.11
MAXREC

integer

MAXREC

expression.

Keyword

specifies the
access
file.
The
specifier is ignored

maximum

number of records permitted in
a
direct
default
is
no
maximum number of records.
This
for other types of files.

AUXILIARY

the

This keyword has
MAXREC

INPUT/OUTPUT STATEMENTS

following

form:

9.1.12

integer

expression.

NAME Keyword

The
NAME specifies the name of the file to be connected to the unit.
name can be any file specification accepted by the operating system.
file
The appropriate PDP-11 FORTRAN IV user's guide describes default
name

conventions.,

1f the file name is stored in a variable, array, or array element, the
<characters terminated by an ASCII null
ASCII
of
consist
name must
character

byte).

This keyword has
NAME
fln

the

following

form:

fln

An array name, variable name, array element name, Or alphanumeric
literal.

9.,1.13

NOSPANBLOCKS Keyword

It
only.
on disk
NOSPANBLOCKS is used for sequential files stored
If any
specifies that records are not to cross disk block boundaries.
record exceeds

the size of a disk block,

This keyword has

the

following

an error occurs.

form:

NOSPANBLOCKS

9.1.14

READONLY Keyword

READONLY prohibits writing by the program to a file.
This keyword has

the

following

form:

READONLY

9,1.15

RECORDSIZE Keyword

RECORDSIZE specifies the logical

record

length.

the
specifies
RECORDSIZE
If the file contains fixed-length records,
If the file contains variable-length records,
record.
each
of
size
RECORDSIZE specifies the maximum length for any record.

You must specify RECORDSIZE when you create a file
records,

with

fixed-length

AUXILIARY

This

keyword

has

the

INPUT/OUTPUT

following

RECORDSIZE

expression.

STATEMENTS

form:

integer

The

value

the

value

the

records

are

depends

formatted,

the

length

the

records

are

unformatted,

storage

units

(four

9.1.16

SHARED

Keyword

SHARED

specifies

than

one

Sequential

that

the

file

executing

may

only

program may

have

write

only

one

This

keyword

has

FORM

(see

Section

the

number

1length

the

9.1.9).

characters;

number

If
if

numeric

bytes).

program

files

the

shared

following

opened

for

shared

access

disk,

and

simultaneously.

they

are

stored

guide

for

access.

form:

SHARED

See

the

appropriate

information

9.1.17

TYPE

specifies

the

file

status
already

1If

you

specify

'SCRATCH',

when

the

file

closed.

will first try
'NEW', thereby

typ

FORTRAN

must

created.

This

PDP-11

keyword.

user's

keyword

the

typ

One

the

'OLD';
if
creating a

has

TYPE

the

of file
exist.

the
new

following

alphanumeric

opened.
1If you specify 'OLD',
you specify 'NEW', a new file is

If
a

new

you

file

created

UNIT
The
be

UNIT

'UNKNOWN',

file is not found,
file.
The default

the system
is 'NEW'.

keyword

system

will

use

form:

literals

'OLD',

'NEW',

specifies the logical unit to which a file is
unit
specification must appear in the list.

UNIT

the

'SCRATCH',

Keyword

connected

This

and

specify

unknown.

9.1.18

additional

Keyword

TYPE

deleted

this

has

the

logical

following

integer

unit

expression.

when

form:

the

OPEN

to
be
Another

statement

connected.
file cannot
executed.

AUXILIARY

INPUT/OUTPUT STATEMENTS

CLOSE
9.2

STATEMENT

The CLOSE statement disconnects a file from a unit.
The CLOSE statement has

the following

form:

DISPOSE

(UNIT=u],

|[,ERR=8])

DISP
u

logical

unit

number.

]

»
]
of
disposition
An alphanumeric literal that determines the
1Its values are 'SAVE', 'KEEP', 'DELETE', and 'PRINT'.
file.

The

label

executable

the

statement.

If you specify either 'SAVE' or 'KEEP', the file is retained after the
If you specify 'PRINT', the file is submitted to the
1is closed.
unit
after
deleted
is
file
the
On some systems,
line printer spooler.
For scratch
If you specify 'DELETE', the file is deleted.
printing.
is
for all other files, the default
files, the default is 'DELETE';
The disposition specified in a CLOSE statement supersedes the
'SAVE'.
disposition specified in the OPEN statement, except that a file opened
file cannot be saved or printed, nor can a file opened
scratch
a
as
for read-only access be printed or deleted.

For

example:

CLOSE (UNIT=1,DISPOSE="'PRINT"')

This statement closes the file on unit 1

and

submits

the

file

for

the

printing.

E'
,ERR=99)
CLOSE (UNIT=J,DISPOSE='DELET

This statement closes the file on unit J and deletes it.

REWIND
9.3

REWIND STATEMENT

The REWIND statement

beginning of

the

file.

repositions

open

The REWIND statement has the following form:
REWIND

A logical

unit

number.

sequential

file

AUXILIARY

The unit
magnetic

number

must

tape.

For

REWIND

This

You

must

direct

refer to
example:

open

STATEMENTS

sequential

file

disk

statement

currently

INPUT/OUTPUT

repositions

open

not

logical

wunit

file.

issue

REWIND

statement

access.

for

file

the

beginning

that

open

for

BACKSPACE
9.4

BACKSPACE

STATEMENT

The BACKSPACE statement repositions an open
sequential
file
at
beginning
of
the
preceding record.
When the next I/0 statement
the unit is executed, that record is available for processing.
The

BACKSPACE

The

statement

BACKSPACE

unit

logical

unit

magnetic

has

must

refer

tape.

For

example:

must

not

form:

open

sequential

file

disk

the

open

for

This statement repositions
beginning of the preceding
You

following

for

number.

number

BACKSPACE

direct

the

issue

append

the open
record.

BACKSPACE

file

statement

access.

1logical

for

file

wunit

that

FIND
9.5

FIND

The

STATEMENT

FIND

This

statement positions a direct access file on a
particular record.
No data transfer takes place.

statement
FIND

the

following

form:

(u'r)

logical

The

has

direct

unit

number.

access

record

number.

specified

unit

AUXILIARY INPUT/OUTPUT STATEMENTS

The record number cannot be less than 1 or greater than the number

the

the file.

for

records defined

The associated variable of the file,
record

direct access

number.

specified,

1is

set

ENDFILE
9.6

ENDFILE

STATEMENT

The ENDFILE statement writes an end-file record to the specified unit.
This statement has the following
ENDFILE

form:

unit

A logical

number.

You can write an end-file record only to sequentially accessed
sequential organization files containing variable-length or segmented
records.

For

example:
ENDFILE

This statement outputs an end-file record to logical unit 2,

DEFINE FILE
9.7

DEFINE FILE STATEMENT

The DEFINE FILE statement

describes

direct-access

files

sequential

The OPEN statement
‘that are associated with a logical unit number.
The DEFINE FILE
this.
to do
way
(Section 9.1) is the preferred
statement establishes the size and structure of the direct access
file.

The DEFINE FILE statement has the following form:
DEFINE FILE u
u

An integer
logical

records

An integer
length,

constant
number.

[,u(m,n,U,asv)]

integer

...

variable

that

specifies

the

An integer constant or integer variable that specifies the number

of
n

unit

(m,n,U,asv)

the

file.

constant

in 16-bit words,

integer

variable

Specifies that the file is unformatted

only acceptable entry in

that

(2 bytes), of each record.

this position.

(binary);

specifies

the

1is

the

this

AUXILIARY

INPUT/OUTPUT

STATEMENTS

asv

An
At
of

integer variable, called the asociated variable of
the
file.
end of each direct access I/0 operation, the record number
the next higher-numbered record in the
file
1is
assigned
to

the

asv.

DEFINE FILE specifies that a file containing m fixed-length records of
n
1l6-bit
words
each exists, or is to exist, on logical unit u.
The
records in the file are numbered sequentially from 1 through m.
DEFINE

FILE

statement

must

that

refers

executed
to

the

before

specified

the

first

direct-access

DEFINE
FILE
also
establishes
the
integer
variable
asv
as
associated variable of the file.
At the end of each direct access
operation, the FORTRAN I/0 system places in asv the record
number

the
the
the

record immediately
associated variable
file
(unless
it

For

example:

1/0

file,

the
I/0O
of

one just read or written.
Since
to the next sequential record in
by an assignment, input, or FIND
statement),
direct
access
1I/0
statements
can
perform
sequential
processing of the file by using the associated variable of the file as
the record number specifier.

DEFINE

This

FILE

statement

file

1000

words

long.

1000,

and

this file,
the record

(1000,48,U,NREC)

specifies that logical unit 3 is to be connected
fixed-length records;
each record is forty-eight

The
are

following the
always points
1is
redefined

records

are

unformatted.

the integer
immediately

numbered
After

each

variable NREC
following the

sequentially

from

direct

I1/0

access

to
a
16-bit
through

operation

will contain the record
record just processed.

number

APPENDIX

CHARACTER

A.1l

The

FORTRAN CHARACTER

FORTRAN

character

The

letters

The

numerals

The

following

Character
A

consists

through
0

through

special

of:

and

Space

Equal

sign

Plus

Minus

through

characters:

Name

(TAB

SETS

SET

set

Character

tab

Name

Comma

Period

Apostrophe

Quotation

Asterisk

Dollar

Slash

Exclamation

(

Left

Colon

)

Right

sign
sign

parenthesis

mark

sign
point

parenthesis

Other printing characters can appear in a FORTRAN
statement
only
as
part
of a Hollerith constant.
Any printing character can appear in a
comment.
See Table A-1l.

A.2

ASCII

CHARACTER

SET

Table A-1 is a table representing the ASCII character set.
of
the
table are hexadecimal digits (0 to 7), and to the
table are hexadecimal digits (0 to F).
To determine
the
value

ASCII

character,

locate

the

ASCII

character

use the row number as the unit's position digit,
number as the 16's position digit.
For example,
of the equal sign (=) is 3D.

and
the

At the top
left of the
hexadecimal
the

table,

use
the
column
hexadecimal value

CHARACTER

Table

A-1:

ASCII

SETS

Character

Set

Columns

NUL
SOH
STX
ETX
EOT
ENQ

A.3

NOL

DLE

SOH

DCl

STX

DC2

ETX

DC3

)

EOT

DC4

ENQ

NAK

o
7

ACK
BEL

SYN
ETB

6
7

F
G

v
W

£
g

v
W

8
9
A

BS
HT
LF

CAN
EM
SUB

8
9
:

H
I
J

X
Y
Z

h
i
j

B
C

VT
FF

ESC
FS

(
)
*
+
,

;
<

K
L

[
\

k
1

X
y
z
{
J

]

DEL

Null
Start of Heading
Start of Text
End of Text
End of Transmission
Enquiry

DLE
DCl1
DC2
DC3
DC4
NAK

ACK

Acknowledge

SYN

BEL

Bell

ETB

Data Link Escape
Device Control 1
Device Control 2
Device Control 3
Device Control 4
Negative Acknowledge
Synchronous Idle
End of Transmission Block

Backspace

CAN

Cancel

HT
LF
VT
FF

Horizontal Tabulation
Line Feed
Vertical Tab
Form Feed

EM
SUB
ESC
FS

End of Medium
Substitute
Escape
File Separator

CR
SO
SI
SP

Carriage Return
Shift Out
Shift In
Space

GS
RS
us
DEI,

Group Separator
Record Separator
Unit Separator
Delete

RADIX-50

Radix-50 is
characters
character

CONSTANTS

AND CHARACTER

set

subset

characters

Character

.
(Unassigned)
0 -9

and

the

their

ASCII Octal
Equivalent

Space
S

SET

a special character data representation in which up
to
3
can
be
encoded
and
packed
into 16 bits.
The Radix-50

The Radix-50

ASCII

character

corresponding

- 132
44

56
60

code

Radix~50 Value
(Octal)

40
101

set.

- 32
33

34
35
- 47

values

are:

CHARACTER

SETS

Radix-50 values are stored, up to 3 characters per
word,
them into single numeric values according to the formula:
((i

where

Thus,

the

§)

and

maximum

47*50*50

Radix-50

the

code

Radix-50

value

is:

constant

has

packing

represent

47*50

the

values

Radix-50

characters.

174777

following

form:

nRclc2...cn
n

unsigned,

characters

nonzero

integer

constant

that

states

the

number

follow.

character

from

The maximum number
include

any

character

only
of

that

valid
in

and

character

12.

appear

Radix-50

DATA

valid

Radix-50

of characters

spaces

constants
Examples

the

1in

set.

The

the

character

character).

You

count

must

(the

space

string
can

use

Radix-50

statements.
invalid

valid

Invalid

4RABCD

4RDKO:

Radix-50

(the

constants

colon

not

are:

Radix~-50

character)

6 RATOAAA

When

Radix-50

element,
type

the
the

contains
characters

constant

number

component

fewer
(0

more
are

(see

bytes

bytes)

bytes
not

assigned

bytes

than

are

Table

the

appended

than

used.

that

the

can

numeric

variable

assigned

depends

2-2).

1length
on

length

the

Radix-50

component,

right.

the

or
on

component,

the

array
data

constant

ASCII

null

the

constant

the

rightmost

APPENDIX

FORTRAN

B.1l

EXPRESSION

LANGUAGE

Table

Type

Arithmetic

SUMMARY

OPERATORS

Table B-1 lists the expression
descending precedence.

Data

B-1l:

Operator

* ok

operators

Expression

each

data

type

order

Operators

Operation

Operates

Exponentiation

Upon:

Arithmetic
expressions

*,/

Multiplication,
division

+,-

Addition,
unary

Relational

.GT.

subtraction,

plus

Greater

and

minus

than

Arithmetic
expressions

.GE.

LLT.
.LE,

Logical

Greater

than

equal

Less

than

Less

than

equal

.EQ.

Equal

.NE,

Not

.NOT.

.AND.

and
are
.OR,

both

only
both

A.OR.B
either

true

A.AND.B

operators

priority)

equal

relational

have

equal

.NOT.A
only

or logical
(all

true

if A
true

and

false

Logical

integer

expressions

and

is true if
A or B or

are

true

(continued

page)

FORTRAN LANGUAGE SUMMARY

Table B-1

(Cont.):

Expression Operators

Operates Upon:

Data Type

Operator

Operation

Logical
(cont.)

.EQV.

A.EQV.B is true if and
only if A and B
are both true or A
and

.XOR.

B.2

are

A.XOR.B
only
B is
true

both

is true

.EQV. and .XOR.
have equal
priority

false

if and

if A is true and
false or B is
and A is false

STATEMENTS

e the PDP-11
The following list summarizes the statements availablin
including the general form of each statement.
FORTRAN IV language,
The
The statements are listed alphabetically for ease of reference.
"Manual Section" column indicates the section of the manual that
describes each statement in detail.

Effect

Form
ACCEPT

See READ, Formatted Sequential
See READ, List-Directed

Manual

Section
7.3.1
7.3.3

3.1, 3.2

Arithmetic/Logical Assignment
v=e

A variable name,

An expression.

or an array element name.

Assigns the value of the arithmetic or
logical expression to the variable.

6.2.1

Statement Function
f(lpl,pl...]1)=e
f

A symbolic name.

An expression,

symbolic

name.

Creates a user-defined function having the
When
variable p as a dummy argument.
referred to, the expression is evaluated

using the actual arguments in the function

call.

FORTRAN

Form

LANGUAGE

SUMMARY

Manual
Section

Effect

ASSIGN

3.3

label

integer

executable

variable

statement.

name.

Associates the statement label s
integer variable v for later use
assigned GO TO statement.

BACKSPACE

9.4

integer

expression.

Backspaces

logical

BLOCK

DATA

nam

one

unit

record

the

currently

open

file

[nam]

symbolic

Specifies
BLOCK

CALL

with the
in an

name.
the

DATA

subprogram

that

follows

subprogram.

f[(fa]l,[al]l...)]

4.5
6.2

£
a

subprogram

name

expression,

entry

array

point.

name,

procedure

name.

Calls the subroutine
name specified by £,

arguments

subprogram with the
passing the actual

replace

the

subroutine

dummy

definition.

(pl,pl...)
P

9.2

One

UNIT

the

forms:

DISPOSE
DISPOSE
DISPOSE
DISPOSE
ERR

following

=
=
=
=

'SAVE'
'KEEP'
'DELETE'
'PRINT'

integer
label

expression.
an

executable

Closes the specified
be abbreviated DISP.

statement.

file.

DISPOSE

can

FORTRAN

LANGUAGE

SUMMARY

Manual

Section

Effect

Form

COMMON

[/[cbl/]

nlist

[[,]/[cb]/nlist]...
block

common

nlist

list

array

one

names,

separated

Reserves

one

5.4

name.

variable

array

names,

declarators

commas.

blocks

storage

space under the name specified to contain
the variables associated with that block
name.

CONTINUE

Causes

DATA

nlist/clist/[[,]
nlist

list

processing.

nlist/clist/]...
of

one

array names,
separated by

expressions

clist

variable

or array
commas.

must

names,

element names
Subscript

constant.

A list of one or more constants separated
by commas, each optionally preceded by j*,
where j is a nonzero, unsigned integer
constant.

Initially stores elements of clist
in the corresponding elements of
nlist.

DECODE

(c,f,b[,ERR=s]) [list]
An

list

integer

format

variable

expression.

specifier.
name,

element

name.

label

I/0

list.

Reads c¢
assigns
list,

FILE

name,

executable

characters

values

converted

specification

DEFINE

array

statement.

from buffer
the elements

according

array

and

the

format

u(m,n,U,v)[,u(m,n,U0,v)]...

integer

variable

integer

constant.

integer

variable

integer

constant.

FORTRAN

Form

LANGUAGE

SUMMARY

Manual
Section

Effect

integer

variable

integer

integexr

variable

name.

constant.

Defines the record structure of a
direct access file where u is the logical
unit number, m is the number of fixedlength records in the file, n is the
length in words of a single record, U is

a fixed argument,
variable.

DIMENSION

and

the

associated

a(d)[,a(d)]...

a(d)

array

5.3

declarator.

Specifies

storage

space

requirements

for

arrays.

[,]

el,e2[,e3]
label

4.3

A variable

el,e2,e3

Numeric

executable

statement.

name.

expressions.

Executes

the
v

loop

performing

Set

Execute all statements
statement number s

Evaluate

Repeat

the

following

steps:

steps

through

v+e3

through

for

the

following

iterations:
MAX

ENCODE

(1,

INT((e2

el)/e3)

(c,f,b[,ERR=s]) [list]

format

variable

integer

7.6

expression.

specifier.
name,

element

name.,

label

list

I/0

list.

array

name,

executable

Writes ¢ characters
contains the values
the list, converted
specification f.

array

statement.

into buffer b, which
of the elements of
according to format

FORTRAN

LANGUAGE

SUMMARY

Manual

Form

Section

Effect

4.9

END

Delimits

ENDEFILE

program

unit.

integer

Writes
unit

expression.
end-£file

record

logical

END=s ,ERR=s

label

executable

statement.

Transfers control on end-of-file or error
condition.
This is an optional element
in

each

the

type

program

of
to

I/0

statement

transfer

and

allows

statement

number s when an end-of-file (END=)
error (ERR=) condition occurs.

EQUIVALENCE

(nlist)[,(nlist)]...

nlist

A list of two or more variable names,
array names, Or array element names,
separated by commas.
Subscript
expressions must be constants.
Assigns each of the names in
the same storage location.

EXTERNAL

nlist

v[,v]...

subprogram

Defines

FIND

the

name,
names

specified

subprograms.

(u'r)

integer

expression.

integer

expression.

FORMAT

(field

Positions
unit u to

the
the

specified

file on
record

logical

specification,...)

8.1

Describes the format in which one or
more records are to be transmitted;
statement

label

must

present.

8.8

FORTRAN

LANGUAGE

SUMMARY

Manual

[typ]

FUNCTION

nam[*n] [([p[,pP)...])]

typ

A data

nam

A data

type

symbolic

length

function

program

name

(p).

can

included.

specifier.

name.

names

subprogram,
and

any

optional

indicating

dummy

type

argument
specification

label

Transfers

name.

symbolic

the

6.2'2

specifier.

type

Begins

Section

Effect

Form

(slist)[,]

slist

executable

control

statement.

statement

number

list

labels

one

separated

integer

Transfers

more
by

statement

commas.

expression.
control

the

statement

specified by the value of e (if e=1,
control transfers to the first statement
label;
1if e=2, control transfers
to

the

second

number

less
of

transfer

statement

than

statement
takes

slist

integer

label,

etc.).

greater

labels

than

present,

the
no

place.

list

variable
one

separated

control

recently

associated

sl,s2,s3

expression.

name.

Transfers
statement,

(e)

[[,] (slist)]

labels

statement

commas.
the

with

statement

most

ASSIGN

FORTRAN

LANGUAGE

SUMMARY

Manual

Effect

Form

label

Transfers

control

the

less

control

than

statement.

statement

value

(if

transfers

e equals 0, control transfers
if e is greater than 0, control

transfers

(e)

executable

depending
sl;
1if
to s2;

Section

s3).

expression.

Any

executable

logical

statement

Executes the statement
expression has a value

IMPLICIT

typ

except

IF.
if
of

the logical
true.

(al,al...)[,typ(al,al...)]...

typ

A data

Either

type

specifier.

a single
alphabetical

The

element

letter, or two letters
order separated by

represents

range of)
letter (s)
initial letter of a

the

variable

single

(or

whose presence as the
variable specifies

that

data

type.

OPEN (par[,par] ees)
par

keyword

specification

one

the

forms:

key
key
key

value

keyword,

Depends

described

the

keyword,

below.
as

described

below.

Keyword
ACCESS

Values
'SEQUENTIAL"
'DIRECT!

'APPEND'
ASSOCIATEVARIABLE

BLOCKSIZE

BUFFERCOUNT

CARRIAGECONTROL

'FORTRAN'
'LIST!
'NONE'

DISPOSE

'SAVE'

DISP

'PRINT'

'DELETE'

'KEEP'

following

FORTRAN

LANGUAGE

SUMMARY

Manual
Section

Effect

Form

Keyword

Values

ERR

EXTENDSIZE

FORM

'"FORMATTED'

'UNFORMATTED'
INITIALSIZE

MAXREC

NAME

NOSPANBLOCKS

READONLY

RECORDSIZE

SHARED

TYPE

'OLD!

'NEW'

'SCRATCH"
"UNKNOWN''
UNIT

array

element

name,
name,

numeric

label

variable
or

name,

alphanumeric

array
literal.

expression.

integer

executable

variable

statement.

name.

Opens a file on the specified logical
according to the parameters specified
the keywords.

PAUSE

unit
by

[disp]

disp

PROGRAM

4'7

A decimal digit string containing
digits, an octal constant, or an
alphanumeric literal.
Suspends

program

display,

one

execution

prints

five

the

specified.

See

WRITE,

Formatted

See

WRITE,

List-Directed.

Sequential.

7.3.2

7.3.4

nam

5.10

symbolic

Specifies

name.
a

name

for

READ

(u,f[,END=s][,ERR=s])
[list]

READ

f[,list]

and

one

the

f[,list]

integer
format

expression.

specifier.

main

program.
7.3.1

FORTRAN

LANGUAGE

SUMMARY

Manual

Form

Effect

label

list

I/0

list.

Section

executable

statement.

Reads one or more logical records from
unit u and assigns values to the elements
in the list.,
The values are converted

according

format

specification

READ (u[,END=s] [ ,ERR=s]) [list]
u

integer

label

list

I/0

list.

7.3.5

expression.
an

executable

statement.

Reads one unformatted record from
unit u, and assigns values to the
elements in the list.

logical

READ(u'r[,ERR=s8])[list]

7.4.3

integer

expression.

integer

expression.

label

list

I/0

list.

executable

statement.

Reads record r from logical unit u and
values to the elements in the list,.

READ

(u,*[,END=s][,ERR=s])list

READ

*,list

7.3.3

*,list

integer

Denotes

list-directed

label

list

I/0

list.

Reads

one

expression.

formatting.

executable

assigns

records

statement.

from

logical unit u and assigns values
the elements in the list.
The
values are converted according to
data type of the list element.

to
the

FORTRAN

Form

LANGUAGE

SUMMARY

Effect

RETURN

4.6

Returns

from

REWIND

control

the

current

the

calling

program

subprogram.

9.3

integer

expression.

Repositions logical unit u
beginning of the currently

STOP

Manual
Section

to the
opened

file.

[disp]
disp

A decimal digit string containing
digits, an octal constant, or an
alphanumeric literal.

one

five

Terminates program execution and prints
the display, if one is specified.

SUBROUTINE

nam[([p[,p)l...])]

nam

symbolic

name.

symbolic

name.

Begins
the

Type
typ

subroutine

program

names

TYPE

6.2.3

name

and

subprogram,
any

dummy

indicating
argument

(p).

See

WRITE,

Formatted

See

WRITE,

List-Directed.

Sequential.

Declaration

5.2

vi,v]...

typ

One

7.3.2

7.3.4

the

following

BYTE

LOGICAL
LOGICAL*1
LOGICAL*4
INTEGER

INTEGER*2
INTEGER*4
REAL

REAL*4
REAL*8
DOUBLE

PRECISION

COMPLEX

COMPLEX*8

data

types:

FORTRAN

LANGUAGE

SUMMARY

Manual

Section

Effect

Form
v

A variable name,

array name,

function

function entry name, or an array declarator.
The name can optionally be followed by a data
type length specifier (*n).
The

symbolic

specified

names

data

(v)

are assigned

the

type.

VIRTUAL a(d)[,a(d)]}...
a(d)

5.5

array

declarator.

Specifies storage space for arrays
outside normal program address space.

WRITE

(u,f[,ERR=s])[list]

f[,list]

TYPE

f[,list]

integer

7.3.2

expression.

format

label

list

I/0

list.

specifier.
an

executable

statement.

Writes one or more records to logical unit
u, containing the values of the elements
in

the

list.

according

WRITE

The

values

format

are

converted

specification

(ul[,ERR=s])[list]
integer

7.3.6

label

list

I/0

list,

Writes
unit u
in

WRITE

executable

statement

label.

one unformatted record to logical
containing the values of the elements
list.

the

(u'r(,ERR=s])

expression.

[list]

7.4.4

integer

expression.

integer

expression.

label

list

I/0

list.

Writes record
the values of

executable

statement

r to logical
the elements

B-12

unit u
in the

label.

containing
list.

FORTRAN

Form

LANGUAGE

SUMMARY

Manual
Section

Effect

WRITE(u,*[,ERR=s])1list
PRINT
TYPE

7.3.4

*,list
*,list
u

integer

expression.

Denotes

list-directed

label

list

I/0

list.

formatting.

executable

statement.

Writes one or more logical records to logical
unit u containing the values of the elements
in the list.
The values are converted
according to the data type of the list
element.

B.3
Table
table

LIBRARY FUNCTIONS
B-2 lists the FORTRAN IV library functions.
refer to notes which follow the table.

Table

B-2:

FORTRAN

Form

Real absolute value
Integer absolute value

DABS (X)

Double

CABS (Z)

FLOAT(I)
IFIX (X)
SNGL (X)
DBLE (X)
REAL (Z)

AIMAG(Z)
CMPLX (X, Y)

precision

Complex to Real,

IDINT (X)

value

Complex to Real conversion,
obtain real part
Complex to Real conversion,
obtain imaginary part
Real to Complex conversion
CMPLX (X, Y)=X+i*Y
functions

return

largest

the

the sign

integer

Argument

Result

Type

Real
Integer

Double

Integer

Real

Complex

Integer to Real conversion?2

the argument

AINT (X)
INT (X)

absolute

absolute valuel

Real to Integer conversion?
Double to Real conversion2
Real to Double conversion?2

Truncation

Library Functions

Definition

ABS (X)
IABS (1)

Superscripts

Real

Real
Double
Real

Integer
Real
Double

Complex

Real

Complex

Real

Complex

Real
Real
Double

Real
Integer
Integer

= largl

Real to Real truncation3
Real to Integer truncation3
Double to Integer truncation3
Remainder functions return the remainder
when the first argument is divided by
the

second.

(continued

page)

FORTRAN

Table

B-2

LANGUAGE

(Cont.):

Form

SUMMARY

FORTRAN

Real

DMOD (X, Y)

Double precision remainder

remainder

Integer

remainder

Maximum value
largest value
list;

AMAXO(I,J,¢40)
AMAX1(X,Y,...)
MAX0(I,J,...)
MAX1(X,Y,...)
DMAX1(X,Y,...)

Argument

Result

Type

Definition

AMOD (X, Y)
MOD(I,J)

Library Functions

Real

Double

Integer
Real
Integer
Real
Double

Real
Real
Integer
Integer
Double

Integer

Real

Real
Integer

Integer

functions return the
from among the argument

arguments.

Real maximum from Integer list
Real maximum from Real list
Integer maximum from Integer list
Integer maximum from Real list
Double maximum from Double list
Minimum value functions return the smallest value from among the argument list;
2

arguments.

AMINO(I,J,...)

Real

AMIN1(X,Y,...)
MINO(I,Jd,...)

Real minimum of Real list
Integer minimum of Integer

MIN1(X,Y,¢..)
DMINL(X,Y,...)

Integer

list

Integer minimum of Real list
Double minimum of Double list
The

SIGN(X,Y)
ISIGN(I,J)
DSIGN(X,Y)

minimum

transfer

(sign

the

value

first

sign

second

functions

Real
Double

Integer
Double

Real
Integer
Double

Real
Integer

Real
Integerx

return

argument)

argument).

Real transfer of
Integexr transfer
Double precision

sign
of sign
transfer

sign

Positive difference functions return the
first argument minus the minimum of the
two

DIM(X,Y)
IDIM(I,J)

arguments.

Real positive difference
Integer positive difference
Exponential
of e raised

functions return the value
to the argument power,

EXP(X)
DEXP (X)

CEXP(2)

eXx
eX
e?

Real
Double
Complex

Real
Double

ALOG (X)
ALOG10 (X)
DLOG (X)
DLOG10 (X)
CLOG(Z)

loge(X)
logq g (X)
loge (X)
logyg (X)
loge (2)

Real
Real
Double
Double
Complex

SQRT (X)
DSQRT (X)

Square
Square

Real
Double

CSQRT (Z)>

root
root

of
of

Real argument
Double precision

Square root of Complex argument

argument

Complex

(continued

Complex

page)

Table

B-2

FORTRAN

LANGUAGE

SUMMARY

(Cont.):

FORTRAN

Library

Functions

Form

SIN(X)

Real

DSIN(X)

Double

CSIN(2Z)6

sine

precision

sine

Complex sine

COS (X)

Real

DCOS (X)

Double

precision

Complex cosine

TANH (X)

Hyperbolic

ATAN (X)

Real

DATAN (X)

Double

ATAN2 (X,Y)

Real

arc

cosine

Type

tangent

precision

arc

tangent

arc

tangent

(X/Y)
.

Double precision arc tangent of

CONJG(2Z)

Complex conjugate,
CONJG(Z)=X-1i*Y

RAN (I,J)9

Returns a pseudo-random number of
uniform
0

absolute

value

Real

Double

complex

Complex

Real

Double

Real

over

the

number,

range

Complex

Real

Double

Real

(X/Y)

Z=X+i*Y

distribution

Real

Double

Complex

DATAN2 (X,Y) 7,8

The

Result

Type

Complex

cosine

Cccos(z)®

Argument

Definition

Real

Double

Complex

Integer

Real

the

value:

(X,Y),

real

(X2+Y2)1/2

Functions

provide

statements.
DBLE with a

without
3.

[x]

is
the

For

example

The

5.
The
than

than

the

one

data

implied

type

another

conversion

type

assignment

real argument and
the
return the value of the

function
argument

[5.7]

equals

ALOG,

argument

and

DLOG,
CLOG

integer
whose
magnitude
sign is the same as

whose
[-5.7]

equals

ALOGl0,

must

not

and
be

does
that

not
x.

-5.

DLOGl0

must

greater

(0.,0.).

The argument of SQRT and DSQRT must be greater than or equal to 0.
result of CSQRT is the principal value with the real part greater
or

equal

The

DATAN2
8.

defined as the largest
magnitude
of
x and

The

When

value

with

the

argument

SIN,

radians.

conversion

effect

The function SNGL with a
double precision argument

argument

principal
6.

cause

same

conversion.

exceed

than

that

the

The

argument

result
is

real

DSIN,

ASIN,

the

part

imaginary

part

COS,

treated

DASIN,

1is

DCOS,

modulo

ACOS,

greater
TAN,

the

than
and

result

equal

DTAN

1is

must

the

0.
be

2*pi.

DACOS,

ATAN,

DATAN,

positive

when

ATANZ2,

and

radians.

The

result

equal

of
to

ATAN2
0.

The

and

DATAN2

result

undefined

both

a(2)

arguments

is
are

less
0.

The argument for this function must
be
an
1integer
variable
or
integer
array
element.
The
atrgument should initially be set to 0.
The RAN function stores a value in the argument that it later uses
to
calculate
the
next
random
number.
Resetting
the
argument
to 0
regenerates

different

the

random

sequence.

number

Alternate

sequences.

starting

values

generate

INDEX

field

descriptor,

8-10

8-11,

8-18
Absolute

value

constant,
ACCEPT

and
B-2,

keyword,

Access

modes,

B-9

9-2,

7~2

Array

arguments,

format

9-3,

7-3

9-9
2-10,

6-1

in,

dummy,

6-3

Array

size,

Array

storage,

record

length,

9-4,

2-13,

8-20

7-5

8-21

2-12

character

set, A-1
characters, 8-22

9-8

null

octal

Allocating

2-13, 6-3
8-3, 8-5, 8-9

storage

elements,

9-8
equivalents of
characters, A-2
statement,

4-2,

All-zero

statement

Alphanumeric

American

National

FORTRAN

label,

literals,

1-6

7-17,

'APPEND'

initial

l-byte

value,

9-3,

9-5

2-9

storage,

list,

4-9

2-9,

actual,

the

dummy,

6-1

CALL

2-9,

6-3,
function

6-1

dummy

references

values

list

values

variables,

and

array

2-14,

expression,

elements,

Assignment

operator,

used

3-4
Associated

variables,

as,

6-3

9-3,
3-1
2-16,

4-3

2-14,

Arithmetic/loyical

2-9,

9-13

keyword,

9-2,

9-5
dummy

arguments,

Asterisk

2-15,

6~-4
statement,

5-12

2-8
statements, 3-1

2-15

2-14,

7-5
arrays,

Assignment

statement,

operators,

elements,

arithmetic/logical,

Associating

assignment

5-13

3-1

ASSOCIATEVARIABLE

6-1

5-9

name,

Arithmetic
elements,

program

B-2

4-9

6-5

5-11
subprogram,

5-5

locations,

6-2

6-1

and

6-1

actual
6-2

(*),

Auxiliary

1-3, 8-3, 8-4
specifications, 9-1
1/0 statement, 9-1

BACKSPACE

statement,

Attribute

4-4

B-1

assignment,

B-2

9-1,

9-11

B-3
2-1,

2-2,

adjustable,
type

dimensions
4-9,

Base

2-11

elements,

2-15

2-13,

6-3

Basic

component,

an,

2-12

Basic

real

in,

ansubscripted,

5-14,

5-3
2-12

7-5

to
to

common

elements

Assignment,

statement,

4-9,

symbolic

6-3

and

4-2

5-14

storage

values,

2-3

Arguments

associating

3-4,

statement,

values

COMMON,

Access

actual,

LOGICAL*]1l

2-8,

8-12

spaces,

Argument

blocks,
character,

Assigning

1-1

2-19

Apostrophe

3-3

Radix-50

4-3

Standard

X3.9-1966,

A-2

4-3

Assigned

2-8

character,

ASSIGN

5-3

data

2-12,

formats

ASCII

field,
Allocation, 2-3

Array,

7-6

specifications,

references,

run-time

All-blank

equivalent, 5-9 to
5-10
multidimensional, 2-10, 7-7

9-5

Addition, 2-14
Adjustable arrays,

5-12

8-23

6-3

Area,

statements,

transmitting,

7-8

Array

Actual

Append

2-11,

5-13

7-2

shared,

.AND.,

5-3

2-10,

making

Access,

Actual

DATA

7-3

sequential,

2-11,

2-1,

2-14

Input

ACCESS

direct,

elements,

2-4

statement,

also

declarator,

Array

integer

B-10, B-13
formatted sequential,
list-directed, 7-10
See

Array

5-4
2-13,

2-13

constant,

2-14

2-5

Binary

data,

Binary

operator,

2-8,

Blank.

See

character

Index-1

7-1

Space

2-14

INDEX

Commas, 7-11
as a field

Blank common block, 5-4
boundaries, crossing disk,
Block, common, 2-2, 5-4 to
See also Common blocks
data subprograms, 2-2
size, physical, 9-5
BLOCK DATA statement, 5-13

5-5

indicator,

1-6
1-6

line,

1-7

1-6,

assigning

storage,

5-4

dimensions,

byte,

5-9

9-2,

9_3 i

Boundary, word, 5-5
BUFFERCOUNT keyword,

5-4

Bypassing input records,
Byte boundary, odd, 5-9
Byte,

zero,

CALL

statement,

4_9 r

of,
control

8-24,

8-19

l1-6

1-3,

6-8, B-3
discussion

Carriage

9-6

9-8

(letter),

5-11

4-9
character,

Character

set

ASCII,

A-1,

FORTRAN,

RADIX~-50,

editing,

Complex

type,

Plus

8-13
(+),

tab,

external
8-13

1-3

integer,

2-4

integer,

2-4

of, stored
8-10

form, FORTRAN,
descriptor,

values,

7-11

2-1

and

DATA

2-6

2-9

Constants,

statement,

storage

2-2

Continuation
field, 1-6
indicator,
lines, 1-6
CONTINUE

5-12

carriage,

80,

1-5,

8-18,

iteration,

format,

one through 72, 1-5
seventy-three through

5-4

1-6

statement,

1-4

2-10

8-22

locations,

Control

8-14

Column

number,

2-10,

1-5

B-3

(:)

2-5

Contiguous

Classes of symbolic names,
CLOSE statement, 9-1, 9-10

Colon

2-5

octal integer, 2-4
repetition of, 7-11
unsigned integer, 2-4,

character-per-column

Coding

4-2

7-11

positive

truncated,

1-5

maximum number
variables,

2-4,

2-6
negative
real,

8-18

formatting,

statement,

negative double precision,

l1-6

special,

statement execution,

logical, 2-7, 7-11
magnitude of a real,

position

space,

8-2, 8-11
input, 7-15
integer, 2-4

transferred,

72, 1-7
position of, in the
record, 8-12 to
printable, 1-4, A-1l

8-15

4-4

A-3

FORTRAN

Conditional

A-1

2-8

2-16

statement,

8-18

of,

8~14

2-3,

absolute value of an integer, 2-4
complex 2-6 to 2-7, 7-11
double precision, 2-6
Hollerith, 1-1, 1-3, 2-7, 7-11,

ASCII, 8-22
carriage control, 8-18, 8-24
colon (:), 8-14
dollar sign ($), 8-13 to 8-14,
number

data

Components,

Characters

apostrophe,

2-7,

2-17
to 8-15

2-17

A-2,

complex operations,
Complex value, 8-14
COMPLEX*8, 2-3

A-2

1-3,

2-6

Complex data

Constant,

count,

Character

B-4

5-5,

7-11

Computed

9"2'

keyword,

constant,

Complex

5-14

2-1

9-6

CARRIAGECONTROL
9-3, 9-6

5-10

and,

extending, 5-10
initial values in,

2-13

allocation,

elements

5-5

blank, 5-4
EQUIVALENCE

Bound

Boundary

LOGICAL*1l

5-5

5-4

2-2,

Common blocks,
to

Blocks

separator,

8-23
Comments, 1-3,

9-8

8-24

4-6

8-21

in DO
Control

loops, 4-6
statements,

Control

transfer

1-5

Index-2

4-4,

6-4,

4-1

methods,
6-6,

6-8

4-1

INDEX

Control variable, 4-5, 7-7
Conversion,
double precision, 2-17

rules

for

DECODE

Default

assignment

statements,

group

2-7

8-17,

8-23

DEFINE

4-6

repeat, 8-1
Crossing disk

(letter),

1-6,

exponent

field

2-6
indicator,

Defined

descriptor,

8-8,

8-7,

8-2,

8-7

8-12
editing

between

external

editing
format

internal

form,

complex,
for input

of,

7-1

statement,

5-12
Data

type,

7-2,

7-7

7-17,

2-9,
to

2-3,

array,

8-9

5-3
default

5-1

integer,

2-2,

READ

length

specifier,
list

the

5-2

logical,

2-3,

rank

2-17

of,

real,

2-3,

2-9,

specifications,
5-3

storage

factor

and,

1-2,
8-15

5-1,

5-2

5-3

Declarator

adjustable
array,

array,

2-11,

dimension,

implied,

7-6

4-6,

5-3,

4-7

9-3

9-6

4-8,

7-10

range,

statement,

executions

sign

4-5

the,

4-6

4-6,

7-6,

($),

8-13

8-14,

negative
Dummy

type,

2-17

constant,

6-3,

association

2-9,

integer,
Dummy

array

4-9,

5-8,

6-5

with

arguments,

2-10,

2-6

2-17

arguments,

6-1

2-15,

2-6

operations,

8-23

2-3,

8-8

constant,

2-10

statements,

7-16

list,

conversion,

type,

Declaration

7-16

loop,

8-17

explicit

8-18
Double precision,

Declaration,

7-15

Dollar

1-6

8-16,

7-16

2-16,

2-3

B-5

5-3

:
requirements,

7-15

file default,
control, 4-6

iteration

6-11,

7-11

5-2

5-4,

statement,

unformatted,

2-16

Debugging statement,
Decimal point, 8-15,
scale

2-6

2-10,

'DIRECT' ACCESS value, 9-3
Disconnecting a file, 9-10
DISPOSE (DISP)
keyword, 9-2,

2-16

2-5,

5-3

9-13

unformatted,

5-2

2-3,

elements,

8-22,

statement,

WRITE

2-4

8-21,

6-3

statement,

Disposition,

double precision, 2-3,
by implication, 2-10

8-14

2-11

2-13,

unformatted,

2-12

rules,

8-14

statement,

complex,

2-3, 2-16
declaration statement,

2-19

9-12

2-16

(colon),

9-6

I1/0

2-10, 2-16 to 2-17, 6-2, 7-14
of an arithmetic expression,
of

B-5

Direct access, 7-3
files, 7-1, 9_7,

5-1,

2-8

2-9

9—].,
to

B-5

5-13
2-2

variable,

DIMENSION

8-2

2-11,

B-4

variable,

8-14 to 8-15
and output,

7-12

9-13,

declarator,

8-1

7-10,

Dimension
adjustable, 6-3
in an array, 5-1

and

integer, 2-10
logical, 2-14
magnitude, and G formats,
rounding numeric, 7-9
transfer,

8-17

8-23

Data

transmission

statement,

$ (dollar), 8-13 to
field, 8-2 to 8-18,

8-16

alphanumeric,

data

'DELETE' DISPOSE value, 9_3,
Delimiting periods, 2-7,
Descriptor

8-23

field

for

2-3

FILE

9-12

block boundaries, 9-8

8-18

repeat,

iteration,

7-17

data names, 5-1
field descriptors,

Default

character,

DATA

allocations

Default

3-2

7-4,

8-4

types,

Count

statement,

7-18,

actual

6-1

6-2

6-3
size,

6-3

6- 3
5- 6

2-11

field
8-7,

‘Index-3

descriptor,
8-15,

8-16

8-2,

8-6

INDEX

Editing,

complex

data,

8-14

Extended
using

Editor,

text,

1-6

Elements

arithmetic,

2-14,

2-15

array,

2-10,

2-11,

2-1,

See

also

assigning
base,

Array

values

FORTRAN

Embedded

2-14

list,
1-2,

program,

7-17

7-4,

1-2,

4-11,

all-blank,

B-6

external,

2-18
sign,

Even

byte

boundary,

5-16

statement,

8-9
8-4

8-13

8-12

2-14,

2-15,

4-10

default

logical,
4-4,

2-14,
6-4

mixed-mode,
relational,

statement

subscript,

FIND

2-16,

for

values,

statement,

character
8-18,

Fixed-length
FORTRAN

2-19,

2-17

to 2-12,

2-18

6-5
5-4,

8-17

8-22

8-18

8-23,

8-24

9-1,

9-11

record,

8-24

records,

9-13

Form

8-16

8-13

B-7

2-14

functions,

2-11

8-15,
8-17

a, 7-2 to 7-3
attributes of, 9-1
seqguential organization, 7-2
Filling array elements, 5-12

output,

2-18

1-1
2-14,

8-19

File, 7-2
access to

First

8-23

8-1,

8-10

specification

6-4

complex, 2-18
defined, 2-1, 2-13
in I/0 1list, 7-5

general trules
Field width

Explicit type declaration, 2-10
Exponent, 2-5, 2-6, 8-5
in an external field, 8-23
Exponent field indicator, 8-6
Exponentiation, 2-14, 2-15

L,
0,

4-11

Expressions,
arithmetic,

8-20,

to 8-11, 8-18
8-7 to 8-8, 8-16
8-6 to 8-7, 8-15,

5-4,

8-21
to

external

8-2, 8-5 to 8-6,
8-2, 8-8 to 8-9,
8-11 to 8-12
8-2 to 8-3

Execution
of a condition statement, 4-4
of the DO range, 4-6
of a formatted I/0 statement,

termination

8-23

(zero-length),

F,
G,
H,
I,

1-2,

statements,

8-9

8-23

of,

2-19

5-5

Executable

part

8-10
8-2,
8-2,

9-6

5-5

Exclamation point, 1-3
Executable program, 1-2,

8-5,

A,
D,
E,

7-4,

of operators,
order, 2-16

8-17,

statement label, 1-6
termination, short, 8-19
Field descriptors, 8-2 to 8-18,
8-21, 8-22, 8-23

B-6

to 8-20, 8-23
sequence number, 1-7
statement, 1-7

9-6 to 9-7
specification,
7-3

8-5

1-6

8-22

separators,

3-1
EQUIVALENCE statement, 2—9 7
5-1, 5-8 to 5-10, B-7
.EQV., 2-19
ERR keyword, 9-2, 9—3'

Evaluation
Evaluation

8-3,

fractional
null

condition,

8-2,

2-7

continuation,

statement,

9-7,

8-23

Field

B-6

Error

block, 5-10
9-2, 9_3’

field

8-22
.FALSE.,

B-7

ERR=

keyword,

descriptor,
8-6, 8-15, 8-17
Factor, scale, 8-15

7-3 to 7-4
record, 7-4, 9-12
transfer of control on, 7-4
END= specification, 1-1, 7-4,
statement,

9-7

common

8-2

statement,

ENDFILE

4-8

separators, 8-19 to 8-20, 8-23
EXTERNAL statement, 5-1, 5-11, B-7

1-3

2-20

spaces,

file,

loop,

field
constructing, 8-22 to
procedure names, 5-11

7-5

End-of-file
condition,

.EQ.,
Equal

the

9-7
External

7-18

END

2-15

logical,
ENCODE

Extending
Extending

EXTENDSIZE

elements

range,

to B-2

B-1

operators,

Expression

8-15

7-5

*n,

2-3

r*,

7-11

coding,

1-4

r*c, 7-11
readable character, 7-1
FORM keyword, 9-2, 9-3, 9-7

Index-4

9-12,

INDEX

Format

effect

magnitude

Hollerith constants, 1-1, 1-3, 2-4,
2-7 to 2-9, 7-11, 8-2, 8-11
data type rules for, 2-8 to

data

on,

8-8

reversion,

8-16,

rules

summarized,

for,

8-21

8-22

2-9
Hollerith

data,

5-2,

8-10,

8-20

8-24
run-time,

8-20

8-21

8-23

8-24

8-1,

8-18

Format

control,

Format

separators,

I
to

8-19

See

descriptor, 8-2
Input/Output

statement,

specifiers,

FORMAT

statement,

8-1,

8-22

summary

'FORMATTED'

7-1,

7-8,

8-24,

B-6

FORM

7-3

7-3,

codes,

value,

7-10,

See

also

Implicit
8-24

number,

9-3,

IMPLICIT

I/O0.

Input;

Output

'FORTRAN'

CARRIAGECONTROL

value,

Increment

9-3,

9-6
character

FORTRAN

language

summary,

FORTRAN

library

functions,

B-13

also

set,

1-3,

A-1

B-1

6-9

B-14

Processor-defined

line

formatting,

program

Inner

1-4

elements,

X3.9-1966,

National

allocation,

part

7-1

1-2,

statement,

6-5,

2-3

FORTRAN

8-23

7-1

2-9,

sequential,

B-14.

field

formats,

6-5,

functions
B-2

8-2,

magnitude

on,

2-18

statement,

See

repeat

.GT.,

2-18

(letter),

field

count,

8-8

4-3,

8-17,

7-14

statement;

Input;

7-3

2-15

2-4,

value

negative,

2-4

unsigned,

8-22

an,

2-4

2-10
for,

logical

8-3

operator,

2-19

2-20

B-7

8-23

Output

2-3,

absolute

variables,
Integer

2-10,

operations,

INTEGER*2,

2-3

INTEGER*4,

2-3

Internal

8-11

sequential,

components,

2-2,

rules

7-12

7-16

READ

also

data,

data

2-7

descriptor,

also

constant,

8-9
4-1

access,

7-19.

Integer,

and

.GE.,
Group

7-10

direct

7-7.

8-17

effect

statement

6-9
See also

descriptor,

7-7

7-9

7-5, 8-23
format control interaction with, 8-21 to 8-22
sequential, 7-7

2-2,

6-6,

8-23

devices, 7-3
lists, 7-4 to

6-6

6-7
library,

Input/output

B-8

statement,

8-22

statement

6-4,

Processor-defined

8-9,

7-11
rules,

5-11

subprogram,

4-8

statement

7-15

Functions
B-13

FORMAT

See

8-12

Index-5

7-1

9-7

loop

unformatted

6-9

9-3,

8-19

list-directed,

4-11,

9-7

9-2,

external,

field,

2-1,

arguments,

6-8,

4-5

field,

formatted

6-6

statement,

Function
6-6

4-7
DO

statements,

6-4

record,

constant,

9-1

processor-defined,

FUNCTION

2-10,

7-6

keyword,

loop,

Input,

1-1

the

reference,

2-9,

4-5
allocation,

unformatted

Function

7-1

B-8

lists,

space

also

American

standard,

positioning,

used

5-2,

parameter,

See

8-23
Function

file

unit,

7-8

1-6
parameter,

Initial
Initial

1-3

Fractional

to
DO

Initiator,

1-7

Four-byte

7-3,

INITIALSIZE

functions

FORTRAN

B-8

Indicator,

FORTRAN

type

statement,

Formatted

See

Data

logical

5-1
Implied

FORTRAN

B-7

8-3

numbers, 2-3, 2-6
Implication, data type by, 2-10.

9-7

See

4-4,

Imaginary

Format

Format,

field

I1/0.

6-3
2-17

representation,

2-20,

INDEX

Iteration
DO iteration control, 4-6
Newton-Raphson method, 6-7

LOGICAL*1l,
array,

2-3

5-5,

elements,
COMMON,

'KEEP'

DISPOSE

value,

Keywords in the
9-1 to 9-9

OPEN

9-3,

LOGICAL*2,

9-6

LOGICAL*4,

statement,

Loop.

See

field

Label,

descriptor,

statement,

7-3
field,

list,

1-5,

1-6,

Language

loop

1-2,
4-10

terminating,
3-4
FORTRAN,

Leading

spaces,

Length

specifier,

8-2,

B-1

MAXREC

and

keyword,

A-3

value,
9-2,

9-3,

9-7

9-8

Memory

type,

requirements

types,

5-3
lower-

2-2

4-11

Radix-50

Maximum

8-5

data

1-3

letter,

6-8,

2-18

2-3,

2-3

Main program,
RETURN and,

summary,

Letters,

2-3

2-18

.LT.,

8-10

1-6
1-2

reference,

.LE.,

8-9,

5-5

Lowercase

5-6

assigning

for

data

2-3

Minus, unary, 2-14
Mixed-mode expressions, 1-1
Multibuffered 1/0, 9-6
Multidimensional arrays,
processing, 7-7
Multiplication, 2-14

uppercase,

1-3

Library

functions, FORTRAN,
6-9, B-13 to B-14,
See also Processor-defined
functions

Line,

1-2

comment,

1-6,

continuation
formatting a
List

*n,

1-7

1-7, 1-9
FORTRAN,

4-9

elements,

assigning

1/0,

7-4

simple

DO,

7-6

7-5,

1/0,

1-7

values

to,

See

Input;

7-5

'NEW'

constant,

Logical

data

type,
Logical

Logical

of,

Logical

operators,

B-1

2-7,

9-4,

2-6

9-9

iteration method,

«NOT.,

].-'2,

keyword,

2-9

character,

ASCII,

9-8

7-15

record,

7-11

values, repetition
zero-length field,
2-18

Number
complex,

2-6

2-20,

of, 9-15
8-20, 8-23

2-7

of dimensions in
implicit logical

4-4

2-19

statements,

Null

value,

statement,

value,

9-8

7-11

arrays, 2-10
unit, 7-3,

7-8

B-2

Logical record length, 9-8
Logical unit, 7-1
number, 7-3
Logical values, 2-14, 2-18,
3-3

value,

NOSPANBLOCKS

4-4

5-11

5-1

8-9

Logical

TYPE

Nonexecutable

literals

2-3, 2-16, 3-3
elements, 2-20
expressions, 2-14,

2-19,

5-4

6-7

Output

3-3
Logical

block

'"NONE' CARRIAGECONTROL
9-3, 9-6

Locations. See Storage
Logical assignment statement,

transfer

Names

Newton-Raphson

1/0.

Alphanumeric

9-8

2-11

List-directed output formats
(table), 7-13
Literals, alphanumeric.

See

9-2, 9-3,
block, 5-4

.NE,, 2-18
Negative constants, 2-4,
Nested DO loops, 4-7

8-21

9-6

List-directed

common

external procedure,
symbolic, 2-1, 2-2,

statement label, 1-5, 1-6
'LIST' CARRIAGECONTROL value,

9-3,

keyword,

Named

common

argument,

7-5
implied

1-4

2-3

NAME

logical
of
2-20,

unit,

7-3

blocks,

9-7

statement,

1-6

Numeric data,
Numeric

Index-6

value,

rounding,
2-14

7-9

INDEX

field descriptor, 8-4
Octal integer constant,

Octal

values, transfer
'OLD' TYPE value, 9-4,
One-byte storage area,
One

(1)

character,

2-4
of,
9-9

8-4

See

2-10

OPEN

9-1

9-4,

B-8

keywords,

9-1

Procedure,

Operators

arithmetic,

9-9

B-1

carriage

evaluation

of, 2-19
exponentiation, 2-14
expression, B-1 to B-2

logical,

2-19
of,

relational,
role

of,

unary,

2-19

Order

B-1

7-17
termination

Ordering

2-18,

elements

progression,

rules,

statement,

4-7

1-7

4-8,

loop

sequential,

7-10
list-directed,
unformatted

7-9

Output

WRITE

access,

TYPE

5-16,

3-4,

5-4,

5-5, 6-6,
block data,

6-7,

6-8,

6-9

5-13

5-14

subscript,

also

Parentheses,

2-20,
PAUSE

REWRITE

8-23

field

2-15

7-11,

8-24

READ

2-16,

2-18,

8-17

4-10,

delimiting,

B-9

2-7,

2-18,

See

block
also

Plus

(+)

Plus,

unary,

Point,

character,

See

also

Precision,
also

9-5.
8-18

2-14

8-15,

8-16, 8-23
constant, 2-4.

integer

Precedence,
See

size,

Block

decimal,

Positive

descriptor,

A-2

data

Constant

2-15,

loss
Data

9-3,

A-3
2-17
B-9,

B-10

sequential,

7-7

sequential,
See also Input
Read-only file, 9-6
keyword,

Real

data

Real

constant,

type,

REAL*4,

2-3

Record,

7-2

end

in,
first

9-2,

2-3,

2-5

7-14.

2-5

2- 6

7-11

external,

2-19

character

8-12, 8-13
character of

position

output,

8-18, 8-24
initiator, 8-19
length

Double

value,

8-13

unformatted

of, 7-9.
type storage

requirements;

precision
'PRINT' DISPOSE

2-18,

type,

statement,

READONLY

2-19

Physical

order

list-directed, 7-10 to 7-11
unformatted direct access,
7-15 to 7-16

4-5.

8-1,

1-7

5-12

Radix-50,
Rank,

Constants

statement,

Periods,

4-11

formatted

statements,

See

of,

physical
size of,
9-6

B-9

statement

rules,

1-3

5-5

5-1,

Progression,

statement;

1-2,

5-4,

1-2

structure,

7-14

FORTRAN,

1-2,

Parameter

unit,

7-13

sequential,

format

2-2,

statement,

2-12,

statement;

7-11
function,

PROGRAM

7-16

7-15
See also

2-10,

Program

end

7-12

direct

unformatted

arrays,

main, 1-2, 2-2
terminating, 4-11

7-7

formatted

FORTRAN,

5-12

keyword

Program

executable,

also

control;

6~-9

B-1

2-14

loop,

Output,

of,

Processor-defined

subscript

2-12,

See

2-20,

5-11

9-6.

Carriage

4-5
multidimensional

2-14

.OR.,

control,

also

iterative,

2-15

2-17

arguments,

CARRIAGECONTROL

B-2

precedence

Outer

5-11

subprogram

See

A-1

1-3

names

Processing

2-14,
2-8

assignment,

1-4,

computing,

external,

to
'

Output

characters,

Procedure

9-4

also

Printing

B-9

examples,

B-12,

sequential, 7-9 to
7-10
list-directed, 7-12 to 7-13.

2-3

array,

B-9,

formatted

8-18

One~dimensional
statement,

statement,

B-13

spaces

Index-7

9-8

end
9-8

of,

9-9

8-20

beginning

of,

7-11

7-8

INDEX

(Cont.)
Record,
terminator, 8-1,
transmitting.

See

Input;

SHARED

Output

zero-filled,
RECORDSIZE,

9-4,

9-8

9-9

Reference

array,

2-12,

2-13,

7-5

arguments,

used

function,

6-9

statement

Reference,
Relational
2-17

label,

Relational

3-4

function, 2-1, 6-4
expressions, 2-14,

2-17

count, 8-1
specifications,

8-17, 8-23
Representation,

2-18,

2-20,

8-21

8-16,

format,

8-22
REWIND

statement,

9-1,

attribute,

9-10

Scratch

file,

9-6

TYPE value,

'"SCRATCH'

Separators

external

field,

summary,

8-19

8-1

format
8-18

specification,
to 8-19

7-8,

7-14

FORMAT

separators,

format,

8-1,

8-18

length,

type

2“3'

format,

7-1,

7-3

assignment,

3-1

Statement

9-9
8-20,

ACCEPT,

7-8,

B-10,

B-13

arithmetic

3-2

See

8-1,

ASSIGN,

3-3,

assigned

BACKSPACE,

TO,
9-1,

4-2

4-3

9-11,

B-3
BLOCK

5-14,

DATA,

5-1,

5-13

4-9,

5—11’

6-7'

6_8,

B-3
CALL,

B-3
CLOSE,

Input;

statement,

5-1

data

Specifiers,

Output

READ

B-6

5-3

9-5

Sequential

9-7,

statement

Specifier,

Sequence number field, 1-7
Sequential access, 7-2
Sequential file, 7-2
'SEQUENTIAL' ACCESS value,
1I/0.

also

statements in BLOCK DATA
program units, 5-13

7-11

Sequential

9-6

1-1,

statements,

9-3, 9-6
8-17, 8-22

9-4,

8-23,

8-19

1-7

8-23
field,

slash,

B-7

See

for subptogram arguments, 6-2
Run-time formats, 8-20 to 8-21

'SAVE' DISPOSE value,
Scale factor, 8-15 to

7-4,

1-1, 7-4,
format, 8-1.

END=,

assignment

8-24
ordering,

8-22 to
statement

width,

field

ERR=,

statements, 3-2
for format statements,

1-3

8-24

7-9

Rules

for

8-5

2-10

determining

Rounding numeric data,
Rule, 1/0 list, 8-23

record,

9-1

type,

data

9-11, B-11
See also Output

conversion

8-2,

9-7

5-9

trailing, 8-2
Special characters,
Specification

B-11

Reversion,

beginning

7-11
leading,

Resuming program execution, 4-10
RETURN statement, 4-10, 6-6,
6-8,

4-11

allocation, initial,
character, 8-18

7-1

8-19

8-18,

Space

sharing storage,
Spaces, 7-11
appended, 2-9

group,

internal,

9-9

7-11

line,

Source

B-1

Repeat
Repeat

9-4,

terminator,

record

separator,

2-18

operators,

9-2,

block, 9-5
Slash (/), 7-11, 8-1,
8-19, 8-21
consecutive, 8-19

5-11

processor-defined

keyword,

Sharing storage space, 5-9
Short field termination, 8-19
Signed quantity, 2-4
Simple 1I/0 list, 7-5
Size, dummy array, 6-3
Size, specifying the physical

7-16

9-2,

7-9

Sequential WRITE statement,
to 7-10, 7-16
Shared access, 9-19

8-19

7-7

9-1,

9-10,

COMMON,

2-9,

2-10,

5-5,

5-11,

B-4

computed

Index-8

TO,

B-3

4-2

5-1,

5-4

INDEX

Statement,

(Cont.)

Statement

field,

4-9,

Statement

function,

CONTINUE,
control,

baTa,

4-1

2-9,

5-13,

2-11,

7-4,

7-18,

5-1,

5-12

7-17

9-1,

9-12

9-13,

5-3

5-4,

B-5
ENCODE,

END,

4-6,

7-4,

7-18,

list,

B-5

7-17

ENDFILE,

9-1,

EQUIVALENCE,

FORMAT,

B-6

5-8

5-10,

ment,

7-8,

formatted.

See
6-4,

9-12,

8-1,

B-6

8-22

functions,

2-2,
to

IF,

4-4,

B-7

2-9,

logical

IF,

9-1

4-10,

PRINT,

7-9

7-13,
READ,

B-8

5-2,

B-9
to

7-10

B-9

7-11,

B-10.

9-1,

9-10,

sSTOoP,

4-11,

terminal,
TYPE,
type

6-7

to
to

9-11,

6-8,

7-10,

5-3,

7-12
2-10,

B-1l1

B-11

B-11
to

7-13

See

Input;

5-6

5-8,

7-10,

7-12

7-9

also

TO,

7-15,
WRITE

5-1,

2-1,
Statement

4-4

6-1

governing,

6-2

procedure

name

block

data,

dummy

arguments,

6-1

and,

as,

6-6
in,

6-7

6-6

6-3

Output

6-9

statement,

6-4

B-12

subroutine,

7-13,

6-7

See

also

2-2,

FORTRAN,
conditional,

array

5-7

5-8

SUBROUTINE

B-11

Index-9

6-6
4-9,

6-9.

virtual

6-4

User-supplied

subprograms;

7-7

6-3,

6-7

2-2,

B-13.

5-11

4-11

function,

statement;

execution,

6-3

2-2

supplied

1-7

6-1

rules

user-written,

4-2

B-12,

components,

7-3

arguments,

FUNCTION

output

Statement

unit,

2-15,

using

END

B-12

unformatted.

See

2-7

function,

VIRTUAL,

5-1

2-3

9-9

1-2,

B-13

7-14

defined,

2-3

5-9

arguments

unconditional

WRITE,

sharing,

2-10

type,

6-6

declaration,

5-2

data

program

6-8

5-2

5-3

contiquous,

Subprogram,

4-9

7-9

B-11

2-12

String,

B-11

SUBROUTINE,

5-1,

2-3

Structure,

B-11
REWIND,

array,

actual

6-6,

B-13
2-10,

elements,

l-byte,

units,

statement;

4-10,

1-5

area,

unit,

Input

RETURN,

tab,

space,

B-9

7-8,

8-22

5-3

locations,

B-13

7-16,

rules,

blocks, 5-4
location, 5-8

7-12

5-16,

READ

B-2

requirements,

B-12,

also

5-1

3-3

7-10,

5-1,

7-7

See

9-4,

of,

declaration,

allocating

2-10,

B-9

B-9,

7-15

Stop,

2-10

Storage

4-4

PAUSE,

PROGRAM,

6-6

of
1-2

summary

B-7

arrays,

5-1

type

6-5,

assign-

1-2

specification,

Output

4-3,

for

nonexecutable,

6-8,

5-3, B-8
logical assignment,
OPEN,

rules

summary

Input;
to

subprogram,

8-24

4-1

DATA

3-2

6-6

B-7
GO

1-7

5-14

executable,

B-6

iMpLICIT,

establishing

FUNCTION,

TO,

BLOCK

conversion

5-11,

9-11

7-3,

8-24,

5-13

format,

9-1,

rules,

1-2
assignment,
3-1

B-6

9-12,

5-1,

1-6

Statements,

B-6
FIND,

3-4

ordering

B-6
EXTERNAL,

4-3,

1-2

Statement
to

2-9,

1-6,

1-6

number,

4-11,

1-5,

Statement

B-=5

1-2,

6-6,

1-6

reference,

5-1,

6-5,

7-3

all-zero,

B-5

4-5

label,

6-8,
field,

FILE,

DIMENSION,

DO,

Statement

B-4

DEFINE

1-7

B-2

B-4

DECODE,

B-4

System-

subprograms

referencing,

statement,

6-7

6-8,

INDEX

operators, 2-14
Unconditional GO TO statement,

Subroutine subprogram, 4-9,

Unary

6-9

6-7

Subscript expressions, 2-11 to

4-2

Subscript progression, 5-12, 7-5
Subscript order,

2-14

Subtraction,

9-7

Unit
logical,

FORTRAN language, B-1

generic function names, 6-14

of rules for format statements,
8-24

of statements, B-2 to B-13

Suspending program execution, 4-10
2-11

2-2

classes,

2-9 to

2-2,

2-1,

Symbolic name,

types of,

defining data

5-2 to

overriding length attributes
5-3

statement,

separator,

value

Terminal

7-11

4-9

Terminating program execution, 4-11
Termination of processing, 7-11
Termination, short field, 8-19
Terminator, record, 8-1, 8-19
8-18

(/),

slash

1-6

editor,

Text

to 8-19

spaces,

8-2

Transfer of control,

6-5, 6-6, 6-8

Transmitting records.
See

7-6

by implication, data,

value

Data

assigning
2-20,

4-5

TYPE statement, B-11 to B-13
formatted sequential,
7-10

7-9

list-directed 7-12 to 7-13
also

the,

2-1,

2-9

2-2,

2-11

assigning

2-10

type

B-18

variables,

declaration, explicit, 2-10
specification, data, 2-10
Type declaration statement, 2-10,
5-1, 5-2 to 5-3, B-11 to B-12
TYPE keyword, 9-2, 9-4, 9-9

See

3-1

to variables and

in dimension declarators,

Type

7-1l1

array elements, 5-12
default field descriptors,
8-17 to 8-18

control,

Truncated constant, 2-9
Two-dimensional array, 2-10
See

Values

Variable

True, 2-7
.TRUE., 2-7

data.

Value separators,

3-3

Output

Input;

6-3

2-9
A-3

initial, 5-14
logical, 2-14, 2-18,

Transmitting array elements by
lists,

of the variable,
Radix-50, A-2 to

in common blocks,

also GO TO statements

implied DO

in form r*c, 7-11
integer, 2-20
null, 7-11
numeric, 2-14

assigning

7-4.

See

2-12

6-4

assigning,

Three-dimensional arrays, 2-10
Trailing

2-13,

8-5

4-5

loop,

8-22

Unsubscripted array, 2-12
4-9, 5-14, 7-5
Unsubscripted array name,

complex, 8-14 to 8-15
constant, 7-11
evaluating magnitude of,

T field descriptor, 8-12 to 8-13
TAB character, 1-5 to 1-6
Tab

7-2

Value

library