Digital PDFs

DEC-S8-OS78A-A-D

December 1977

381 pages

Original

15MB

Document:	o F020080 0S/78 User’s Manual
Order Number:	DEC-S8-OS78A-A-D
Revision:
Pages:	381
Original Filename:	https://svn.so-much-stuff.com/svn/trunk/pdp8/src/dec/dec-s8-os78a/dec-s8-os78a-a-d.pdf

OCR Text

08/78
User’s Manual
Order No. DEC-SS—OS78A-A-D

08/78
User’s Manual
Order No. DEC-SB—OS78A-A-D

September 1977
This document is the user's manual for the 05/78 V1

operating system. Its purpose

is to acquaint new users with the operating system designed for the DECstation 78

minicomputer. This manual presents the background material for getting on the air,
and a detailed description of the 08/78 commands that are used to direct computer

operations. It also describes the 08/78 Editor, PAL8 assembly language, and the two

high-level languages supported by the system, BASIC and FORTRAN IV. Command
examples and demonstration programs are contained throughout the manual.

SUPERSESSION/UPDATE INFORMATION:

This is a new manual.

OPERATING SYSTEM AND VERSION:

08/78 V1

To order additional copies of this document, contact the Software Distribution

Center, Digital Equipment Corporation, Maynard, Massachusetts 01754.

digital equipment corporation mognard, mossochusetts
.

First Printing, September 1977
The information in this document is subject to change without notice and should not be construed as a commitment

by Digital Equipment Corporation. Digital Equipment Corporation assumes no responsibility for any errors

that may appear in this document.
The software described in this document is furnished under a license and may be used or copied only in accordance
with the terms of such license.

Digital Equipment Corporation assumes no responsibility for the use or reliability of its software on equipment
that is not supplied by DIGITAL.

The postage

prepaid READER’S COMMENTS form on the last page of this document requests the user’s critical

evaluation to assist us in preparing future documentation.

The following are trademarks of Digital Equipment Corporation:

DIGITAL

DECsystem-l 0

DEC

DECtape

OMNIBUS

PDP

DIBOL

DECUS

EDUSYSTEM

OS/ 8
OS/ 78

UNIBUS

FLIP CHIP

PHA

COMPUTER LABS

FOCAL

RSTS

COMTEX

INDAC

RSX

DDT

LABS

TYPESET-S

DECCOMM

DECSYSTEM-20

TYPESET~ 10

MASSBUS

TYPESET-l 1

CONTENTS
Page
PREFACE
CHAPTER

CHAPTER

...................................................................

INTRODUCTION

1.1

DECSTATION 78 MINICOMPUTER SYSTEM

1.2

08/78 SOFTWARE

GETTING ON THE AIR

2.1

STARTING THE SYSTEM

...............................................

............................

..............................................

...........................................

1-1
l-1
1-1

2-1

.....................

2—1

........................

2-1

...................

2.1.1

Loading the Diskettes

2.1.2

Calling the Keyboard Monitor

2.1.3

Using the Keyboard

2.1.4

.....................................

2-1

...........................................

2-1

...........................

2-2

........................................

2—3

Typing Convention and Symbology

2.2

MAKING A BACKUP COPY

2.3

DEMONSTRATION PROGRAMS

2.4

NAMING DEVICES AND FILES

......

....................................

.....................................................

2-3

Devices

File Names and Extensions

.......................................

ENTERING MONITOR COMMANDS

..................................

2.51

08/78 Command Format

2.5.2

Incorrect Commands

2.5.3

Correcting Errors

2.5.4

2.5.6.2

Using Input/Output Options
Remembering Previous Arguments
Using Wildcards
Wildcard Input File Specifications
Wildcard Output File Specifications

2.5.7

Indirect Commands (@ Construction)

2.5.8

Asterisk (*)

2.6

USING DEFAULTS

2.7

WHERE TO GET MORE INFORMATION

2.8

FILES AND FILE-STRUCTURED DEVICES

2 .5 .6

2.5.6.1

........................................

...........................................

............................

..................

..............................................

2—4
2—4

2-5
2-6

2-7
2-7
2-8
2-9

..................................

2-10

.................................

2-10

................................

2-1 1

.................................................

2-11

..............................................

...............................

.............................

2—12
2—12
2-13

.........................................

2-13

..............................................

2—14

File—Structured Devices

2.8.2

File Directories

2.8.3

File Types

2.8.4

File Formats

..................................................

................................................

2.9

08/78 COMMAND SUMMARY

08/78 COMMANDS

3.1

ASSIGN COMMAND

3.2

BASIC COMMAND

3.3

COMPARE COMMAND

3.3.2

.................

..................................

2.8.1

3.3.1

2-3
2-3

2.4.2

2.5.5

.....................................

2.4.1
2.5

CHAPTER

..................

......................................

..............................................

.............................................

..............................................

...........................................

Comparing Part of a Line
Making Changes While Using the COMPARE Command

........................................

3.4

COMPILE COMMAND

3.5

COPY COMMAND

3.5.1

File Transfer

3.5.2

Examples of Copy Commands

3.5.3

Predeletion

....................

............................................

...............................................

................................................

.....................................

.................................................

iii

2—14

2-14
2—15

3-1
3—2
3—3
34
3—5

3—5
3—7

3-8
3-8
3—8
3—9

CONTENTS (CONT.)
Page
3.5.4

Postdeletion

3.5.5

Considering the Date on a Transfer

3.5.6

File Protection During a Transfer Operation

3.5.7

Terminating Copy Operations

................................................

3-11

...............................................

3—15

..............................................

3-17

CREF COMMAND

3.8

DATE COMMAND

3.9

DEASSIGN COMMAND

3.10

DELETE COMMAND

3.11

DIRECT COMMAND

3.12.1

3.12.2
3.12.3
3.12.4

3-10

3-14

3.7

3.12

3-9

............................................

3.6

3.11.2

...........................

3-9

.....................................

CREATE COMMAND

3.11.1

.................................

...........................................

3-18

.............................................

3-19

.............................................

Choosing the Directory Format

3-22

....................................

3-22

DUPLICATE COMMAND
Changing Devices Before and After Executing the DUPLICATE Command

..........................................

Performing a Read Check
Transfer Without Checking for Identical Contents
Check for Identical Contents Without Transferring

3.13

EDIT COMMAND

3.14

EXECUTE COMMAND

3.15

GET COMMAND

3.16

HELP COMMAND

3.17

LIST COMMAND

3.18

LOAD COMMAND

.......

3-24

........................

3-24

.......................

3-25

...........................................

................................................

...............................................

..............................................

3.18.1

Absolute Binary Files

3.18.2
3.18.3

Loading Programs in Specified Areas
Editing or Patching a SAVE File

3.18.4

Executive After Loading

3.18.5

Clearing Memory After Loading

3.18.6

Relocatable FORTRAN Binary Files

..........................................

................................

...................................

MAP COMMAND

3.20

MEMORY COMMAND

3.21

ODT COMMAND

3.22

PAL COMMAND

3.23

R COMMAND

3-24

........................................

...............................................

3.19

3-21

............................

Considering the Date in a Directory Listing

........................................

...................................

........

........................

...............................................

............................................

...............................................

3—26
3-28

3-29
3-30
3-32

3-34
3-34
3-34
3-35
3-35
3-35

3-35
3-36
3-39
3-40

................................................

341

..................................................

3-42

3.24

RENAME COMMAND

............................................

343

3.25

RUN COMMAND

...............................................

3-44

3.26

SAVE COMMAND

.............................

3.27

SET COMMAND

3.27.1

ECHO

‘
..................

................................................

345

3—48

3.27.3

3-48
ARROW ................................................... 3-48
SCOPE .................................................... 3—49

3.27.4

WIDTH

3.27.5

HEIGHT m

3.27.6

PAGE ..................................................... 3-50
ESC ...................................................... 3-50

3.27.2

3.27.7
3.27.8

....................................................

.................................................

3-49

.................................................

3-50

3.27.9

PAUSE .................................................... 3-50
READ ONLY ............................................... 3-50

3.27.10

COL n

....................................................

3—51

CONTENTS (CONT)
Page
3.27.11

3.27.12

INIT xxxxx

......................................................

................................................

3.28

START COMMAND

3.29

SQUISH COMMAND

3.30

SUBMIT COMMAND

3.30.1
3.30.2

3-51

.............................................

3-53

............................................

3-54

............................................

3-55

.........................

3-55

..................................................

3-56

Processing and Terminating a Batch Input File
Spooling

.........................................

3-57

..............................................

3-5 8

3.31

TERMINATE COMMAND

3.32

TYPE COMMAND

3.33

UA, UB, UC COMMANDS

3.34

ZERO COMMANDS

.........................................

3-60

.............................................

3—61

..........................

4-1

...............................................

4-1

................................................

4-1

................................................

4-1

CREATING AND EDITING YOUR PROGRAM

CHAPTER

3-51

4.1

INTRODUCTION

4.2

PAGE MAKEUP

4.3

TEXT BUFFER

4.4

CALLING THE EDITOR

..........................................

4-2

.........................................

4-2

4.5

MODES OF OPERATION

4.6

EDITOR KEY COMMANDS

4.7

SPECIAL CHARACTERS

........................................

4-2

.........................................

4-3

...............................................

4.7.1

DELETE Key

4.7.2

Period (.)

4.7.3

Slash (/)

4.7.4

LINE FEED Key

4.7.5

Right-Angle Bracket (>)

4.7.6

Left-Angle Bracket (<)

..................................................

4-3

...................................................

4-4

.............................................

4-4

........................................

4—4

.........................................

4-4

...............................................

4—4

4.7.7

Equal Sign (=)

4.7.8

Colon (z)

4.7.9

Tabulation (TAB)

4.7.10

ESCape Key

4.7.11

Lower Case Characters

..................................................

............................................

................................................

4-5
4—5
4—5

.........................................

4-5

...........................................

4-5

.............................................

4-5

4.8

EDITOR COMMANDS

4.8.1

Input Commands

4.8.2

List Commands

4.8.3

Output Commands

4.8.4

Editing Commands

4.8.5

Search Commands

4.9

CREATING A NEW FILE

4.10

PAGING FILES

4.11

SEARCHING THROUGH FILES

..............................................

............................................

.........................................

................................................

....................................

4.11.1

Single Character Search

4.11.1.1

nS Command

4.11.1.2

m,nS Command

4.11.1.3

8 Command

4.11.2

Character String Search

4.11.2.1

IntraBuffer Character String Search

4.11.2.2

InterBuffer Character String Search

.........................................

...............................................

..............................................

................................................

.........................................

.................................

4.12

EDITING AN EXAMPLE PROGRAM

4.13

EDITOR ERROR MESSAGES

4.14

SUMMARY OF EDITOR COMMANDS AND SPECIAL CHARACTERS

.................................

......................................

...........

4-6
4-7

4-9
4-9

4-10
4-10

4—11
4—11
4—11

4—1 1
4—1 1
4—1 2

4-12
4-14

4-15
4-18
4—19

CONTENTS(CONTJ
Page
CHAPTER

..........................................

5-1

...............................................

5-l

THE PAL8 ASSEMBLER

INTRODUCTION

CREATING AND RUNNING AN ASSEMBLY PROGRAM

u2
5m
m4

am
5%

54
54

..........................................

5-5

....................................

5-6

.........................................

5-6

...........................

5-7

.......................................

5-7

................................................

5-8

PAL8 OPTIONS

CHARACTER SET

STATEMENTS

Labels

Instructions

Operands

Comments

...................

............................................

Creating Program
Assembling a Program
Loading and Saving a Program
Executing the Program
Getting and Using a Cross-Reference Listing
Obtaining a Memory Map
a

..............................................

5-10

....................................................

5-10

................................................

5-10

..................................................

5-10

.................................................

FORMAT CHARACTERS

.........................................

Form Feed

Tab

Statement Terminators

.................................................

......................................................

NUMBERS

SYMBOLS

.........................................

...................................................

Permanent Symbols

User-Defined Symbols

Current Location Counter

Symbol Table

Direct Assignment Statements

Symbolic Instructions
Symbolic Operands

w
5%

wz
war
5%2
53m
53M
5%5
5m

...........................................

.........................................

.......................................

...............................................

....................................

..........................................

...........................................

EXPRESSIONS

................................................

Operators
Special Characters
Dot (.)
Double Quote (“)
Parentheses ( ) and Brackets [ ]
Angle Brackets (< >)
Dollar Sign (S)

..................................................

............................................

....................................................

............................................

...................................

..........................................

..............................................

INSTRUCTION SET

.............................................

Memory Reference Instructions

5m2

Microinstructions

sma

Autoindexing

...................................

.............................................

...............................................

PSEUDO-OPERATORS

...........................................

Indirect and Page Zero Addressing

5M2

Extended Memory

ans

Resetting the Location Counter
Reserving Memory

sn4

.................................

............................................

...................................

............................................

Relocation Pseudo-Operator

sue

Suppressing the Listing
Controlling Page Format

5U]

5-9

................................................

.....................................

.........................................

........................................

5-10
5-10
5-10

5-11
5-1 1
5-11
5-12

5-12
5-12
5-12
5-13
5-14

5-15
5-15
5-15
5-15
5-18
5-19

5-19
5-19
5-20
5—2l

5-21
5-21
5-22
5-25

5-25
5-25

5-26
5—26
5-27
5-27

5-27
5-28

CONTENTS (CONT)
Page
5.11.8

Altering the Permanent Symbol Table

5.11.9

Conditional Assembly Pseudo- Operators

5.11.10

Radix Control

5.11.11
5.11.12

Entering Text Strings
End-of—File Signal

5.11.13

Use of DEVICE and FILENAME Pseudo-Operators

5-29

...............................................

5 29

..........................................

5-30

............................................

5-30

LINK GENERATION AND STORAGE

5.13

TERMINATING ASSEMBLY

5.14

ERROR MESSAGES

5.15

PALS PERMANENT SYMBOL TABLE
BASIC

......................

................................

.......................................

.............................................

................................

......................................................

6.1

INTRODUCTION

6.2

MAJOR COMPONENTS OF OS/78 BASIC

6.3

BASIC INSTRUCTION REPERTOIRE

6.4

CALLING BASIC

6.5

BASIC EDITOR COMMANDS

6.5.1

Using the BASIC Editor

6.5.2

BASIC Editor Commands

6.5.2.1

NEW Command

6.5.2.2

OLD Command

6.5.2.3

LIST/LISTNI'I Commands

6.5.2.4

SAVE Command

6.5.2.5

RUN/RUNNH Commands

6.5.2.6

NAME Command

6.5.2.7

SCRATCH Command

6.5.2.8

BYE Command

6.5.3

BASIC Control Keys

6.5.3.1

Correcting Typing and Format Errors (DELETE, CTRL/U)

6.5.3.2

Eliminating and Substituting Program Lines (RETURN)

6.5.3.3

Interrupting Program Execution

6.5.3.4

Controlling Program Listings at the Terminal Console

6.5.4

(CTRL/S, CTRL/C and CTRL/O)
Resequencing Programs (RESEQ)

6.6

...............................................

........................................

.......................................

............................................

.....

.....................................

..............................................

..........................................

6.6.5.2

Simple Variables
Subscripted Variables
Arithmetic Operations
Operators
Priority

6.6.5.3

Parentheses

.................

...................

...................................

..................................

DATA FORMATS ACCEPTABLE TO BASIC

6.6.3

............................

..........................................

..................................................

.............................................

..........................................

.........................................

..................................................

...................................................

.................................................

6.6.5.4

Rules for Exponentiation

6.6.5.5

Relational Operators

6.6.6

6.6.6.3

String Information
String Character Set
String Conventions
String Concatenation

6.6.7

Format Control Characters

6.6.8

Files

6.6.6.2

.......................................

.............................................

Numbers

6.6.6.1

......................................

..............................................

6.6.2

6.6.5.1

.................................

..............................................

Numeric Information

6.6.5

..............................

...............................................

6.6.1

6.6.4

5—28

.............................

5.12

CHAPTER

...............................

.......................................

..........................................

............................................

...........................................

.........

...................................

..........................................

......................................

.....................................................

vii

5-30
5-31

5-32
5-32
5-34

6-1
6-1
6-1

6—1
6-2

6-2
6-2

6-2
6-2
6-4
6-4
6-5
6—6
6-6

6-6
6-6

6—7
6-7
6-7

6-7
6-7
6-7

6-8
6—8
68
6-9
6—10
6-10

6-10
6-10
6-1 1

6-11
6-12
6-12
6-12
6 -13
6-14

6-14
6-15

CONTENTS (CONT.)
Page
6.7

BASIC STATEMENTS

...........................................

6—15

..............................

6-15

.........................................

6-16

6.7.1

Statement Line Numbers (Sequencing)

6.7.2

The PRINT Statement

6.7.3

Information Entry Statements (DATA, READ)

6.7.3.1

DATA Statement Format

6.7.3.2

READ Statement Format

.........................

6—18

.......................................

6—18

.......................................

6-18

..............................................

6-20

6.7.4

LET Statement

6.7.5

Loops (FOR and NEXT Statement)

6.7.5.1

FOR Statement Format

6.7.5.2

NEXT Statement Format

6.7.6

Control Statements (GOTO, IF-THEN, GOSUB, RETURN)

6.7.6.1
6.7.6.2
6.7.6.3

Branch to Subroutine (GOSUB Statements)

6.7.6.4
6.7.7

Return from Subroutine (RETURN Statement)
Program Termination Statements (END, STOP)

6.7.7.1

END Statement Format

................................

6-21

........................................

6-21

.......................................

6—24

.................

6—24

Unconditional Branch (GOTO Statement)

............................

6-24

Conditional Branch (IF-THEN Statement)

............................

...........................

.........................

........................................

6-25
6-26
6-28
6-28
6-28
6-28

6.7.7.2

STOP Statement Format

6.7.8

The INPUT Statement

6.7.9

The REMark Statement

6.7.10

Ancillary Statements (DIMension, RESTORE, DEFine,
RANDOMIZE and CHAIN)

6—31

6.7.10.1

DIMension Statement Format

6-31

6.7.10.2

RESTORE Statement

6.7.10.3

DEFine Statement

6.7.10.4

RANDOMIZE Statement

6.7.10.5

CHAIN Statement

6.7.11

File Handling Statements

6.7.11.1
6.7.11.3

FILE# Statement
PRINT# Statement
INPUT# Statement

6.7.11.4

RESTORE# Statement

6.7.11.5

CLOSE# Statement

6.7.11.6

IF END# Statement

........................................

.........................................

........................................

......................................

6.7.11.2

6.8

BASIC FUNCTIONS

....................................

..........................................

............................................

.......................................

............................................

.......................................

............................................

...........................................

.........................................

...........................................

.............................................

6.8.1

Arithmetic Functions (ABS, INT, EXP, RND, SGN, SQR)

6.8.1.1

BASIC ABS Function

6.8.1.2

BASIC EXP Function

..................

..........................................

6—28
6-30

6-33
6—34

6—35
6-36
6-36

6-36
6-37
6—39

6-40
6-41

6—42
642
6-43
6-43
6-43

6.8.1.3

BASIC INT Function

6.8.1.4

BASIC RND Function

6.8.1.5

BASIC SGN Function

6.8.1.6

BASIC SQR Function

6.8.2

Trigonometric Functions (ATN, COS, LOG, SIN)

6.8.2.1

BASIC ATN Function

6.8.2.2

BASIC COS Function

6.8.2.3

BASIC LOG Function

6.8.2.4

BASIC SIN Function

6.8.3

String Handling Functions (ASC, CHP$, DATS, LEN, POS,
SECS, STRS, TAB, VAL)

6-45

6.8.3.1

BASIC ASC Function

6-45

..........................................

.........................................

..........................................

.......................

.........................................

..........................................

.........................................

..........................................

.......................................

..........................................

viii

6—43
6-43
6-44
6-44

6-45
645
645
6-45

6-45

CONTENTS (CONT)
Page
.........................................

6-47

.........................................

6—47

..........................................

6-48

6.8.3.2

BASIC CHR$ Function

6.8.3.3

BASIC DAT$ Function

Esm

BASIC LEN Function

6.8.3.5

BASIC POS Function

6.8.3.6

BASIC SEG$ Function

6.8.3.7

BASIC STRS Function

..........................................

.........................................

6.8.3.8

BASIC TAB Function

6.8.3.9

BASIC VAL Function

6.8.4

Display Console Control Function (PNT)

6.8.5

Trace Function

..........................................

.........................................

.............................

..............................................

6.9

SUMMARY OF BASIC EDITOR COMMANDS

6.10

SUMMARY OF BASIC STATEMENTS

6.11

SUMMARY OF BASIC FUNCTIONS

6.12

BASIC ERROR MESSAGES

7.1.2

Using the EXECUTE Command

7.1.3
7.1.4

Compiling
Loading

7.1.5

Run—Time System

.............................................

...................................

.................................................

...................................................

............................................

System Library
08/78 FORTRAN IV OPERATION
FORTRAN IV Compiler
Using the COMPILE Command
Examples of Compilation
Compiler Error Messages

..............................................

...................................

........................................

...................................

.......................................

........................................

7.2.2

FORTRAN IV Loader

7.2.2.1

Examples of Loading

7.2.2.2

Loader Error Messages

7.2.3

FORTRAN IV Run-Time System (F RTS)

7.2.3.1

Run-Time System Error Messages

7.2.4

FORTRAN IV Library: Library Functions and Subroutines

7.3
7.3.1
7.3.2

7.3.3
7.3.4
7.3.5
7.4

.........................................

..........................................

.........................................

Executing a Program
Compiling a Program
Creating a Loader Image File
Running Subprograms
Specifying I/O Devices

............................

..................................

RUNNING OS/ 78 FORTRAN IV PROGRAMS

.................

...........................

..........................................

.....................................

.........................................

....................................

FORTRAN IV SOURCE LANGUAGE

7.4.1

Constants

7.4.1.1

Integer Constants

7.4.1.2

Real Constants

7.4.1.3

Octal Constants

7.4.1.4

Logical Constants

7.4.1.5

Hollerith Constants

7.4.2

Variables

6—57

7-1

Source Programs

7.2.1.3

..................................

7-1

7.1.1

7.2.1.2

6-55

............................................

SYSTEM OVERVIEW

7.2.1.1

6—53

............................................

7.1

7.2.1

6-53

6-57

Run-Time System Error Messages

7.2

6-51
6-52

6-57

6.12.2

7.1.6

6-50

........................................

Compiler Error Messages

OS/78 FORTRAN IV

..................................

6—49
6-49

........................................

6.12.1

CHAPTER

...........................

................................

6~48

6-48

.................................

..................................................

.............................................

..............................................

............................................

...........................................

..................................................

7—1
7-1

7—2
7—2
74
74
7-4

7—5
7-5

7-6
7-6

7-8
7-9
7-9
7-10
7-15
7-16
7-22
7-22
7-23
7-24
7—24

7-27
7-27

7-28
7-28
7-28
7-29
7-29

7-29
7-29

CONTENTS (CONT)
Page
7.4.2.1

....................................................

7—30

.................................................

7-30

................................................

7-30

7.4.3

Arrays
Subscripts
Expressions

7.4.3.1

Arithmetic Expressions

7.4.3.2

Logical Expressions

7.4.4

Assignment Statements

7.4.4.1

Arithmetic Statements

7.4.4.2

The GO TO Assignment Statement

7.4.5

Control Statements

7.4.5.1

GO TO Statements

7.4.5 .2

IF Statements

7.4.5.3

DO Statement

7.4.5.4

CONTINUE Statement

7.4.5.5

PAUSE Statement

7.4.5.6

STOP Statement

7.4.5.7

END Statement

7.4.6

Data Transmission Statements

7.4.6.1

FORMAT Statement

7.4.6.2

DEFINE FILE Statement

7.4.6.3

Input/Output Statements

7.4.6.4

Device Control Statements

7.4.7

Specification Statements

7.4.7.1

Storage Specification Statements Dimension Statement

7.4.7.2

Type Declaration Statements

7.4.8

Subprogram Statements

7.4.8.1

Functions

7.4.8.2

FUNCTION Statements

7.4.8.3

Subroutine Subprograms

7.4.8.4

RETURN Statement

7.4.8.5

BLOCK DATA Statement

7.4.8.6

EXTERNAL Statement

7.4.9

08/78 FORTRAN IV Statement Summary

7.4.2.2

CHAPTER

BATCH

.........................................

7-30

...........................................

7-32

.........................................

7-34

.........................................

7-34

.................................

7-36

...........................................

7-36

............................................

7-36

...............................................

7-38

...............................................

7-39

.........................................

............................................

7-42

..............................................

7-42

....................................

7-42

..........................................

7-42

.......................................

7-54
7-54

...................

.....................................

........................................

..........................................

.......................................

........................................

............................

.....................................................

8.2

BATCH MONITOR COMMANDS

8.3

THE BATCH INPUT FILE

8.4

BATCH ERROR MESSAGES

8.5

RESTRICTIONS UNDER 08/ 78 BATCH

.................................

....................................

.........................................

DEBUGGING A PROGRAM
9.1

ODT FEATURES

9.2

CALLING AND USING ODT

9.3

ODT COMMANDS

.......................................

...............................

.......................................

...............................................

.......................................

..............................................

9.3.1.1

Special Characters
Slash(/) Open This Location

9.3.1.2

RETURN

9.3.1.3

LINE FEED

9.3.1.4

Circumflex (A)

............................................

Close Location
'4

....................................

......................................

Close Location, Open Next Location

.....................

7-54

7-59
7-59
7-59

7-60
7-61
7-62
7-62
763
7-63

8-1
8-1
8-2

8-4
8-6
8-7

9-1

9—1
9-1

9-2
9-2
9—2

9-3
9-3

Close Location, Take Contents as Memory

—

Reference Instruction and Open Effective Address
9.3.1.5

Underline (_)

9.3.2

Illegal Characters

7-50

........................................

..................................................

7—50

......................................

BATCH PROCESSING UNDER OS/78

9.3.1

7-42

.............................................

8.1

CHAPTER

7-41

Close Location, Open Indirectly

.......................

........................

.............................................

9-4
94
9-4

CONTENTS (CONT)
Page
9.3.3

Control Commands

9.3.3.1

fnnnnG

9.3.3.2

fnnnnB

9.3.3.3

9.3.3.4

9.3.3.5

9.3.3.6

9.3.3.7

nnnnC

9.3.3.8

—

...........................................

Transfer Control to Program at Location nnnn of field f

Remove Breakpoint

9—5

......................................

9—4

........................................

9-5

..................................

9-5

Continue nnnn +1 Times from Breakpoint

......................

9-6

........................................

9-6

.........................................

9-6

...........................

9-6

..................

9-6

.........................................

9-6

....................................................

9-7

nnnnW

—

Word Search

9.4

ERRORS

9.5

PROGRAMMING NOTES

9.6

ODT COMMAND SUMMARY

.........................................

......................................

APPENDIX A

CHARACTER/CONTROL CODES AND SPECIAL SYMBOLS

APPENDIX B

USEFUL MATHEMATICAL SUBROUTINES

............................

B-l

UNSIGNED FRACTIONAL MULTIPLICATION SUBROUTINE

UNSIGNED INTEGER DIVISION SUBROUTINE

UNSIGNED FRACTIONAL DIVISION SUBROUTINE

.........................

C-l

............................................

................................

..............................................

C.2.1

FETCH Device Handler Function Code

C.2.2

LOOKUP Permanent File Function Code

C.2.3

ENTER Output (Tentative) File Function Code

C.2.4

The CLOSE Function

C.2.5

Call Command Decoder (DECODE)

Function Code = 5

C.2.6

CHAIN Function

C.2.7

Signal User ERROR

C.2.8

Lock USR In Memory (USRIN)

Function Code

..........................

C.2.1l

RESET System Tables

Function Code

.....................

............................

...................

.............................

Function Code

Dismiss USR From Memory (USROUT)

....................

Function Code

Function Code
13

1 l
=

...............

..............

...........................

.......................................

D.1

COMMAND DECODER CONVENTIONS

COMMAND DECODER ERROR MESSAGES

CALLING THE COMMAND DECODER

COMMAND DECODER TABLES

Input Files

...............................

Ascertain Device Information (INQUIRE)

Output Files

...........................

C.2.10

THE COMMAND DECODER

C.2.9

D.4.2

B-4

............................................

Direct and Indirect Calling Sequence

D.4.1

B-3

C-l

C.1.2

APPENDIX D

B-2

........................................

Cl .1

USR FUNCTIONS

...............

B-5

Standard USR Call

C.2

...................

......................

USER SERVICE ROUTINE
CALLING THE USR

9-7
A-l

UNSIGNED INTEGER MULTIPLICATION SUBROUTINE

9-7

.................

B.1

APPENDIX C

..........................................

9.3.3.13

9.3.3.11

9—4

........................................

9.3.3.12

9.3.3.10

9-4
9-4

Open C(AC) Location
Open C(L) Location
Continue From a Breakpoint
~

............

.....................

Set Breakpoint at Location nnnn of field f

—

Open Data Field
F
Open Current Field
M
Open Search Mask
M LINE FEED
Open Lower Search Limit
M LINE FEED LINE FEED
Open Upper Search Limit

9.3.3.9

...............................

............................

................................

....................................

................................................

.................................................

C-1
C-3
C—3
C—3
C-5

C-6
C-7
C-8

C—9
C—lO
C-10
C—11

C—ll
C-12
D-l

D-l
D—2
D-2
D-3
D-3
D—4

CONTENTS (CONT)
Page
D.4.3

Command Decoder Option Table

D.4.4

Example

D.5
D.5.l
D.5.2

..................................

..................................................

SPECIAL MODE OF THE COMMAND DECODER

Calling the Command Decoder Special Mode
Operation of the Command Decoder in Special Mode

...........................

F-l

USING DEVICE HANDLERS

CALLING DEVICE HANDLERS

F.2

OS/78 DEVICE HANDLERS
Line Printer (LPT, LQP)

INDEX

....................................

F~3

........................................

F~3

File—structured Devices (SYS, DSK, RXAO, RXAl, RXA2, RXA3)

F.2.3

Terminal Handlers (TTY, SLU2a SLU3)

F.2.4

Multiple Input Files

G-2

ERROR MESSAGES

.....

........

............

F-3

.............................

F-4

...........................................

F—4

OS/78 ERROR MESSAGE SUMMARY
SYSTEM HALTS

F-l

.......................................

F.2.2

G-l

D-7

.......................................

APPENDIX F

APPENDIX G

D-7

E-l

OS/78 DEFAULT FILE NAME EXTENSIONS

GLOSSARY

.....................

D—S
D-7

...........................

APPENDIX E

F.2.l

.........................

D-S

................................

...............................................

.............................................

.........................................................

G-l
G-l
G-2

Glossary-1

...........................................................

Index-1

FIGURES
FIGURE

6—1

BASIC Editor Commands and Related Uses

7-1

Creating a FORTRAN IV Program
Executing a FORTRAN IV Program
Compiling a FORTRAN IV Program
Loading and Executing a FORTRAN IV Program
Main and Subprogram for Calculating the Volume of a Regular Polyhedron
Nested DO Loops

7-2

7-3
7-4
7-5

7-6

..............................

....................................

...................................

..................................

..........................

.........

...............................................

6-3

7-1
7—2

7-3
7-3
7-25

7-40

TABLES
TABLE

2-1
2-2
2—3
3-1

3-2
3-3
3-4

3-5

OS/ 78 Assumed Extensions
CCL Options
08/ 78 Command Summary
COMPARE Options
COMPARE Error Messages
COPY Options
COPY Error Messages
CREF Options

........................................

..................................................

........................................

..............................................

.........................................

2-5

2-9
2-15
3—6
3—6

.................................................

3-1 1

............................................

3—13

.................................................

3-15

xii

TABLES (CONT)
Page
3-6

CREF Error Messages

3-7

DELETE Options

............................................

...............................................

3-16
3-20

..........................................

3-20

...............................................

3-23

3-8

DELETE Error Messages

3-9

DIRECT Options

3-10

DIRECT Error Messages

3—11

DUPLICATE Options

3-12

DUPLICATE Error Messages

3-13

HELP Error Messages

............................................

3-23
3-25

.........................................

3-25

.............................................

3—30

..........................................

3-31

...................................................

3-33

3—14

Available 08/78 Help Files

3—15

LIST Options

3—16

LIST Error Messages

3—17

MAP Options

3-18

MAP Error Messages

..............................................

3-33

...................................................

3-38

..............................................

3-38

..............................................

3—43

................................................

3-45

3-19

RENAME Options

3-20

Job Status Word

3-21

SET Command Attributes

3-22

4-2

SETErrorMessages.............................................: 3-52
3-56
SUBMIT Options
3-58
TYPE Options
TYPE Error Messages
3-59
4-2
Editor Options
4—3
Editor Key Commands

4—3

Editor Error Codes

Editor Command and Special Characters

5—1

PAL8 Options

5-2

Use of Arithmetic Operators

5—3

PALS Error Codes

6-1

Decimal/Character Conversions

7-1

Standard FORTRAN IV File Extensions

7—2

03/ 78 FORTRAN IV Operational Process

3-23
3-24

3-25
4—1

.........................................

3-48

...............................................

.................................................

.............................................

.................................................

...........................................

..............................................

................................

.................................................

........................................

..............................................

......................................

................................

...............................

7—3

FORTRAN IV Compiler Run-Time Options

Compiler Error Messages

7-5

Loader Run-Time Options

7-6

Loader Error Messages

7—7

Run-Time System Options

.............................

4-19

4-19
5—8
5-16
5-32
6-46
7-4
7—5

7-6

..........................................

7-7

.........................................

7-9

............................................

.........................................

7-8

Run-Time System Error Messages

7—9

Logical Operators and Their Meanings

7-10

Relational Operators and Their Meanings

7-11

Truth Table for Logical Expressions

7-12

Conversion Rules for Assignment Statements

....................................

7—13

Numeric Field Codes

7-14

Conversion Under G Format

7-15

Device Control Statements

7-16

FORTRAN IV Statement Summary

8—1

BATCH Run-Time Options

.................................

...............................

...................................

.............................

.............................................

........................................

.........................................

...................................

.........................................

8-2

BATCH Monitor Commands

8-3

BATCH Error Messages

9-1

ODT Command Summary

Summary of USR Functions

........................................

...........................................

.........................................

........................................

xiii

7-10
7-14

7-15
7-32
7-32
7-34
7—35

7—45
7-46
7-54
7-63
8—2
8-3
8—6
9-8
C—2

PREFACE

This manual describes the OS/ 78 Operating System for the DECstation 78 Minicomputer system.

Chapter 1 briefly describes the system and the contents of the software provided on the master diskettes. The basic
instructions on how to use the 08/78 Operating System are contained in Chapter 2. This chapter should be read in
its entirety prior to using 08/ 78 to gain familiarity with the operating characteristics and features of the system.
The commands that direct computer operations are given in alphabetical order in Chapter 3.
The OS/ 78 Editor, a program that provides the facility for creating and modifying ASCll source files, is described in

Chapter 4. PAL8, a symbolic assembler, is described in Chapter 5. The two high-level programming languages,
BASIC and FORTRAN IV, that fun under OS/78 are described in Chapters 6 and 7, respectively. Chapter 8 describes
BATCH, and Chapter 9 contains information on changing and correcting programs using ODT (Octal Debugging
Technique).
For advanced users, information on the use of device handlers, the User Service Routine, and the Command Decoder
is contained in the appendices. Also included in the appendices are useful mathematical multiplication and division

subroutines as well as an alphabetical summary of all OS/78 error messages.
Related handbooks and manuals that will be helpful to the 08/78 user are as follows:

Introduction to Programming (DEC-08-XINPA-A—D)

05/8 Handbook (DEC-SS—OSHBA-A-D)
05/8 Software Support Manual (DEC-SS—OSSMB-A-D)
DECstation/ 78 User’s Guide (EK-VTX-78-TM-OP—001)
DECstation/ 78 Maintenance Manual (EK«VTX—78—TM-OO l)

CHAPTER 1

INTRODUCTION TO 08/ 78

The OS/78 operating system is a practical and reliable operating system designed for the DECstation 78 Minicom-

puter system. The operating system permits the use of the VT-52 video terminal, dual-drive floppy disk system,
either an LA78 or LQP78 line Printer, and all available memory up to 16K. In addition, two auxiliary terminals and
a second

pair of floppy disks can be supported by the system. OS/78 offers a versatile set of keyboard commands

that supervise a comprehensive library of system programs. These programs allow you to develop programs using

PAL8 assembly language, BASIC and FORTRAN IV. A BATCH stream processor is also included, providing batch—
mode (unattended) system operation. This chapter briefly describes both the DECstation 78 Minicomputer and the

OS/78 software.
1.1

DECSTATION 78 MINICOMPUTER SYSTEM

The DECstation 78 Minicomputer system consists of four units:
a

PEN ?

terminal,

the dual disk drives,

the central processor, and
the main memory.

In addition, a line printer (either an LA78 or LQP78), a second pair of disk drives, and up to two auxiliary terminals

supported, through the serial asynchronous interface ports. These ports are suitable for primary and secondary communications applications, terminals, and the attachment of a variety of devices.
can be

The console terminal provides the means for interacting with or “talking” to the computer. It serves as the primary
source

of OS/78 command inputs.

The dual-disk drives include two disk drives that allow the use of flexible disks called diskettes or floppy disks. The
diskettes perform two functions:
1.

The 03/78 software that allows the computer system to be run is prerecorded on them.

They provide space to store programs and files that have been created.

The central processor unit (CPU) is located within the terminal. It is the “heart” of the computer system. The CPU
executes instructions from main memory and controls all peripheral devices.

The main memory is the main storage area in the computer from which instructions are fetched and executed, and
in which data may be stored and manipulated.

For more information on the DECstation 78 Minicomputer System, refer to the

DECstation 78 User’s Guide.

1.2 08/78 SOFTWARE
OS/78 can run programs written in three languages: PAL8 Assembly Language, BASIC, and FORTRAN IV. The
manipulation of program source and data files is done by typing in the 08/78 commands at the terminal. The software to do this is provided on the two master diskettes.
Only one of the two diskettes is used (in the left-hand
drive) as a system device. This leaves the right-hand drive free for using blank diskettes to store programs and data.

l-l

Introduction to OS/ 78
Diskette 1 contains the following Commonly Used System Programs (CUSPS) or Program Names:
ABSLDRSV

CCL.SV

DIRECT.SV

PIP.SV

PAL8 .SV

SRCCOM.SV

CREF .SV

RXCOPY.SV

FOTP.SV

BATCH.SV

EDIT .SV

BASIC .SV

BITMAP.SV

BLOAD.SV

SET.SV

BASIC.AF

BC OMP.SV

BASIC .FF

BRTS .SV

BASIC .SF

RESEQ .BA
HELP .SV
HELP .HL
In addition, there are several demonstration programs on Diskette 1 that begin with the name “DEMO”. These
programs are discussed in Section 2.3 of Chapter 2.

Diskette 2 contains the following Commonly Used System Programs (CUSPS) or Program Names:

PIP.SV

CCL.SV

DIRECT.SV

EDIT.SV

FOTP.SV

F4.SV

BATCH.SV

PASS20.SV

PASS2.SV

RALF.SV

PASS3 .SV

FRTS.SV

LOAD.SV

RXCOPY.SV

FORLIB.RL

SET.SV

HELPISV

ABSLDRSV

HELPHL

Essentially, the PAL8 assembler, BASIC, and utility programs are located on Diskette 1, while the FORTRAN IV
system programs are located on Diskette 2. It is not necessary to know the meaning of these program names since
they are automatically called up when required by 05/78.

1-2

CHAPTER 2
GETTING ON THE AIR

This chapter provides the basic information needed to use the OS/78 operating system. You should read the entire

chapter prior to using the system to become familiar with the features and capabilities of 05/78.
2.1

STARTING THE SYSTEM

2.1.1

Loading the Diskettes

The diskettes that contain the system programs are loaded by simply sliding them, label side up, into the open drive

slots. The system diskette is loaded into Drive 0, which is the left-hand slot.
After insertion, close the door; the latch will “click” when the door is closed properly.

Open the doors by pinching

the door latch between your thumb and index finger.
Since the diskette is thin and flexible, exercise caution when handling and storing it. Hold it by the edges and do

Also, use only felt-tip pens when writing on
diskette labels. Store them in a dust-free area, and do not expose them to magnetic devices, direct sunlight, heat, or

not touch the exposed portions that show through the protective cover.

humidity.
NOTE

The system diskette may be placed in Drive 1 and used

there, if necessary.

Calling the Keyboard Monitor

2.1.2

Once a system diskette is properly loaded into Drive 0, the 08/78 Monitor is called by pressing the START button
on the

right-hand side of the console. After a brief pause to read in the 08/78 Monitor, an OS/78 command

summary is displayed on the terminal. The system will then display a dot (.) on the terminal screen. The dot
means that OS/78 is ready to accept a command. You can now type any OS/78 command.

2.1.3

Using the Keyboard
The keyboard on the terminal allows data to be entered into the computer. Most of the keyboard is identical to a
typewriter keyboard. Unlike a typewriter, pressing a key does not automatically display a character on the keyboard screen. Pressing a key causes the character selected to be sent to the Central Processing Unit (CPU). The
program running would normally then transmit this character back to the terminal, thereby displaying it. Consequently, typing a character will have no effect if 08/78 is not running, or if 05/78 is busy running a program that
is not waiting for input from the keyboard.
Certain keys have special functions when used in 08/78 commands.

They are as follows:

RETURN key
This key is used to terminate OS/78 command lines, Editor command lines, program statements, and responses to
program input queries.

DELETE key
This key is used to correct typing errors made while entering Editor command lines, Monitor command lines, and

program statements.

2-1

Getting on the Air
LINE FEED key
This key is used while entering Monitor command lines. It causes the Monitor to display the command line as it has
been entered so far as a convenience check.

CTRL keys
This key is always used in combination with another key. The CTRL key is held down while the other key is
In this manual, these combinations are shown by using a slash (/) between CTRL and the other key such as

pressed.

CTRL/U, CTRL/C. The effect of these combinations is as follows:

CTRL/C

Terminates execution of the currently active program, if any, and returns control to the Monitor.
It also returns control from a BASIC program to the BASIC editor.

CTRL/O
CTRL/S
CTRL/Q
CTRL/U

Stops the display of characters on the screen.
Suspends terminal output of a program or monitor command but does not terminate the
program or command.
Resumes the terminal output that was suspended by CTRL/S.
Deletes a line typed to the Monitor or Editor.

The CTRL key commands require no terminator; the system performs the function as soon as you type the

command.

ESCape key
This key performs certain functions when using the 08/78 Editor and running FORTRAN IV and PAL8 programs.
The function of this key is described in those sections where its use is applicable.
2.1.4

Typing Convention and Symbology

One convention will be used throughout this manual to indicate what the system will display and what you should

type. Anything that is not underlined must be entered (typed-in) through the keyboard. Everything that the system
displays will be underlined. For example:

LDATE
NONE
O

In this example, the system displays the monitor dot, you type DATE (and then press the RETURN key). The

system displays NONE, and returns to the MOnitor dot (.).
The following keyboard characters will be designated by symbols to indicate specific operations that should be

performed when using 08/ 78. They are as follows:
Key

Symbol

Function.

LINE FEED

Perform a line feed

TAB
ESC

ESC

DELETE

DEL

Skip up to eight spaces
Performs special functions when using
PAL8, FORTRAN and the 08/78 Editor
Delete a character
NOTE

In this manual all command and input lines are followed

by a carriage return (pressing the RETURN key) unless
otherwise specified by using one of the above symbols.

2-2

Getting on the Air
2.2

MAKING A BACKUP COPY

It is important that a backup copy be made of the master diskettes.

NOTE
The master diskettes should never be used for running
the system or altered in any way.
Make a backup copy by using the following procedure.

y—A

Place the master diskette in Drive 0, the left-hand slot, and a blank diskette in Drive 1.

Start the system by pressing the START button.

In response to the monitor dot (.), type the DUPLICATE command followed by a carriage return (pressing
the RETURN key) as follows:

_._IIUF'L

R'Xfil :iRXAO:

The system will duplicate the contents of the diskette in Drive 0 onto the diskette in Drive 1, and when

completed, will return to the Monitor dot.

2.3

Remove both diskettes and label the diskette taken from Drive 1 with the appropriate markings.

save

Repeat Steps 1 through 4 for the second master diskette.
Store the master diskettes in a safe place.
Choose one of the copies of the two system diskettes and place it into Drive 0. This diskette will be the

Push the START button and you are ready to use OS/78.

system diskette.

DEMONSTRATION PROGRAMS

A series Of programs are provided on Diskette 1 to demonstrate some system capabilities of 08/78. These programs
are run via a

self-explanatory batch stream. To initiate this demonstration, start the system and type the following

in response to the Monitor’s dot (.):

LSUBM I T

DEMO/H

The batch stream controlling the demonstration repeats it every three minutes until a CTRL/C is typed.
When the Monitor dot appears, type

LSET
Lear

TTY
TTY

enaow
N0 ESC

These commands restore 08/7 8 to its state prior to the demonstration. Once you become familiar with the system,
you may wish to remove the files comprising the demonstration to make room for other programs. To do so, type

LHEL
2.4
2.4.1

DEMO??.*

NAMING DEVICES AND FILES

Devices

Each device on the system must be referred to by name. The device names that are recognized by the system are
as follows:

Getting on the Air
DSK

Default output device; usually same as SYS

SYS

The diskette inserted in Drive 0, upon which the system programs reside.

RXAO

The diskette inserted in Drive 0

RXAl

The diskette inserted in Drive 1

RXA2

The diskette inserted in Drive 2

RXA3

The diskette inserted in Drive 3

TTY

Keyboard/screen

LPT

Line printer

LQP

Letter quality printer

SLU2

Serial port #2

SLU3

Serial port #3

BAT

Batch handler

Each device on the system is recognized by the permanent device name given above and which is built into the

system. However, a particular name can be given to a device called a user-defined device name by using the
ASSIGN command. For example,

LQSSIGN

RXAI

PLACE

causes all future references to PLACE to address the device

RXAl. More than one user-defined device name can be

active at a time. The DEASSIGN command causes OS/78 to eliminate any user-defined device names.

Also, all

such user-assigned names are lost when restarting 08/78 with the START button.
2.4.2

File Names and Extensions

When referring to a diskette, the system will usually expect a file to be specified. Files are referenced symbolically

by a name of up to six alphanumeric characters followed, optionally, by a period and an extension of up to two
alphanumeric characters. The extension to a file name is generally used as an aid for remembering the format of a
file. Some commonly used extensions are given in Table 2-1 and Appendix E.
In most cases, conformance should be maintained to the standard file name extensions established for 08/78. If
is not specified for an output file, some system programs append assumed extensions. Where an extension for an input file is not specified, the system searches for that file name with the default extension. Failing to
an extension

find such a file, a search is then done for the original file without an extension. For example, if PROG were

specified as an input file to PAL8 the assembler for the 08/78 Operating System, the system would first look for
the file PROG.PA (since .PA is the standard extension for PAL8 input files). If PROG.PA were not found, the
system would try to find the file PROG (with no extension). As not all system programs utilize default extensions,
refer to Table 2-1 and the individual system programs for details.
Some programs also accept the characters * and ? in file names.
acters have special meanings to the programs involved.

They are called wild-card characters. These char-

The use of these characters is described in detail in

Section 2.5.6.
Since files are the basic units of the OS/78 system, an understanding of files and their structure is helpful when using
08/78. A description of files and file-structured devices is given in Section 2.8. This section can be skipped since
it is not necessary for properly operating the 08/78 system, but does give an insight into what is happening in the

system.
2.5

ENTERING MONITOR COMMANDS

OS/78 operating system commands are entered at the terminal in response to the dot printed by the Monitor.
Pressing the RETURN key initiates execution of the entered command. Some commands require a particular format
to distinguish between input and output files or to name certain devices. Any errors that are made while
utilizing
these commands result in an error message being displayed by the Monitor. Error messages are summarized in
Appendix G.

2.4

Getting on the Air

Table 2-1

05/ 78 Assumed Extensions

Extension

Meaning

.BA

BASIC source file (default extension for a BASIC input file).

.81

Batch input file (input for BATCH).

.BN

Absol-ute binary file (default extension for ABSLDR and BITMAP input files; also used as
default extension for PAL8 binary output file).

.CM

Command file.

.DI

Directory listing file.

.FT

FORTRAN language source file (default extension for FORTRAN input files).

.LD

FORTRAN load module file (default assumed by run-time system, FORTRAN IV loader).

.LS

Assembly listing output file (default extension for PAL8).

.MP

File containing a loading map (used by the Linking Loader, MAP command).

.PA

PAL8 source file.

.RA

RALF assembly language file (FORTRAN IV).

.RL

Relocatable binary file (default extension for a Linking Loader input file).

.SV

Memory image file (SAVE file); default for the R, RUN, SAVE, and GET commands.

.TM

Temporary file generated by system.

OS/78 commands and the abbreviated forms for each command are summarized at the end of this chapter. Greater
detail on each command is given in Chapter 3.
2.5.1

OS/78 Command Format

An OS/78 command is usually made up of the following parts:
a command

O\U!-PWN:—‘

a list of

word or words,

output devices and files,

left-angle bracket (<),
input devices and files,
a series of options, and
the RETURN or ESCape key.
a

a list of

Parts 2 through 5 above are called the command parameters because they supply information concerning the
devices and files that are affected by the command.
The general format for a command may thus be expressed as follows:

.COMMAND dev:outfil.ex<dev:infil.ex/options (terminator)

2-5

Getting on the Air
where COMMAND is any 08/ 78 command. Both the input and output specifications consist of a device, a file
name, and a file name extension. For each input specification given without explicitly naming a device, the device

associated with the previous input specification is assumed. If no device has been given for the first input specifica-

tion, then DSK: is assumed. Output specifications also consist of a device, a file name, and a file name extension.
Output specifications are optional. If no output device is given, then DSK: is assumed; if no file name is given,
then any or all the files on the designated device are assumed, the exact action depending on the command.
The output specification is to the left of the left-angle bracket (<) and the input specification is to the right of the
bracket.

Many options can be stipulated with commands to perform various operations in a simpler way; these are

discussed in Section 2.5.4. The command line terminator is usually a carriage return (pressing the RETURN key)
that causes the system to begin execution of the command. In some cases, the ESCape key is used, and its use will

be explained where applicable.
The command word and parameters in each command must obey the following rules:
1.

The command word or words may be abbreviated only as indicated in Table 2-3 or in Chapter 3.

No command may have more than three output devices or files, and these must be separated by commas.

The left-angle bracket (<) may be omitted if no output devices or files are specified.

In general, no command may have more than nine input devices or files, separated by commas. DIRECT

Some commands such as COPY and DIRECT may have no more than one output specification.

and COPY commands are limited to five input devices or files.
5.

The file name may not contain more than six characters; the extension may not contain more than two
characters.

In a command line, the device name is followed by a colon (1), and is always separated from any file name

by a colon.
For example, the command line

LDIR

RXA13/E

when executed by pressing the RETURN key, displays the directory of RXAl on the terminal screen. The E option
also causes the system to display any empty files in the order that they are located on the diskette.
The command line

_.CDF'Y

RXQO i ‘tZII-"(Xriiil MARITH . PA to MULT . F'A , UIUIIIE . F'A

when executed, transfers the specified input files ARITH.PA, MULT.PA and DIVIDE.PA from RXAl onto RXAO.
Commands also can be simplified by the use of defaults. The use of defaults is explained in Section 2.6.
2.5.2

Incorrect Commands

If you enter an incorrect command line and then press RETURN, one of the following results will occur:
1.

If the command is not recognized, the system will respond with a question mark and a dot, and then wait

If the command is recognized, execution will begin immediately. If an error is noted in the command line,

for you to enter a correct command.
press CTRL/C as soon as possible to terminate the action and return control to the Monitor.
upon the type of command and the files involved, this may or may not prevent execution.

Depending

Execution of an incorrect command can destroy important files if it occurs when using the following
commands:
COPY

SAVE

SQUISH

DELETE

ZERO

Getting on the Air

Therefore, be extremely careful when using
Any specific cautions that should be exercised when using these commands are fully
described in the command chapter (Chapter 3). Always be sure that a back-up copy of the system diskette
has been made, along with and all other important files that may have been created (Section 2.2).
In fact, the error may actually destroy part of the software.

these commands.

Section 2.5.3 describes how to correct any errors that have been made in the command line prior to executing the
command.
2.5.3

Correcting Errors

Until the RETURN key is pressed, the command line will not be processed by the system. Therefore, the command
line is still available for correction. Always check the command lines before pressing the RETURN key since a line
with errors may cause undesirable results.
To correct typing mistakes, use the DELETE key. This key erases the last character typed. Successive DELETEs

each cause one more character to be erased. The correct character or characters can then be typed.
A command line may be deleted completely before it is entered by typing CTRL/U (produced by holding down the

CTRL key and pressing the U key). This echoes as “U and returns control to the Monitor without processing the
current line. The Monitor dot (.) indicates that the system is ready to accept a new command.
can then be

The command line

Use the LINE FEED key to verify the contents of a command that is in the process of being

retyped.
entered, that is, whatever has been entered in the command line thus far will be displayed.
2.5.4

Using Input/Output Options

In addition to output and input files that are specified in the command line, there are various options that can be

typed in to perform certain functions. Options are either numbers or alphanumeric characters. Numbers used as
options are generally contained in the command line with the equal sign (=) or square brackets ([1) construction.
The alphanumeric option characters are set off from the I/O specifications by the slash (/) character for single
character options, and parentheses for a string of single characters. The usage of the slash, parentheses, equal sign,
and square brackets is explained below. In addition, a series of switch options known as CCL options are available
for use in the command line.
An equal sign (=) followed by an octal number may occur only once in a command line
Equal Sign Construction
and must be followed by a separator character (comma or left-angle bracket), other options, or a line terminator
—

(RETURN or ESCape). For example,

_._I’.I IRECT

8Y8 ‘2 32:13

will list the directory of the diskette in Drive 0 on the screen in three columns.
Slash Construction

A single alphanumeric character is preceded by a slash and can occur
anywhere in the command line (even in the middle of a file name) although the usual position is at the end of the line. For
example,

LCDF‘Y

RXG 1 28135001413

ISZX-iIRXfiIO 2 F' I R‘SEiT o EZX/T

means the same as

lCOF‘Y

R'XI‘T-‘IZI. : SECOND oE;‘(/T'=1:I:;IXAO iii-"IRST o EX

The option /T instructs the Monitor to assign the current date to the

2-7

output file.

Getting on the Air
Parentheses Construction

—

When two or more letter options are used, they may be grouped together inside

parentheses. This construction is also valid anywhere in the command line. For example,
RXAZI. 2 OUTPUT t EX-iISYS: INPUT 0 EX ( GT)

LCOF'Y

means the same as

Ran It 2 OUTPUT . ISEIX-iISSSYSS : INF‘UT . IEX/Q/T

ACOF‘Y

Square Bracket Construction

—

The square bracket construction can only occur immediately after an output file
bracket, a decimal number between 1 and 255, and a close square bracket. The

name and consists of an open square

square bracket construction is generally only used when necessary to optimize file storage.
This construction is used to provide an upper limit on the number of blocks (256 words per block) to be contained
in the output file in order to allow the system to optimize file storage. For example,

I} .T. NNQY L' 1 ‘91] r LISTING [I 200 3 {SOURCE

_._l“-'(‘+L.

The output files are a file named BINARY on device DSK: having a maximum length of 19 blocks, and a file named
LISTIN (only six characters are significant) on the device DSK: with a maximum length of 200 blocks. The input

file is SOURCE on device DSK.
CCL Options

This family of options is of the form

-ex

where ex is one of the options specified in Table 2-2.
In the following example, typing

Lear

TESTmT

will assemble TEST.PA, store the resulting binary program in TEST.BN on DSK, and display the program listing
on the

terminal.

2.5.5

Remembering Previous Arguments

A feature of the system is that it remembers arguments used by the commands COMPILE, CREATE, LOAD, PAL,

EDIT, and EXECUTE, and can use these arguments in other commands. The remembrance for the EDIT command
is located in a separate area from the others. Each time these commands are executed, the command with its argument is remembered in a temporary file.

Therefore, the file name used last can be recalled for the next command

without again specifying the arguments. If, for example, the EXECUTE command

LEXECUTE
was entered

TESTI 0 PA

previously, compilation of TEST.PA would be accomplished by simply typing

LCOMF'ILESI
These commands do not remember arguments typed on previous days.

2-8

Getting on the Air
Table 2-2

CCL Options

Meaning

Option
-L

Send output to LPT.

-LS

Generate a listing file (used with the COMPILE, EXECUTE, and PAL commands). The listing
file is written onto SYS: if no output device is specified and is given a .LS extension. The listing
file name is the same as the file name that immediately preceded the CCL -LS option.

-MP

Generate a memory map (used with the COMPILE, EXECUTE, and PAL commands).

-NB

Do not create a binary file (used with the COMPILE, EXECUTE, and PAL commands).

-T

Send output to terminal.

For example, typing

LfllfiwL
will list the directory of the system device on the device named LPT.
-PA

Selects the PALS compiler when the files extension does not determine it (used with the
COMPILE and EXECUTE commands).

-FT

Selects the FORTRAN lV compiler when the file extension does not determine it (used with the
COMPILE and EXECUTE commands).

2.5.6

Using Wildcards

Certain commands allow Wildcards in file name specifications. These commands are COPY, DELETE, DIRECT,

LIST, RENAME, and TYPE.
Wildcards allow a file name or the extension in a command to be replaced totally with an asterisk or partially with a

question mark to designate certain file names or extensions. The wild characters are particularly useful when doing
multiple file transfers. The asterisk is used as a wild field to designate the entire file name or extension. This is
illustrated in the following examples:
TEST1.*
*.BN

All files with the name TESTl and any extension.
All files with a BN extension and any file name.

*.*

All files (except when used with the DELETE command).

The question mark is used as a wildcard character to designate part of the file name or extension. A question mark
is used for each unspecified character that is to be matched. For example, PR?? matches all files beginning with

PR that are two to four characters long. Other examples are as follows:
TEST2.B?
TES??.PA

All files with the name TEST2 and any extension beginning with B.
All files with a PA extension and any file name from three to five characters long beginning
with TES.

??.??

All files with file names of two characters or less.

The asterisk and the question mark can be specified together in the same command line:

???.*

All files with file names of three characters or less.

Getting on the Air
A specification may not contain embedded *’s. For example, A*B.* is an illegal specification and will produce the

following message:
I LLEGAL

If an * or ? is included in a command other than COPY, DELETE, DIRECT, LIST, RENAME, or TYPE, the

following message appears:
Il...|.-EGm.2.5.6.1

3‘

Wildcard Input File Specifications

This section describes wildcards when specifying input files.

—-

For example, to display the directory entries for all files on RXAI with file names from two to four characters long
and beginning with PR, type

LIIIF.‘

RXQ].2F'R'??.*

To display entries for all files on RXAO with extensions ending with R, type

LIIIl-"x'

RXAO:*.?R

The following is a list of other illegal wildcards:
A*.BD

Illegal because the asterisk should replace the entire file name or extension

ABC.*D
.BA

Illegal because the file name cannot be omitted; it must be specified in some form.

RXA?:C.BA

Illegal because wildcards may not be used in device names.

2.5.6.2 Wildcard Output File Specifications

—

This section describes the use of wildcards when specifying output

files.
The wildcard asterisk may be used in output files, but the question mark is illegal. If no file name is specified, then

*.* is assumed.
For example,

_._[IIZL.

* . BN‘iIX . F'A

is legal. This command deletes any file with a .BN extension if the file and a .PA extension exist.

Another example would be

LCOF‘Y

'1'Xf-51Uk e BK‘iSYS:

F'A

where all files with a .PA extension would be copied from the system device onto RXAl, having the same file name
but given a .BK extension.

The question mark may not be used in output file entries. For example,

LCDF‘Y

RXAZI. :NEH . 'E”?-:ZIO|...II . BA

is not allowed.

2-10

Getting on the Air
Greater detail on using wildcards is given in the 03/8 Handbook.

Wildcards, when used with the COPY and DELETE commands, is an extremely powerful feature, but is capable of

destroying needed files if not used carefully. Therefore, the following two steps are recommended:
1.

Keep a backup copy of the system diskette and all other important files
Use the / Q option when running COPY or DELETE with wildcards. This allows you to decide what to
do with each file involved before it is acted upon by the system. Type Y to allow the operation, N to
disallow it.

2.5.7

Indirect Commands (@ Construction)

When many file names and options are to be included in a single command, they can be put into a file and thus
need not be typed each time they are required. This is done by using the @ file construction, which may appear

anywhere within the argument (file and option list) portion of a command. The @ construction is of the form
@dev:file.ex
If dev: is omitted, DSK: is assumed. The word file must be a file name. If the extension is omitted, .CM is assumed.
Indirect command files are not permitted in the following commands:

ASSIGN

RUN

DEASSIGN

SAVE

GET

ODT

START

DATE

The information in the specified file is then put into the command string to replace the characters @dev:file.ex.
For example, if a file FLIST.CM is created using the Editor, and contains the string

FILEB,FILEC/L,FILED
then the command

LCDMF‘ I LIE

F' I LEA ! F' I LEB ! F I LEG/l... , F I LEI! , F ILJ‘EZ

could be replaced by the command

LC UMF' I I... IEEI

F I I... EA , WEI... I EST 5' F' I I...IESIJ'(.'.

Carriage returns and line feeds within the file are ignored, but nulls are not; the nulls signify end-of—line. Command
files may not exceed one block in length. If a command line is more than 512 characters in length, the following
message is printed:

C) [I] M M A N III

I... I N Iii

(I) U E R F' L. (I) w

The construction @dev:file.ex may not be immediately followed by a letter or digit (it would be absorbed as part
of the .ex). The @dev:file.ex may be followed by an apostrophe C) which would then be ignored.
2.5.8

Asterisk (*)

Under certain conditions, 08/78 may display an asterisk. This is a function of a special program called Command

Decoder, and signifies a request to enter another list of files for the current system command. Except in cases
where its application is specifically described, just type CTRL/C in response to the asterisk to return to the OS/78
Monitor. The Command Decoder is described in Appendix D.

Getting on the Air
2.6

USING DEFAULTS

The amount of typing necessary for entering commands may be reduced by taking advantage of the default ability

of the system. A default is a command parameter that when not typed is assumed by the system. Defaults are

generally used to avoid typing device names.
1.

The general default device assumed by the Monitor is DSK: (usually equivalent to SYS: or RXAO:), the
diskette in Drive 0.

For all output files and input files, the default device is DSK:. For example, the following two file

descriptions mean the same to the Monitor:
Command FILEA.AB,FILEB.CD<FILEC.EF

Command DSK:FILEA.AB,DSK:FILEB.CD<DSK:FILEC.EF
When a device name is explicitly given for an input file, any additional input files will default to the last

explicit device name:
Command TTY:<FILEA, FILEB, RXAl :FILEC, FILED
Command TTY:<DSK:FILEA,DSK:FILEB,RXA1:FILEC,RXA1:FILED
In this description,
FILEA

was assumed to be on DSK:

FILEB

was assumed

FILEC

was stated to be on

FILED

was assumed to be on

because it was the first input file.

to be on DSK: because the preceding file (FILEA) was on DSK: (same as

RXAO:).
RXAl :.
RXAl: because the preceding file (F ILEC) was on RXA12.

Some of the individual commands assume special default devices. The DIRECT and TYPE commands
assume the

terminal as the output device. The following command line will display the directory of the

diskette inserted in Drive 1 on the terminal screen.

LII! I R 135: {I} T

R Xi”: .1. 1

Note that the left-angle bracket is not needed in the above command line because no output files are

specified. However, commands such as CREF (cross reference program) and LIST default to the LPT.
Thus, if a LPT is not available, the output of the CREF command must be sent to a file or to the
terminal (using the CCL option-T).
2.7

Specific cases and use of options are discussed in Chapter 3.

WHERE TO GET MORE INFORMATION

Many times more information is desired such as the function of certain commands, their format, and the particular
options that can be used with them. A file named HELP.HL is available to provide this information. The information will be displayed on the terminal by using the HELP command and giving the command name for which
information is desired as an argument. For example,

LH if}: I... l3‘

[:3 [I] M F' a I‘i' I35

2-12

Getting on the Air
will display the following information on the terminal:
SRCCOM

CALLING

C(IlMMfiNIIS:
PA , IIEU : INF ILEQ t Pfi

COMPQR

DEV 2 OUTFII. Ff . F'F1~=ZIIIEU 2 INF ILE ll.

+COMPAR

OUTFILE.PQ%INFILEIoPAvINFILE2.PA

/FILES

DSN

SUITOHES:
fB

COMPARE BLANK
DON’T COMPARE

LINES

/C
lS

OON’T

TQBS

/T
/X

CONVERT TABS TO SPACES
DON’T COMPARE OR PRINT

ERRORS:
?0

INSUFFICIENT

INPUT

T2
T3

INPUT ERROR
OUTPUT FILE

OUTPUT

CAN’T

(SLASHED)

COMPARE

COMMENTS
SPACES
ON OUTPUT
COMMENTS

CORE
FILE

ERROR

fiNfl

(OR

LESS

THAN

INPUT

FILES)

FILE 2
TOO LHRGE

ERROR
OPEN

OUTPUT

FILE

If a line printer is attached to the system you can get a hard copy printout by outputting the desired HELP text to

the line printer as follows:

LHELF‘
or

LF‘T 3 'fiF‘AL v COMPARE

(by using the CCL option -L)

LHELF‘

PAL ! COMF‘ARE”L

The HELP files for PAL and COMPARE will be printed on the line printer.

2.8

FILES AND FILE-STRUCTURED DEVICES

Files are basic units of the 08/78 system, and an understanding of file structure is helpful for its use. A file is any
collection of data to be treated as a unit. The format of this data is unimportant; for example, 05/ 78 can manipu-

late several standard formats, including ASCII files, binary files, and memory image files. The important consideration is that the data forms a single unit within the system.
2.8.1

File—Structured Devices

A file-structured device is one that can be logically divided into a number of 256-word blocks and has the ability
to read and write from any desired block.
a diskette is the

Diskettes are file-structured devices but a terminal is not.

Currently,

only file-structured device supported by 08/78.

All OS/78 file-structured devices must be logically divided into these 256-word blocks.

Hence, 256 words is con-

sidered the standard OS/78 block size. A given file consists of one or more sequential blocks of 256 words each

(consecutively numbered). A minimum of one block per file is required, although a single file could occupy all of
the blocks on a device.

OS/78 blocks are conventionally given in octal. File lengths (number of blocks per file) are conventionally given in
decimal.

2-13

Getting on the Air
2.8.2

File Directories

To maintain records of the files on a device, blocks 1 through 6 are reserved for the file directory of that device.

Thus, file structured devices are also called directory devices. Block zero is unused, except on the system device,
in which case it contains the system bootstrap, a program that reads in the 08/78 Monitor.

Entries in this directory inform the system of the name, size, and location of each file, including all empty files and

the tentative output file, if one exists. Six blocks are always allocated, though all are not necessarily active at any

given time.
2.8.3

File Types

Three types of files exist in the 08/78 system:
1.

Empty File
An empty file is a contiguous area of unused blocks.

deleted,
2.

or when

Empty files are created when permanent files are

the ZERO command is used.

Tentative File
A tentative file is a file that is open to accept output and has not yet been closed.
can be

open

Only one tentative file

any single device at one time. Tentative files never appear in directory listings.

Permanent File
A permanent file is a file that has been given a fixed size and is no longer expandable. A tentative file

becomes permanent when it is closed.
To further understand file types, consider what occurs when a file is created.

Normally, in creating a tentative file,

the system first locates the largest empty file available and creates a tentative file in that space. This establishes the
maximum space into which the file can expand. The user program then writes data into the tentative file. At the

end of the data, the program tells the system to close the tentative file, making it a permanent file. When the file
is closed, whatever space remains at the end of the tentative file is allocated to a new, smaller, empty file.

2.8.4

File Formats

08/78 and relevant system programs use three standard file formats:
1.

ASCII and Binary Files

ASCII files are packed as three 8-bit characters into two words as follows:

WORD 1

WORD 2

CHARACTER 3

CHARACTER 1

(BitsO-S)

(BitsO- 7)

CHARACTER 3

CHARACTER 2

(Bits 4

(Bits 0-7)

7)
3

Memory Image (.SV Format) Files

A memory image file consists of a header block followed by the actual memory image. The header block is tradi-

tionally called the Core Control Block (CCB).
The CCB uses only the first 128 words of the 256 word block reserved for that

purpose.

The CCB is a table of information containing the program’s starting address, areas of

memory into which the program is loaded, and the program’s Job Status Word. (Refer to Section 3.26 for a description of the Job Status
Word.) The CCB is created at the time the program is loaded into memory and is written onto the .SV file by the
SAVE operation.

2.14

Getting on the Air
When a program is loaded, the Starting Address and Job Status Word are read from the CCB and stored in memory.
The CCB itself is saved in words 200 (octal) through 377 (octal) of block 37 on the system device. It is placed there

by using the GET and RUN commands.
2.9

05/78 COMMAND SUMMARY

A summary of all OS/78 commands is given in Table 2-3. The table shows the abbreviated form of the command

acceptable to the system, command format, and a brief functional description of each command.
Table 2-3
Command
ASSIGN

08/78 Command Summary

Abbreviated

Format

Function

.AS permdev user-deV-name

Makes a new user-defined device name

equivalent to the permanent device name.
COMPARE

COMPA

.COMPA dev:output.ex<

devzfilel .ex.dev:file2.ex
COMPILE

COM

Compare two source files line by line and

display all their differences.

.COM file .ex <fi1e.ex/options

Produce binary files and/or compilation

or for

listings for specified program files. This

multiple input files

command will chain to either the PAL8

.COM A,B<C,D

assembler or BASIC compilers dependent
upon the filename extension. Under

FORTRAN IV, only one input file can
be specified.

COPY

COP

.COPY dev:file.ex <dev:fi1e.ex,...

Transfer files from one 08/ 78 device to
another.

Up to five input files can be

specified.
CREATE

CREA

.CREA file.ex

Open a new file for editing.

CREF

.CREF file.ex

Assemble and produce a cross-reference

listing from the source file. Defaults to
LPT, thus the -T option must be used to
output to terminal.
DATE

Enter the date into the system. If no

.DA dd-mmm-yy

argument is given, print current date
on terminal or NONE if no date was

specified.
DEASSIGN

.DE

DELETE

DEL

.DEL dev:fi1e.ex,.../options

Eliminates all previous user-defined names.
Delete one or more files.

Up to five input

files can be specified, separated by
commas.

same

DIRECT

DIR

All files must reside on the

device.

Produce a listing ofan OS/78 device

.DIR devz/options

directory.
Continued on next page

2-15

Getting on the Air
Table 2-3 (Cont.)

Command
DUPLICATE

Function

Format

Abbreviated
DUPL

OS/78 Command Summary

.DUPL outdev:<indev: / options

Reproduce the contents of the input
device onto the output device (floppies

only).
EDIT

Open an already existing file for editing.
Using the input/output construction
allows a copy of the original file to be

.ED file
or

.ED file <file

retained, while naming a new output
file.
EXECUTE

EXE

.EXE file.ex,fi1e .ex/ options
or

.EXE file .ex<file .ex

Produce binary files and/or compilation
listings for the specified program files,
load the binary file, and execute the
program. Using the input/output format
executes the output file while producing
from the input file. Under FORTRAN.

IV, only one file can be specified.
.GE dev:file.ex

Load memory image files (.SV format)
into memory from a device.

HELP

.HELP command

Display instructional information on
the specified OS/78 command.

.LI dev:file.ex,...

List the contents of the specified files

GET

HELP

LIST

LOAD

.LO dev:file.ex

LPT.

Run one of the OS/ 78 loaders for either

PAL8, BASIC or FORTRAN IV de-

pending upon the extension of the
filename.

MAP

.MAP dev:file.ex<dev:file.ex

Produce a memory map of the specified

input file.
MEMORY

MEM

ODT

PAL

.MEM n

Display the number of fields available
or actually being used by 08/78.

.OD

Load and start the CDT (Octal Debugging Technique) system.

.PAL dev:file.ex<dev:file.ex

Run the PAL8 assembler and assemble

the specified source file.
Load memory image files from the

.R file.ex

system device (making it equivalent to
a RUN SYS file .ex command

except

that the system scratch area is not

affected) and start it at the starting
address. If .ex is omitted, .SV is assumed.

Continued on next page
2-16

Getting on the Air
Table 2-3 (Cont)

0S/78 Command Summary

Abbreviated

Format

Function

RENAME

REN

.REN dev:new.ex<dev:old.ex

Change the name of the file from the
input name to the output name.

RUN

.RU dev:file.ex

Load a memory image file, move its
Core Control Block to the system scratch

Command

area, and start the program at its starting

address. This command is equivalent to
3 GET and

START command, and must

be used to “run” a program that does
not reside on SYS:.

The colon after dev

in the command line may optionally be

replaced by a space.
SAVE

.SA dev:file.ex

Save an image of the program currently
in memory on the device specified. If
.ex is

SET

.SET dev[no] attribute(argument)

omitted, .SV is assumed.

Modify the handler for either the line
printer or terminal by the attributes or
arguments specified in the command
line.

SQUlSH

.SQ dev:<dev:

Eliminates all embedded empty files on
the device.

START

.ST fnnnn
OI

.ST

Start the program currently in memory
at location nnnn in field f.

Start the program currently in memory
at the starting address specified in the

Core Control Block.

SUBMIT

.SU dev:file.ex<dev:file.ex

Run the BATCH program where the

output is the optional spooling file and
the input is the BATCH input file.
TERMINATE

TER

.TERMINATE

Cause the VT-78 to enter the VT-52
emulator mode.

TYPE

.TY dev:file.ex,...

Display the specified files on the
terminal.

.UA (argument)

Remember the argument where the argument must be a legal OS/ 78 command.

UA with no argument executes the last
remembered argument.

Continued on next page

Getting on the Air
Table 2-3 (Cont)
Command

Abbreviated

OS/78 Command Summary

Format

Function

.UB (argument)

Similar to UA

.UC (argument)

Similar to UA

.ZERO dev:

Zero the specified device directory,

ZERO

deleting any files that may exist on
the device.

CHAPTER 3

OS/ 78 COMMANDS

08/78 system software supports a powerful set of commands that allows you to direct computer operations. Each
OS/78 command has a standard format that should be well-understood prior to using the command.
The format for each command is shown, including one or more examples. Each command description is followed,
when applicable, by the options that can be used with the command, and the error messages that could result where

using the command.
In response to the dot produced by the 08/78 Monitor, type the command line and those options that can be used
with the command to perform a specific function. Press the RETURN key to enter the command into the system

for execution. Characters that are underlined are system responses; characters that you type are not underlined.

Depending upon the command, arguments may or may not be specified. Many of the commands have default
options that are automatically inserted in the command line by the system if an argument is not explicitly specified.
Default options may also vary with each command. The use of defaults is described in Chapter 2, Section 2.6.

3-1

OS/ 78 Commands

ASSIGN

3.1

ASSIGN COMMAND

The ASSIGN command assigns a user-defined name to one of the available permanent devices such as SYS or TTY.
Format:

.ASSIGN

permdev user-dev—name

Example:

_o_AS

RXA 1

DEU2

In this example, device RXAl is assigned the symbolic name DEV2 and can now be addressed by both RXAl: and

DEV21. Note that the colon is not used with both the permanent device name and the user-defined name in the
ASSIGN command. However, when referenced in most other commands the colons must be used with the userdefined name.
If a user-defined name is not specified in the command line, any one that may have existed is removed and only the

permanent device name is valid.

Example:
_: AS

RXA 1

3.1.1

DEV?

Previous ASSIGN command
User-defined name becomes invalid

User-Defined Name

The user~defined name that is assigned to a permanent device can be from one to four characters long. If a three or
four character name is used it becomes a single word when translated into internal representation. As a result, it is

possible that this code could be identical to some other internal code used by the system.
The ASSIGN command is particularly useful if an LQP line printer is attached to the system instead of an LA78
line printer. Since several of the 08/78 commands automatically send output to the LA78 line printer (LPT),

assigning LPT to LQP allows output to be sent to the LQP78 line printer.

Example:

Les

LGP

LPT

ASSIGN processing is done entirely by the Monitor.

0S/_78 Commands
3.2

BASIC COMMAND

The BASIC command requests execution of the BASIC editor.

Format :
.BASIC

Example :

LBASIC
NEU

0an~

For additional information on BASIC, refer to Chapter 6.
This command runs CCL.SV and BASICSV.

BASIC

08/ 78 Commands

COMPAR E

3.3

COMPARE COMMAND

The COMPARE command compares one source input file to another source input file to see if they are identical or

different. Differences are sent to the output file.
Format:

.COMPARE RXAl :DIFVER<SYS:VERSl .FT,VERSZ.F T

Example:

LCUMF‘Q

RX!) 2|. 3 l".| I F'UEEIR'SSYE; 3 VIEWS 1 . FT 9 VIEEIRSES'Z’. . PT

In this example, every source line in files VERSI .FT and VERSZFT are compared and the results of the comparison
are sent to the

file named DIFVER on device RXAl. The file extension .LS is automatically added to DIFVER.

Along with the necessary arguments for the command line are COMPARE options that can restrict or enhance the
result of the COMPARE operation. The COMPARE program is commonly used to print all the editing changes
between two different versions of a program. This makes it a useful documentation and debugging tool.
Before the first two lines of input are compared, the current version number of the COMPARE command is sent to
the output device followed by a header for each input file. The header is the first line in the file which is usually the

title, date, file name, or description of that file.

Example :
SRCCOM

U49

/THIS

'2.)

/THIS

THE

FIRST

SIEPONII

SOURCE

COMPARE

FILE

When the COMPARE command is executed, lines from the first file that are either identical or different with lines
from the second file are sent to the output device until there are three consecutive lines that agree. The first line
sent to the output device is preceded by the numeral one, a closing parenthesis, and a 3-digit decimal number. The

numeral one represents the first file specified in the command line and the 3-digit decimal number is the page that
the proceeding difference group is located on. Each succeeding line is also preceded by the numeral one and

closing parenthesis. Out of the three lines that agree, only the first line is sent to the output device which is followed by a group of asterisks on the next line. This ends the first difference group for file 1.

Example :
tl)002
1)
1)
II.)
1)

o
.

53-4512:
.5.

1)
1)
1)
1)

1)
1)

M?”

ZTZLH
34

OS/ 78 Commands
The COMPARE command then sends all lines from the second file that both agree and disagree with lines from the
first file to the output device until there are three consecutive lines that agree.
NOTE
To change the number of consecutive lines that interrupt

processing, use the =k option. By specifying the =k
option, any octal number can be used to set the number
of consecutive lines that separates one difference group

from another.
The first line sent to the output device is preceded by the numeral two, a closing parenthesis, and a 3-digit decimal
number. The numeral two represents the second file specified in the command line and the 3-digit decimal number
is the page the proceeding difference group is located on. Each succeeding line is also preceded by the numeral two

and closing parenthesis. Out of the three lines that agree, only the first line is sent to the output device which is
followed by a group of asterisks on the next line. This ends the first difference group for file 2.

Example:
2)0022
2)
2)
2)
2)
2)
2)
2)

2)
2)
23)

2)
********

231m<ce><to~b

If all the lines on both files agree (disregarding the header lines), a message “NO DIFFERENCES” is sent to the out-

put device.
3.3.1

Comparing Part of a Line

The /C option compares all source lines in one file with source lines in another file but disregard all assembly comment lines. In addition to disregarding comment lines in a comparison the output of the comment lines can be

restricted by using the /X option.

By specifying the /X option, all comments are ignored while the comparison

takes place; furthermore, when /X is specified comments are not sent to the output device.
If the format of one file is different from that of another because of tabs and spaces, and a comparison is needed, use
the /S option. By specifying the / S option, the COMPARE command compares all source lines but disregards all tabs
and differing spaces.

3.3.2

Making Changes While Using the COMPARE Command

Normally, blank lines are disregarded when comparing two files. If a blank line is to be considered as a valid input
line instead of a carriage return in a comparison, use the /B option. By specifying the / B option, a blank line is
considered to be a valid line containing all blanks.
NOTE
When the /S and /B options are used together, a blank

line is not taken into consideration.

This command runs CCL.SV and SRCCOM.SV.

3-5

OS/ 78 Commands
Table 3-1

COMPARE Options

Ignore all comment fields during the comparison
(assumes assembly language source files are being

(comment
fields)

compared)

’

(blank
lines)

Consider a blank line as a valid input line containing

(tabs and
spaces)

Ignore all tabs and spaces during the comparison.

(tabs to
spaces)

Convert all tabs from the input file to spaces on the

(output of

Ignore all comment fields during the comparison

comment

and do not send any comments to the output device.

blanks instead of a carriage return.

output device.

fields)
Table 3-2
‘20

COMPARE Error Messages

The differences between the two files are too large for an effective compare.

NOTE:
If the large number of differences is caused by comment

lines, use the /X option to alleviate this problem.
‘21

There is an input error on file 1, or two input files are not specified in the
command line.

There is an input error on file 2.

'23

Ran out of room on the output device.

There is an output error.

An output file cannot be created.

3-6

OS/ 78 Commands
3.4

COMPI LE

COMPILE COMMAND

The COMPILE command assembles PAL source input files to produce absolute binary output files, compiles FOR-

TRAN source input files to produce relocatable binary output files and compiles, loads and executes BASIC source

input files.
Format:
.C OMPILE outdev:file .ex<indevzfile .ex/ options

Along with the necessary arguments for the command line are several PAL, BASIC, and FORTRAN options that can
modify the result of the COMPILE operation. For PAL, BASIC, and FORTRAN options, see the appropriate
language chapter.
When the COMPILE command assembles a PAL input file, the absolute binary output file created usually has a .BN
extension (the default for PAL8).

Whereas, when the COMPILE command compiles a FORTRAN input file, the

relocatable binary output file created usually has an .RL extension (the default for FORTRAN IV). Both absolute
and relocatable binary output files may be used as input files in the LOAD command.
An absolute binary file is an independent stand-alone program while a relocatable binary file is a dependent program
which needs other system programs to aid it in its execution.
The extension of the first input file determines which compiler will be used. The .PA uses PAL8, .FT uses FOR-

TRAN, .BA uses BASIC. If the extension is not specified as part of the file name in the command line, all assembler
and compiler extensions are compared with the unspecified extension until a match is made. Then the appropriate

assembler or compiler is called and executed. If an extension other than the ones listed in Appendix E is used,
specify the correct assembler or compiler extension as a CCL option in the -ex portion of the command line.

Example:

LCUH

R'XA 1 :COMF‘fiACCTS o 03""FT
NOTE
If no output file is specified, the first input file specified
in the command line is used as the default

option.

For

example,
_. [I [I] M I33 I I... E

F' I I... IS:

The COMPILE command can recall arguments from a previous COMPILE, LOAD, EXECUTE, or PAL command.
This allows the COMPILE command to be typed without any arguments.

However, it must be used on the same day
previous day.

because the COMPILE command does not remember arguments typed on a
NOTE

Refer to the appropriate language chapter for any error
messages received as a result of the COMPILE command.
This command causes execution of the CCL.SV program and either PAL8.SV, F4.SV, or BCOMP.SV.

OS/ 78 Commands

COPY

3.5

COPY COMMAND

The COPY command transfers files from one device to another. It also allows you to change the file’s name.

Format:
.COPY outdev:fi1e.ex<indev:file1 .ex...fi1e5.ex/options
When the COPY command is executed, all Specified input files are copied onto the output device. A message informing you of the files copied followed by the name. of each file copied is displayed until copying is done.

Example:

LCOF'Y

R'Xfil:-=ZIF'INEVTEQ|\'VROSE

FILES

COF'IEID:

TEAK
PINE
ROSE
3.5.1

File Transfer

When the COPY command is executed, each input file is copied in its exact format to the output device in the same
order they are found on the input device. No changes are made during file transfer and as a result, any file, whether
memory image, binary or ASCII, can be transferred. During the transfer, the original creation date is transferred and
included in the directory. Along with the necessary arguments for the command line, COPY options can be speci-

fied that can change and enhance the result of a file transfer.
If input files are to be transferred to the output device in the exact order specified in the command line, use the /U

option.
If the majority of the input files on the input device are to be transferred to the output device, use the /V option
and specify the files that are not to be transferred. The COPY command then transfers all the files that are not
specified in the command line. Wildcards can be used with the COPY command (described in Section 2.5.6).
3.5.2

Examples of Copy Commands

The following are legal COPY command strings. When the operation is completed, control returns to the Monitor.
Some of the examples show the use of wild cards.

Example 1:

LCUF'Y

RXA1:<:IMATH . PA

This command string transfers file MATHPA from the device DSK to RXAl.

Example 2:

LCDF‘Y RXfil 3‘39: 9 B y C v [I y E
This command string transfers the files A, B, C, D, and E from the system device to RXAI, producing a list of those
files copied.

Example 3:

_._COF'Y

LF‘T : {2* . FT ’ )k 0 BA/U

This command string lists all FORTRAN files, then all BASIC files on the line printer (same as LIST).

3-8

OS/ 78 Commands
Example 4:

:COF‘Y

LF‘T 3 {IF-:XAO 2 )k 0 SU , )k . BN/U

This command string copies from RXAO to the line printer all files other than memory image (.SV) and

binary (.BN).
3.5.3

Predeletion

Before the COPY command copies the designated input files to the output device, the output device is checked for
files having the same name as the input files. If an identical file name is found, that file is deleted before file transfer
is performed. This process is known as predeletion and, by default, is automatically done each time the COPY

command is executed.
If there are no files with the same name on the output device, each file is transferred to the smallest available empty
space that it can fit in.

Predeletion is advantageous because it creates space for a new file on the output device by deleting the old file before file transfer. In many cases, the output device has received as many files as it can hold and therefore may not
have enough room for a new tentative file.

By first deleting the old file, a sufficient amount of space could be
created for the new file. Predeletion normally places the new file in the space occupied by the file being replaced.
Therefore, if the length of the input file is the same, or shorter than the length of the deleted file, that empty space
Otherwise, the new tentative file is placed somewhere else on the device leaving a gap where the old file was
deleted. This alters the order of the files on the output device after transfer.

is filled.

If predeletion is not desired, use the /N option in the command line. When the /N option is specified, postdeletion
takes place.
3.5.4

Postdeletion

If the /N option is used with the COPY command, the designated input files are copied to a tentative file on the

output device without checking for files having the same name. When the transfer operation is completed, the
tentative file is closed.

Closing a tentative file makes it a permanent file and, at the same time, deletes any old files

having the same name. This process is known as postdeletion and takes place only when the /N option is specified.

Example:

_.CUF'Y

Fx'Xf—Yl 3*1RXAO 2 ACCTOé/N

A file named ACCT06 on device RXAO is copied to a tentative file on device RXAl. After file transfer is completed, the file is closed and any other file on device RXAI with a filename of ACCTO6 is deleted.
The primary advantage of postdeletion is that if an input/ output error
(due, for example, to a bad diskette) occurs
on reading an input file, the
corresponding output file will be preserved instead of deleted. Always use the /N
option when copying important files to backup diskettes to avoid losing both file and backup file in the event of an

input/output error.
3.5.5 Considering the Date on a Transfer
When files are transferred from one device to another, their creation dates are also transferred. These dates along
with their new or original file names are included in the output

directory. If it is desired to assign the current date

to an output file and disregard the creation date, use the /T

option. This option allows the COPY command to
transfer files from one device to another, and append the file names and the current date to the

output directory.

To transfer all input files that have the current date, use the /C

option. When the /C option is used, the COPY
command copies only those files that were created or changed on the current date and
the filenames and

appends

their creation dates to the output directory.

05/78 Commands
However, the transfer of all input files that have a creation date other than the current date is done by using the
/0 option. By specifying the /0 option in the command line, the COPY command copies only those files with a
date other than the current date from one device to another, and adds the file names and their creation dates to the

output directory.
3.5.6

File Protection During a Transfer Operation

When transferring a large number of files, use the / F (failsafe) option. When the / F option is Specified and an input

file is encountered which will not fit on the output device, a message is printed on the terminal informing you to
dismount the current diskette and mount a new diskette on the same drive. To continue the COPY operation, type
device would not
any character. If the / F option had not been specified, then files which did not fit on the output
get copied. Instead, a warning message would be printed:
NO ROOM, SKIPPING filename

Example

_.(:or-=-Y

RXM:-=::RXA0:*.FT/F

FILES COPIED:
UERS.9.FT
VER812.FT
UER810.FT
UER811.FT
VERS7.FT

MOUNT

UUTF‘UT

VOLUME:

VER85oFT
VERSéoFT
UER88.FT
When the / F option is used, all output directories on new devices that might be used for continued COPY operations
should be emptied (zeroed). This is done by using the ZERO command before the COPY operation begins. It is

possible to continue COPY operations to a device with already existing files. However, if the length of the next file
to be transferred exceeds the lengths of the empty spaces on the output device, another message for a new device to

be mounted is printed on the terminal.

If many files are being transferred at one time, especially ones with similar names, each affected file specified in the
command line can be verified as the correct file for the operation by using the / Q (query) option. When the / Q

Option is specified, the first input file name followed by a question mark (7) is printed on the terminal. At this time,
a decision must be made on whether or not it is the correct file for the operation. If yes, type a Y and that file is
transferred to the output device. If no, type any other character and that file is ignored. After your file is either
transferred or ignored, the next input file name followed by a question mark is printed on the terminal. This

questioning continues until all affected files in the command line have been processed.

3—10

OS/ 78 Commands
Example:
8Y8 2 *ZIR'XWli

LCDF'Y
FILES
MAIN

ESSESIM .-, W) it MAIN 4. FT/Q

COF‘IED:
FT'i’Y

ASSEMJi'fiiw

In this example, the file named ASSEM is not transferred.
3.5.7

Terminating Copy Operations

The only control character used to abort COPY operations is CTRL/C.
When CTRL/C is typed during 3 COPY operation, file transfer terminates immediately and control returns to the
Monitor. Because file transfer is terminated before a successful completion, some requested files may not have

been transferred.

Example :

LCOF'Y

RXAI 2-:IIRXA02TABLEoFT

er
FILES

COF‘IEI‘J:

This command causes the execution of batch the CCL.SV and FOTP.SV programs.
Table 3—3

COPY Options

Option

Meaning

Transfer all input files that have the current date to the output

(current date)

device.

A file will not fit on the output device. The message:

(failsafe)
MOUNT NEW DEVICE:
is printed on the terminal. Dismount current device and mount
new device on same

unit. Continue the COPY operation by

typing any character. The new device must have a valid OS/78

directory on it.
/T

Transfer all specified input files to the output device and at the

(today’s date)

same

time, change the creation date on the output files to the

today’s date.
/U
(in exact order)

Find and transfer the designated input files in the exact order

they are specified in the command line, not the order they are
found on the input device.

Transfer all files on the input device that are not specified in the

(other than the specified

command line to the output device.

files)

Continued on next page

0S/ 78 Commands
Table 3-3 (Cont.) COPY Options

Meaning

Option

/W
(current version
number)

Before any specified input files are transferred to the output
device, print the current version number of the program used by

/N
(no predeletion)

After each file is transferred as a tentative file, any existing file

the COPY command.

having the same name is deleted and the tentative file becomes
permanent. This is postdeletion.

/0
(other than the current date)

Transfer only those input files with dates other than the current

Each time a file name followed by a question mark is printed on

(query the user)

the terminal, type a Y (yes) to transfer the file or any other

date.

character (no) if the file is not to be transferred.
NOTE

All output directories on devices that could be used for
continued COPY operations (/ F) should have valid

OS/78 directories on them. One way to accomplish this
is to ZERO a diskette before execution of the COPY
operation. This diskette can then be used as an overflow
diskette for any files that do not fit on the original
diskette.

(See the ZERO command).

3-12

OS/ 78 Commands
Table 3-4

COPY Error Messages

Meaning

Message
BAD INPUT DIRECTORY

The directory on the specified input device does not exist or is
not valid.

BAD OUTPUT DEVICE

This message appears when a non-file structured device is speci-

fied as the output device.
BAD OUTPUT DIRECTORY

The directory on the specified output device does not exist or is
not valid.

ERROR ON INPUT DEVICE
SKIPPING (file name)

The file specified is not transferred, due to a disk-reading error,
but any previous or subsequent files are transferred and indicated
in the new directory.

ERROR ON OUTPUT DEVICE
SKIPPING (file name)

ERROR READING INPUT

The file specified is not transferred, but any previous or subse-

quent files are transferred and indicated in the new directory.
A disk-reading error occurred on the input device.

DIRECTORY
ERROR READING

A disk-reading error occurred on the output device.

OUTPUT DIRECTORY

ERROR WRITING

A disk—reading error occurred on the output device.

OUTPUT DIRECTORY
ILLEGAL *

An * was entered as an embedded character in a file name, e.g.,

TMP*,BN (see Section 2.5.6 on wild cards).
ILLEGAL ?

A ? was entered in an output specification (see Section 2.5.6
on

NO FILES OF THE
FORM xxxx
NO ROOM, SKIPPING

(file name)
SYSTEM ERROR-C LOSING

wild cards).

No files of the form (xxxx) specified were found on the current

input device group.
No space is available on the output device to perform the

transfer. Predeletion may already have occurred.

Directory format or system has become corrupted.

FILE

3-13

05/ 78 Commands

CR EAT E

3.6

CREATE COMMAND

The CREATE command calls and runs the 08/78 Editor to allow a new file to be opened. Once opened, the new

file is ready to receive any text typed in. Only one output file specification is allowed.
Format:

.CREATE outdev:file .ex/ options
After pressing the RETURN key, the number sign (#) is displayed on the terminal, indicating that the Editor is

ready to receive Editor commands. The CREATE command also allows the Editor options. (See Chapter 4 for a
detailed explanation of the 08/78 Editor).

Example :

l'fl'l‘

CREATE

RXAI ZRUNI .F'A

In this example, the CREATE command opens a file named RUN1.PA on output device RXAl

Each time a CREATE command is executed, the accompanying arguments (device and file name) are remembered in
a temporary area. These remembered arguments allow the EDIT command to be typed without any arguments,
allowing further editing of a file previously created (on the same day). However, the EDIT command does not recall
any files without an assigned date. Therefore, the DATE command should be used prior to the CREATE command.

This command causes the execution of both the CCL.SV and EDIT.SV programs.

3-14

CREF

OS/ 78 Commands
3.7

CREF COMMAND

The CREF command assembles a PAL program and produces a cross-reference listing usually on LPT.

Format:

.CREF outdev:file.ex<indev:file .ex/ options

Example:

LCR‘EF

F‘ROGA

The cross-reference listing contains the source program with the assembled instructions followed by the crossreference table which contains every user-defined symbol and literal. These listings serve as an aid in writing, debugging and maintaining assembly language programs. CREF is the same as using the PAL command with the /C
option specified. Note that if no output device is specified, LPT is the default output device.
This command causes the execution of the PAL8.SV,CCL.SV, and CREF.SV programs.

Table 3-5

CREF Options

Meaning

Option Code
/P

Disable pass one listing output. The output is reenabled when
S is encountered. Thus the $ and symbol table are printed if the

/P option is used.
/U

Disable pass one listing output and the symbol table.

Do not process literals. For programs with too many symbols
and literals for CREF, this option may create enough space for
CREF to operate.

Do not eliminate the intermediate file CREFLS.TM that is out-

put from the assembly and used as input to produce the CREF

listing.
/M

Cross—reference mammoth files in two major passes. Pass one
processes the symbols from A through LGnnnn; pass two processes the symbols from LHnnnn through Z and literals. This

permits significantly large files to be cross—referenced.

3-15

OS/ 78 Commands
Table 3-6

CREF Error Messages

Error Messages

SYM OVERFLOW

Meaning
More than 896 (decimal) symbols and literals were encountered

during a major pass. Try again using the /M option.
ENTER FAILED

Entering an output file was unsuccessful
specified to a read only device.

OUT DEV FULL

The output device is full (directory devices only).

CLOSE FAILED

CLOSE on output file failed.

INPUT ERROR

A read from the input device failed.

2045 REFS

More than 2044 (decimal) references to one symbol were made.

HANDLER FAIL

This is a fatal error on output.

3-16

—

possibly output was

OS/ 78 Commands
3.8

DATE

DATE COMMAND

The DATE command enters the date specified in the command line into the system.

Format:
.DATE dd-mmm-yy
where
dd

is a 2-digit decimal number representing the day,

mmm

is a 3-character alphabetic string representing the month which must be the first three letters of

the month, and
is a 2-digit decimal number representing the last two digits of the year.

Example:

Lin-VHS.

23""MAY-"77

If the DATE command is typed without any arguments after the date has already been entered by a previous DATE

command, the current day of the week and date are displayed.

Example :

LIN)
LITA

II.

JUN“- 7 7‘

NEIIIINIEEESIIIQY

JUNE

1 7

1977

If the current date is entered into the system via the DATE command after booting, the only directory entry dates
that are valid are those for the current year and seven years preceding the current date.

Any dates prior to this time

will be printed incorrectly.
The DATA command is used to date directory entries when creating or updating, to date directory listings, and to
date program output listings.

If the date is specified incorrectly,

an error message is

displayed.

Example:

Lnn

23w05w77

BAH

DATE

This command causes the execution of CCL.SV and is processed by the Monitor.

3-17

OS/ 78 Commands

DEASSIG N

3.9

DEASSIGN COMMAND

The DEASSIGN command removes all user-defined names that were previously assigned to one or more of the permanent devices.

Format:

.DEASSIGN

Example :

LAB
Lm:

SYS

DEUl

In this example, the system device can no longer be referenced as DEVI.
The Monitor performs the DEASSIGN function.

3-18

Only SYS, DSK or RXAO are valid.

OS/ 78 Commands
3.10

DELETE

DELETE COMMAND

The DELETE command deletes all specified files.

Format:

.DELETE indev:filel.ex...fi1e5.ex/options
In response to the dot produced by the Monitor, type the files that are to be deleted in command line and press
the RETURN key. This will delete the files from the specified device. Wildcards are permitted.

Only one device

may be specified.

Example:

LEIEI...

r(X1431:Rli-IF'OR'T

F I I... IE: 53

[IE l... E T IE [I 2

FT , SECUF: . FT 9 ASST 8T

SEISEICUR . F T
Ft E2. F'OR T I, F T

:1} $3 $3 I 8 T
A message followed by the name of each file deleted is displayed on the terminal.

NOTE
If no filenames are specified, *.* (meaning all files) is
the default input specification.

This command deletes all copies of your system FORTRAN files which are on RXAl.

Example:
6

1'] I33. I...

R‘Xf-h 1 '5; >06.

«.

*-=IISS:‘-YSES 3 *2 . FT

An advanced format that can be used with the DELETE command is

.DELETE outdev:file.ex<indev: file 1,.../ options
If an output specification is given, then this command works differently. Instead of deleting the input files speci-

fied, a list is made up of those file names which match the given file specification, and that are on the input device.
This list is then transformed to an output list by applying the wildcards specified in the output specification. The

resulting list is then used to delete files from the output device specified.
This command causes the execution of both the CCL.SV and FOTP.SV programs.

OS/ 78 Commands
Table 3-7

DELETE Options

Delete only those specified files having the current date.

Delete only these specified files with dates other than the current date.

Delete all files that are not Specified in the command line.

Each time a filename followed by a question mark is printed on the terminal, type a Y (yes)
to delete the file or any other character

(no) if the file is not to be deleted.

Displays a log of all files matched on the output device to be displayed, but no files are actually
deleted.
Table 3-8

DELETE Error Messages

BAD OUTPUT DEVICE

Self—explanatory. This message usually appears when a nonfile structured device is specified as the output device.

BAD OUTPUT DIRECTORY

The directory on the specified output device is not a valid

OS/78 device directory.
DELETES PERFORMED

More than one input device was specified with the /D option

ONLY ON INPUT

when no output specification (device or file name) was

DEVICE GROUP 1

included.

CAN’T HANDLE
MULTIPLE DEVICE
DELETES
ERROR READING

Self-explanatory.

OUTPUT DIRECTORY
ERROR WRITING

Self-explanatory.

OUTPUT DIRECTORY
ILLEGAL *

An * was entered as an embeded character in a file name, for

example, TMP*.BN.
ILLEGAL ?

A ? was entered in an output specification.

NO FILES OF THE

No files of the form (xxxx) specified were found on the cur-

FORM xxxx

rent input device group.

3-20

OS/ 78 Commands
3.11

DIR ECT

DIRECT COMMAND

The DIRECT command sends the directory listings for file-structured devices to an output device. The terminal
is the default output device.

Format:
.DIRECT dev:listfi1e<dev:filetype,.../options

Example:
.

DIRECT

Fx'Xi‘hil.§

Along with the necessary arguments for the command line are DIRECT options that can limit or expand the result
of the DIRECT operation. The standard directory listing has the following format:
filenameex

nnn

dd-mmm-yy

(filename with extension)

(number of

(day, month, year)

blocks used
in decimal)

NOTE

If the date is not entered into the system via the DATE

command, directory listings will not contain file creation
dates.
If no output device is specified, the output defaults to the terminal.
on

LPT,

use the CCL

However, if you want the directory printed

option -L in the command line.

Example:

LIZ! I R

R'Xfi 1 3 ml...

is equivalent to

iIIIR

I... F' T z ~22: RXr—fi 1 z

Wild cards are permitted in the input specification but are not allowed in the output specification.

Example:

LEI I F:

SYS t ‘IK t F'T

As a result, only those files with FT extensions in the directory are displayed on the terminal. For example,

3.13 I FR'ECT

lii'Xfi 1 '£ b1i‘-l?’?'§”i’

will display all the files from the directory on RXAl that have a file name two to six characters long beginning with

WN, and having any extension.

3-21

0S/ 78 Commands
To save the directory listing and store it in a file, specify the file name and device that it is to be stored on.

Example:

Lilli-1‘
3.11.1

Fix-Q1 ZITIIREEICT

[II-SENS:

Considering the Date in a Directory Listing

Use the /C option to display the names of the files that have the current date on the terminal. To display only those

file names from the directory with a date other than the current date, use the /0 option in the command line.
3.11.2

Choosing the Directory Format

To display just the file names from the directory, use the / F option in the command line. The / F option displays

only the file names, and omits all block lengths and dates.
The /E option displays file names with their extensions, lengths, and dates, and empty files with their lengths in
the exact order they are on the diskette.

When the /M option is specified, each empty file and its file length is displayed. Do not mix the /M option with
the other options.
The =n option displays the list of file names in columns. The number of columns is determined by the decimal
number specified as n. The values 1 through 7 can be substituted for n. For use on the terminal, values 1 through
3 are suggested. Values more than 3 are normally used for the line printer. Two columns is the default value. This
value can be changed via the SET command. See the SET command for additional information.
To display a specific file name and all file names from that point on, use the /R option in the command line. This

option should not be mixed with other options.
If there is a large number of file names from the directory to be displayed, use the /V option.

By specifying the

/V option, all file names that are not specified by the command line are displayed.
The starting block numbers (octal) of each file in the directory are displayed by using the /B option. The starting

block numbers are in the form of nnnn and displayed following the file name and extension.
This command causes the execution of both the CCL.SV and DIRECT.SV programs.

3-22

OS/ 78 Commands
DIRECT Options

Table 3-9

Include the starting block numbers (octal) for each file in the directory listings.

List only files with the current date, that is, the date entered with the most recent DATE

command,

Include empty file spaces in the directory listing.

List only the file names and omit the lengths and dates.

List only the empty files from the directory.

List the directory in the number of columns specified by n. The n is a number ranging from

1-7.

List only those file names from the directory with dates other than the current date.

List the file name specified in the command line and the names of any files stored on disk follow-

ing that specified file name.
List each input specified as a separate directory listing with the file lengths and dates in the exact

order specified in the command line.

List all the files on the input device that are not specified in the command line.

Display the current version number of DIRECT.
Table 3-10

DIRECT Error Messages

Message
BAD INPUT DIRECTORY

Meaning
This message occurs when the directory is non-existent or not
in legal format.

DEVICE DOES NOT HAVE

This input device is a non-directroy device, for example, TTY

A DIRECTORY

LPT. DIRECT can only read directories from file-structured

devices.

EQUALS OPTION BAD

The =n option is not in the range 0-7.

ERROR CLOSING FILE

System error.

ERROR READING INPUT

A disk—read error occurred while trying to read the directory.

DIRECTORY
ERROR WRITING FILE

A disk-write error occurred while trying to write the output file.

ILLEGAL *

An asterisk (*) was included in the output file specification or an

illegal

was included in the

input file name.

ILLEGAL ?

A question mark ('2) was included in the output file specification.

NO ROOM FOR

The output device does not have sufficient space for the direc-

OUTPUT FILE

tory listing file to be written.

3-23

OS/ 78 Commands

DUPLICATE

DUPLICATE COMMAND

3.12

The DUPLICATE command copies or transfers the entire contents of one diskette to another diskette.
Format:

.DUPLICATE outdev:<indev: /options
In addition to the necessary arguments in the command line, DUPLICATE options can be used to affect the
DUPLICATE operation.

Example:

LDUF‘L I CAVE".

RXA II. :-=::r~‘:><no 2

The contents of input device RXAO is copied onto output device RXAl.
3.12.1

Changing Devices Before and After Executing the DUPLICATE Command

You can only duplicate from RXAO to RXAl or RXAl to RXAO. Driver RXA2 and RXA3 is not supported by the
DUPLICATE command. Since the Monitor resides on the system device, the system device must remain on line
when interacting with the Monitor and any OS/78 system programs.

If you want to transfer the contents of a diskette containing only files (one that does not contain a system), use the

/P option. The /P option pauses before and after its execution of the DUPLICATE command. A ready message
followed by a question mark is displayed. This pause provides time to remove the system device and mount a device
onto which the contents are to be transferred, to or from. To start the DUPLICATE operation, type a Y after the

question mark and press the RETURN key. After the DUPLICATE operation is completed, a message is displayed
asking if the Monitor is remounted. The second pause provides time to remove the new device and remount the
system device so control can return to the Monitor. After remounting the system device, type a Y after the question
mark and press the RETURN key, to return control to the Monitor.

Example:
R X («i 0 i <1 Fl“ X A ll. Z x’ F'
1 [I U
REfiIIY’PY

III S

M {I} N I T [I] F:

R I}: M U U N T E: [I 'F’ ‘r'

3.12.2

Performing a Read Check

To check the integrity of a diskette, use the / R option and specify only the input device. By specifying the /R

Option, every block of the specified device is read and checked for bad sectors. If bad sectors exist, a message with
the device, track number and sector number is displayed. If none exist, control returns to the Monitor.

Example:

lIIIU

-=2IRXfiIl3/F:

INPUT

3.12.3

DE!)

READ

ERROR

Tl-‘(fiCKH‘n'i

Sliiltnrr

Transfer Without Checking for Identical Contents

To transfer the contents of one device to another without performing any checks, use the /N option. By Specifying
the /N option, the contents of__the input device is transferred or copied to the output device. If /N is not specified
and the DUPLICATE operation is completed, the contents of the output device is compared to that of the input

device‘to assure accuracy.

3-24

OS/ 78 Commands
3.12.4

Check for Identical Contents Without Transferring

To check the contents of devices to see if they are identical without transferring, use the /M option. By specifying

the /M option the contents of both devices are read and checked for a match. If they do match, control returns to the
Monitor. However, if they differ in any way, a message, the device name, the track number, and the sector number of
the sectors or blocks that do not match are displayed.

Example :
RXAII.$<ZI-1'XfiOt/M
COMPARE: ERROR TRACK

_._I|U

l"il"l

SECTOR

This command causes the execution of both the CCL.SV and the RXCOPY.SV programs.
Table 3—11

DUPLICATE Options

Meaning

Option

Pause and wait for user response before and after data transfers to/from diskette.

Copy the contents of one device to another but do not check them for identical contents unless
otherwise specified.

Check both devices for identical contents and list the areas that do not match but do not perform
a

transfer unless otherwise specified.

Read every block on the input device and list the bad sectors but do not perform a transfer unless

otherwise specified.

Display the current version number of the program invoked by the DUPLICATE command.
Table 3-12

DUPLICATE Error Messages

Meaning

Message
N0 INPUT DEVICE

CAN’T LOAD INPUT DEVICE

No input device is specified.

The name of the input device specified in the command line
is not a permanent device name.

CAN’T LOAD OUTPUT DEVICE

The name of the output device specified in the command line
is not a permanent device name.

COMPARE ERROR

When using the /M option all the areas that do not match are

printed as COMPARE ERRORS.
INPUT DEVICE READ ERROR

Bad data on input device; bad sectors.

OUTPUT DEVICE READ ERROR

Bad data on output device; bad sectors.

OUTPUT DEVICE WRITE ERROR

Hardware disk-write error.

DEVICE IS NOT RX

Either the input or output device specified is not one of the

following:

3-25

RXAO:

SYS:

RXAl:

DSK:

OS/ 78 Commands

EDIT

3.13

EDIT COMMAND

The EDIT command calls the 08/78 Editor which allows the editing of already existing files. These files must
contain ASCII source code.

Format:

.EDIT outdev:file.ex<indev:file.ex/options-ex
In response to the dot produced by the Monitor, type the necessary command line and press the RETURN

key. The file is now open and the output file is ready for modification. After pressing the RETURN key, the
number sign (#) is displayed on the terminal, informing you to proceed with an Editor input command which is
almost always a read (#R) command. (See Chapter 4 for a detailed explanation of the 08/78 Editor).
Example :
l° IEEIIII T

RXA 1 2 F I LE1 . FT-fiin'XAO 2 TABLE . FT

[#1
In this example, the EDIT command opens output file FILE1.FT an output device RXAl: and Opens input file
TABLE.FT on input device RXAO:.

NOTE
Since no output device is specified, the default option

(DSK:) is used as the output device.
The EDIT command can recall arguments from a previous CREATE or EDIT command. This allows the EDIT
command to be used without any arguments. However, the command must be used on the same day the file was
created because the command does not remember arguments that were typed on a previous day. However, if you

specify both an input and an output file, only the arguments up to but not including the left-angle bracket (<) are
remembered.

Example 1 :
rawauwmy7

inn
1rm

TUESDAY

JUNE

LCREATE

MATH.PA

17,

1977

35
LEDIT

1R
(MATH.PA is remembered by
the EDIT command)

3-26

OS/ 78 Commands
The above example shows a sequence of commands in creating and editing of file MATH.PA. The EDIT command is

given without any argument since it remembers the file MATH.PA specified with the CREATE command.
If the date has changed and a command line is typed containing the EDIT command without any arguments, the
error message

BAD RECOLLECTION will be displayed.

Example 2:

1!: I] f. I
an

1'1}?

F."

f.) j

v.-

er:
LEDIT
Since the second EDIT command had no arguments, the previous argument up to the left-angle bracket (<) is

recalled. Thus, the second EDIT command is equivalent to:

innit

netwr

This command causes the execution of both the CCL.SV and EDIT.SV programs.

3-27

OS/ 78 Commands

EXECUTE

3.14

EXECUTE COMMAND

The EXECUTE command performs the following functions:

1.
2.
3.

assemble/compile, link, load and execute a source program;
link, load and execute an already assembled or compiled program; or
execute an already linked and loaded program.

Format:

.EXECUTE outdev:file.ex<indev:file.ex/options-ex
To use the EXECUTE command with a source file, the input device must be specified. Then that source file is

assembled or compiled, linked and loaded into memory and executed. The file extension used determines the

compiler or assembler called.
If the file name extension is not a standard extension, use the correct assembler or compiler extension as a CCL

option in the -ex portion of the command line as follows:

Example:

LIEEXIE

RXfil 2NABS o IN—~F'A

If the EXECUTE command is used with an already assembled or compiled program, the input device must be

specified. When the EXECUTE command is processed, the binary file is loaded into memory and executed. If the
file name extension is not specified in the command line but is specified in the directory, the correct extension is
automatically added.
The EXECUTE command can remember arguments from a previous COMPILE, LOAD, PAL, or EXECUTE
command. For the EXECUTE command to recall remembered arguments, the EXECUTE command must be used
on the same

day the remembered arguments were stored in the temporary file. The reason for this is that the

command does not remember arguments typed on a previous day.

NOTE

For‘Options acceptable to the EXECUTE command,
refer to the appropriate language section.
This command causes the execution of CCL.SV and one of the following: PAL/8.5V and ABSLDR.SV, F4.SV and

LOAD.SV, or BCOMP.SV and BLOAD.SV.

3-28

OS/ 78 Commands
3.15

G ET

GET COMMAND

The GET command loads memory image files with .SV extensions from an input device back into the same

location in memory from which it was saved.
Format:

.GET indev: file.ex

Example:

16!}: T

F: X (-10 ‘3 JG BUN T

If you specify a file name with no extension, an .SV extension is assumed.

Example:
:

[I-IiIEEI T

RX r3: ll. 2 F" I LIES 3

In this example, F ILE3.SV is loaded into memory.

The specified input file along with its Job Status Word is loaded into memory. The entire Core Control Block is
sent to the system device where it can be referenced and maintained (see Section 2.8.4).

The GET command is commonly used to load save files from a device into memory. These save files can be executed
either by the START or EXECUTE command.
In addition, they can be debugged by the 08/78 octal debugging program (see ODT command).

NOTE
The colon used with the permanent device name can be

replaced with a space.
The Monitor performs the GET function.

3—29

05/ 78 Commands
3.16

HELP COMMAND

on the terminal. These HELP files
The HELP command accesses a specified HELP file and displays its contents
line (see Table 3-14).
contain useful information about the OS/ 78 program you specified in the command

Format:
.HELP outdev:file.ex<argument
where

argument is usually an OS/78 command.
NOTE

If no output device is specified, the terminal (TTY) is
assumed.
For the current HELP files available, see the list containing all 05/ 78 HELP files below.
If no output file extension is specified the .HL extension is assumed. There is a HELP file which contains all OS/78
commands. This file is displayed by typing the HELP command without any arguments as follows:

iHELP
If a HELP file for a specific command is desired, type the name of the command for which the information is
desired. For example,
iH

E I... F'

F' A I...

To obtain a list of all legal arguments for HELP, type the HELP command followed by an asterisk or type the HELP
command followed by ‘HELP’ as follows:

:HELP

LHELP

HELP

Table 3-13

HELP Error Messages

Meaning

Message
NO HELP

There is no HELP.HL file on the system device.

NO TTY HAND

There is no handler in the system for the terminal.

NO HELP FILE

The input file specified does not exist as a HELP file.

READ ERR

An error occurred while reading the specified HELP file.

WRITE ERR

An error occurred while writing to the output file.

FETCH ERR

Cannot load handler for the specified device.

DEVICE FULL

The output device becomes full when writing or is already full before output
'

begns

3-30

OS/ 78 Commands
Table 3-14

Available OS/78 Help Files
LIST

ABSLDR
ASSIGN

LOAD

BASIC

MAP

COMPARE

ODT

COPY

PAL

CREATE

CREF

RENAME

DATE

RUN

DEASSIGN

SAVE

DELETE

SQUISH

DIRECT

START

DUPLICATE

SUBMIT

EDIT

TERM

TYPE

GET

ZERO

3-3 1

OS/ 78 Commands

LIST

LIST COMMAND

3.17

The LIST command prints the contents of the specified input files on LPT.

Format:
.LIST indev;file l .ex...file5 .ex/ options

Example:

ALI

PROD . Bfi

Along with the necessary arguments for the command line are options that affect the LIST operation.
When you use the LIST command, the contents of each input file is listed on LPT in the same order the file names
are

specified in the command line.

Example:

_._L I

FUN-511 3 F7 L. 2|. . FT 9 Phil? t FT r MASFL. Q FT

Although the format specifies that only five files can be used in the command line, more than five files can be listed
on the line printer through the use of wild cards.
3.17.]

Using LIST Options

You can put a restriction on the files you want by specifying the LIST options in the command line. By specifying
the /C option, all the specified input files with the current date are listed. If the /0 option is used, all specified input

files with dates other than the current date are listed. If the /V option is used all files other than those you spec—
ified in the command line are listed. If you do not know which files you want to go through the directory, file by

file, use the / Q option. By specifying the /Q option, each file name followed by a question mark is printed on the
terminal. Then you must decide whether or not that file is a desired file for the LIST operation. If yes, type a Y
and that file is listed. If no, type any other character and that file is ignored. After your file is either listed or ignored,
the next file name followed by a question mark is printed on the terminal. This questioning continues until all

specified input files have been processed.
Example:

LLIST*.PA/G
TOT . PA?Y

MASFIL

F'A?N

UF'IIT06 . F'A?N
FINTOT .F'A?Y
As a result, the files named MASFIL.PA and UPDTO6.PA are not listed.

This command causes the execution of both the CCL.SV and FOTP.SV programs.

3-3 2

OS/ 78 Commands
Table 3-15

LIST Options

List only those files having the current date.

List only those files having a date other than the current date.

List all the files not specified in the command line.

Each time a file name followed by a question mark is printed on the terminal, type
a Y

(yes) to list the files or any other character (No) if the file is not to be listed.

Table 3-16
BAD INPUT DIRECTORY

LIST Error Messages

The directory on the specified input device is not valid

OS/78 device directory.
BAD OUTPUT DEVICE

Self-explanatory. This message usually appears when a
non-file structured device is specified as the output device.

ERROR ON INPUT DEVICE,

The file specified is not transferred, but any previous or

SKIPPING (file name)

subsequent files are transferred and indicated in the new
directory.

ERROR ON OUTPUT DEVICE,
SKIPPING (file name)

ERROR READING INPUT

The file specified is not transferred, but any previous or

subsequent files are transferred and indicated in the new
directory.

Self-explanatory.

DIRECTORY
ILLEGAL

An * was entered as an embeded character in a file name,
for example, TMP*.BN.

ILLEGAL ?

A ? was entered in an output specification.

NO FILES OF THE

No files of the form (xxxx) specified were found on the

FORM xxxx

current input device group.

3—33

OS/ 78 Commands

LOAD

LOAD COMMAND

3.18

The LOAD command loads either an absolute binary file with a .BN extension or a relocatable FORTRAN binary

file with an .RL extension into memory.
Absolute Binary Files

3.18.1

Absolute binary files are loaded into memory and are ready for execution.
Once a program file is loaded into memory, the following three things can be done.
1.

Start the program, using the START command.

Save the program, using the SAVE command.

Examine or debug the program, using the CDT command.

Format:
.LOAD indev: file 1 .ex...file9.ex/options

Along with the necessary arguments for the command line are options for absolute binary file that can affect the
LOAD operation.

Example:

LLOAII

FLIS O RN 9 l‘i'lJNSESEIQ é BN
NOTE
Use the CDT command to examine memorylocation after
the LOAD command is entered and your files are loaded.

If all concatenated programs with .BN extensions located in a single file are to be loaded into memory, use the /S

option in the command line. By specifying the /S option, not only the first program found with a .BN extension
in a single file but all programs with .BN extensions are loaded into memory.
3.18.2

Loading Programs in Specified Areas

Absolute binary files can be loaded into a field other than the one they would normally load into by using /n

option, where n is an octal number 0 through 3 which represents a field. By specifying the /n option, all specified
input files are loaded beginning with the indicated field at a loading address equal to the address specified in the
file.
Both the field and the starting address can be specified by using the =fnnnn option
where
f
nnnn

is an octal number 0 through 3 which represents the field, and
is a 4-digit octal number 0000 through 7777 which represents a starting address.

NOTE
If fnnnn is all zeros, the default option (field 0 starting address

0200) is used.
If locations 0-1777 in field zero are not being used by the program, use the / 8 option in the command line. By

specifying the / 8 option, the contents in these locations are not saved nor is the Command Decoder (system
program) reloaded. In effect, these unnecessary operations are not performed.

334

OS/ 78 Commands
Iflocations 0-1777 in field one are not being used by the program, use the /9 option in the command line. By
specifying the / 9 option, the contents in these locations are not saved nor is the User Service Routine (system pro—

gram) reloaded. In effect, these unnecessary operations are not performed.
3.18.3 Editing or Patching a SAVE File
Changes can be made to the contents of files after they are already assembled and saved on a device by using the /1
option in the command line. By Specifying the /1 option, SAVE files are loaded and treated as absolute binary files.
This allows patches to be made to an already saved file without reassembling the entire program.
Execution After Loading

3.18.4

If execution of the program is to follow immediately after the loading of the program use the / G option in the command line.

By specifying the /G option, execution takes place immediately after loading is completed. Control is

transferred to the executing program and remains there until completion or until a transfer to the Monitor from
within the program takes place.

3.18.5

Clearing Memory After Loading

If you have already loaded a specific number of files and want to change either some or all of the files loaded into
memory, use the / R option in the command line.

By specifying the / R option, the portion of memory where your

files were loaded is reset or cleared to appear as though nothing had been loaded.
When absolute binary files are loaded using the LOAD command, both the CCL.SV and ABSLDR.SV programs are
executed.

3.18.6

Relocatable FORTRAN Binary Files

Since relocatable FORTRAN binary files are dependent programs, both files and system programs or subroutines
are loaded

and linked in memory.

Format:
.LOAD outdev: file.ex<indev;file.ex/options-ex

After files are loaded, the system library FORLIBRL is searched for programs or subroutines that are referenced by
the program. When the subroutines are found, they are loaded and linked to the program.

If you have your own library file and have referenced several subroutines from that file, use the / L option in the
command line.

By specifying the / L option, the first input file in the command line is used as a library file and any
remaining input files in the command line are ignored. The directory is searched for the files which, once found, are

loaded in memory and linked to your program.
NOTE
A FORTRAN program can be compiled, linked, and exe-

cuted by using the COMPILE command with the /L and

/G options. This eliminates the need for using the LOAD
and EXECUTE commands.
If you have more than nine FORTRAN input files to load, use the /C (continue) option in the command line.

specifying the /C option, the remaining files are accepted as valid input and should be continued on the next entry
line.
If a symbol map output file is specified in the command line with a .RL file and you want the system symbols also

included in the map, use the /S option in the command line.

By Specifying the /S option, all system symbols are

included in the symbol map and marked by a preceding number sign (#).

When relocatable binary files are loaded and linked using the LOAD command, both the CCL.SV and LOAD.SV
programs are executed.

3-35

OS/ 78 Commands

MAP

MAP COMMAND

3.19

The MAP command produces a map of all the memory locations used by the specified absolute binary files. The
LPT line printer is the default output device for this command. Terminal output may be requested through the use
of the CCL option -T.
Format:

outdev:file.ex<indev:filel.ex.

.MAP

.file9.ex/options-ex

In addition to the necessary arguments are MAP options that can affect the MAP operation.
NOTE
If a filename without an extension is specified as the out-

put file, an .MP extension is automatically appended. Input extensions default as described for the LOAD
command.
MAP Output

3.19.1

The output of the MAP operation is a series of lines, each of which is comprised of a string of digits. Each digit

represents a single memory location, and can have the value 0, l, 2 or 3. The value means the following:

WNHO

means that

the location was not loaded into.

means that

the location was loaded into once.

means that the

location was loaded into twice.

means that the

location was loaded into three or more times.

Appearance of a 2 or 3 may imply a programming error (for example, two separate routines are each trying to
load values into the same location).
Each line of digits represents 64 (decimal) or 100 (octal) memory locations. All lines are blocked in groups of two,
whereby, each group represents a page. If the resulting map is sent to the terminal, it is bordered on top by a set of
octal digits which range from 00-77. Each pair of octal digits corresponds to one memory location (see following

example).
It is also bordered down the left hand column by a set of octal digits which are multiples of 64(decimal). Each set of
octal digits corresponds to the first memory location on that line. To determine the address (relative to zero) of a
specific location in memory, find the line containing that location. Take the set of octal digits bordering that line
and the pair of octal digits that specific memory location falls under and add both values together.
The following is an example of output on a terminal:

BITMAF‘

11-" 113511-11

00000000111. 1 1.. .1. 1.1.13.2.

~..

01.3.54336'7()122\51176:701.

54 r J/ )1."3511' 56.70 .1. 234367011. 154567021.21345670113334567

00000

’3 .755535153 3444444445303itiqcrciéééééééé77777777

'1000000000000000000000000000000000000000000000

00100
0011200

00400

005500

3-36

OS/ 78 Commands
00600
007 00
01000
01000
01 100

1 11111111111111 11.111111.111111111111.1.111111111111111111 1 111 1 1 1 111
1 1 1 1 1 1 1 1 1 1 1 1 1 2|. 1 1 .'l. 1 1 1. 1 1 1. 1 1 1 II. 1 1 1 1 1 1 1 1 1 1 .1. 1 1. 1 .'l. .1 1 1 .‘l. 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
-

01200
01300
01400
01500
01600
01700

In this example, the location resulting from the addition is 322.

The following is an example of output directed to a line printer:

RITMAP
00000

FIELD

22222222

0
21111111

10000000

00000000

00100
00000

7.70777?
11-22.12

doooonoo
111212

ovoofiov?

nonnnono

21221211

boooaooo

deadwood

poopooop
2211111“

onwonnnw

. .0..ng 22-

00300

"I
22222222

22212222

22111111

11111111

11122221

00400

75707797
~L~g~~h~

....~...,.,.._._._

A.~..«A.......~

M...~._._~.....

’9’9’37'3'9‘3’3

‘.”3’7"”?"’
’2’
~A...-.~.._‘-~

00:00
d

’7’1'3’3’.
'9’)?
1..'...A'.A--....~.

999'39777
hA—kb‘b‘n‘afiuk

A..~..~.~~.~.A.

79777797

97779979
4.aVra.~.a..a_._...

7977')???
A‘Mfifihhkh-

3979')???

’3’3'9’3'3'3'3’3

..................
22
M

00600

11111111

00700

11111111

11101111

11111111

01000
01100
01200

01300
01400

01500
01600

01700

If any of the Specified input files contain more than one program, only the first binary program in each file is used
in the MAP command.

However, by using the /S option in the command line, all binary programs in the specified
input files are used in the MAP command. If there is more than one program being mapped, and you want all maps
stored in one memory field, use the /n option, where n is an integer representing a field number. By specifying the
/n option, all the maps of programs in the specified input file are constructed in the field you specified as n.
If more than nine input files are to be specified in the MAP command, specify only nine input files, then press the

ESCape key. This causes the Command Decoder to print an asterisk (*) which allows you to continue to specify
more input files, each terminated by the RETURN key. If it is realized at this point that a wrong input file was
specified, use the /R option at the end of that line. By specifying the /R option, memory is reset to look as though
nothing has been read in. This is shown in the following example; the second ESCape causes the command to be
executed.

3-37

OS/ 78 Commands
Example:

LMQP (-1vaCvUvaF'vaHvI @
35“]
ER

il../Fx‘@
If only a terminal is available and no line printer, the style of output on the terminal can be changed to look as
though it came from a line printer by using the /T option. Similarly, the style of output on the line printer can
be changed to appear as though it came from the terminal.

This command causes the execution of both the CCL.SV and BITMAP.SV programs.
Table 3-17

MAP Options

Option Code

Meaning

Reset the map just constructed in memory to look as though nothing has been

read in.

Instead of mapping only the first binary file contained in the specified input

files, map every absolute binary file in the specified input files.

/n
/T

Confine the construction of all maps to the field you specify as n.
If output is sent to the terminal, change the format of the map to that of the

line printer and if output is sent to the line printer, change the format of the
map to that of the terminal.

Table 3-18

MAP Error Messages

Message
I/O ERROR

Meaning
An 1/0 error occurred in input file number n.

FILE #n
BAD INPUT,

A physical end of file has been reached before a logical end of file, or extrane~

FILE #n

ous characters have been found in

BAD CHECKSUM,

File number n of the input file list had a checksum error.

binary file n.

FILE #11

NO INPUT

No binary file was found on the designated device.

ERROR ON

Error occurred while writing on output device.

OUTPUT
DEVICE

NO /I

Cannot produce a map of an image file.

3-38

OS/ 78 Commands
3.720

MEMOR Y

MEMORY COMMAND

The MEMORY command is used to find out the highest memory field available to the OS/ 78 system in hardware
or to limit

that value in software.

Format:
.MEMORY n

.MEMORY

where
is an octal number representing the number of fields (4K) available to be used by OS/ 78. It is in the

range of 0-3.

Example:

LMEMDR‘Y
16K

MEMORY

The following table lists the values of n and their meanings.
11

memory

all available memory (16K)

12K

16K

To find the amount of memory actually being used by OS/78, type the command with no argument.
.

MEN

“BK/16K

MEMORY

In this example, a 16K system has been restricted to only 12K of available memory. This was done by using a

MEMORY 2 command.
If all available memory is being used, the total amount of memory is printed.

Example:

LMEM
16K

MEMORY

This command causes the execution of the CCL.SV program.

3-39

ODT

3.21

OS/ 78 Commands

ODT COMMAND

The ODT command allows you to debug a program currently in memory, control its execution, and make alterations to the program by typing ODT instructions at the terminal.

Format:
.ODT
The current program in memory is available for examination and modification of locations in memory and controlled
execution through the use of DDT instruction breakpoints. For more detailed information on ODT, see Chapter 9.

The Monitor processes the CDT command.

Example:

Lonr

3-40

OS/ 78 Commands
3.22

PA L

PAL COMMAND

The PAL command assembles a PAL8 source file, producing an absolute binary file with a .BN extension.

Format:
.PAL outdev: file.ex<indev: file.ex/options-ex
or

.PAL bin,list<input specification
to output a binary and a listing file

.PAL bin,list,cref<input specification
to output a binary, listing, and CREF file.

Example:

LF’AL

ASSEHZI. . F'A

If a file name is specified without an extension in the command line, .PA is assumed.
The PAL command can recall arguments from a previous COMPILE, LOAD, or EXECUTE command. (See Chapter 5

for information on the PAL8 assembler.)
This command causes the execution of both the PAL8.SV and CCL.SV programs.

3-41

05/ 78 Commands

3.23

R COMMAND

The R command loads the specified SAVE file in memory at its starting address and initiates execution.
NOTE

The difference between the R and RUN command is that
the R command assumes the program is on device SYS,
while the RUN command makes no such assumption. In

addition, the RUN command writes on SYS during execution to save the program’s Core Control Block for
‘

subsequent use in saving and/ or starting the programs.
The Core Control Block is not transferred to the system
device when the R command is executed.
Format:
.R file.ex
Note that a device name does not have to be given.

Example:

TEES T

When a file name is specified without an extension, a .SV extension is assumed.
The R command is most commonly used on programs that are not going to be resaved. It is suggested that the RUN
or GET

commands be used with programs that will eventually be updated (using, for example, DDT) and saved.

The Monitor processes the R command.

3-42

RENAME

0S/78 Commands
3.24

RENAME COMMAND

The RENAME command changes the name of the input file to the name specified as the output file. The same device must be specified in both the input and output specifications.
Format:

.RENAME dev:file.ex<dev:file.ex
Execution of the RENAME command changes the file name, and a message followed by the old file name is displayed
on

the terminal.

Example:
USSR 1 FILE . F'A'iIDSK 2 RECORD . F‘A

LRIZZN
FILES

RIENK-‘IMEIII:

RECORII.F'A
This new file name replaces the old file name in the input directory. The creation date and the contents of the input

file remain the same.
Wildcards may be used with this command.

Example:

lliiilEEINfiMlEEI

>l< . 1-2 h"; ~211>l< v F 'l'

renames all files with

.FT extensions to .BK extensions.
Table 3-19

RENAME Options

Option Code

Meaning

Rename the input file only if it has the current date.

Rename the input file only if it has a date other than the current date.

Rename all files other than the one specified in the command line.

After renaming the file, change the date of the new output file to today’s

date.

3-43

RUN

3.25

05/ 78 Commands

RUN COMMAND

‘

The RUN command loads the specified memory image (.SV) file in memory, transfers its Core Control Block onto
the system device, and then initiates execution at the program’s starting address.

Format:
.RUN indev: file.ex

Example:

LRUN

RXAJ. iF'ROG .

If a file name is Specified without an extension, the extension “.SV” is assumed.
NOTE
The RUN command is equivalent to a GET command followed by a START command.
See the R command for further information.

NOTE
‘

The colon used with the permanent device name can be

replaced with a space.
The Monitor processes the RUN command.

344

OS/ 78 Commands
3.26

SAVE

SAVE COMMAND

The SAVE command saves the portion of memory being used. The memory used is indicated by the contents of the

Core Control Block saved on SYS in a Monitor area or as explicitly specified by you. This new memory image file is

given a default .SV extension if an extension is not specified.
Format:
.SAVE input devzfilenameex fnnnn-fmmmm,fpppp,.

.;fssss=cccc

where

fnnnn-fmmmm

fnnnn is a S-digit octal number representing the beginning address of a contiguous portion
of memory to be saved. The beginning address consists of a field number (f) followed by
the memory location within the field,

fmmmm is a 5-digit octal number representing the ending address of the portion in
memory. The ending address consists of the same field number as the beginning address

(0 followed by the memory location within the field;
NOTE

The beginning and ending address must have the same

field number.

fpppp

Crossing field boundaries is not permitted.

fpppp is a 5-digit octal number representing the address of one location in memory. This
represents saving the entire page in which the given location occurs;

;fssss

;fssss is a 5-digit octal number representing the starting address of the program to be
saved. The starting address consists of a field number (f) followed by the memory location within the field; and

=CCCC

=cccc is a

4-digit octal number representing the contents of the Job Status Word.

The Job Status Word indicates what parts of the file use memory and how as shown in Table 3-20.

Table 3-20

Job Status Word

Bit Condition

Meaning

Bit 0:1

File does not load into locations 0-1777 in field 0, (0000-1777).

Bit l=l

File does not load into locations 0-1777 in field I, (10000-11777).

Bit 2=l

Program must be reloaded before it can be restarted because it modifies itself
during execution.

Bit 3=l

Program being run will not destroy the BATCH monitor.

Bit 4=1

A memory image file that was generated through the LINKER contains overlays.

Bit 5

Reserved for OS/ 78 system programs

Bits 6-9

Unused, and reserved for future expansion.

Bit 10:1

Locations 0-1777 in field 0 need not be saved when calling the Command De-

coder overlays.
Bit ] l=l

Locations 0-1777 in field 1 need not be saved when calling the USR.

3-45

OS/ 78 Commands
NOTE
If these arguments are not specified in the command line,

the arguments are automatically taken from the current
Core Control Block, and stored by the last GET, RUN,
LOAD or EXECUTE command.
Restrictions on Arguments of the SAVE Command

3.26.1

There are several restrictions on the arguments used with the SAVE command. They are as follows:
1.

The output device must be explicitly specified in the command line. It does not default to DSK.

LSESAUE
2.

1353K : FI LE. 1 0 PA

The beginning and ending addresses of an area in memory (fnnnn-fmmmm) must both be located in the
same

field.

Crossing field boundaries is not permitted.

Example:
.SAUE
3.

RXAIHTLZ

10077m-10122

RXAltFLQ

10].'56-~101'77

If a field number is not specified in the starting and ending addresses of an area in memory, field zero is

Otherwise, the field number must be specified.

LEMUE

20055-~20643

Therefore, if another area on the same page is
specified, an error message is sent to the terminal informing you that page is already saved.

used.

EXAMlz'lm.BN

Once an area on a page is specified, that entire page is saved.

LSEiAUlI-E
LSSAUE
4.

8Y8

SYSS

F SELF. 1

’

0M1 77 , 201 “377

Saving the area between locations 7600-7777 in any field is permitted. However, locations 7600-7777 in
fields 0 and 1 should not be saved because the system resident Monitor code resides there. Therefore, if
the area between 7600-7777 is going to be saved, limit it to fields two and three and even then only if
necessary (due to BATCH).
If you specify an address on a page having an odd page number, that page cannot be saved without the

previous page also being saved. This is automatically done by the system.

Example:

LEAVE

IIISK

F'ROGZL

1-2434

In this example, the location is on page 12 which is even.

Thus, the contents of page 12 are saved.

Example:

LSAUE

[ISK

F'RO(32

2634

in this example, the address 2634 is on page 13 which is odd.

Therefore, in order to save page 13, page 12

is also saved. This means that the effective area being saved is 2400-2777.

3—46

OS/ 78 Commands
NOTE
The colon used with the permanent device name can be

replaced with a space.
The Monitor processes the SAVE command.

3-47

05/ 78 Commands

SET

SET COMMAND

3.27

The SET command modifies the operating characteristics of 08/78, according to options that are specified.
Format:

.SET device [NO] attribute [argument]
where

indicates the handler of 08/78 that is to be modified, such as:

device

SYS, TTY, or LPT;
indicates that the feature specified is to be turned off; NO cannot

[NO]

be used with every attribute;

attribute

is the feature to be modified (See the SET command attributes); and

[argument]

is a variable (numeric or alphabetic) that is supplied by you.

Table 3-21

SET Command Attributes

TTY

SYS

LPT

READONLY

INIT xxxxx

WIDTH

Any Device
=

READONLY

LC
PAGE

ECHO
SCOPE
HEIGHT m

PAUSE n
COL n
ARROW
ESC
3.27.1

ECHO

When used with the NO modifier, this attribute removes TTY character echoing from the TTY handler.
This does not affect monitor I/O.

‘

Format and Example:

LSIET

TTY

NC]

ECHO

To restore character echoing, use the ECHO attribute in the command line.

Example:

LSET
3.27.2

TTY

ECHO

ARROW

When used with the NO modifier, this attribute forces each control character to be printed on the terminal
in the form:

3—48

OS/ 78 Commands
where :

represents the CTRL key

represents the control letter

Format and Example:

:SSSEIT

TTY

(Al-TRON

For a CTRL C character, a T C is printed on the terminal.
To remove this attribute, use the ARROW attribute in the command line.

Example:

:SESEIT

TTY

ARROW

SCOPE

3.27.3

When an input line of characters is typed on the terminal and echoed on the screen, this attribute erases
the last character echoed on the screen each time the DELETE key is pressed.

Format and Example:

LSET

TTY

SCOPE

To remove this attribute, use the NO modifier in the command line.

Example:

LSET
3.27.4

TTY

SCUF‘E

WIDTH = n

This command changes the width of the terminal to the argument you specify as n.
Format:

.WIDTH TTY WIDTH

where

is a decimal number and a multiple of 8 which is in the

range of 001-255

NOTE
The 11 must not be 128.

Also, the NO modifier is not

permitted to be used with this attribute.

Example:

L8 [1' T

TTY

N 1' [I T H ==== 7 13

This command can also be used with the LPT handler, in which case, LPT replaces TTY in the command string.

3-49

05/ 78 Commands
3.27.5

HEIGHT m

This attribute changes the number of lines that are printed on the terminal between pauses. Twenty four
lines is the default value of m.
Format and Example:

LSES‘ET
3.27.6

TTY

HEIGHT

PAGE

When used with the NO modifier, this attribute removes the CT RL/S and CTRL/Q features (See Section 2.1.3).
Format and Example:

SET

T TY

NC}

F‘AG E

To restore the CTRL/ S and CTRL/ Q features, use the PAGE attribute in the command line.

lSET
3.27.7

TTY

F'AGE

ESC

This attribute restrains the printing of the dollar sign ($) on the terminal when the ESCape key is pressed.

However, ESC does allow the terminal to receive its octal value (0338) which puts the system into a
different mode. Once it is in this mode, moving the cursor and erasing the terminal screen are some of
the options that can be performed.
Format and Example:

LSE‘T'

TT Y

ESC

To remove this attribute, use the NO modifier with the ESC attribute.

LSEEET
3.27.8

TTY

ESC

PAUSE

This attribute changes the pause time between terminal output frames from 3 seconds (its default) to the
decimal number you specify as n.

Format and Example:

_._$3l3§T

TTY

PAUSE

5
NOTE

If you specify zero as n, or use the NO modifier,
no pause

3.27.9

takes place.

READ ONLY

This attribute causes the device specified to become a read-only device.

Therefore, any output sent to this device will cause an error message informing you that the output
device is a read-only device.

3-50

0S/ 78 Commands
Format and Example:
9C1]

REQIIIO

TTY

To remove the READONLY attribute, use the NO modifier in the command line.

1539-15 T

T TY

N (I)

RlEIriilllO
NOTE
The READONLY attribute remains in effect only
until the next time you push the START button
at which time its original status is restored.

3.27.10

COL n

This attribute changes the number of columns used to print the directory from two (its default) to the
decimal number you specify as n.
Format and Example:

Lear

TTY

COL

To remove the value you specified as n and return to its default value of 2, you must specify 2 as the

argument n.

LSET
3.27.11

TTY

COL

This attribute causes the TTY or LPT handler to accept lower case characters on input.
Format and Example:

LSET

TTY

To remove this attribute, use the NO modifier in the command line.

LSET
3.27.12

TTY

NC)

INIT xxxxx

This attribute causes the system device to execute the command you specify as xxxxx when the
START button is pushed. This command name can contain a maximum of five characters excluding
a

carriage return.

Format and Example:

LSET

8Y5

INIT

HELF‘

To execute the command contained in the file INIT.CM, type:

LSET

SYS

INT. T

If you want the system to print the monitor dot immediately after the START button is pushed, type
the following command:

LSET

8Y8

INIT

3-5 1

OS/ 78 Commands
Table 3-22

SET Error Messages

? SYNTAX ERROR

Incorrect format used in SET command or NO specified when not allowed.
? UNKNOWN ATTRIBUTE FOR DEVICE dev
An illegal attribute was specified for the given device.
? CAN’T

—-

DEVICE IS RESIDENT

No modifications are allowed to the system handler.
? CAN’T

—

OBSOLETE HANDLER

The handler has an old version number.
? CAN’T

—

UNKNOWN VERSION OF THIS HANDLER

The version of the handler is not one recognized, possibly because it is a newer version.
? ILLEGAL WIDTH
A width of 0 or a width too large was specified; or for the TTY, a width of 128 or one not a

multiple of 8.was specified.
? NUMBER TOO BIG

The number specified was out of range.
? CAN’T

—

DEVICE DOESN’T EXIST

A nonexistent device was referenced.
? I/O ERROR ON SYS:
An [/0 error occurred while trying to read or rewrite the handler.

3-52

0S/ 78 Commands

START

START COMMAND

3.28

The START command initiates execution of the program that is currently in memory either at the specified

starting address in the command line or the starting address specified in the current Core Control Block.
Format:

START fnnnn

where
fnnnn

is a 5-digit octal number representing the starting address of the program. The starting

address consists of a field number (f) followed by the memory location (nnnn) which
is within the field. If the starting address is not specified in the command line, the

program is started at the address specified in the Core Control Block.

Example:

LSi‘TAFx‘T

105555

In this example, this command line will execute a program starting in field 1 at location 555.

The Monitor processes the START command.

3-5 3

OS/ 78 Commands

SOUI SH

3.29

SQUISH COMMAND

The SQUISH command eliminates any embedded empty files on the input device.
Format:

.SQUISH indev:
Whenever the SQUISH command is used, a message followed by a question mark that asks if this is the correct
device for the SQUISH operation is displayed.

Example:

LEN}

RXA II. 2

ARIES

YOU

SLJRE’PY

If yes, type a Y and the files are compressed.

L8 [J

SSYS :

FIRES

Y 01.1

81.1le '?N

If no, type any other character and the operation is ignored.

If the device is a system diskette, the system is preserved during the transfer. In order to eliminate the system,
use

the ZERO command (See ZERO command). If no output device is specified, then the specified device in

the command line is itself squished.

Example:

LSQ

8Y8:

ARE

YOU

SURETY

In this example, all embedded empty files on the system device are eliminated.

NOTE
A diskette error during a SQUISH operation will

leave the entire contents of the diskette corrupted
in a non-obvious way.

Therefore, this command

should not be used unless a backup copy of both
the system and other files is available.

This command causes the execution of both the CCL.SV and PIP.SV programs.

3-54

OS/ 78 Commands
3.30

SUBMIT

SUBMIT COMMAND

The SUBMIT command provides batch processing with the option of spoofing to output files.

Format:

.SUBMIT spool devz<DSKtbatch file.ex/0ptions

Example:

LSU

RXAO 2 BTCH I N

If no input file extension is specified,.BI is assumed.

Along with the necessary arguments for the command line are SUBMIT options that can affect the result
of the SUBMIT operation.
The SUBMIT command is commonly used to run multiple programs and sequences of system commands
that require little or no interaction with the user or operator.
3.30.1

Processing and Terminating A Batch Input File

The SUBMIT command starts processing your batch input file by printing a job header. The job header is

printed on the LA78 line printer if it is available. Note that the LQP78 line printer is not supported in batch
processing. If an LA78 line printer is not available, the terminal is used as the output device to print a log
of batch operations. Processing continues with the 08/78 commands and BATCH monitor commands
following the $108 batch monitor command and terminates with the $END batch monitor command.
Termination can also Occur from an error in the 08/78 commands. However, batch processing is allowed
to continue in spite of errors if the /E option is used in the command line. By specifying the /E option,
all errors are treated as non-fatal errors so that batch processing can continue uninterrupted.
If the ESCape key is used to terminate an OS/78 command in the batch file, it will cause your batch run
to terminate unless the /E option is specified.

To terminate all the commands in a batch file, use the

RETURN key. If CTRL/C is typed during batch processing, the batch run is terminated and control
returns to the Monitor.

If you want just the $JOB, and $MSG batch monitor commands, and all batch results sent to the output
device, disregarding 08/78 commands and comment lines, use the /Q option in the command line. If you
have a LA78 and want all batch output sent to the terminal instead, use the / T option in the command line.
By specifying the /T option, batch output is sent to the terminal only.

If you want your batch input file to be processed without interruption, use the /U option in the command
line. By specifying the / U option, any errors are nonfatal and your batch input file is processed without

interruption. Anything you type on the terminal except CTRL/C during batch processing is ignored.
If you want your batch input file to be processed without echoing on the terminal and without sending'the $JOB
and $END batch monitor commands to the terminal and BATCH log, use the /H option in the command
line.

By specifying the /H option, echoing, $J OB, and $END batch monitor commands are suppressed.

Statements that are not suppressed are as follows:
1. Any 05/78 command messages and user program messages sent to BATCH
2. User program messages sent to the terminal

3. Batch run-time error messages
4. $MSG statements

3-55

OS/ 78 Commands
NOTE
Use either the /Q

[H option but not both in a

command line.
3.30.2

Spooling

The SUBMIT command also provides optional spooling of output files.
to slow-speed devices (such as a line

Spooling is a process whereby output

printer) is placed into queues on fast-speed devices (such as a diskette)

until the slow speed devices are available for processing. One advantage of using spooling is a more efficient
use

of slow-speed devices, memory, and the central processor unit.

To use the spooling option during batch processing, specify an output device, the system disk as the input
device and the batch file in the command line. The output device specified as the spool device must be a

file-structured device (that is, diskette). Therefore, TTY and LPT should not be used. If no output device
is specified, spoofing does not take place.

Example:

_._EiU

RXAIL :~=IIF.'XQO 3 BATCHX
NOTE
Your batch file must be on your system device (SYS).

When spooling is specified, all output to a non-file structured device is intercepted and stored in temporary
device. Every file that is spooled to your spool device is given a name. The first non-file
files on the

spool

structured output file is called BTCHAI. The second is called BTCHA2 and so on up to and including
BTCHA9. If you have more than nine input files the names continue with BTCHBO up to and including
BTCHZ9.
See Chapter 8 for a detailed explanation of OS/ 78 Batch.
This command causes the execution of both the CCL.SV and BATCH.SV programs.

Table 3-23

SUBMIT Options

Meaning

Option

IE
/Q

Treat all errors as non-fatal errors so batch processing continues uninterrupted.
Send only the $JOB and $MSG batch monitor commands, and all results from

batch processing to the output device.

/T
[U

Send all output from batch processing to the terminal only.
Do not interrupt batch processing and ignore anything typed at the keyboard

with the exceptions of CTRL/C.

Process the batch input file without echoing and without sending the $103
and $END batch monitor commands to both the terminal and batch log.

3-56

OS/ 78 Commands
3.31

TERMINATE

TERMINATE COMMAND

The TERMINATE command causes the system to emulate a terminal without any knowledge of disk drives,
processor or memory.

Example:

LTER
This command causes the execution of the CCL.SV program.

3-57

05/ 78 Commands

TYPE

TYPE COMMAND

3.32

The TYPE command displays the contents of the specified input files on the terminal screen.
Format:

.TYPE indevzfilel .ex...fi1e5 .ex/options

Along with the necessary arguments for the command line are TYPE options that affect the result of the
TYPE operation.
'

The TYPE command displays the contents of each input file on the terminal in the same order the file
names are

specified in the command line.

Example:

_._TY

SYS 3 MASFL

Although the format specifies that only five file names can be used in the command line, more than five files
can be typed on the terminal by using wildcards.
3.32.1

Using TYPE Options

Use the /C or /0 options if the date is to determine what files are to be displayed.

By specifying the /C

If the /0 option is used, all specified

option, all the specified input files with the current date are typed.
input files with dates other than the current date are typed. The /V option displays all files other than the
ones that were specified. The /Q option allows each file name to be displayed followed by a question mark.
If that file is the correct file for the TYPE operation, type a Y and that file is displayed on the terminal.
If not, type any other character and that file is ignored. After the file is either typed or ignored, the next
file name followed by a question mark is displayed on the terminal. This questioning continues until all
specified input files have been typed.
NOTE
Use the LIST command to print files on the LA78

line printer (LPT).

This command causes the execution of both the CCL.SV and FOTP.SV programs.
Table 324

TYPE Options

Type only those files with the current date.

Type only those files with a date other than the current date.

Type all files that were not specified in the command line.

Each time a file name followed by a question mark is printed on the

terminal, type a Y (yes) to type the file or any other character (no)
if the file is not to be typed.

3-58

05/ 78 Commands

Table 3-25
BAD INPUT DIRECTORY

TYPE Error Messages

The directory on the specified input device is not valid OS/78
device directory.

BAD OUTPUT DEVICE

Self-explanatory. This message usually appears when a non-file
structured device is specified as the output device.

ERROR ON INPUT DEVICE,

The file specified is not transferred, but any previous or

SKIPPING (file name)

subsequent files are transferred and indicated in the new
directory.

ERROR ON OUTPUT DEVICE,
SKIPPING (file name)

ERROR READING INPUT

The file specified is not transferred, but any previous or

subsequent files are transferred and indicated in the new
directory.

Self-explanatory.

DIRECTORY
ILLEGAL *

An * was entered as an embedded character in a file name,

e.g., TMP*.BN.
ILLEGAL ’.’

A ? was entered in an output specification.

NO FILES OF THE

No files of the form (xxxx) specified were found on the

FORM xxxx

current input device group.

3-59

05/78 Commands

UA, UB, UC
3.33

UA, UB, UC COMMANDS

The UA, UB, and UC commands are three separate commands that save their specified arguments in a temporary

Only one argument is permitted with each command and this must be a single OS/78 command. These
three commands have the capability to recall their arguments when the command that saved the OS/ 78 command
is typed without any arguments. Once recalled, the 08/78 command is executed.
area.

Format:

.UA
.UB

OS/78 command output

.UC
In response to the dot produced by the Monitor, type one of the commands along with the 08/78 command
’

you want saved.

Pressing the RETURN key saves the specified OS/78 command in a temporary area. There are 3 areas, one for
each command. To recall and execute the stored OS/78 command, type the command that saved it, but do
not specify any arguments.
Example:

LUA

COF‘Y

RXAI:£DSK:RECALL

In this example, the 08/78 command, COPY and its arguments, is stored in a temporary file through the use of

the UA command.

Example :
_0 U A

In this example the 05/78 command previously specified with the UA command and stored in a temporary

file is recalled and executed.
These commands are most commonly used with 08/78 commands that are repeated throughout a batch file.

The use of these instructions decreases the amount of input
can be saved at one

time.

The Monitor processes the UA, UB, and UC commands.

3-60

required. Note that only three OS/78 commands

OS/ 78 Commands
3.34

ZERO

ZERO COMMANDS

The ZERO command clears the directory on a device, deleting any and all files stored on that device.

Use this command with extreme caution, since it could destroy the contents of a system diskette.
Use it only for diskettes that are used to store programs and data files. If used on a system diskette, it

will destroy the system and convert it to a data diskette.
Format:

.ZERO dev:
This command causes the execution of both the CCL.SV and PIP.SV programs.

Example :

.LZERD

RXAl:

By using the DIRECT command on the device just zeroed, a blank directory will appear showing the

availability of 487 free blocks.
If the specified device is device SYS, a message followed by a question mark is printed on the terminal asking
if this is the correct device for the ZERO operation. If yes, type a Y and the directory is zeroed. If no, type
any other character and the ZERO operation is ignored. Remember that if a Y is typed, the system on the

specified device will be destroyed.
NOTE
Never issue a ZERO SYS:
erase

the system diskette.

command, as this will

CHAPTER 4

CREATING AND EDITING YOUR PROGRAM

4.1

INTRODUCTION

The Editor is a program that allows you to create and modify ASCII source files. These files may contain assembly

language programs, FORTRAN and BASIC programs, or any other information that has the format of character
strings.
The Editor is a very helpful tool; however, it must still be told precisely what to do. Its operation is directed by

typing commands in the form of a single letter or a letter with arguments. Most commands are executed by pressing
the RETURN key directly after the command line.
PAGE MAKEUP
The page of text is the fundamental unit into which the Editor considers a file to be divided. A page of text consists of all the ASCII characters between two form feed characters, not including the second form feed. In other

4.2

words, a page begins after the form feed of the previous page, and ends with a form feed. A page is usually a physical page of a program listing.
A line of text consists of all the characters between two carriage retum/line feeds, not including the second carriage

return/line feed. All lines are implicitly numbered in decimal, starting with 1. This implicit enumeration is continually updated by the Editor to account for line insertions, moves, and deletions. For editing and listing purposes,
each line is referred to by its current implicit decimal line number.
4.3

TEXT BUFFER

The text buffer is a storage area in memory that holds a text file one page at a time to allow the operations to be

performed to create or modify the text. The editing commands will affect only the information that is placed into
the text buffer.

The text buffer has room for approximately 5600 (decimal) characters. When text has been input so that there is
room

ing.

for only 256 (decimal) more characters in the buffer, the Editor causes the terminal buzzer to ring as a warn-

From this point on, whenever a carriage return is detected during text input, the buzzer again sounds and the

Editor prompts with the number sign (#), signifying it needs an editing command.

Inputting one line at a time

may continue until the absolute end of the buffer is encountered. At this point, no more text can be placed in the

buffer, a question mark “‘2” is printed, and the Editor prompts with the number sign (#), indicating it has returned
to the command mode. If text is segmented into pages as explained later in this chapter, there should always be
sufficient room in the text buffer.
The Editor reads one page of text at a time from an input file into the text buffer where the page becomes available
for editing. Then the Editor writes the text buffer onto the output file.

lowing functions:

fiePJNr‘

Locate text to be changed (string and character searches).
Execute and verify changes.
Output a text page to the output file.
List an edited page on the console terminal or line printer.

4-1

Editing commands then perform the fol-

Creating and Editing a Program
4.4 CALLING THE EDITOR
The commands CREATE and EDIT call and run the Editor.

Type

.CREATE filenameex
or

.EDIT filename.ex
CREATE is used to open a new file for the first time, that is, create a completely new file. EDIT is used to open

already existing file that is to be modified or corrected. Thus, typing either command in response to the monitor
dot (.) followed by a carriage return will call the Editor for your use.
an

The two valid I/O specification options for the Editor are given in Table 4-1.
Table 4-1

Editor Options

Meaning

Option
/B

Convert two or more spaces to a TAB when reading from input device.

Delete the old copy of the output file (if one exists) before opening the new

output file on the device. If /D is not used, the old copy of the output file
is not deleted until all data has been transferred to the new file by an E or
Q command.
4.5

MODES OF OPERATION

It is important to know the difference between the Editor command mode and the text mode. In the command

mode, the Editor interprets all input typed on the keyboard as commands to perform some operation or allow

operation to be performed on the text stored in the buffer. In text mode, the Editor interprets all typed
input as text to replace, to be inserted into, or to be appended to the contents of the text buffer.
some

Immediately after being loaded into memory and started, the Editor is in the command mode, and the text buffer
is empty.

When the Editor enters the text mode (through the use of the Insert, Change or Append commands), corrections,
additions or insertions may be made to the text.

Typing a form feed (CTRL/ L combination) will terminate the

text mode by instructing the Editor to return to command mode. The Editor answers by prompting with a num-

ber sign (#) indicating that it has returned to the command mode.
The RETURN key is important in both the command and text modes. In the command mode, pressing the RE-

TURN key allows the Editor to execute the command just typed. In the text mode, pressing the RETURN key
after typing a line of text causes the line of text to be entered into the text buffer. A typed line is not actually

part of the buffer until it is terminated by the RETURN key.
Editor key commands and special characters and their effect in the two Editor modes are described in Sections 4.6
and 4.7.
4.6

EDITOR KEY COMMANDS

The key commands in Table 4-2 allow you to transfer between modes. The commands are
produced by pressing
the CTRL key and the appropriate character key at the same time.

4.2

Creating and Editing a Program
Table 4-2

Editor Key Commands

Mode in Which

Command

Used

Meaning

CTRL/C

Text and Command

Returns control to the Keyboard Monitor. All text
that has been edited is lost.

Modes

CTRL/C should be

used with utmost caution, since no output file will

be stored.

Stops the listing of text.

Text Mode

CTRL/D

Returns control to

Command Mode.

CTRL/ L

Text Mode

CTRL/U

Text Mode

Returns the Editor to Command Mode.

Typing CTRL/U while entering text from the keyboard causes all text typed so far in the current
line to be deleted. A carriage return/line feed is

generated and the line may be retyped. (The command is equivalent to typing the DELETE key back
to the beginning of the line.)

When used in command, mode, CTRL/U cancels the

Command Mode

entire command. The Editor performs a CR/LF

and remains in the command mode (equivalent to

DELETE key).
4.7

4.7.1

SPECIAL CHARACTERS
DELETE Key

The DELETE key on the terminal keyboard is used in error recovery in both the command and text modes. In
the text mode, typing the DELETE key erases the last typed character.

Repeated deletions erase the characters

from right to left up to, but not including, the CR/LF, which separates the current line from the previous one. For

example,

uRITE<6.25)-1a)
will be entered in the buffer as

“RITE ( 4 r 1 5)
When used in command mode, DELETE is equivalent to a CTRL/U combination and cancels the entire command.
The Editor then prints a question mark (7), performs a CR/LF, and waits for another command to be typed.
4.7.2

Period (.)

The Editor assigns an implicit decimal number to the line on which it is currently operating. At any given time the
period, which represents this decimal number, may be used as an argument to a command. In the following exam-

ple, the L command is used since it allows text to be listed. Typing
#.L

4-3

Creating and Editing a Program
means list

it .

the current line.

means list

Typing

1 y o +1L

the line preceding the current line, the current line, and the line following it, and then update the current

line counter to the decimal number of the last line printed. The current line counter, represented by the period, is

generally updated as follows:
1.

After an R (Read page) or A (Append) command, the period is equal to the number of the last line in the
buffer.

After an I (Insert) or C (Change) command, the period is equal to the number of the last line entered.

After an L (List) or S (Search) command, the period is equal to the number of the last line listed.

After a D (Delete) command, the period is equal to the number of the line immediately after the

deletion.

4.7.3

After a K (Kill) command, the period is equal to 0.

After a G (Get and list) command, the period is equal to the number of the linedisplayed by the G.

After an M (Move) command, the period is not updated and remains whatever it was before the command.

Slash U)

The symbol slash (/) has a value equal to the decimal number of the last line in the buffer. It may also be used as an
argument to a command. For example,

#10 y /L
means

list from line 10 to the end of the buffer.

4.7.4

LINE FEED Key

Typing the LINE FEED while the Editor is in the command mode is equivalent to typing

and will cause the Editor to display the line following the current one and to increment the value of the current line
counter (dot) by one.

LINE FEED does not perform this function while in the text mode.

4.7.5

Right-Angle Bracket (>)
Typing the right-angle bracket (>) while in command mode is equivalent to typing

and will cause Editor to echo > and then display the line following the current line. The value of the current line
counter is increased by one so that it refers to the last line

displayed.

4.7.6

Left-Angle Bracket (<)
Typing the left-angle bracket (<) while in command mode is equivalent to typing
OM11and will cause Editor to echo < and then print the line preceding the current line. The value of the current line
counter is decreased by one so that it refers to the last line

printed.

4.7.7\ Equal Sign (=)
Use the equal sign in conjunction with either the line indicator period (.) or slash

(/). When typed in the command
preceding it. In this way the number of the
current line may be found (.=nnnn), or the total number of lines in the buffer
(/=nnnn).
mode it causes the Editor to display the decimal value of the argument

4.4

Creating and Editing a Program
4.7.8

Colon (2)

The colon performs exactly the same function as the equal Sign (=).
Tabulation (TAB)

4.7.9

The Editor simulates tab stops at eight space intervals across the terminal screen. When the TAB key is typed, the

Editor produces a tabulation. A tabulation consists of from one to seven spaces, depending on the number needed to
bring the carriage to the next tab stOp. Thus, the TAB key produces neat columns and increases the legibility of
programs and other text.

4.7.10

ESCape Key

The ESCape key is used to signal an intrabuffer character search when in the Editor command mode. Where typed,
it echoes as a dollar sign ($) on the terminal screen.

In the Editor text mode, typing the ESCape key echoes as a

dollar Sign, but it is stored in the file as an ESCape character (033).

4.7.11

Lower Case Characters

Releasing the keyboard CAPS LOCK key
Upper case character can then by typed by pressing either SHIFT

Lower case characters can be entered into a file in the Editor text mode.

will enter all alphabetic characters in lower case.

key. Locking the CAPS LOCK key will again generate all upper case characters.
4.8

EDITOR COMMANDS

Commands to the Editor are grouped under five general headings:

Input, List, Output, Editing and Search. Explana-

tion of the five kinds of commands is given in Sections 4.8.1 through 4.8.5. Error messages generated while using the

Editor are discussed in Section 4.13.
The Editor contains an automatic text collector that reclaims buffer space following the use of a D(delete),

S(search), or C(change) command. The text collector removes the deleted text and updates the necessary pointers.
If a full buffer condition is reached, lines of text may be output using the P command, for example, and then these
lines can be deleted from the buffer.
A command directs the Editor to perform a desired operation. Each command consists of a single letter, preceded

by up to two arguments. The letter indicates the command function; the arguments usually specify which numbered
line or lines of text are affected. Commands to the Editor must take one of the following forms, where X repre—
sents any command letter.

The commands discussed in the following pages can each be given whenever the Editor prints a number sign (#) at
the left margin, Editor commands are given in upper case characters only. These commands are of the general form

#nX
or

#m,nX
where m and n represent the line number designation (m must always be less than n), and X represents the command
letter. Editor commands use decimal numbers.

NOTE
4

Except where specified, each line must be terminated by
pressing the RETURN key.

4.8.1

Input Commands
Input commands allow text to be entered into the text buffer from the terminal.

4-5

Creating and Editing a Program
Command
A

Meaning

Format

Append the following text being typed at the keyboard until a form feed (ASCII 214 or CTRL/L) is
found. The form feed returns control to command

mode. Text input following the A command is appended to whatever is presently in the text buffer.
#1

Insert whatever text is typed before line I of the text

buffer. The form feed (CTRL/L) terminates the
insertion process and returns control to the command

mode.
NOTE

Special characters, including lower case letters may be
input to the file. The ESCape character is echoed as a
dollar sign ($) for readability.
#nl

Insert whatever text is typed (until a form feed is

typed) before line 11 of the text buffer.
#R

Read one page from the input device specified to the
EDIT or CREATE commands and append the new
text to the current contents of the buffer.

If no input

file was indicated or if no input remains, a question

mark (?) is printed and the Editor returns to the

command mode.
NOTE
In these commands, the Editor ignores ASCII codes 340

through 376. These codes include the codes for the
lower case alphabet (ASCII 341-372). The Editor returns
to the command mode only after the detection of a form

feed or when the text buffer becomes full.
4.8.2

List Commands

List commands display all or any part of the contents of the text buffer on the terminal to permit examination of the
text.

Command
L

Format

Meaning

List entire contents of the text buffer on the terminal.

#nL

List line n of the text buffer on the terminal.

#rn,nL

List lines m througn n of the text buffer on the

terminal.
The Editor remains in command mode after a list command and the value of the current line counter is updated to be

equal to the number of the last line printed.

4-6

Creating and Editing a Program
Comm and

Meaning

Format

Get and list the next line that has a label associated
with it. A label in this context is any line of text
that does not begin with one of the following:

space

/
TAB
RETURN

(ASCII 240)
(ASCII 257)

(ASCII 211)
(ASCII 215)

At the termination of a G command, control returns
to the command mode with the current line counter

equal to the line just listed.
#nG

Get and list the first line that begins with a label,

starting the search at line n.
#B

Print the number of available memory locations in
the text buffer. The Editor returns the number of

locations on the next line. To estimate the number of
characters that can be accommodated in this area,

multiply the number of free locations by 1.7.
4.8.3

Output Commands

The format and meaning of the output commands are as follows:

Command
E

Format

Meaning

Output the current buffer and transfer all remaining
pages of input to the output file, close the output file
and enter it in the directory. When this buffer is full,

the text is output to the indicated output file. The P

command automatically outputs a form feed after
the last line of output, and returns control to the

Monitor.
NOTE
If the E command is not used to close a file after editing,
any changes, additions or corrections will not appear in

the output file.

Thus, the E command should usually
always be the last command used in an editing session

(also see Q command).
P

Write the entire text buffer to the output file.

#nP

Write line n of the text buffer to the output file.

#m,nP

Writes lines m through n, inclusive, to the output file.

4-7

Creating and Editing a Program
NOTE
The P command automatically appends a form feed to its

output, thus producing a page of text. This command
allows you to paginate your listing. However, if the K
command is not used after a P command, the text remains
in the buffer and is again output with the new text read
in before the next P command, thus resulting in an

additive effect.
Command

Format

Meaning
Kill the buffer. All text is deleted from the text
buffer.
NOTE

The Editor ignores the commands nK or m,nK to prevent
the buffer from accidentally being destroyed if the user
means to

type a List command (m,nL).
Immediate end-of-file. The Q command causes the
text buffer to be output.

The file is then closed

(entered into the directory with the current date as
its creation date); returns to the Monitor.
#N

Write the current buffer to the indicated output file

and read the next logical page. The N command is
equivalent to a P, K, R command sequence.

#nN

Write the current buffer to the output file, kill the

buffer, and read the next logical page. This is done n
times until the nth logical page is in the text buffer.
Control then returns to command mode.

The N command cannot be used with an empty text

buffer since there is no text to be written. If the
buffer is empty when the N command is attempted,
a

question mark (?) is printed. For example, to read

in the fourth page of a file, give the commands

#R (to read the first page)
and

#3N (to read three more pages)
#V

The V command causes the entire text buffer to be

listed on the line printer. The V command only works
with the LA78 line printer. It does not work with the

LQP78 line printer.
#nV

List line n of buffer on the line printer.

#m,nV

List lines m through 11, inclusive, on the line printer.

4-8

Creating and Editing a Program
4.8.4

Editing Commands

The following commands permit deletion or alteration of text in the buffer.

Command

Format

Meaning
Change the text of line n to the line(s) typed after the

#nC

command is entered (typing a form feed terminates
the text input). The C command is equivalent to a D

command followed by an I command.
Delete lines m through n and replace with the text

#m,nC

line(s) typed after the command is entered. (Typing
CTRL/L indicates the end of the changed lines.)
The C command utilizes the text collector in altering
text.

#9an

Delete line 11 from the buffer.

#m,nD

Delete lines m through n from the buffer.

#nY

Read (Yank) in n pages from the input file into the
text buffer, without writing any output.

For example,

#SY
reads through four logical pages of input, deleting them
without producing output. The fifth page is read into

the text buffer and control automatically returns to
command mode.

NOTE
This command should be used with caution, since it

irrevocably deletes the contents of the text buffer.
M

Move lines m through n directly before line p in the

#m,n$pM

text buffer.

The $ character represents the dollar sign

and is typed using the dollar sign key (SHIFT/4).
The old occurrence of the moved text is removed.

This command can move one line but it needs three

arguments. This is done by specifying the same line
number twice. For example,

#6,6$2IM
moves line

4.8.5

6 in front of line 21.

Search Commands

The search commands provide the ability to set the current line indicator designated by a period (.) in order to

make changes at specific lines.

Creating and Editing a Program

Meaning

Format

Command
S

Character search command (Section 4.11.1).

Interbuffer search command for character strings

(Section 4.11.2).
Follows a string search. Look for next occurrence of

the string currently being sought.

ESC($)

4.9

(Section 4.1 1.2).

#$TEXT’

Perform a intrabuffer character string search for the

#”

string TEXT. Following a string search, #” causes a
search for the next occurrence of the string.

CREATING A NEW FILE

Suppose you want to create a simple FORTRAN program that will compute and print the powers of two. The
following program will be used as an example.
C

FORTRAN

DEMONSTRATION

AND PRINT
DIMENSION A(lé)
COMFUTE

PONERS

THO

URITE(4915)
15

FORMAT<1H v’PONER
H0
20 leylé

THO’)

A(N)m2.#*N
20

CONTINUE
WRITE (4725)

FDRMAT<1H

(NvA(N)yNw1916)

y’2**

’yIRy’m’vF10.1)

STOP
END
Use CREATE to open a new file called POWERFT.

LCREQTIE

Type

F'ULJERJ-TT

The Editor will then indicate that it is ready to accept an editing command by printing a number sign (#).
Now type the A command (an I command could also be used) to append text onto the empty text buffer:

it)
Start typing the program, using the TAB key to make your program easy to read. At the end of every line, press the
RETURN key. After typing the program, return to the Editor command mode by typing a CTRL/L. Most editor

commands automatically go back to the command mode to allow the next editor command to be typed in. However, the I(Insert), A(Append) and C(Change) commands (classified as text mode commands) require that a CTRL/L

be typed to return to the command mode. A final command required to close the file and return to the Monitor is
the E(Exit) command. The execution of the E(Exit) command closes out the file and places it on the device (DSK:).
Verify this by typing the 08/78 command DIRECT, which displays the program just created in the directory
listing. Note that the new file is given a creation date only when the date is entered using the 08/78 DATE command. The date must be reentered every time the system is booted up.
4.10 PAGING FILES
To paginate a file, either at a convenient point for dividing a program, or when the capacity of the text buffer is

approached, transfer that part of the program already typed and clear the text buffer for the next portion of your
program. Do this by using the P(Paging) command. Typing #P writes the entire text buffer to the output file and

4-10

Creating and Editing a Program

places a FORM FEED character after the last line of text in the output buffer. The FORM FEED character allows
the Editor to distinguish between pages of text. Before typing in the next page, be sure that there is no text in the
buffer. This is done by using the K(Ki11) command, which deletes any text in the buffer. Then type the A(Append)
command, and continue typing your program. Upon completion, type the E(Exit) command to output the current
buffer, transfer all input to the output file, and close the output file.
SEARCHING THROUGH FILES

4.11

Two types of searches are available to provide the ability to move within a program to insert changes and delete portions of a program. The first is the standard character search that searches for a single character (Section 4.1 1.1), and

the second is the character string search that searches for a combination of characters (Section 4.1 12),

Single Character Search

4.11.1

The single character search command is one of the most useful functions in the Editor. It is also structured somewhat

differently from other Editor commands.
The single character search is of the form

#nS
or

#m,nS

where m and n represent line numbers (In must always be less than n) and S initiates the search. This search command
sets up to search the entire text buffer or the line or lines specified by the command.

4.11.1.1

nS Command

The nS command searches line n for the character specified after pressing the RETURN

key which enters the command. The line, up to and including the character being searched for, is then displayed,
output stops and all or any combination of the following operations may be undertaken. At this point the following
options are available.
1.

Delete the entire portion not yet displayed and terminate the line and the search by pressing the RETURN

key.
2.

Delete from right to left one of the printed characters each time the DELETE key is typed.

0.)

Insert characters after the last one printed simply by typing them.

Insert a carriage return/line feed, thus dividing the line into two, by pressing the LINE FEED key followed

by CTRL/ L.
Continue searching the line to the next occurrence of the search character by typing CTRL/L. When
printing stops all options are again available.
Change the search character in the line and continue searching by typing CTRL/G(BELL) followed by
the new search character. This allows all editing to be done in one pass.
Type CTRL/G(BELL) twice to terminate the command.
NOTE
The usual form of the character search command is #5,

followed by the RETURN key and then typing the
desired character to be located. This form of the com-

mand is used to modify the current line.
4.11.1.2

m,nS Command

‘

The m,nS command searches lines m through it inclusive in the same way as described

above. The only difference is that pressing the RETURN key deletes the entire undisplayed portion of the line and
terminates that line, but the search continues on the next line.

4.11.1.3

S Command

—

single character, as above.

By typing S with no arguments, the entire buffer may be searched for occurrences of a

Creating and Editing a Program
Character String Search

4.11.2

The character string search can identify a given line in the buffer by the presence in that line of any unique combina—
tion of characters. This search returns the line number as a parameter that can be used to further edit the text.

Two types of string search are available: intrabuffer search and interbuffer search.

4.11.2.1

IntraBuffer Character String Search

—

The intrabuffer search scans all text in the current buffer for a

specified character string. If the string is not found, a question mark (?) is displayed and control returns to the
Editor command mode. If the string is found, the number of the line that contains the string is put into the current
line indicator (.) and the Editor waits for another command to be issued. A command that uses this number may
also be used.

Thus, searching for a character string in this manner furnishes a line number that can be used in conjunction with
other Editor commands. This provides a useful framework for editing since it eliminates the need to count lines or

search for line numbers by listing lines.
An intrabuffer search is signaled by pressing the ESCAPE key (which echoes as a dollar sign) in response to the

Editor’s number sign (#). The string to be found is then typed. The string can be up to 20 characters long. The
search string cannot be broken across line boundaries, that is, it cannot contain a carriage-return/line-feed combination.

Typing a single quote (’) terminates the character string and causes the search to be performed beginning at

line 1 of the text buffer. Use of the double quote (”) instead causes the search to begin at the current line +1.

(Use

of the single quote (’) and double quote (”) as command elements prohibits their use in the search string.)

The FORTRAN program previously created will be used as an example in illustrating the features of character string
search. First, open the file by typing

LEDIT

F'ONER 0 FT

Then, in response to the number sign (#) sign, type

=_I:F{
which instructs the Editor to read in the program into the text buffer and return to the command mode.

Typing

indisplays the program on the terminal. The Editor again returns to the command mode (#). To list the line that contains WRITE, type

=I= fl} U [23: I T 152'.

’

l...

where $ represents the ESCape key.

The search begins with line 1 and continues until the string is found. The current line counter is set equal to the
line in which the string WRITE occurred, and the L command causes the line to be displayed as

U133: I TIE:

( 4 p 113 J

Control returns to the command mode, awaiting further commands. The next reference to WRITE can be found
by

typing

4-12

Creating and Editing a Program
In this case, the double quote (”) is a command that causes the last string searched for to be used again, with the
search beginning at the current line +1. It is not necessary to enter the search string again. The command may be

used several times in succession. For example, in a larger program the fourth occurrence of a string containing the
characters WRITE is found by typing

1‘JSHRIIITIEEI

I...

This command lists the line that contains the fourth occurrence of that string. The L (List) command (or any other

command) can be given following either the single quote C) or double quote (”). The command causes the line to be
listed when and if it is found.

The properties of the commands single quote C) double quote (”) allow for easy and useful editing, as the following

example illustrates. To change DIMENSION A(l6) to DIMENSION A(20) type the following commands:

1131:"ORTRAN

’

$131M" C

IIIMENSESIUN

(“130)

These instructions first cause the Editor to start at line 1 and search for the line beginning with FORTRAN. A
search is then made for DIM starting from the line after the line containing FORTRAN. When this string is found,
the line number of the line containing the string DIM becomes the current line number. The C(Change) command

allows the line to be changed to the correct statement.

CTRL/L.

Control is returned to the Editor command mode by typing

Use the TAB key to line up new or changed text that is being input to match the format that was originally

used.

Since this search feature results in a line number, any operations that can be done by

explicitly specifying a line num-

ber can be done by specifying a string instead. For example,

195 F‘ 0 R T Ii: .4» N

’

+ 33 L

will list the fifth line after the first occurrence of the text FORTRAN in the text buffer

:20

N11: 1 v 15)

Typing

fflil‘TORTRIfiN

’

$130

will list all lines between the two labels, inclusive.
To list the line in which the format statement 25 appears the second time, type

1: 9r} 12’. '1'";

The single quote (’) accounts for the first 25 encountered in the WRITE statement. The double quote (”) then seeks
the second line that contains the number 25.
When both strings and explicit numbers are used, strings should be used first. For example, the commands

iii-+110

’

L-

will not list the next line after the string DO occurs. The correct syntax is

193130

’

i- 1 I...

4-13

Creating and Editing a Program
The interbuffer search scans the current text buffer for a charac-

4.11.2.1

InterBuffer Character String Search

ter string.

If the string is not found, the current buffer is written to the output file, the buffer is cleared, and the

—

next buffer is read from the input device. The search then resumes at line 1 of the new buffer.

tinues until either the string is found or no more input is left.

This process con-

Ifinput is exhausted, control returns to the command

mode with all the text having been written to the output file. If the string is found, control returns to the command
mode with the current line equal to the number of the line containing the first occurrence of the string. For exam-

ple, a command to find the character string WRITE may appear as follows:
in

guanE'

'_o==M
The J command initiates an interbuffer search. The Editor automatically displays the dollar sign ($), and then the
character string that is sought is typed-in and terminated by the apostrophe. The search proceeds, and when the

string is found, control returns to the command mode. Now type the .= construction to find the number of the line
in the current buffer on which the string is contained, or any other desired editing commands.

To find further occurrences of the string WRITE, type the F command. The F command uses the last character

string entered to search the buffer starting from the current line count + 1:

£122
g.moooa
This example causes a search for the string WRITE starting at the current line + 1. If no output file has been

specified, the J or F command read the next input buffer without attempting to produce any output.
NOTE
Use the I command for interbuffer searches only. After
the J or F command has processed the entire input file,
execute either an E or Q command to close the output

file.
The following two commands may be used to abort the string search command, once given:

Command

CTRL/U

Explanation
A CTRL/U will return control to the Editor command mode if executed while en-

tering text in a string search command; the string search command is ignored, as in
the following example:

§DIM
lit-

The interbuffer search for the characters DIM was aborted by the user typing

before terminating the string with single quote C) or double quote (”).
DELETE

Pressing the DELETE key while entering text for a string search causes the text so
far entered to be ignored and allows a new string to be inserted. The Editor displays
a dollar sign ($) in response to the DELETE
key, as shown in the following

example:

.szITE-

$19

444

Creating and Editing a Program
4.12

EDITING AN EXAMPLE PROGRAM

It is often necessary to edit a closed file, that is, to change the code, correct overlooked errors, or insert additional

information.
This section illustrates how to use the various editing commands in editing an already existing assembly language proand when executed displays
gram. The example program is shown below. It is written in PAL8 assembly language
HELLOL Note that the example has the errors deliberately left in it.

#200
MONADRu7600
QTARTr

CLA

CLL

/CLEAR

BUFADR
RNTR

/SET
/FOR

/CLEAR

TLS
TAB
DCA

NEXTv

/NO:
/GET

.ml
I PNTR

JMP
TAD

RNTRy

TERMINAL

AND

LINK

CHARACTER

TAH

/GET

SKA
JMP

CLA

JMP

SET

CHECK AGAIN
CHARACTER

/DONE YET?
/CLEAR ACCUMYLATOR

PNTR

FLAG

PNTR
CLL

ISZ
CLA

BUFAUR

/PRINT

TLS

LINK

UP POINTER
GETTING CHARACTERS

/SKIP

TFF

ACCUMULATOR AND
TERMINAL FLAG

ANOTHER

CHARACTER

/JUMP ON ZERO ANU CLEAR
/RETURN TO MONITOR

MON

/GET

REAflY

ANOTHER

/BUFFER ADDRESS
/POINTER

RUFF
RUFF

BUFFy

215$212"H9“E$'L§'L3'09'!?0

MUN;

MONAHR

/MONITOR

ENTRY

POINT

You have checked the above program and want to make the following changes:

Insert a comment line at beginning of program.

99).“?

Correct TFF to TSF in line with label NEXT.
Correct the spelling of accumulator where it appears the second time.

Transpose lines JMP I MON and JMP NEXT.
Add a comma after label BUFADR.

Typing the EDIT command and the file called SAMPLEPA will open that file and cause the Editor to prompt with
the number Sign (#) thereby indicating it is ready to receive the editing commands:

It‘l .

EDIT

SSAMF‘LEJZ'A

If you are changing a part of a program, but want to save your original file, type the following commands:

LRENAME SAMPLE.BK‘iISAMF'LEoF'A
LEIIIT SAMPLE F'A-=ZISAMF'LE BK
.

The file that is changed will be output as SAMPLEPA, while SAMPLE.BK will remain unchanged.
At this point, the text buffer is empty.
case

Type the R (Read) command to read in the first page of the file, and in the
of this example, the entire program. After this command is executed, the Editor prompts with the number

sign (#) for another command.

Use the L command to display the program.

4-15

Creating and Editing a Program
Simply type the I(Insert) command and the text to be in—

The first Change is the addition of the comment line.
serted as follows:

iI
/F\‘OUTINE

TYF'E

MESSAGE

This command inserts whatever text is typed before line 1 of the text buffer, and in this example, becomes line 1.
The RETURN key is typed after the line to allow the Editor to recognize this text as ‘a separate line.
then typed to return to the Editor command mode.

CTRL/L is

Verify that this is now line 1 by typing the command

i1L
which displays:

/F.'0UTINE

TCI

TYF'E

MESSAGE

Your next change is that of correcting TFF to TSF.
Use the string search ESCape command to find and display the line to be corrected as follows:

EESTFF

’

Use the single character search option to make the correction in the line by typing.

followed by a RETURN key.
The line number can be left out since the period is equal to the number of the last line searched for the L command. Now type the search character which is F, The line will be displayed up‘ to the F. Delete the F by pressing
the DELETE key and type in the correct character S. Since this completes all changes in this line, type CTRL/G

(BELL) followed by another CTRL/G (BELL) to return to the command mode.
The next change to be made is the correction of the word ACCUMYLATOR to ACCUMULATOR. This example
will illustrate the ability of the search commands to accomplish repetitive searches for identical character strings.
The intrabuffer search commands are shown here to illustrate their use.
The intrabuffer search uses the ESCape key and the single quote C) and double quote (”) constructions.

Typing

E$CLA"L
will list the line containing the second occurrence of CLA, that is,

CLA

CLL

/CLEAR

ACCUMYLATOR

AND

LINK

Note that the incorrect word still appears. The word can be corrected by typing the single character search com-

mand S and using the DELETE key. The period construction is used with the S command since the period is equal
to the number of the last line listed.

Type
1.8

4-16

Creating and Editing a Program
Then type a Y, the character that is to be searched for. Use the DELETE key to erase this character and type the
correct character U.

Since this concludes the corrections on this line, get back to the Editor command mode by typing CTRL/G twice.
Note that if only a carriage return were typed after making the correction, control would return to the command
mode, but the remainder of the line after the corrected character to the end of the line would be deleted.

Verify that the proper correction was made by typing
#.L
which will list the corrected line

CLA

CLL

/CLEAR

ACCUMULATOR

ANI]

LINK

The next correction is to transpose the two lines JMP I MON and JMP NEXT.
Use the character string search command J since this command gives the number of the line to be moved.

Type

Use the M(Move) command to transpose the line by typing

16y16$18M
This command moves line 16 (JMP I MOV) before line 18, or after line 17 (JMP NEXT), which in effect performs
the transposition. It also deletes the initial occurrence of the moved line, now making JMP I MON the 17th line.

Type 17L to verify this.
The next correction requires the addition of a comma after the label BUFADR. The C(Get) command allows you
to get the next line that has a label associated with it after the line that is equal to the current line counter.

Thus,

typing

£6
will get the line starting with BUFADR and display it on the terminal as follows:

BUFADR

BUFF’

/BUFFEF\'

ADDRESS

Then make the correction by adding a Comma after the label, using the change or character search commands

previously described.
When all the

changes-have been made, display the corrected file on the terminal. Type

#L
and the entire contents of the text buffer will be displayed. Remember to close out the file and return to the
Monitor by typing the E command.

During the editing of a file, the output device specified in the command string may become full before the editing
process is complete.

4-17

Creating and Editing a Program
If this happens and a write is attempted on that device, an error occurs. The output file is closed, and the following
message is displayed:

FULL

i
A new output device and file must be indicated before continuing any further editing. Since the contents of the
text buffer are retained through this procedure, no text will be lost if this error occurs.

NOTE
An output file must be supplied following the
asterisk (*), and terminated with a left~ang1e bracket (<)
and the RETURN key. However, specifying an improper output device will cause a fatal error that will
destroy the output file.

Assuming the output device is valid, the Editor will continue the operation that filled the old file, putting all output
into the new output file. After editing is completed, the output files should be combined. The entire process may
appear as follows:

LEIIIT

OUT-=33 I N

iY
it)
$ST R I NG

’

FULL

Device DSK: is full; RXAl : is specified as the new output device, and editing
continues.

1 [I
3E
.

Device RXAl: has become full; remove and insert new diskette. RXAl : is

FULL

ERX

1 30UT3-fi'.

specified as the output device, and editing continues.

At this point, the output “file” is the series of files

DSK:OUT, RXA1:OUT2, and RXA1:OUT3. When output
Therefore, the output files should never be

is split like this, the split may have occurred in the middle of a line.

edited separately since the split lines will then be lost.
4.13

EDITOR ERROR MESSAGES

Two types of error messages, minor and major, are generated when an error is made while running the Editor.
Minor errors, such as incorrect format in a command string or a search for nonexistent information, cause the

Editor to display a question mark. For example, if a command requires two arguments, and only one is provided,
the Editor will display a question mark (?), perform a carriage return/line feed, and ignore the command as typed.

Similarly, if an illegal or unrecognized command character is typed, the error message ? will be displayed, followed
by a carriage return/line feed; the command will be ignored. However, if an argument is provided for a command
that does not require one, the argument may be ignored and the normal function of the command performed. The
following examples illustrate minor errors that can be encountered while using the Editor.
Message
L

Explanation
The buffer is empty. Nonexistent information is requested.

’?

7 y SL

The arguments are in the wrong order. The Editor cannot list backwards.

’?

4-18

Creating and Editing a Program

Explanation

Message

This command requires two arguments before the $; only one was provided.

1 7551 OH
?

Nonexistent command letter.

H
'?

Major errors cause control to return to the Monitor and may be due to one of the causes listed in Table 4-3. These
errors cause a message to be printed in the form
rIn

where n is an error code listed in the table and‘C indicates that control has passed to the Monitor. These errors
will generally result in complete loss of the output file.

Table 4-3

Editor Error Codes

Error Code

Meaning

Editor failed in reading a device. Error occurred in device handler; most
likely a hardware malfunction.

Editor failed in writing onto a device; generally a hardware malfunction.

File close error occurred. For some reason the output file could not be

closed; the file does not exist on that device.
File open error occurred. This error occurs if the output device is a read-only

device or if no output file name is specified on a file-oriented output device.

Device handler error occurred. The Editor could not load the device handler for

the specified device. This error should not normally occur.
4.14

SUMMARY OF EDITOR COMMANDS AND SPECIAL CHARACTERS

The command and special characters discussed in this chapter are summarized in Table 4-4.
Table 4-4
Command

Editor Command and Special Characters

Format

Meaning

Append the following text being typed at the keyboard until a
CTRL/L (form feed) is typed. The form feed returns control to the
command mode. Text input following the A command is appended
to whatever is present in the text buffer.

List the number of available memory locations in the text buffer.
The Editor returns the number of locations on the next line. To

estimate the number of characters that can be accommodated in this
area, multiply the number of free locations by 1.7.

#nC

Change the text of line n to the line(s) typed after the command is
entered (typing a CTRL/L terminates the input).

Continued on next page
4-19

Creating and Editing a Program

Table 4-4 (Cont.)
Command

Editor Commands and Special Characters

Format

#rn,nC

Meaning
Delete lines m through 11 and replace with the text line(s) typed after
the command is entered.

(Typing CTRL/ L indicates the end of the

inserted lines.)
D

#nD

Delete line 11 from the buffer.

#rn,nD

Delete lines m through n from the buffer.

Output the text buffer and transfer all remaining pages of the input
file to the output file, closing the output file and returning to the
Monitor.

Follows a string search. Look for next occurrence of the string

currently being sought (by the J command).
G

Get and list the next line that has a label associated with it. A

label in this context is any line of text that does not begin with one
of the following:
space

/
TAB

RETURN

(ASCII 240)
(ASCII 25 7)
(ASCII 211)
(ASCII 215)

At the termination of a G command, control goes to the command
mode with the current line indicator (.) equal to the line just listed.

#nG

Get and list the first line which begins with a label, starting the
search at line n.

Insert whatever text is typed before line 1 of the text buffer.

Typing CTRL/ L terminates the entering process and returns control to the Editor command mode.

#111

Insert whatever text is typed (until a CTRL/L is typed) before line n

of the text buffer.
J

Interbuffer search command for character strings (see Section 4.11.2

describing the InterBuffer Character String Search).
K

Kill the buffer. Delete all text from the text buffer.
NOTE

The Editor ignores the commands nK and m,nK to prevent
the buffer from accidentally being destroyed if the user
means to type a List command

(m,nL).

Continued on next page

4-20

Creating and Editing a Program
Table 44 (Cont)
Command
L

Editor Commands and Special Characters

Format

Meaning

List entire contents of the text buffer on the terminal.

#nL

List line 11 of the text buffer on the terminal.

#m,nL

List lines In through 11 of the text buffer on the terminal. Control then
returns to command mode.

#m,n$pM

Move lines m through n directly before line p in the text buffer. The $
character represents typing the dollar sign key (SHIFT/4). The old
occurrence

of the moved text is removed.

Write the current buffer to the output file and read the next page.

#nN

Write the current buffer to the output file, kill the buffer, and read

the next page. This action is done n times until the nth page is in the
text buffer. Control then returns to command mode.

The N command cannot be used with an empty text buffer. A question
mark (?) is printed if this is attempted.
P

Write the entire text buffer to the output file. The P command auto-

matically outputs a FORM character (214) after the last line of
output.
#nP

Write line n of the text buffer to the output file and a FORM
character.

#m,nP

Write lines m through n, inclusive, to the output file and a FORM
character.

Immediate end-of—file.

Q causes the text buffer to be output and the

file closed.
R

Read one page from the input device and append the new text to the
current contents of the text buffer.

In no input file was indicated or

if no input remains, a question mark (?) is displayed and control
returns to the command mode.

Character search command (see Section 4.11.1).
The V command causes the entire text buffer to be listed on the line

printer.

#nV

List line n of the text buffer on the line printer.

#m,nV

List lines In through n, inclusive, on the line printer.

#nY

Read (Yank) in a logical page from the input file, without writing
any output. For example,

#SY
Continued on next page
4-21

Creating and Editing a Program
Table 4-4 (Cont.)
Command

Editor Commands and Special Characters

Format

Meaning
reads through four logical pages of input, deleting them without

producing output. The fifth page is read into the text buffer and
control automatically returns to the command mode.

$(ESC)

#$TEXT’

Perform a character string search for the string TEXT.

#”

string search, #” causes a search for the next occurrence of the
string (see Section 4.11.2 describing the IntraBuffer Character
String Search).

Following

.=or.:

Typing these characters obtains the current line number (.=) and the

/=or/ :

last line number in the text buffer (/=). The number is printed by the
Editor immediately’after the user types the equal sign.
character is equivalent to the equal sign.)

Equivalent to .+1L, list the next line in the text buffer.

Equivalent to .— 1L, list the next line in the text buffer.
Equivalent to .+1L, list the next line in the text buffer.

LINE FEED Key

Print the current Editor version number.

NOTE
If the search fails when using the F or I command,

further commands will cause the system to prompt
with a ?. The file must be closed, and then reopened

again.

4—22

(The colon

CHAPTER 5

THE PAL8 ASSEMBLER

5.1 INTRODUCTION

PAL8 is the 08/78 Operating System assembler. It is used to generate binary object files from source programs
written in the PALS assembly language. Once a source program is edited and stored on a diskette, it is ready for

assembly.
PAL8 is a two-pass assembler. The source program is read in pass 1 and an internal symbol table is produced.

Any
symbols used in the source program are defined and added to the PAL8 permanent symbols during pass 1.
During pass 2, the assembler reads the source file again and generates the binary code using the symbol table definitions created during pass 1 and continues defining symbols as well. The binary file that is output may be loaded
into memory as the “current” executable program by the LOAD command. Absolute binary format consists of
8-bit bytes, containing field setting commands, address setting commands, and sequential data words. An optional
third pass may be involved if a program listing is desired. During pass 3, the assembler reads the source file a final
time and generates the assembly listing as an ASCII (character string) file. The listing consists of the source statement together with its current location counter and the generated code in octal. The first 40 (decimal) characters
of the first line of each page of the listing contains a title, the assembler version number, the date and the listing
new

page number.
Use the 08/78 command PAL to call the assembler. It can also be called by the commands CREF and EXECUTE;
use

of these commands is explained in this chapter.

The PAL command specifies the binary and listing output devices and file names, the input devices and file names,
and any options selected by the user. From one to nine input files may be specified. The typical way to assemble,
load, and then run a program called PROG is as follows:

LF‘AL
lLOAD
LSAUE
.

F'R‘OG

—

SYS

F'R'UG

FROG

—

Assemble the program
Load the program into memory
Save the program
Run the program

The long form of the command string is
.PAL DEV:BINARY,DEV:LISTING,DEV:C REFLS<DEV21NPUT,.../ OPTIONS
If the extension to the file name is omitted, the following extensions are assumed

.PA for input files.
.BN for binary output file.

.LS for listing output file.
.TM for intermediate CREF file.
If an assembly or CREF listing is not desired, omit the listing file or CREF file,

respectively.

For example, to assemble, load, and run a PAL8 program named SAMPLE, which is stored on diskette unit 1,

LF‘AL

RXA128AMF‘LE/GMT

5-1

type

The PAL8 Assembler

After assembly the program is loaded and run (since the / G was specified) with the starting address assumed
to be location 0200 in field 0; the binary file is stored on the DSK: device as SAMPLE.BN. The CCL option -T

displays the assembled program listing on the terminal.
If a binary file is not desired, specify the -NB option at the end of the command line (NB stands for No Binary).
For example, to get a listing only, type

LF‘AL

SAMF‘LE-mLS-“NB

The -LS option indicates that a listing should be produced.
The assembler displays any error messages encountered in the program on the terminal, even when a listing is not

produced. Typing CTRL/O at the keyboard during an assembly suppresses the display of error messages. However,
messages are still printed in the listing file (if any) and occur immediately before the line that is in error.
For example, the command line

LF‘AL
causes PAL8

SAMF'LE/SWLS
to assemble SAMPLEPA (or SAMPLE), generating DSK: SAMPLE.BN and putting the listing into the

file SAMPLE.LS on the default device DSK. The / S option suppresses listing of the symbol table.

The command line

LF‘AL.

BIN‘riSAMF‘LE

F'fi/GwéOO

assembles SAMPLE.PA, putting the binary output into a file named BIN.BN, and then loading the file BIN.BN and

starting it at 600. The construction =600 is an option that specifies the starting address.
Assembly can be terminated at any time by typing CTRL/C on the keyboard, and any output files being stored
Otherwise, PAL8 always returns to the Monitor upon completion of assembly.

will be deleted.

A source program may be broken down into multiple files. There files may then be assembled by specifying them
as

input files separated by commas. For example,

LIT-”AL

F'ARTI 9 PART? , F'ARTTS

assembles a three-part program. This technique is useful when it is desired to assemble two programs that are
identical except for a few lines at the beginning of the programs. There different lines can be broken out into a

“prefix file”.

For example, two different file assemblies may be generated by

LPN...

F'RFX 1 9 FILE

4 F‘ If?! I...

F'FCF" X 233 9 F' I LIE:

and

Up to nine input files may be concatenated together in an assembly specification.
If more than one input file is specified, and output files are desired but not explicitly specified, the name of the

first input file is used for the output file names. For example,

LPAL

9,3

produces the binary file A.BN.

5-2

The PAL8 Assembler

If a file name other than the first input file is desired for the binary file name, use the -NB option after the last input
file name not desired as the binary file name. For example,

LPAL

NBvB

produces DSK: B.BN and

LPN...

A w B r C----NB 9 [I v E y F

produces DSK: D.BN.
If a -LS option is specified, it must appear immediately after an input file name. This is the name that will be used

for the name of the listing file. For example,

LPAL

fivaLS

produces DSK: B.LS while

lF-‘fil-

fim-LS , 13

produces DSK: A.LS.
A CCL -L or —T option used with a PAL or COMPILE command outputs the listing on the line printer and terminal,

respectively.
Note that the PAL command normally produces a binary file even when a name is not given.

LPN...

Thus, typing

L. PT 2 ~23 file

produces a binary file.
The PAL command assumes the input file type (extension) is .PA when none is specified.
L F‘ m...

Thus, the command

T 13:”. ST

looks for a file DSK: TEST.PA. If no file with .PA is found on DSK:, then a file test without any extension is
looked for next. It is good practice to include a .PA extension with any PAL8 source files to remind you what type

of source file it is.
The COMPILE and EXECUTE commands may also be used to invoke PAL8. These commands search the directory

of the specified device for the file given with the command, and if one is found with a .PA extension, PAL8 is
invoked. For example,

LCDMF‘ ILE

TEST

will run PAL8 if TEST.PA is found. An unusual extension may be explicitly specified by typing

LF‘r-fil...

TEST

which will assemble DSK: TEST.XX. To specify PAL8 as the processor in the COMPILE command, include -PA
as a CCL

option in the command line as follows:

LCDMF‘ILIEEC

TEST

XX-"F‘fi

5-3

The PAL8 Assembler

The EXECUTE command is similar to the COMPILE command except that the EXECUTE command is supported

by the /G option.
If an argument is not given with a PAL, or COMPILE, or EXECUTE command, the argument used with the last such

command is assumed when that command is used again.
5.2

CREATING AND RUNNING AN ASSEMBLY PROGRAM

The following steps demonstrate the procedure for creating and running a PAL8 program.

5.2.1

Creating a Program

Create the assembly language source file by calling the Editor as follows:

LCR‘EATE

SfiMF‘I...I?-§

F'A

Since a new program is being created, only a single file name need be specified. The OS/ 78 Editor will then display
a number

sign (#) to indicate it is ready to accept a command. (See Chapter 4 for a detailed discussion of the

08/78 Editor.)
Type the A (Append) command to allow the Editor to accept text. Then type in the program, one line at a time
with each line followed by the RETURN key.

EA
/RUUTINE

TYPE

MESSAGE

$200
M0NAURR7600
START;

NEXT,

CLA
TLG

CLL

/CLEAR
/CLEAR

TAO

RUFAOR

/8ET

OCA

PNTR

/EOR

TGE

.wl

/NO;

TAO
TLS

/GET A
/PRINT

POINTER

CHECK

ISZ

PNTR

/OONE

CLL

/CLEAR

TAO

/GET

GZA

CLA

/JUMP

JMP
JMP

NEXT
I MON

/GET READY
/RETURN TO

EUFAOR,

BUFF

PNTRv
RUFF.

BUFF

MON.

MONAOR

AGAIN

CHARACTER
A CHARACTER

CLA

PNTR

LINK

GETTING CHARACTERS
/GNIP IF TERMINAL FLAG SET

JMP

PNTR

ACCUMULATOR ANO
TERMINAL FLAG

YET?
ACCUMULATOR

ANOTHER
ON

/BUFFER

ANO

LINK

CHARACTER

ZERO

ANO

CLEAR

TO PRINT
MONITOR

ANOTHER

AOORESG

/POINTER
2153212‘-H;-E;-L;-L3-O;'!;o
/MONITOR

ENTRY

POINT

Now type a CTRL/ L to terminate input. This command returns you to the Editor command mode.

Type the L (List) command in response to the Editor’s number sign (#) to list the text that was inserted into the
text buffer.

When you are satisfied that the input is correct, type the E (Exit) command to store the file and return to the
Monitor.

5-4

The PAL8 Assembler

5.2.2

Assembling a Program

Now assemble the source program just created. Use the command:

LPAL

SAMPLE-LS

This command creates a binary file SAMPLE.BN and produces a listing called SAMPLE.LS on the diskette

(performed by the CCL -LS option). Use the TYPE command to display the listing on the terminal or the LIST
command. to print the listing on a line printer.
The assembly listing produced by PAL appears as follows:
/RUUTINE

TYPE

/ROUTINE
0200

llmfiAYM77

PALS”U10A

MESSAGE
TYPE

PAGE

MESSAGE

*200

MONADRm7éOO

7600
00200

7300

00201

$046

TLS

00202
00203

1216
3217

TAD
DCA

00204

6041

00205
00206

5204
1617

JMR
TAD

00207

6046

TLS

00210
00211

2217
7300

182
CLA

RNTR
CLL

/UONE YET?
fCLEAR ACCUMULATOR

00212

1617

TAD

/GET

00213

7640

82A

OLA

/JUMR

00214

5204

JMP

ffiET

READY

00215

3631

JMR

/RETURN

00216
00"“

0220
0220

BUFAURv
PNTRv

RUFF

0215

HUFFV

2159212§"H§'E§"L$'L§“O§"!$0

MON!

MONADR

START:

CLA

NEXTv

CLL
BUFAUR
PNTR

/CLEAR

ACCUMULATOR

/CLEAR

TERMINAL

/SET
/FOR

TSF

/N03

/GET

BUFF

MON

TERMINAL

FLAG

SET

ANfl

LINK

CHECK AGAIN
A CHARACTER

/PRINT

RNTR

LINK

UP POINTER
GETTING CHARACTERS

/SKIP
.m1
I PNTR

AND

FLAG

CHARACTER

ANOTHER

CHARACTER

ZERO
TO

AND

CLEAR

ANOTHER

MONITOR

/BUFFER AUDRESQ
/POINTER

0212
0310
030$
0314

0000
7&00

/ROUTINE

TYPE

ERRORS

DETECTED:

LINKS

GENERATED:

BUFADR

0216

BUFF
MON

0220
0231

HDNAHR

7éOO

NEXT
PNTR

0204
0217

START

0200

/MONITOR

MESSAGE

ENTRY

PALBWUlOA

POINT

11wMAYm77

The PAL8 Assembler

The first column of the listing gives the field number and octal address. The second column is the assembled object
code. The symbol table is printed at the end followed by the number of errors detected and links generated. Link

generation is described in Section 5.12.

Each error generates on error message. The error messages are described in

Section 5.14.
If errors have been detected, the program has been written or typed incorrectly. Check it again.

The COMPILE command may also be used to assemble the program by typing

LCDMF‘ILE

SAMPLE

Several options are available with the PAL command. The options are described in detail in Section 5.3.
5.2.3

Loading and Saving a Program

Load the binary file generated by assembling SAMPLE.PA into memory by typing

iLOAD

SAMPLE

The SAMPLE program is now the “current” memory image.
If more than nine input files are to be specified, specify the nine input files in the command line, then press the

ESCape key. This causes the Command Decoder (see Appendix D) to print an asterisk (*). Then type the additional
files in response to the asterisk, terminating each file by a RETURN key.
Since programs in memory image format can be executed directly, it is desirable to save this format of your
program. Do this by typing

LSAUE

SYS

SAMPLE

The memory image fomat of SAMPLE is now both in memory and on the system device as a new file called

SAMPLESV.
5.2.4

Executing the Program

Since the program now resides on the diskette and in main memory, it can be executed, by typing

LE; TAR T
Otherwise, the file SAMPLESV can be loaded into memory from SYS: and run by typing

SAMPLE

The program will be executed and HELLO! will appear on the screen.

Also, the program can be assembled, loaded and run by simply typing

LEIXECUTE

SAMPLE

This command produces the binary file SAMPLE.BN, loads it into memory, and starts it running.

The PAL8 Assembler

Another method that is available to assemble, load and run a PALS assembly language program is the use of the /G

option with the PAL command. Typing

LF‘fiL

SAMPLING

assembles the input file SAMPLEPA, loads the binary file, and executes the program. Also, the command

EEQMF‘I...EI/G

_._l...f.mfl

will load the binary file SAMPLE.BN and execute it.
5.2.5

Getting and Using a Cross-Reference Listing

The Cross-Reference Program (CREF) is an aid in debugging assembly language programs by providing the ability
to pinpoint all references to a particular symbol.

Generate the CREF listing by using the CREF command or the PAL command with the /C option.

:F' t-‘i I...

Typing

SESAMF‘LJE/ CulfC-S

will produce a binary file and the CREF listing as a file called SAMPLE.LS.

Using the TYPE or LIST command

will display the listing on the terminal or print it on a line printer, respectively. Further information on CREF is

given with the discussion of the CREF command in Chapter 3.
The output of CREF is identical to the PAL8 assembler output except that the CREF program numbers each line
in decimal and generates a cross-reference table after the listing that has the following format:

BUFADR
RUFF

18¢

MON

29%

MONAHR
NEXT

3t
8%

PNTR

START

201

16
12

19%

The cross-reference table contains every user-defined symbol and literal, sorted alphabetically. If literals are used,
each literal is indicated by an underline followed by the field and address at which the literal occurs. For each

symbol and literal there appears a list of numbers that specify the lines in which each is referenced. The line where
the symbol is defined is followed by a #.
5.2.6 Obtaining a Memory Map
Many times it is desirable to obtain a map of a program showing memory locations used by the given binary file.
Generate the map by using the MAP command. Typing

iMAF‘

SAMF‘LE~L

will print the map on a line printer from SAMPLE.BN, and typing

LMAP

SAMPLE

will display the map on the terminal (in effect, equivalent to the command MAP SAMPLE -T).

5-7

The PAL8 Assembler

The input file must always be a binary file (.BN extension). Use the command

VIN”

M :5) F' F I I... ~71: $33 ("a H I” l... IE:

to place the map in the file MAPFILMP.

It can be displayed on the terminal or printed on a line printer by using

the commands TYPE and LIST, respectively. The map for the program SAMPLE is shown below.

BITMAP

FIELD

000000001II.11!.1|.1].Ii2531323253222123331333323134444444451" .J 5556666666677777777
012134536701223456701213456701234567012313456701 21343670123456701234567
00000
001.00
0017300

111111 11111311:I.:I.1:1.:I.:I.:l.:l.:l.:l.:l. :I.:I.00000000000000OO0000000000000000000000

00300
00400
00500
00600
00700

The output is a series of lines that are made up of a string of digits. Each digit represents a single memory location,
and can have a value of 0 to 3. A 0 means that the location is empty while a 1 means that the location was loaded into
into once. The appearance of a 2 or 3 means that a location was loaded into two (2) or three or more times (3) and
same
may imply a programming error in that two or more separate routines are each trying to load values into the
once
which
is
into
correct.
0231
loaded
0200
shows
locations
location. The example program
being
through
memory

Further information on the MAP command is given in Chapter 3.
5.3

PALS OPTIONS

The command string typed for the PAL command may include several options. The options available are listed
in Table 5-1.

Table 5-1

PALS Options

Option

Meaning
This option makes the operator ! a 6-bit left shift instead of an inclusive OR (A!B equals

AAIOO B). This allows you to pack two 6-bit ASCII characters into a 12-bit word. This
effect applies for the entire assembly.

Run SYS:CREF.SV after assembly. The third output file specified (optional) is the temporary

output file passed to CREF. The second output file is the listing file to be produced. If no
third output file is given, SYS:CREFLS.TM is assumed and will be deleted after use. The /C

option supersedes the /G and /L options if specified in the same command string.
/E

Enable error messages ifa link is generated. The LG error message is generated as well as
the link being flagged.

Continued on next page
5-8

The PAL8 Assembler

Table 5-1 (Cont)

PAL8 Options

Meaning

Option

Disable extra zero fill in TEXT pseudo-op. If the text in the TEXT pseudo-0p contains an even

number of characters, no word of zeros will be added to the end.

Load the binary file into memory and begin execution at the indicated starting address. If no
starting address is indicated, start at 200.

Generate nonpaginated output. Headers (including page numbers and page format) are suppressed.

Do not list lines containing code in conditional brackets which is conditionalized out.

Used in assembling very large programs; allows more space in Field 1 to be used for symbol table

storage.

Load the resulting binary file into memory but do not start it.

Generate the symbol table, but not the rest of the listing.

Disable origining to 200 after a FIELD pseudo-op. The origin remains what it was before the FIEED
pseudo-0p.

Omit the symbol table normally generated with the listing.

Output a carriage return/line feed in place of form feed character(s) in the program listing.

Do not remember the number ofliterals that were previously stored on a page after origining off
page and then back on again.

When the /L or /G option is specified, you can also include any option in the command line for the LOAD

command, such as

starting address option. If no address is specified, 00200 is assumed. If no binary output file

is specified (by using the CLL -NB option) with a /L or /G, a temporary file SYS:PALSBN.TM is created and loaded.

5.4

CHARACTER SET

PAL8 programs are composed of physical lines containing assembly language mnemonics that indicate processor in-

structions, user-defined symbols, comments, listing control characters and pseudo—operators (assembler directives).
The following characters are used to specify these components.

hwp -i

The alphabetic characters A through Z.
The numeric characters 0 through 9.
The characters described in following sections as special characters and operators.

Characters that are ignored during assembly such as LINE FEED and FORM FEED.

all other characters are illegal (except when used in a comment) and cause the error message
IC nnnn
to be printed during passes l and 2 where nnnn represents the octal location at which the
illegal character occurred.
(As assembly proceeds, each instruction is assigned a location determined by the current location counter. When an
illegal character or any other error is encountered during assembly, the value of the current location counter is displayed in the error message.)

5-9

The PAL8 A ssembler

5.5

STATEMENTS

PAL/8 source programs are usually prepared on the terminal (using the Editor) as a sequence of statements. Each
statement is written on a single line and is terminated by a carriage return. There are four types of elements in a

PALS statement which are identified by the order of their appearance in the statement and by the separating (or

delimiting) character that follows or precedes the element. These types are as follows:
1

label
instruction

operand

comment

A statement must contain at least one of these elements and may contain all four types. The assembler interprets

and processes the statements, generating one or more binary instructions or data words, or performing an assembly
process.
5.5.1

Labels

A label is the symbolic name created by the programmer to identify the location of a statement in the program. If

present, the label is written first in a statement. It must begin with an alphabetic character, contain only a alphanumeric characters, and be terminated by a comma; there must be no intervening spaces between any of the
characters and the comma. A label may be of any length but only the first six characters are significant. If a label
is the only element on a line, it identifies the location of the next program location. For example,

A,
B,0
5.5.2

Defines A as 00200

Defines B as 00200 and stores 0000 at location 00200

Instructions

An instruction may be one or more of the mnemonic machine instructions or a pseudo-operation that directs

assembly processing. (Assembly pseudo-ops are described in Section 5.11.) Instructions are terminated with zero
or more spaces (or tabs) followed by a semicolon, slash, or the end of the line.
5.5.3

Operands
Operands are the octal or symbolic addresses of an assembly language instruction or the argument of a pseudooperator, and can be any legal expression. In each case, interpretation of an operand depends on the instruction
or the pseudo-op. Operands are terminated by a semicolon, slash, or the end of the line.
5.5.4

Comments

Comments are arbitrary strings of characters that begin with a slash (/). Comments do not affect assembly pro-

cessing or program execution but are useful in the program listing to record information for later analysis or debugging. The assembler ignores everything from the slash to the next carriage return.
It is possible to have only a carriage return on a line, resulting in a blank line in the final listing. [Such a line is

ignored; the current location counter is not incremented.
5.6

FORMAT CHARACTERS

The following characters are useful in controlling the format of an assembly listing to improve readability.

They

allow a neat readable listing to be produced by providing a means of spacing through the program.
5.6.1

Form Feed

The form feed character causes the assembler to output blank lines (or a form feed character iflisting on the line
printer) in order to skip to a new page in the output listing during pass 3; this is useful in creating a page-by-page
listing. The form feed is generated by the Editor P(Page) command. The Pseudo-op EJECT may also be used to
form pages in the assembly listing (see Section 5.11.7).

5-10

The PAL8 Assembler

5.6.2

Tab

Tabs are used in the body of a source program to separate fields into columns. For example, a line written

[~30 :1

T Mil

LUUF'

TOTAL] Mfi I N

is much easier to read if tabs are inserted to form

TrfitII

[30 :1

/’ MA I N

'l'U‘T'rfitl...

LOUF'

Each occurrence of a tab character causes PAL8 to output enough spaces to move to the next text column. Each
text column is 8 characters wide.

5.6.3

Statement Terminators

Each statement is terminated by the carriage return/line feed character combination produced by the Editor when
the RETURN key was pressed during the Insert or Append modes. The semicolon (;) may also be used as a statement
terminator and is considered identical to a carriage return except that it will not terminate a comment. For

example,
TAI]

fir

x’THIEEB

Tfifl

C(IlMMIEZN'H-

1-3

The entire expression between the slash and the end of the line is considered a comment. Thus in this case the
assembler ignores the TAD B. If, for example, a sequence of instructions to rotate the contents of the accumulator
and link six places to the right is desired, it can be written as follows:

RTR

R T F:
RTR

However, as an alternative, all three instructions can be placed on a single line by separating them with the special
character semicolon and terminating the entire line with a carriage return. The above sequence ofinstructions can

then be written

H "I" R 'v F: T R a? R T R
These multistatement lines are particularly useful when setting aside a section of data storage. For example, a
4~word block of data could be reserved by specifying either of the following:
I... II." EST 5'

:l. ; 133 9 I5 6 4

L. I ST ,

3
4
5.7

NUMBERS

Any sequence of digits delimited by either a SPACE, TAB, semicolon, or end of the line forms a number. PAL8
initially interprets numbers in octal (base 8). This can be changed to decimal using the pseudo-op DECIMAL
(Section 5.11.10). Numbers are used in expressions.

5-11

The PAL8 Assembler

5.8

SYMBOLS

A symbol is a string of alphanumeric characters beginning with a letter and delimited by a nonalphanumeric character.

Although a symbol may be any length only the first six characters are significant. Since additional characters are
ignored, symbols which are identical in their first six characters are considered identical.
5.8.1

Permanent Symbols

The assembler symbol table initially contains definitions of the symbols for all computer instructions and PAL8

pseudo-ops. These symbols are permanently defined by PAL8 and need no further definition by the user; they are
summarized in Section 5.15. For example,
HLT
5.8.2

This is a symbolic instruction assigned the value 7402 in its permanent symbol table.

User-Defined Symbols

All desired symbols not defined by the assembler (in its permanent symbol table) must be defined within the source
program.
User symbols may be defined in two ways:
1.

As a statement label. Labels are assigned a value equal to the current location counter.

As an explicitly-defined symbolic value.

(For example, A

33.)

Permanent symbols (instructions, special characters, and pseudo-ops) may not be redefined as a label or symbolic

value. The following examples are legal labels:

ADDR,
TOTAL,
SUM,
A1,
The following are illegal labels:

AD>M,
7ABC,
LA BEL,
D+TAG,
LABEL,
5.8.3

(contains an illegal character)
(first character not alphabetic)
(contains embedded spaces)
(contains a legal but non-alphanumeric character)
(not immediately terminated by a comma)

Current Location Counter

As source statements are processed, PAL8 assigns consecutive memory addresses to the instructions and data words

of the object program (binary and listing) being produced.
The current location counter contains the address in which the next word of object code will be assembled and is

automatically incremented each time a memory location is assigned. A statement that generates a single object
program storage word increments the location counter by one. Another statement might generate six storage words,
incrementing the location counter by six.
The location counter is set or reset by typing an asterisk followed by an expression giving the address in which the
next program word is to be stored.

The expression may include symbols, but every such symbol must have been
defined at some previous point in the current source file(s) being assembled. If the origin is not set by the user,

PAL8 begins assigning addresses at location 200.

5.12

The PAL8 Assembler

The symbol TAG in the following example is assigned a value of 0300, the symbol B a value of 0302, and the

symbol A a value of 0303.
/SET

*300
CLA

TfiG’

JMP

CURRENT

LUCQTION

COUNTER

300

B y

A v

13th

If a symbol is defined more than once as a label, the assembler will display the “illegal definition” error message.
ID address

where address is the octal value of the location counter at the second occurrence of the symbol definition. The

symbol is not redefined. PAL8 error conditions are described in Section 5.14. For example,
7K "(5 0 0

STN‘QTr
(33 [I] N T I N 9

T041]

.6;

XII [Z 15%
.J M 53

l: 0 U N T E: F:
I... [SE A U E

\JMF'

‘3 T AR T

by
CUUNTERv

W74
o

START»

one

CLL

\"I

The symbol START would have a value of 0300; the symbol CONTIN would have a value of 0302', the symbol A

would have a value of 0304; and the symbol COUNTER (considered COUNTE by the assembler, because the
assembler uses only the first six characters of a symbol) would have a value of 0305. When the assembler processes
the next line, it will display the error message:
ID COUNTE+0001
PAL8 will also display an error message if reference is made to an undefined symbol. For example,

)l‘. L7 1 7 0
{15 ,

C y

T (“I

C I... If)

C Mfi

H l... T
\J HF”

(a 1

This would produce the “undefined symbol” error message

US A+0003
since the symbol A1 has not been defined.

Symbol Table
Initially, the assembler’s symbol table contains the definitions of the computer instructions and PAL8 pseudo-ops;
these are PALB’s permanent symbols. As the source program is processed, user-defined symbols along with their
12-bit binary values are added to the symbol table. Entries in the symbol table are listed in alphabetic order at the
end of the listing file produced, if any.
5.8.4

5-13

The PAL8 Assembler

During pass 1, if PAL8 detects that the symbol table is full (in other words, there is no more memory space in which
to store symbols and their associated values), the “symbol table exceeded” error message is displayed as follows:

SE address
and control returns to the Monitor. The number of symbols defined in the program may be reduced by using relative addressing for example, JMP START+3), It is also possible to segment a program and assemble the segments
separately, taking care to define correct links between the segments. PAL8’s symbol capacity is 2621 (decimal)
symbols of which 96 are permanent. Where PAL8 is run under BATCH, 2357 Symbols can be defined. (Use of the
/K option expands these to 2971 and 2719 respectively, at the expense of slower assembly time.)

Instructions concerning altering the permanent symbol table are discussed in Section 5.11.8.
5.8.5

Direct Assignment Statements

New symbols with their assigned values may be inserted directly into the symbol table by using a direct assignment
statement of the form

SYMBOL=VALUE
VALUE may be a number or expression. No spaces or tabs may appear between the symbol to the left of the equal
sign and the equal sign itselfbut they may appear (and are ignored) after the equal sign. The following are examples
of direct assignment statements:
A=6
EXIT=JMP 1 0
C=A+B

CO=JMSI [.1
All symbols to the right of the equal sign must already be defined, except that symbols are allowed to be undefined

during pass 1.

For example,

A=B

B=3

During pass 1, A will equal 0, since it is undefined thus far. During pass 2, A will equal 3, since B is given the value 3
The use of the equal sign does not increment the location counter; it is an instruction to the

at the end of pass 1.

assembler itself rather than a data value.
A direct assignment statement may also equate a new symbol to the value assigned to a previously defined symbol.

For example,
BETA=17

GAMMA=BETA
The new symbol GAMMA is entered into the user’s symbol table with the value 17. The value assigned to a symbol
may be changed as follows:
ALPHA=5
ALPHA=7
The second line of code shown changes the value assigned to ALPHA from 5 to 7.

5-14

The PAL8 Assembler

Symbols defined by use of the equal sign may be used in any valid expression. For example,
3% $3 0 O
A ==== 1 0 (TI

/ I! O E. 8

B3400
A+B

/II|(3ES NUT UPDATE CUFL‘R‘ENT LOCATIIIIN [COUNTER
/ TH E 'J A L. U If: {’3 O 0 I ‘3 IQ S S E M B L. E [I A T L O C 2'3 0 0

T A [I

/THE

N0T

U F‘ I] A T l?

V {A l... U E:

II. .'|. O 0

CURR ENT

l... U C r3) T I (I) N

A S 8 E M B L. E II

(i! T

C O U N T E F:

LUC

13 01

If the symbol to the left of the equal Sign is in the permanent symbol table, the “redefinition” diagnostic
RD address
will be displayed as a warning, where address is the value of the location counter at the point of redefinition. The
new value will

be stored in the symbol table. For example,

CLA=7600
will cause the diagnostic
RD+2OO
Whenever CLA is used after this point, it will have the value 7600.

5.8.6

Symbolic Instructions

Symbols used as instructions must be predefined by the assembler or defined in the assembly by the programmer.
If a statement has no label, the instructions may appear first in the statement and must be terminated by a space,
tab, semicolon, slash, or carriage return. The following are examples of legal instructions:
TAD
PAGE
ZIP

(a mnemonic maching instruction)
(an assembler pseudo-0p)
(an instruction defined by the user)

5.8.7

Symbolic Operands
Symbols used as operands normally have a value defined by the user. The assembler allows symbolic references
to instructions or data defined elsewhere in the program. Operands may be numbers or expressions. For example,
TOTAL, TAD ACI+TAG
The values of the two symbols AC] and TAG (previously defined in the program) are combined by a two’s complement add.

5.9

(See Section 5.9.1 on Operators.) This value is then used as the operand address.

EXPRESSIONS

Expressions are formed by the combination of symbols, numbers, and certain characters called operators, which
cause specific arithmetic operations to be performed. An expression is terminated by either a comma,
carriage
return, or semicolon.

5.9.1

Expressions are evaluated by a strict left-to-right scan.

Operators

Seven characters in PALS act as operators:
+

Two’s complement addition

Two’s complement subtraction
A

Multiplication (unsigned, 12-bit integer)
Division (unsigned, 12-bit integer)

5-15

The PALS Assembler

Boolean inclusive OR

Boolean AND

Space
(or TAB)

Treated as a Boolean inclusive OR except
in a memory reference instruction

No checks for arithmetic overflow are made during assembly, and any overflow bits are lost from the high-order end.
For example,
7755+24

will give a result of l.
The operators plus (+) and minus (~) may be used freely as unary (prefix) operators.

Multiplication is accomplished by repeated addition. No checks for sign or overflow are made. All 12 bits of each
factor are considered as magnitude. For example,

3000'?
will give a result of 6000.
Division is accomplished by repeated subtraction. The quotient is the number ofsubtractions that are performed.

The remainder is not saved and no checks are made for sign. Division by 0 will arbitrarily yield a result of 0. For

example,
7000%1000
will yield a result of 7. This example could be written as:

—1000%1000
The answer might be expected to be ~l (7777), but all 12 bits are considered as magnitude and the result is still 7.

Use ofthe multiplication and division operators requires more attention to sign (on the part of the programmer)
than is required for simple addition and subtraction. Table 5-2 contains examples of expressions using arithmetic

operators.
Table 5-2

Expression
7777+2
7776—3

Use of Arithmetic Operators
Also Written as

—

1+2

-2-3

Result
+1

7773 or-S

0‘2

2‘0

10007

7000 Or

07012

12%0

1000

7777%l

—l%1

7000%1000

-1000%1000

1%2

7777 or-l
7

5-16

The PALS Assembler

The ! operator causes a Boolean inclusive OR to be performed bit by bit between the left-hand term and the right-

Giving the /B option changes the memory of “l” throughout the assembly to become a 6-bit left shift
of the left term prior to the inclusive OR of the right. According to this interpretation,

hand term.

If A=1 and B=2
then

A!B=0102
Under normal conditions A!B would be 0003.
The & operator causes a Boolean AND to be performed bit by bit between the left and right values.

SPACE is an operator that has special significance depending on the context in which it is used. When the symbol
preceding the space is not a memory reference instruction as in the following example
SMA CLA
it causes an inclusive OR to be performed between them. In this case, SMA=7500 and CLA=7600. The expression

SMA CLA is assembled as 7700. When SPACE is used following pseudo-operators it merely delimits the symbol.
When it is used after memory reference operators it has a special function explained below.
User-defined symbols are treated as non-memory reference instructions. For example,
A=1234

=77
A B
stores a data value of 1277 (octal), the same as A!B.

If data values are generated, the current location counter is incremented. For example,
B- 7 ;A+4;A— B

produces three words of information; the current location counter is incremented after each expression.

The

statement

HLTC LA=HLT CLA

produces no information to be loaded (it produces a value for “HLTCLA” in the symbol table) and hence does not
increment the current location counter.

In the program

$4713}. 2!.

TEMP 9
TENS}. ,

the location counter is not incremented after the line TEMP,; the two symbols TEMP and TEM2 are assigned the
same

value, in this case 4721.

5-17

The PAL8 Assembler

Since a CPU instruction has an operation code of three bits as well as one indirect bit, one page bit, and seven
address bits, the assembler must combine memory reference instructions in a manner somewhat differently from the
way in which it combines operate or IOT instructions. The assembler differentiates between memory reference
instructions and user-defined symbols. The following symbols are the memory reference instructions:
AND

0000

TAD

1000

182

2000

DCA

3000

JMS

4000

JMP

5000

Logical AND
Two’s complement addition
Increment and skip if zero
Deposit and clear accumulator
Jump to subroutine
Jump

When the assembler has processed one of these symbols, the space or tab following it acts as an address field
delimiter. In the example,

#4100
\JMF' A
CLA

A has the value 4101, JMP has the value 5000, and the space acts as a field delimiter. These symbols are represented
in binary as follows:
A:

100001000001

JMP:

101 000 000 000

The seven address bits of A are taken, for example,

000 001 000 001
The remaining bits of the address are tested to see if they are zeros (page zero reference); if they are not, the current
page bit is set:

000 011000 001

The operation code is then ORed into the JMP value to form
10101100000]
or, in octal

5301
In addition to the above tests, the page bits of the address field are compared with the page bits of the current
location counter. If the page bits of the address field are nonzero and do not equal the page bits of the current location counter, an out-of—page reference is being attempted and the assembler will take action as described in Section

5.12 on Link Generation and Storage.

5.9.2

Special Characters

In addition to the operators described in the previous section, PAL8 recognizes several special characters that serve

specific functions in the assembly process:
=

equal sign
comma

asterisk

5-18

The PALS Assembler

dot

“

double quote

()
[]
/

parentheses
square brackets
slash

;

semicolon

angle brackets

dollar sign

The equal sign, comma, asterisk, slash, and semicolon have been previously described. The remaining special
characters are described in the following sections.

5.9.2.1

Dot (.)

The special character dot (.) represents the current location counter. It may be used in any

——

expression (except to the left of an equal sign), and must be separated from other symbols by a space or other
operator. For example,
*200
JMF‘ 0 +2
is equivalent to JMP 0202.

Also,

#300
. +2400
will produce in location 0300 the quantity 2700. Consider

X140
PRINTS-JMS

I o

2200
The second line G’RINT=JMS I .) does not increment the current location counter; therefore, 2200 is placed in location 140 and PRINT is placed in the user’s symbol table with an associated value of 4540 (the octal equivalent of

JMS I .). This technique is useful in creating “global” subroutine calls.

Large buffers may be defined by using a format such as the following:

BUFFER!
NEXT;
5.9.2.2

X11200

/BUFFEF\‘

/FIRST

*.+4OO

/IIEFINE

/F‘F\'UGF\'AM

Double Quote (”)

LOCATION
UORII OF BUFFER
A

401

UORII

BUFFER

CONTINUES

When a double quote (”) precedes an ASCII character, PALS assembles the 8-bit

ASCII equivalent of the character.

(ASCII codes are listed in Appendix A.) For example.

OLA

TAD

("f—i)

The constant 0301 is placed in the accumulator when these two instructions are eventually executed. The character must not be a carriage return or one of the characters that is ignored on input (discussed at the end of this

section).
5.9.2.3

Parentheses ( ) and Brackets [

]

—

Left and right parentheses, ( ), enclose a current page literal (closing

member is optional).

5-19

The PAL8 Assembler

$13300

Clint-'5
TAD

INDEX

TQD
Ill: A

(.13)
I N 1"] EX

The left parenthesis is a signal to the assembler that the expression following is to be evaluated and assigned a word
in the literal area of the current page. This is the same area in which the indirect address linkages are stored. In

the above example, the quantity 2 is stored in a word in the literal area beginning at the end of the current memory

page. The instruction in which the literal appears is given the address of the literal. A literal is assigned to storage
the first time it is encountered; subsequent references to the same literal from the current page are made to the same
location. The use of literals frees symbol storage table and makes programs much more readable. Literal allocation
starts with the last location on the page and works towards the first location.

If the literal area reaches the

instruction/data area, a PE (Page Exceeded) error message is generated and assembly continues.
If square brackets ([ and ]) are used in place of parentheses, the literal is assigned to page zero rather than the
current page.

This enables a value to be referenced from any address within the field. For example,

3k 2?. 0 0
T A Iil [i 2’. Z]

>l’. 5‘} (l t)
T A Ill l: 12’. I]

The closing member is optional. Literals may contain any expression.

NOTE
Literals can be nested, for example:

$13300
T913

(TIM?

(30))

This type of nesting may be continued to as many as six

levels, depending on the number of other literals on the
page and the complexity of the expressions within the
nest. If the limits of the assembler are reached, the error
messages BE (too many levels of nesting) or PE (too
many literals) will result.
5.9.2.4

Angle Brackets (<>)
Angle brackets (<>) are used as conditional delimiters. The code enclosed
in the angle brackets is to be assembled or ignored, depending on the definition of the symbol or value of the
expression preceding the angle brackets. (The IFDEF, IFNDEF, IFZERO, and IFNZRO
pseudo-operators are used with
angle brackets and are described in Section 5.11.9.)
—

5-20

The PAL8 Assembler

NOTE

Programs that use conditionals should avoid angle
brackets in comments as they will be interpreted as
beginning or terminating the conditional.
Dollar Sign ($)

5.9.2.5

—

unconditional end-of—pass.

The dollar sign ($) character is optional at the end of a program and is interpreted as an
It may however occur in a text string, comment, or double quote (”) term, in which case

it is interpreted in the same manner as any other character. This feature is provided for compatibility with older

PDP-8 assemblers, and its use is not recommended.
Other Characters

—»

The following characters are handled by the assembler for the pass 3 listing, but are otherwise

ignored:
FORM FEED

Used to skip to a new page

LINE FEED

Used to create a line spacing without causing a carriage return

Nonprinting characters include

TAB

Spaces to next character position
Spaces to next “tab column” of 8 characters

RETURN

Terminates each line

BELL (CTRL/ G)

Sounds the terminal buzzer

SPACE

5.10

INSTRUCTION SET

The instruction set for PALS includes processor instructions, input/output (I/O) instructions, and assembler
instructions (pseudo-ops). The processor instructions are further divided into two basic groups of instructions:
memory reference and microinstructions. (See Section 5.15 for detailed listing of instructions.)

5.10.1

Memory Reference Instructions

Memory reference instructions have the following format:

OPERATION

lNDIRECT

ADDRESS

CODE
I

ADDRESSING—j

MEMORY PAGE

Memory Reference Bit Instructions
Bits 0 through 2 contain the operation code of the instruction to be performed. Bit 3 indicates whether the
memory reference is indirect. Bit 4 indicates whether the instruction is referencing the current page rather than
page zero. Bits 5 through 11 (7 bits) specify an address. Using these seven bits, 200 octal (128 decimal) locations
can be

directly specified; the page bit increases accessible locations to 400 octal or 256 decimal. A list of the

memory reference instructions and their codes is given at the end of this chapter.
In PAL8, a memory reference instruction must be followed by one or more spaces and tabs, an optional I and/or

Z designation, and any valid expression.

5-21

The PALS Assembler

When the character I appears in a statement between a memory reference instruction and an operand, the operand,
is interpreted as the address (or location) containing the address of the operand to be used in the current instruction.

Consider:
TAD 40

which is a direct address statement, where 40 is interpreted as the location on page zero containing the quantity to
be added to the accumulator. References to locations on the current page and page zero may be done directly. For

compatibility with older PDP-8 assemblers, the symbol Z is also accepted as a way of indicating a page zero reference,
as

follows:
TADZ40

This is an optional notation, not differing in effect from the previous example. Thus, if location 40 contains 0432,
then 0432 is added to the accumulator when the code is executed. Now consider:
TAD I 40

which is an indirect address statement, where 40 is interpreted as the address containing the address of the quantity
to be added to the accumulator.

Thus, if location 40 contains 0432, and location 432 contains 0456, then 456 is

added to the accumulator when the instruction is eventually executed.

NOTE
Because the letter I is used to indicate indirect addressing,
it may not be redefined as a label or variable. Likewise

the letter Z, which is sometimes used to indicate a page
zero

5.10.2

reference, may not be redefined.

Microinstructions

Microinstructions are divided into two classes: operate and Input/Output Transfer (IOT) microinstructions. Operate
microinstructions are further subdivided into Group 1, Group 2, and Group 3 designations.

NOTE
If an illegal combination of rnicroinstructions are

specified, the assembler will perform an inclusive OR
between them, resulting in an unexpected Operation.
For example,
CLL SKP is interpreted as SPA

(7100) (7410)
5.10.2.1

(7510)

Operate Microinstructions
Operate instructions are divided into three groups of micro instructions.
is
it
to
combine
instructions
within a group, it is not logically possible to combine instructions
Although
possible
from different groups.

—

Group 1 microinstructions perform clear, complement, rotate and increment operations on

the Accumulator and Link registers, and are designated by the presence of a 0 in bit 3 of the machine instruction
word.

5-22

The PAL8 Assembler

CLA

CLL

CMA

CML

BSW

IAC

ROTATE AC AND L RIGHT

ROTATE AC AND L LEFT
ROTATE 1 POSITION IF A 0,2 POSITIONS IF A I

(BSW IF BITS 8,9 ARE OI
LOGICAL SEQUENCE:

1- CLA, CLL

2- CMA, CML

IAC

RAR, RAL, RTR, RTL, BSW

Group 1 Operate Microinstruction Bit Assignments
The following constants can be produced in the accumulator by a single instruction:

Constant

Instruction

CLA

CLA IAC

CLA CLL CML RTL

CLA CLL CML IAC RAL

CLA CLL IAC RTL

CLA CLL CML IAC RTL

100*

CLA lAC BSW

2000

CLA CLL CML RTR

3777

CLA CLL CMA RAR

4000

CLA CLL CML RAR

5777

CLA CLL CMA RTR

6000*

CLA CLL CML IAC RTR

7775

CLA CLL CMA RTL

7776

CLA CLL CMA RAL

7777

STA (=CLA CMA)

Instructions that are starred (*) must not beused on software to be transported onto old (non-omnibus) PDP-8

computers.

Group 2 microinstructions check the contents of the Accumulator and Link and, based on the check, continue to or
skip the next instruction. Group 2 microinstructions are identified by the presence of a l in bit 3 and a 0 in bit 11
of the machine instruction word.

5—23

The PAL8 Assembler

1
l

CLA

SMA

SZA

OSR

SNL

HLT

REVERSE SKIP SENSING OF BITS 5,6, 7 IF

LOGICAL SEQUENCE:

SET—J

1(BIT8IS 0)‘SIVIA OR SZA OR SNL
(BIT 8 IS 1)

SPA AND SNA AND SZL

2- CLA
3

OSR, HLT

Group 2 Operate Microinstruction Bit Assignments
Group 3 microinstructions reference the MO register. They are differentiated from Group 2 instructions by the
presence of a l in bits 3 and ll 0f the machine instruction word.

OPERATION
CODE7
l

‘I

CLA

MQA

MOL

0
I

CONTAINS A 1 TO
SPECIFY GROUP 3

CONTAINS A 1 TO SPECIFY GROUP 3

Group 3 Operate Microinstruction Bit Assignments

Group 1 and Group 2 microinstructions cannot be combined since bit 3 determines either one or the other. Within
Group 2, there are two groups of skip instructions. They can be referred to as the OR group and the AND group.
0R Group

AND Group

SMA

SPA

SZA

SNA

SNL

SZL

The OR group is designated by a 0 in bit 8, and the AND group by a 1 in hit 8. OR and AND group instructions
cannot be combined with each other since bit 8 determines either one or the other.

If skip instructions are combined, it is important to note the conditions under which a

OR Group

—

skip may occur.

If these skips are combined in a statement, the inclusive OR of the conditions determines

the skip. For example:
SZA SNL
The next statement is skipped if the Accumulator contains 0000 or the link is a 1 or both.

5-24

The PAL8 Assembler

AND Group

—

If the skips are combined in a statement, the logical AND of the conditions determines

the skip. For example:
SNA SZL

The next statement is skipped only if the accumulator differs from 0000 and the link is 0.
5.10.2.2

Input/Output Transfer Microinstructions

—

Input/output transfer microinstructions initiate operation

of peripheral equipment and effect an information transfer between the central processor and the input/output

device(s).
0

._>{

—————-I

DEVICE CODE SELECTION

CONTROL BITS

IOT Instruction Bit Assignments
5.10.3

Autoindexing

Consecutive address references are often necessary for obtaining data values when processing large amounts of data.

Autoindex registers (locations 10-17 of each memory field) are used for this purpose. When one of the absolute
locations from 10 through 17 (octal) is indirectly addressed, the contents of the location are incremented and then
used as an indirect operand address. This allows consecutive memory locations to be adressed, using a minimum of
instructions. It must be remembered that initially these locations (10 through 17 on page 0 of each field) must be
set to one less than the first desired address. No incrementation takes place when locations 10 to 17 are addressed

directly. For example, if the instruction to be executed next is in location 300 and the data to be referenced is on the
the page starting at location 5000, autoindex register 10 can be used to address the data as follows:
013376

137'?

TAD

(4777)

/:==5000"~1

01-377

3010

[ICA

0300

1410

TAU

10
I 21.0

/SET UF‘ AUTO INDEX
/INCF:EMENT TO 5000

/'BEFC]F:E

/’ faDflF-tESSS

4 3-77 I?

(Literal Area of Page 1)

037 I7

USE:

INDIRECT

Note that the Data Field must be set to the field of the data being referenced, in this case, Field 0.

When the instruction in location 300 is executed, the contents of location 10 will be incremented to 5000 and the
contents of location 5000 will be added to the contents of the accumulator. If the instruction TAD I 10 is

executed again, the contents of location 5001 will be added to the accumulator, and so on.

5.11

PSEUDO-OPERATORS

Pseudo-operators are used to direct the assembler to perform certain processing operations or to interpret subsequent
coding in a certain manner. Some pseudo-ops generate storage words in the object program, other pseudo-ops
direct the assembler on how to proceed with the assembly. The pseudo-ops are defined in the permanent symbol
table.

5.11.1

Indirect and Page Zero Addressing

The pseudo-operators I and Z specify the type of addressing to be performed. These were discussed in Section 5.10.1.

The PAL8 Assembler

5.11.2

Extended Memory

The pseudo-op FIELD instructs the assembler to output a field setting so that it may assemble code into more than
This field setting is output during pass 2 in the object binary file and is recognized by the LOAD
command which in turn causes all subsequent information to be loaded into the field specified by the expression.

one memory field.

The form is
FIELD n
where n is an integer, a previously defined symbol, or an expression whose terms have been defined previously

within the range 0 to 3.
This field setting is output to the binary file during pass 2 followed by an origin setting of 200. These settings are
read by the LOAD command when it is executed to begin loading information into the new field.
The field setting is never remembered by the assembler except as the high-order digit of the Location Counter on the
listing. A binary file produced without field settings will be loaded into field 0 when using the LOAD command.
When FIELD is used, the assembler follows the new FIELD setting with an origin at location 200. For this reason,
to assemble code at location 400 in field

FIELD:

1, it would be necessary to write

/CORRECT EXAMPLE

*400
The following is incorrect and will not generate the desired code:

/ INCORRECT

*400
FIELD:

Specifying the /0 option to PAL8 inhibits the origin to 200 after a FIELD pseudo-op, leaving the origin at its
previous value.
5.11.3

Resetting the Location Counter

The PAGE 11 pseudo-op resets the location counter to the first address of page 11, where n is an integer, a previously

defined symbol, or a symbolic expression, whose terms have been defined previously and whose value is from 0 to 37
inclusive. If n is not specified, the location counter is reset to the beginning of the next page of memory.
For example,

PAGE 2 sets the location counter to 00400
PAGE 6 sets the location counter to 01400
If the pseudo-op is used without an argument and the current location counter is at the first location of a page, the
current location counter will not be reset.

In the following example, the code TAD B is assembled into location

00400:

)R 3 37 3’
.J M F“
F' fit (3 IE
T A III

1'5

If several consecutive PAGE pseudo-ops are given, the first will cause the current location counter to be reset as

specified. The rest of the PAGE pseudo-ops will be ignored.

5-26

The PAL8 Assembler

5.11.4

Reserving Memory

ZBLOCK instructs the assembler to reserve n words of memory containing zeros, starting at the address indicated by
the current location counter. It is of the form
ZBLOCK n

For example,
ZBLOCK 40
causes the

assembler to reserve 40 (octal) words and store zeros in them. The n may be an expression. If n=0, no

locations are reserved. A ZBLOCK statement may have a label.
5.11.5

Relocation Pseudo-Operator

It is sometimes desirable to assemble code at a given location and then move it at run time to another location for
execution. This may result in errors unless the relocated code is assembled in such a way that the assembler assigns

symbols their execution-time addresses rather than their load-time addresses. The RELOC pseudo-0p establishes a
virtual location counter without altering the actual location counter. The line
RELOC expr
sets the virtual location counter to expr.

The line

RELOC
resets the virtual location counter back to the actual location counter, terminating the relocation section.

For example, the following program

0200

#200

00200

1205

TAD

00201

7402
1367

HLT
RELOC

01367*

1371

TAD

01370*
01371*

7402
0000

CUDEv

0205
00205
causes the

0000

CHAR
1367

CODE

HLT
0

RELOC
CHAR,

assembler to load the ZERO word at CODE into location 204, but treats it as if it were loading into location

2000. The asterisks after the location values indicate that the virtual and the actual location counters differ for that
line of code. Only the virtual location counter is listed. RELOC always causes current page literals to be dumped.
5.11.6

Suppressing the Listing

The portions of the source program enclosed by XLIST pseudo-ops will not appear in the listing file; the assembled

binary will be output, however.
Two XLIST pseudo-ops may be used to enclose the code to be suppressed in which case the first XLIST with no

argument will suppress the listing, and the second will allow it again. XLIST may also be used with an expression as
an argument. The listing will be inhibited if the
expression is not equal to zero, or allowed if the expression is equal
to zero.

XLIST pseudo-ops never appear in the assembly listing.

5-27

The PAL8 Assembler

5.11.7

Controlling Page Format

The EJECT pseudo-op causes the listing to skip to the top of the next page. A page eject is done automatically every
55 lines; EJECT is useful if more frequent paging is desired. If this pseudo-op is followed by a string of characters,
the first 40 (decimal) characters of that string will be used as a new title at the top of each page of the listing.
5.11.8

Altering the Permanent Symbol Table

If the system includes one or more optional devices that are to be programmed directly whose instruction set is not
defined in the permanent symbol table (for example, line printer), then the symbol table would have to be altered to
include instructions for these devices.
In another situation, programmed-defined symbols might require more space than available in the symbol table.

Again, the symbol table would have to be altered by removing all definitions not needed in the program being
assembled. For such purposes, PAL8 has three pseudo-ops that can be used to alter the permanent symbol table.
These pseudo-ops are recognized by the assembler only during pass 1. During either pass 2 or pass 3, the assembler
ignores them and they have no effect.
EXPUNGE deletes the entire permanent symbol table, except pseudo-ops.
FIXTAB appends all currently defined symbols to the permanent symbol table. All symbols defined before the
occurrence

of FIXTAB are made part of the permanent symbol table for the current assembly.

To append the following instructions to the symbol table, it is necessary to generate an ASCII file called SYM.PA.
This file contains the following:

CLSK=6131
FIXTAB

/SKIP ON CLOCK INTERRUPT
/SO THAT THIS WON’T BE
/PRINTED IN THE SYMBOL TABLE

The ASCII file is then entered in the PAL8 input designation. The definitions could be placed at the beginning of

the source file, to eliminate the need to load an extra file. Each time the assembler is loaded, the PAL8’s initial

permanent symbol table is restored.
The third pseudo-op used to alter the permanent symbol table in PAL8 is F IXMRI. F IXMRI defines a memory

reference instruction and is of the form:
FIXMRI name=value
The letters FIXMRI must be followed by one space, the symbol for the instruction to be defined, an equal sign, and
the value of the symbol. The symbol will be defined and stored in the symbol table as a memory reference instruction. The pseudo-op must be repeated for each memory reference instruction to be defined. For example,

EXPUNGE
FIXMRI TAD= 1000
FIXMRI DCA=3000

CLA=7200
FIXTAB
When the preceding program segment is read by the assembler during pass 1, all symbol definitions are deleted and
the three symbols listed are added to the permanent symbol table. Notice that CLA is not a memory reference
instruction. This process can be performed to alter the assembler’s symbol table so that it contains only the symbols
used at a given installation or by a given program.

5-28

The PALS Assembler

Conditional Assembly Pseudo-Operators

5.11.9

The IFDEF pseudo—op takes the form

IFDEF symbol <source code>

If the symbol indicated is previously defined, the code contained in the angle brackets is assembled; if the symbol is
undefined, this code is ignored. Any number of statements or lines of code may be contained in the angle brackets.
The format of the IFDEF statement requires a single space before and after the symbol.

For example,
IFDEF A <TAD A
DCA B>
The IFNDEF pseudo-op is similar in form to IFDEF and is expressed as:
IFNDEF symbol <source code>
If the symbol indicated has not been previously defined, the source code in angle brackets is assembled. If the

symbol is defined, the code in the angle brackets is ignored.
The IFZERO pseudo-op is of the form

IFZERO expression <source code>
If the evaluated expression is equal to zero, the code within the angle brackets is assembled; if the expression is

Any number of statements or lines of code may be contained in the angle brackets.
The expression may not contain any embedded spaces and must have a single space preceding and following it.
nonzero, the code is ignored.

IFNZRO is similar in form to the IFZERO pseudo—op and is expressed as
IFNZRO expression <source code>
If the evaluated expression is not equal to zero, the source code within the angle brackets is assembled; if the

expression is equal to zero, this code is ignored.

Pseudo-ops can be nested.

For example,

IFDEF SYM <1FNZRO X2 <...>>
The evaluation and subsequent inclusion or deletion of statements is done by evaluating the outermost pseudo-op
first.

Conditional code that is not assembled can be deleted from the listing output by using the /J option.
5.11.10

Radix Control

Numbers used in a source program are initially considered to be octal numbers. However, the program may change
the radix interpretation by the use of the pseudo-operators DECIMAL and OCTAL. The DECIMAL pseudo-op

interprets all following numbers as decimal until the occurrence of the pseudo-op OCTAL. The OCTAL pseudo-op
resets the radix to octal.

5-29

The PAL8 Assembler

5.11.11

Entering Text Strings

The TEXT pseudo-op allows a string of text characters to be entered as data and stored in 6-bit ASCII. The format

required is the pseudo-0p TEXT followed by one or more spaces or tabs, 3 delimiting character (must be a printing
character), the string of text, and the same delimiting character. If the number of characters in a specified string is
odd, the last word will contain zero in its right half. If the number of characters is even, a final word of zero will be
appended. Either way, a final 6-bit character of zeros is generated providing a convenient “end-of-string” indication.
Note that the / F option prevents the extra word of zero when the number of characters is even. Note also that six
bits is only sufficient to encode the printing characters. For example:
TEXT“123*”

TAG,

The string would be stored as

6162
6352
0000

The / F option inhibits the generation of the extra 6-bit zero word.

Alternatively, the statement “*.-1” may be used

to eliminate the extra zero word (when the number of characters are even).

5.11.12

End-of-File Signal

PAUSE signals the assembler to stop processing the file being read. The current pass is not terminated, and process-

ing continues with the next file.

The PAUSE pseudo-op is present only for compatibility with paper tape assemblers,

and its use is not recommended.

5.11.13

Use of DEVICE and FILENAME Pseudo-Operators

The pseudo-operators DEVICE and FILENAME may be used by calls to the User Service Routine (see Appendix C),
or may be used

for other purposes.

They store 6-bit ASCII strings at the current location. The form for these

pseudo-ops is
DEVICE name
FILENAME nameextension
The name used with DEVICE can be from 1 to 4 alphanumeric characters. These are trimmed to 6-bit ASCII and

packed into two words, filled-in with zeros on the right if necessary. With FILENAME (FILENA is also acceptable),
the name (or name.extension) may be from 1 to 6 alphanumeric characters and the optional extension may be 1 or 2
characters. The characters are trimmed to 6-bit ASCII and packed two to a word. Three words are allocated for the
file name, filled with zeros on the right if fewer than 6 characters are specified, followed by one word for the
extension.

For example,

FILENAME ABC.DA

is equivalent to the following coding:

0102
0300
0000

0401
The symbols DEVICE and FILENAME may not be used as labels since they are predefined pseudo-ops.

5-30

The PAL8 Assembler

5.12

LINK GENERATION AND STORAGE

In addition to handling symbolic addressing on the current page of memory, PAL8 automatically generates “links”

for off-page references. If a direct memory reference is made to an address not on the page where an instruction is
located or on page 0, the assembler sets the indirect bit (bit 3) and an indirect address literal (a “link”) will be stored
on

the current memory page. If the specified reference is already an indirect one, the error diagnostic 11 (Illegal

Indirect) will be generated. In the following example,
#2117

CLA
O

#2600
JMF‘ A
the assembler will recognize that the register labeled A is not on the current page and will generate a link to it as

follows:
1.

In location 2600 the assembler will place the word 5777 (equivalent to JMP I 2777).

In address 2777 (the last available location on the current page) the assembler will place the word
2117 (the actual address of A).

In the listing, the octal code for the instruction will be followed by a single quote (’) to indicate that a link was

generated.

Although the assembler will recognize and generate an indirect address linkage when necessary, the program may
indicate an explicit indirect address by the pseudo-op I. For example,

#2117
A:

OLA

$2600
JMF‘

(A)

The assembler cannot generate a link for an instruction that is already specified as being an indirect reference, since
the computer supports no such instruction format. For the example shown below, the assembler will print the error
message 11 (Illegal Indirect).

$9.117
Av

(ILA

#2600
\JMF'

A
NOTE

The option / E makes link generation a condition that

produces the LG (Link Generated) error message.
The above coding will fail because A is not defined on the page where J MP I A is attempted, and the indirect bit is

already set.

5-31

The PAL8 Assembler

Literals and links are stored on each page starting at page address 177 (relative) and extending toward page address 0
(relative). Whenever the origin is then set to another page, the literal area for the current page is output. There is
room for 160 (octal) literals and links on page zero and 100 (octal) literals on each other page of memory. Literals
and links are stored only as far down as the highest instruction on the page. Further attempts to define literals will
result in a PE (Page Exceeded) or ZE (page Zero Exceeded) error message.
5.13

TERMINATING ASSEMBLY

PAL8 will terminate assembly and return to the Monitor under any of the following conditions:
1.

Normal exit: The end of the source program was reached on pass 2 (or pass 3 if a listing is being

generated).
2.

Fatal error: One of the following error conditions was found and flagged (Section 5.14):
BE

5.14

CTRL/C:

If typed, control returns to the Monitor.

Any partial output files are deleted.

ERROR MESSAGES

PAL8 will detect and flag error conditions and display error messages on the console terminal. The format of the
error message is

COde address
where code is a 2-letter code that specifies the type of error, and address is either the absolute octal address where
the error occurred or the address of the error relative to the last label (if there was one) on the current page. For

example, the code:
BEG,

TAD LBL

%TAD LBL
would produce the error message
IC BEG+0001
since % is an illegal character because of its placement.

In the listing, error messages are output as 2-character messages on the line just prior to the line in which the error
occurred. Table 5-3 lists the PAL8 error codes. Fatal errors cause PAL8 to terminate the assembly immediately

(deleting any output files produced so far) and return to the Monitor.
Table 5—3

PAL8 Error Codes

Error Code
BE

Meaning
A PAL8 internal table has overflowed. This situation can usually be corrected by decreasing

the level of literal nesting or the number of current page literals used prior to this point on
the page. Fatal error:

assembly cannot continue.

Chain to CREF error. CREF.SV was not found on SYS.

Device error. An error was detected when trying to read or write a device. Fatal error:

assembly cannot continue.
Continued on next page

5-32

The PAL8 Assembler

Table 5-3 (Cont)

PAL8 Error Codes

Error Code

Meaning

Device full. Fatal error:

Illegal character. The character is ignored and the assembly is continued.

Illegal redefinition of a symbol. An attempt was made to give a previous symbol a new value
by means other than the equal sign. The symbol is not redefined.

Illegal equals. An attempt was made to equate a variable to an expression containing an

assembly cannot continue.

undefined term. The variable remains undefined.
II

Illegal indirect. An off-page reference was made; a link could not be generated because the
indirect bit was already specified.

Illegal pseudo-op. A pseudo-op was used in the wrong context or with incorrect syntax.

Illegal page zero reference. The pseudo-op 2 was found in an instruction which did not refer
to page zero. The Z is ignored.

The /L or /G options have been specified but the Absolute Loader system program is not

present.
LG

Link generated. This code is displayed only if the / E option was specified to PAL8.

Current nonzero page exceeded. An attempt was made to
l.

Override a literal with an instruction.

Override an instruction with a literal.

Use more literals than the assembler allows on that page.

This can be corrected by decreasing either the number of literals on the page or the number
of instructions on the page.
PH

A conditional assembly bracket is still in effect at the end of the input stream. This is caused

by nonmatching < and > characters in the source file.
RD

Redefinition. A permanent symbol has been defined with =. The new and old definitions do
The redefinition is allowed.

not match.

Symbol table exceeded. Too many symbols have been defined for the amount of memory
assembly cannot continue.

available. Fatal error:
UO

Undefined origin. An undefined symbol has occurred in an origin statement.

Undefined symbol. A symbol has been processed during pass 2 that was not defined before
the end of pass 1.

Page 0 exceeded. This is the same as PE except occurs in page 0.

5-33

The PAL8 Assembler

PAL8 PERMANENT SYMBOL TABLE

5.15

The following mnemonics represent the central processor’s instruction set found in the permanent symbol table

within the PALS Assembler. For additional information on these instructions, refer to the DECstation 78 User’s
Guide.

Mnemonic

Code

Operation

Memory Reference Instructions
AND

0000

TAD

1000

ISZ

2000

DCA

3000

J MS

4000

JMP

5000

[OT

6000

OPR

7000

Logical AND
Two’s complement add
Increment and skip if zero
Deposit and clear AC
Jump to subroutine
Jump
In/Out transfer
Operate

Group 1 Operate Microinstructions
NOP

7000

N0 operation

IAC

7001

Increment AC

BSW

7002

Byte swap

RAL

7004

Rotate AC and Link left one

RTL

7006

Rotate AC and Link left two

RAR

7010

Rotate AC and Link right one

RTR

7012

Rotate AC and Link right two

CML

7020

CMA

7040

Complement the link
Complement the AC

CLL

7100

Clear Link

CLA

7200

Clear AC

Group 2 Operate Microinstructions (1 cycle)
HLT

7482

Halts the computer

SKP

7410

SNL

7420

Skip unconditionally
Skip on nonzero Link
Skip on zero Link
Skip on zero AC
Skip on nonzero AC
Skip on minus AC
Skip on positive AC (zero is positive)

SZL

7430

SZA

7440

SNA

7450

SMA

7500

SPA

7510

Group 3 Operate Microinstructions
MQA
MQL

7501

M0 OR into AC

7421

SWP

7521

Load MQ, clear AC
Swap AC and MO

Combined Operate Microinstructions
CIA

7041

STL

7120

Set Link to l

GLK

7204

Get Link (put Link in AC, bit 11)

STA

7240

Set AC to -1 (7777)

Complement and increment AC

5-34

The PAL8 Assembler

Mnemonic

Code

Operation

Internal IOT Microinstructions
SKON

6000

Skip with interrupts on and turn them off

ION

6001

Turn interrupt facility on

IOF

6002

Turn interrupt facility off

GTF

6004

Get flags

RTF

6005

Restore flag, ION

CAF

6007

Clear all flags

Memory Extension IOT Instructions
CDF

62N1

CIF

62N2

Change the Data Field to N
Change Instruction Field to N at the next JMP 0r JMS instruction

CDF CIF

62N3

Combined CDF and CIF

RDF

6214

Read (OR) the Data Field into bits 6-8 of AC

RIF

6224

Read (OR) the Instruction Field into bits 68 of AC

RSB

6234

Read Instruction Save Field into bits 7-8 and Data Save Field into
bits 10-11 of AC

RMF

6244

Restore memory fields to state prior to last interrupt by loading the
Data Save Field into DF register and the Instruction Save Field
into the IB register and inhibiting interrupts. At the next JMP or JMS,
IB is loaded into the IF register and the interrupt inhibit is removed.

Keyboard (1 cycle)
KCF

6030

KSF

6031

Skip on keyboard flag

KCC

6032

Clear keyboard flag and AC

KRS

6034

OR keyboard buffer into AC

KIE

6035

KRB

6036

Set/clear interrupt enable
Clear AC, read keyboard buffer
Clear keyboard flag

Clear keyboard flag

Terminal (1 cycle)
TSF

6041

Skip on terminal flag

TCF

6042

Clear terminal flag

TSK

6045

TLS

6046

Skip on keyboard or terminal flag
Load terminal, display character, and clear terminal flag

5-35

CHAPTER 6
BASIC

6.1

INTRODUCTION

OS/78 BASIC* is an interactive programming language used in scientific and business environments to solve mathematical problems with a minimum of programming effort. It also is used by educators and students as a problemsolving tool and as an aid to learning through programmed instruction and simulation.
In many respects the BASIC language is similar to other programming languages (such as FORTRAN IV), but BASIC
is aimed at facilitating communication between the user and the computer. BASIC simply requires that you type in
the computational procedure as a series of numbered statements, making use of common English words and familiar
mathematical notations. Because of the small number of commands necessary and its easy application in solving

problems, BASIC is an easy computer language to learn. With experience, the advanced techniques available can be
added in the language to perform more intricate manipulations or to express a problem more efficiently and
concisely.
6.2

MAJOR COMPONENTS OF 08/78 BASIC

The BASIC subsystem has four major components.
1.

BASIC Editor (BASIC.SV)

Compiler (BCOMPSV)

Loader (BLOAD.SV)

BASIC Run Time System (BRTSSV, BASIC .AF, BASIC .SF

The BASIC editor is used to create and edit program source files.

BASIC.FF)

During this process, the editor creates a file called

BASIC.WS containing the current program.
Once the program has been prepared for execution, entering a RUN command causes the editor to chain to the

BASIC compiler. The compiler converts the statements in BASIC.WS into relocatable binary instructions.

Following compilation, the BASIC loader is automatically requested. The loader converts the relocatable binary
data output by the compiler into executable form and loads the result into memory.
The BASIC loader then chains to the BASIC Run Time System (BRTS) which executes the program. The modules

BASIC.AF, BASIC.FF, and BASIC.SF are overlays to BRTSSV.
6.3

BASIC INSTRUCTION REPERTOIRE

BASIC instructions and commands can be grouped in three categories as follows:
1.

BASIC Editor commands that allow you to
Create or modify a program,

Execute a program,
Retrieve a program from diskette, and
Save a program.

*BASIC is a registered trademark of the trustees of Dartmouth College.

6-1

BASIC

BASIC statements, comprising the BASIC language, that are the building blocks used to create and structure BASIC programs.

BASIC Functions, represented by subroutines, that are built into BASIC primarily to facilitate problem
solving activities.

CALLING BASIC

6.4

To enter the BASIC subsystem, type

LBQSIC
in response to the Monitor dot. This command invokes the BASIC editor.

BASIC EDITOR COMMANDS

6.5

Using the BASIC Editor

6.5.1

The BASIC editor incorporates all the tools and capabilities necessary to create, correct, modify, execute, save and
retrieve BASIC programs. After calling BASIC, the BASIC editor responds with the displayed query

Editing is now continued in one of two ways:
1.

Typing NEW with a file name instructs the system to initiate the creation of a new file.
Typing OLD with a file name instructs the system to retrieve an old file containing a previously-generated
program.

Figure 6-1 summarizes user actions implemented by the BASIC editor. This figure shows two arbitrarily selected
file names (MAKER/ALTER). File names may contain no more than six alphanumeric symbols.
The left side of Figure 6-1 shows the types and patterns of activities that you ordinarily pursue after typing the NEW
command. The right side of the illustration shows activities frequently carried out after typing the OLD command.
Note that most of the BASIC editor commands appear on both sides of the figure.
are

Only the sequence in which they

used differs.
BASIC Editor Commands

6.5.2

This section summarizes the BASIC editor commands. Letters that are not required in naming the command are
shown as lower case in the Command/Parameter description. For example, only NE is required to be recognized by

the BASIC editor as the NEW command.

NOTE
The RETURN key must be pressed following each BASIC
editor command.

6.5.2.1

NEW Command

—

The NEW command clears the memory workspace and specifies the name of the pro-

gram that is to be input.

Command

Parameters

NEw

file [.ex]

file .ex

is the new file name and extension of the program about to be typed in. If the extension
is omitted, the editor assigns .BA.

6-2

BASIC

BASIC
N

/
NEW

Miscellaneous BASIC

E’YII OR \6LD

Editor Commands

CTR L/C

FILE NAME
,.

SCRATCH

OIL-D

/
V

BYE

FILE

MAKER

NAI‘ME

ALTER
R A Y

R EADY
CREATE

EXECUTE

Your Program

Your Program Using

Using BASIC

STATEMENTS

FUNCTIONS

REVIEW
Your Program Using
us]-

8”“
RUNNH

‘

LISTNH
Gratis,
CT‘RLIQ

MODIFY
Your

Program
0.

Using

MODIFY
Your Program Using

Cor

Keys

rection{

Keys

g::f::
RETURN

}

Correction

Keys

R EV I EW

388250-

Your Program Using

Add BASIC Statements

PRESERVE
Your Program Using

LISTS”!

CTRL/S

crisis/o}

Key

SAVE

PRESERVE

EXECUTE

Your Program Using

Your Program USIHQ

L ERROR

SAVE

MESSAGE

EXECUTE
Your Program Using
i'
——>

ERROR MESSAGE

——

End of BASIC Session

1. *Keyboard Monitor Command
2.

Shaded areas indicate user
action at the console

Figure 6-1

BASIC Editor Commands and Related Uses

{

mm‘i
“”8"“

BASIC

An alternate method is to type NEW without a file name, followed by the RETURN key. BASIC displays

FILE

NAME

—w

in response to which the file name and extension is typed.

For example, to clear the workspace and name the new program “TESTBA”, type

NEH

TEST

TEST 0 BA

01'

6.5.2.2

OLD Command

—

The OLD command clears the memory workspace, and causes the editor to find a

program on a diskette and place it into the workspace.
Command

Parameters

OLd

dev '. file [ .ex]

devzfileex

is the device, file name, and extension of the program on the disk. If the extension

is omitted, BASIC assumes “.BA”.

Another method is to type OLD without a file name. BASIC displays

F I LE

NAME

--~

in response to which the device, file name, and extension is typed. When no device is specified, the BASIC editor

defaults to DSK (usually

SYS).

For example, to bring TESTBA into the workspace from RXAl type
,

OLD

RXAI 3 TEST 0 BA

R'Xfil 3TEST

The LIST command displays the current program along with a header line,
LIST/LISTNH Commands
the
and
the
revision number of BASIC. The date is displayed only if the current
containing
program name, date,

6.5.2.3

—

date has been entered into the system.
Command

Parameters

LIst

[n]

If n is omitted all program statements in the workspace are displayed. When n is specified, line 11 and all subsequent

lines are displayed. Type CTRL/O to terminate a listing.

BASIC

For example, typing LIST (or L1) displays the program PROGBA:

L I ST

F'RDG

26~JUL~~77

fizz],

FOR

[€er

READY

Typing L1 30 displays line 30 and all subsequent lines:
L I.

F'RUG

END

26~JUL~77

REAIIY
The LISTNH command also displays the program statements in the workspace but without the header.

Command

Parameters

LISTNH

[n]

has the same effects as specified for the LIST command, but does not display the header.

The SAVE command writes the program in the workspace onto the diskette as a
permanently saved file. Do not confuse this command with the Monitor SAVE command.

6.5.2.4

SAVE Command

Command

Parameters

SAve

[devzfileex]

dev

is the device on which you want to store your program.

file .ex

is the file name and extension that the program will have on the diskette. If both are left

out, BASIC will use the current file name and extension of the program in the workspace.
If only the extension is omitted BASIC assigns .BA. If DEV: is omitted, BASIC assumes the
device is DSK.
In the following example, the program “TESTBA” is in the workspace. To store it on RXAl under the same file
name

and extension, type

{SAVE

VOW-11 3 TEST . rm

RXf—‘iil. :TEIST

6-5

BASIC
6.5.2.5

RUN/RUNNH Commands

—

The RUN command executes the program in the workspace. Note that

this command differs from the monitor RUN command.
Command

Parameters

RUn

(none)

Prior to program execution, this command displays a heading consisting of the file name and extension, BASIC
version number and system date, assuming that the current date has been entered in the system.

When program

execution is completed, BASIC displays

READY
The following example shows the program PROG.BA in the workspace displaying the result of a calculation:

RUN
F'ROG

26-~JAN--76

5%)

3 o 179
The RUNNH command executes the program in the workspace, but does not display the header line.

Command

Parameters

RUNNH

(none)

Thus, the only difference between the RUNNH and RUN commands is that RUN prints the header and RUNNH
does not display the header.
6.5.2.6

NAME Command

The NAME command allows the user to rename the program in the workspace.

Command

Parameters

NAme

newfil [ .ex]

newfilex

is the new file name and extension of the program in the workspace. If the extension
is omitted, the editor assigns .BA.

To change the name of the program in the workspace to PROG.BA, type

NAME

F'ROG 0 BA

NA
6.5.2.7

F'RDG

SCRATCH Command

—

The SCRATCH command erases all statements from the workspace, that is,

it clears the workspace.

Command

Parameters

SCratch

(none)

6.5.2.8

BYE Command

—

The BYE command exits from BASIC and returns control to the Monitor from

BASIC.

6-6

BASIC

Command

Parameters

BYE

(none)

Typing BYE before SAVEing a newly created program will delete the program. Whenever the BASIC editor is
waiting for user input, typing CTRL/C performs the same function as BYE.
BASIC Control Keys

6.5.3

This section describes the control keys that are used to correct errors, eliminate and substitute program links, and

control program listings.
Errors made while typing programs at the
Correcting Typing and Format Errors (DELETE, CTRL/U)
deletion
of
the last character typed. One character is
terminal are easily corrected. Pressing the DELETE key causes
deleted each time the key is pressed.

6.5.3.1

—

Sometimes it is easier to delete a line being typed and retype the line rather than attempt a correction using a series

Typing CTRL/ U will delete the entire line currently being worked on and echoe “DELETED” and a
carriage return-line feed. Use of the CTRL/ U key is equivalent to typing DELETEs back to the beginning of the line.

of DELETEs.

To delete a program line that has already been
Eliminating and Substituting Program Lines (RETURN)
entered into the computer, simply type the line number and then press the RETURN key. Both the line number

6.5.3.2

—

and the statement(s) are removed from the program.

Change individual lines by simply retyping them in again. Whenever a line is entered, it replaces any existing line
having the same line number. New lines may be inserted anywhere in the program by giving them unique line
numbers.

Interrupting Program Execution
Program execution may be terminated by typing CTRL/C. BASIC
the
READY
responds by displaying
message allowing you to correct or add statements to the program.

6.5.3.3

NOTE
BASIC responds to CTRL/C with a “READY” message

only if you have already given the RUN command.
Typing CTRL/C when the BASIC Editor is operational
causes a return to

the Monitor.

6.5.3.4

For programs
Controlling Program Listings at the Terminal Console (CTRL/S, CTRL/C and CTRL/O)
a
frame
the
user
to
wish
the
effect
that
occurs
after
exceeding single display
(24 lines)
may
stop
scrolling
typing the
~

LIST/LISTNH command. Three sets of control keys are provided to do this. They are as follows:
1.

CTRL/S keys. Simultaneously pressing these keys suspends listing (scrolling) of the program. However
it leaves the program in a state where you may resume listing.
CTRL/ Q keys. Simultaneously pressing these keys (after having suspended scrolling with CTRL/S),
resumes the listing process.
CTRL/O keys. Simultaneously pressing these keys aborts the listing process and causes the BASIC editor
to display READY.

6.5.4

Resequencing Programs (RESEQ)

If a program is extensively modified, you may find that some portions of the program have line numbers

spaced so
closely together that they do not permit any further addition of statements. Renumbering the lines in the program
to provide a practical increment between line numbers can be done by
using the RESEQ program. Note that the
RESEQ program modifies the line numbers in GOSUB and IF-THEN statements to agree with the new line numbers
assigned to program statements by RESEQ. Line lengths must not exceed 80 characters and programs may not
exceed 350 lines.

6-7

BASIC

Typically, the program would be used as follows:
SAVE DSK:SAMPLE BA

User saves program SAMPLE which requires renumbering.

READY

BASIC is ready for next command.

OLD DSKzRESEQ

User calls for program RESEQ.

READY

BASIC is ready for next command.

RUNNH

User runs RESEQ program.

FILE:DSK:SAMPLE.BA

Program asks for filename. User responds with device, name, and extension of
program to be renumbered.

START,STEP: 100,10

Program asks for a starting line number (START) and for the increment
between line numbers (STEP). User requests that SAMPLE start with line
number 100 and each line be incremented by 10.

6.6

READY

Renumbering is accomplished. BASIC ready for next command.

OLD DSK:SAMPLE.BA

User calls back his program.

READY

BASIC ready for next command.

LISTNH

User gets listing of program SAMPLE for further modification.

DATA FORMATS ACCEPTABLE T0 BASIC

6.6.1

Numeric Information

6.6.2

Numbers

Numbers are expressed in decimal or E (exponential) format.
Decimal

E Type

10.23E27

6.21E+27

+69

—7.232E6

~52

2.211E—3

-3.9265

Examples of numbers in both categories are as follows:

—2.2114E—4

0.123
“0.769
In the decimal format, the decimal point is optional for integers. That is, BASIC assumes a decimal point after the
rightmost digit of an integer. In E type format, BASIC assumes a positive exponent when no plus (+) or minus (-)
’

sign follows the E. Substitute the words “times ten to the power of’ for the letter E when reading E type format
numbers.

6-8

BASIC

Numeric data may be input in either format. Results of computations with an absolute value outside the range
+.000001<N<999999 are always output in E type format. BASIC handles six significant digits as shown by the

following examples:
Value Output by BASIC

Value Typed In
.01

0.0099999

.0099

0.0099

999999

100000

.100000E+007

.0000009

.899999E— 006

Note in the above examples that the nature of the binary numbering system does not permit a completely accurate

representation of certain decimal numbers. Hence they are output as close to the true value as the internal logic of
the computer permits.
BASIC automatically suppresses the printing of leading and trailing zeros in integer numbers and all but one leading
zero in decimal numbers.

As can be seen from the preceding examples, BASIC formats all exponential numbers in

the form:

sign .xxxxxxE(+or~ )n
where x represents the number carried to six decimal places, E stands for “times 10 to the power of”, and n represents the exponential value.

For example,
-

.347021E+009 is equal to

347,021,000, and

.726000E— 003 is equal to 0.000726
All numbers used in BASIC must have an absolute value (N) in the range:

10‘-616<N<10‘+6l6
6.6.3

Simple Variables

A simple variable in BASIC is an algebraic symbol representing a number, and is formed by a single letter or a letter

followed by a digit. For example,

Acceptable Variables

Unacceptable Variables

2C (a digit cannot begin a variable)

AB (two or more letters cannot form a variable)

Values may be assigned to variables either by indicating the values in LET statements, or by
data via INPUT and DATA statements.

inputting the values as

BASIC

Examples:
10 LET 1:53721
20 LET B3=456.9

30 LET X=2OE9
40 INPUT Q
50 DATA 5,6,7
6.6.4

Subscripted Variables

In addition to simple variables, BASIC accepts another class of variables called subscripted variables.

Subscripted

variables provide additional computing capabilities for handling arrays, lists, tables, matrices, or any set of related
variables. Variables are allowed one or two subscripts. A single letter or a letter followed by a digit forms the name
of the variable. The subscript is formed by one or two integers enclosed in parentheses and separated by commas.

Up to 31 arrays are possible in any program, subject only to the amount of memory available for data storage. For
example, an array might be described as A(J) where J goes from 1 to 5 as follows:
,

A(1),A(2),A(3),A(4),A(5)
This allows reference to be made to each of the five elements in the array A. A two-dimensional array A(J,K) can be
defined in a similar manner, but the subscripted variable A must always have the same number of subscripts (that is,
AU) and A(J,K) cannot be used in the same program). It is possible, however, to use the same variable name as both
a

subscripted and an unsubscripted variable. Both A and A(J) are valid variable names and can be used in the same

program. For more information on arrays and the use of subscripts, see the description of the DIM statement

(Section 6.7.10.1).
Arithmetic Operations

6.6.5

BASIC performs addition, subtraction, multiplication, division and exponentiation, as well as
Operators
more complicated operations (explained in detail later in the manual). The five
operators used in writing most
6.6.5.1

—

formulas are:

Symbol
Operator

Meaning

Example

Addition

A+B

Subtraction

A- B

Multiplication

A*B

Division

A/B

A(or**)

Exponentiation
(Raise A to the B Power)

A*B or (A**B)

6.6.5.2

Priority

—

In any given mathematical formula, BASIC

ing order:
1

performs the arithmetic operations in the follow—

Parentheses receive top priority. Any expression within
parentheses is evaluated before an unparenthesized expression.

In the absence of parentheses, the order or
a.

b.
c.

priority is:

Exponentiation
Multiplication and Division (of equal priority)
Addition and Subtraction (of equal
priority)

If either sequence 1 or 2 above does not

clearly designate the order of priority, then the evaluation of

expressions proceeds from left to right.

6-10

BASIC

The expression A+B-C is evaluated from left to right as follows:
1.

A+B

step 1

(result of step 1)-C

answer

The expression A/B*C is also evaluated from left to right since multiplication and division are of equal priority:
1.

A/B

step 1

(result of step l)*C

answer

Parentheses

6.6.5.3

—

Parentheses may be used to change the order or priority because expressions within paren—

Thus, by enclosing expressions appropriately, the order of evaluation can be

thesis are always evaluated first.

controlled. Parentheses may be nested, that is, enclosed by one or more sets of parentheses. In this case, the ex-

pression within the innermost parentheses is evaluated first, and then the next innermost, and so on, until all have
been evaluated. Consider the following example:

A=7*((BA2+4)/X)
The order of priority is:
1

B‘2

step 1

(result of step 1)+4

step 2

(result of step 2)/X

step 3

(result of step 3)*7

answer

Parentheses also prevent any confusion or doubt as to how the expression is evaluated. For example,

A*B‘2/7+B/C+D‘2
((A*B‘2)/7)+((B/C)+D‘2)
Both of these formulas will be executed in the same way.
stand.

However, the second example may be easier to under-

Spaces may also be used to increase readability. Since the BASIC compiler ignores spaces, the two

statements:

10 LET B=DA2+1

lOLETB=DA2+l
are

identical, but spaces in the first statement provide ease in reading.

6.6.5.4

Rules for Exponentiation

The following rules apply in evaluating the expression AAB.

If B=0, then AAB=1

3AO=1

If A=0 and B>O, then AAB=O

If A=0 and B<0, then AAB=0 and a DV error message is displayed

OA2=0
012:0

If B is an integer >9, then

3‘5=3*3*3*3*3=243

If B is an integer <0 then

If B is a decimal (non-integer) and A>O, then AAB=EXP(B*LOG(A))

2‘3.6=eBl“A=e3'61’12

If B is a positive or negative decimal (non-integer) and A>0, the

—3‘2 .6 is illegal.

AAB=A1*A2*A3. .*An, where n=B
A“B=l/A1*A2*A3. .*An), where n=B
.

program halts and an EM error is displayed.

6-11

3“—5=l/243

BASIC

6.6.5.5

Relational Operators

A program may require that two values be compared at some point to discover

their relation to one another. To accomplish this, BASIC makes use of the following relational operators:
=

equal

less than
< or <

>
=

less than or equal to

greater than
> or >

>< or <>

greater than or equal to
not equal to

Depending upon the result of the comparison, control of program execution may be directed to another part of the
program. Relational operators are used in conjunction with the IF-THEN statement.
The meaning of the (2') sign should be clarified. In algebraic notation, the formula X=X+l is meaningless. However, in BASIC (and most computer languages), the equal sign designates replacement rather than equality.

Thus,

this formula is actually translated “add one to the current value of X and store the new result back in the same
variable X”. Whatever value has previously been assigned to X will be combined with the value 1. An expression

such as A=B+C instructs the computer to add the values of B and C and store the result in a third variable A. The

variable A is not being evaluated in terms of any previously assigned value, but only in terms of B and C. Therefore,
if A has been assigned any value prior to its use in this statement, the old value is lost; it is replaced instead by the
value of B+C.

String Information

6.6.6

The previous sections dealt only with numerical information.

However, BASIC also processes alphanumerical
information called strings. A string is a sequence of characters, each of which is a letter, a digit, a space, or some
character other than a statement terminator (backslash or carriage return).
6.6.6.1

String Character Set

The character set recognized by BASIC is as shown below. The decimal code for

each character is also shown.

Code

Number

Number
18 R

0 GI}
l A

19 S

2 B

20 T

3 C

21 U

4 D

22 V

5 E

23 W

6 F

24 X

7 G

25 Y

8 H

26 Z

9 I

10 J

28 \

16 P

[ (left bracket)
(back slash)
29 ] (right bracket)
30
(exponent sign)
31
(underscore)
32
(space)
33 Y (exclamation point)
34
(double quotes)

17 Q

35 #

11 K

“

12 L

13 M

14 N
15 O

”

6-12

BASIC

Code

Number

Number
50 2

36 $
37 %

51 3

38 & (ampersand)

52 4

’

53 5

(apostrophe)

40.(
41 )

54 6

42 *

56 8

43 +

57 9

(comma)
45
(hyphen or minus sign)
46
(period)
47 / (slash or division sign)

58 :

55 7

59 ;

—

60 < (left-angle bracket)

61 =

48 0

62 > (right-angle bracket)

49 l

63 7

6.6.6.2

String Conventions

—

Strings may be used as constants or variables. String constants are enclosed in

quotes. For example,
“THIS IS A STRING CONSTANT”
A string variable consists of a letter followed by a dollar sign ($) or a letter and a single digit followed by a dollar

sign. A93 and A1$ are both legitimate string variable names while 2A$ and AA$ are not.
A string variable may contain at most eight Characters unless it has been dimensioned with the DIM statement.

Quotation marks may be included in strings by indicating two quotation marks in succession. For example, the

string A”B would appear in a program as:
10 LET A$=“A”“B”

The following lines:

20
9‘?

LET A$-='
PRINT #5
END

'GUOTE'

will result in this display:

'GUOTE'
It is important to recognize the real, structural difference between strings and numerical data. This number
2

is not identical to this string:

‘62))

Numerical data may not be used where strings are required, and vice-versa.

6-13

BASIC

String Concatenation
Strings can be concatenated, that is, connected like links in a chain, by using the
For
a
result
as
of the following lines, the next statement executed will be line 460:
ampersand (&).
example,

6.6.6.3

400

LET

A135

'AB'

410

420

LET “Bit!"
LET (31$

“(SD'
'ABCD'

430

CI$3A1$881$

GOTD

460

The ampersand can be used to concatenate string expressions wherever a string expression is legal, with the ex-

ception that information cannot be stored by means of a LET statement in concatenated string variables. That is,
concatenated string variables cannot appear to the left of the equal sign in a LET statement. For example, this
statement is legal
LET A$=B$&C$
while this statement is not:

LET A$&B$=C$
Format Control Characters

6.6.7

In 05/78 BASIC, a terminal line is formatted into five fixed zones (called print zones) of 14 columns each. A program such as:

bur-J

LET

A:2.3

LET

B321

LET

C=15é.75

m .4;

D: 1 134

LET

E123 4

!-"{.0='i F'RINT AvaCrnvF.
H5

END

where the control character comma (,) is used to separate the variables in the PRINT statement, will cause the values

of the variables to be displayed, using all five zones. For example,

RUNNH
12.3

156.75

1.134

N {A

.2)

READY
It is not necessary to use the standard five zone format for output. The control character semicolon (;) causes the
text or data to be output immediately after the last character printed.

The following example program illustrates the use of the control characters in PRINT statements.

READ

15
20

PRINT
PRINT

UATA

4,596

DATA

16,25736

AyBrC9H7E9F
AvByCyDvErF

A;B3C§D9E§F

6-14

BASIC

RUNNH

£2

3.3
4

READY
As this example illustrates, when more than five variables are listed in the PRINT statement, 08/78 BASIC auto-

matically moves the sixth number to the beginning of the next line.
6.6.8

Files

Files are referenced symbolically by a name of up to six alphanumeric characters followed, optionally, by a period
and an extension of two alphanumeric characters. The extension to a file name is generally used as an aid for

remembering the format of a file.
A fixed length file is one which is already in existence. That is, it has been created and CLOSEd. The length of a

fixed length file is equal to the number of blocks in the file and cannot be changed.
A variable length file is a newly created file. Until the file is CLOSEd, it is equal in length to the largest free space
on the

device. When the file is CLOSEd it becomes a fixed length file equal in length to the actual number of blocks

it occupies. Unless the file is CLOSEd, the CHAIN, STOP or END statements will cause a loss of the file.

6.7

BASIC STATEMENTS

BASIC statements are the principal components of BASIC programs. The general format of BASIC statements
is:

Word Type Parameters

xxx

where
xxx is

the line number;

Word Type is the statement (instruction) type; and

Parameters are the variables used in conjunction with the statement type.
Each statement starts with a line number followed by the word statement type.

Spaces have no significance in

BASIC language statements except in messages or literal strings which are displayed or printed out. Thus, spaces
may. but need not, be used to modify a program and make it more readable.

Multiple statements may be placed on a single line by separating each statement from the preceding statement with
a backslash. For
example:
10 LET A=5\LETB=.2\LETC=3\PRINT “ENTER DATA”
All of the statements in line 10 will be executed before BASIC continues to the next line.
ber at the beginning of th entire line is necessary.

Only one statement num-

However, it should be remembered that program control cannot

be transferred to a statement within a line, only to the first statement of the line in which it is contained. This
consideration is important when using control statements or loop statements (see related descriptions).
6.7.1

Statement Line Numbers (Sequencing)

Failure to assign a line number results in the message:
WHAT

6-15

BASIC

Each line of the program begins with a line number of l to 5 digits that serves to identify the line as a statement.
The largest allowable line number is 99999.
A common programming practice is to number lines by fives or tens, so that additional lines may be inserted in a
program without the necessity of renumbering lines already present.

Renumbering a program can be accomplished

by using the RESEQ program (Section 6.5.4).
6.7.2

The PRINT Statement

The PRINT statement is used to perform calculations and display results. It is also used to display alphanumeric

(string) messages.
PRINT Statement Form

Line Number

Statement

Parameter

xxx

expressions

Where expressions may be numbers, variables. strings or arithmetic expressions, separated by commas or semicolons.
When used without an expression, a blank line will be output on the terminal.
In BASIC, a line is formatted into five fixed zones (called print zones) of 14 columns each. If the expressions in a

PRINT statement are separated by commas, each will begin at a 14-column interval. That is, the first expression
will be displayed starting at the first position, the second at the fifteenth position, and so forth. If more than five

variables are involved, the display will automatically continue at the beginning of the next line.
If this format is not desired. you may separate expressions with the semicolon (;), causing the text or data to be out-

put immediately after the last character printed (see Example 3).
If the last expression in a PRINT statement is followed by a comma or semicolon, the next display called for by the
program will begin on the same line (if there is room). This also applies to the question mark displayed by the
INPUT statement (see Examples 2 and 4).

Any algebraic expression in a PRINT statement will be evaluated using the current value of the variables.

Regardless of format (integer, decimal, or E-type), BASIC prints numbers in the form:
sign number space
where the sign is either minus (7) or blank (for plus) and a blank space always follows the number (see Example 3).

Strings and numeric expressions may be combined in a single PRINT statement (see Example 3).
To output an alphanumeric message, enclose the expression in quotation marks. To print a quotation mark (”)

put two quotation marks (“”) in the expression (see Example 5).
The following examples illustrate the use of the PRINT statement.

6-16

BA SIC

Example 1 :
The following lines

LET

fi==2l

550
595

LET
LET

EH22
0:23

9‘?

END

firByC

will cause each variable to begin at a l4column interval as follows:

.2:

:
Example 2:
The following lines

110
1220

LET

A$=3"F“RINTING"

LET

[Mi-'2:

130

9‘15;

140
19‘?

PRINT
ENII

18$

"STRINGS'

will display the following:

PRINTING

Example 3:

STRINGS

The following lines:

ASE

20
99

PRINT
ENE!

'NUMBER“ 99; 'ANU' 36

will cause this display:

NUMBER

AND

Blanks appear before the 5 and 6 because they are positive‘ The blank following the 5 accounts for the blank that

BASIC always generates after a number.

Example 4:
The following lines

“NUMBER

INF'UT

END

YEARS";

will display (the number 9 is typed in by the user) the following:

NUMBER

YEARS

”a"?

6-17

BASIC

Example 5 :
The following lines

PRINT 'MESSAGE'
PRINT 'A"?'
100 PRINT "'QUOTE"'

80
90

199

END

will cause the following display:

AGE

TtTU‘OTE

Information Entry Statements (DATA, READ)

6.7.3

The DATA and READ statements are always used together. The DATA statement is used to set up a list of numeric
or

string values. These values are accessed by the READ statement which assigns those values to variables in a

program.

6.7.3.1

DATA Statement Format

The DATA statement sets up a list of values to be used by the READ

statement.

DATA Statement Form
Line Number

Statement

Parameters

xxx

DATA

Values

where values are numeric and/or string entries separated by commas. The DATA statement serves as a source of

input variables for the program. The variables are accessed for processing by the READ statement.
There may be any number of DATA statements in a program. They are not executed and may be placed anywhere
within the program. However, BASIC treats them all together as a single list.
Both string data and numeric data may be intermixed in a single DATA statement.

String data in 3 DATA list must always be enclosed by quotation marks.
For example, the three DATA statements

30
120
240

are

Dt-“olTA
DATA

DATA

FIRST
2 9
'5 7

SECOND
"
THIRD v

equivalent to the following statement:
13340

6.7.3.2

DATA

FIRST

“

READ Statement Format

13 7
v

SECOND

3 v

THIRD

The READ statement accesses variables defined by the DATA statement and

assigns those values to variables in a program.

6-18

BASIC

READ Statement Form
Line Number

Statement

Parameters

xxx

READ

variables

where variables are names corresponding to values contained in DATA statements. The following examples illustrate
the use of the READ statement.

Variable names must be separated by commas.
A READ statement may contain numeric and string variable names intermixed, to correspond to numeric and alpha—
numeric values in the DATA list.

However, since values are taken from the DATA list in sequential order, you must

insure that the values in the list are in the correct sequence to correspond to the variable names of the same format

(numeric or string).
A READ statement may have more or fewer variables than there are values in any one DATA statement. The

READ statement causes BASIC to search all available DATA statements in the order of their line numbers until
values are found for each variable in the READ. A second READ statement will begin reading values where the first
one

stopped.

The READ statement is always used in combination with the DATA statement.

Example 1:
The following lines

READ
DATA

QvaC
1’293

will set variable A equal to l, B equal to 2, and C equal to 3.

Example 2:
These statements

550

READ

Crfl$rf£yF$(l"§)

(SO

IN-‘ITfi

lily

"AAA“vlflr

"WORD"

will set:

C=5

D$=“AAA”
E=l2

F=$=“WORD”

Example 3'.
The following program will display the first and third variables in 3 DATA list:

'30

.60

REZfiII (Hi ! Bfli v [3‘13
lil'l'wl'INT (W: to C55

DATA

9"?

EINII

“DIES“ v 'XXX" 3: "F‘LAY"

6-19

BASIC

The screen will show:

[I I 8 F‘ L A Y

Example 4:
The following program uses DATA statements to supply both a variable number of scores and variable score values
to an average calculation routine.

100
110

PRINT
PRINT

120

125

READ

127
130

FOR

'NUMBER‘
'OF SCORES'y'AUERAGE'
N1
T0

1x1
READ N

GOTO

135

140
150

LET
FOR

Nxo
5:0
K31

160

PEAn

170
180

LET Sws+T
NEXT K

190

200

890

NEXT
nATA

900

nATA

3,82,88,97

910
920

nATA
nATA

5,66,78,71’82975
4,82,86,100’91

930

DATA

4972982973782

940

DATA

6961;73’67980984979

999

END

999
N

NyS/N
I

RUNNH

NUMBER
OF

SCORES

AVERAGE

8‘?
74.4

Cbbifibi

89.75
77.25
74

REQDY
6.7.4

LET Statement

The LET statement assigns the value of an algebraic expression to a variable. Use of the word LET is optional.

LET Statement Form
Line Number

Statement

xxx

LET

Parameters
v =

expression

6-20

BASIC

where
v is a

variable; and

expression is a number, another variable, a string (enclosed in quotes), or an arithmetic expression.
The following examples illustrate the use of the LET statement.

Example 1:
The numeric variable A is set equal to 5 by the statement
10 LET A

Example 2:
The string variable A$ is set equal to “XYZ” by the statement
10 A$

“XYZ”

Example 3:
An element (3,2) in array A is set equal to an element (1,4) in array B by the statement

10 LET A(3 ,2)

B(1,4)
NOTE
The LET statement does not necessarily imply an equal-

ity. LET means “evaluate the expression to the right of
the equal sign and assign this value to the variable on the

left”. Thus, the statement
L=L+l

means

“set L equal to a value one greater than it was

before”.
6.7.5

Loops (FOR and NEXT Statement)

6.7.5.1

FOR Statement Format

The FOR statement is used in combination with the NEXT statement to

specify program loops.
FOR Statement Form

Line Number

Statement

xxx

FOR

Parameters
V

x to y

LOOP

[STEP 2]

INDEX

INDEX INITIAL TERMINAL STEP
VALUE

VALUE

where
V is a variable that serves as the INDEX for the loop. The index value is incremented
(or decremented) each
time the loop is executed.

6-21

BASIC

is an expression (numerical value, variable name, or mathematical expression) indicating the initial value

of the index, that is, the value it will have before the loop is executed the first time;
y

is the terminal value of the index. When v is “beyond” y, the loop will not be repeated; and

is the step value, that is, it is the value that is added to the index each time the loop is executed. Vari-

able 2 may be assigned a minus value in cases where it is desired to decrement the index. If“STEP Z”
is omitted, a value +1 is assumed.

The x, y, and z values are expressions: these expressions are evaluated upon first encountering the loop. This in-

cludes setting the index equal to the initial value.

Therefore, the program can laterjump back to the same FOR

statement any number of times to re—start the loop. If the x, y, and z values are unchanged, the loop will be re-

peated the same number of times each time the program executes the FOR statement.
A variable used as an index in a FOR statement must not be subscripted.

The block of instructions to be executed repeatedly will immediately follow the FOR statement. After the last of
these instructions there must be 3 NEXT statement whose parameter is the same as the index of the loop.
If the initial value is “beyond” the terminal value, the loop will never execute because an initial check is made of the

starting and terminal values before the loop is executed. “Beyond”, as used here, means that the initial value is not
equal to the terminal value and adding the step value will only increase the difference. If the step value is negative
“beyond” means “less than”. If the step value is positive, beyond” means “greater than”.
“

The value of the loop index can be changed within the loop, thus influencing the number of times the loop is
executed.
A program can have one or more loops within a loop. This is called “nesting loops”, and is often used with sub-

scripted variables.
It is possible to exit from a loop without the index reaching the terminal value by using an IF statement. Control
may transfer into a loop only if that loop was left earlier without being completed.

The following examples illustrate the use of the FOR statement.

Example 1:
The following program

DIM

C(2v3)

FOR

921

FOR B31 T0 2
READ C(BVA)

DATA

19273147596

END

C(B!A)7

Bffi

6-22

BASIC

will display:
H
I-J

outburst—s UNFJFJF‘H
H
m
H

Lines 30 through 60 of the above program represent a loop nested within another loop (line 20 through 70). The
FOR statement on line 30 will be executed three times. Each time, its loop statements (lines 40 through 60) will be

repeated twice.

Example 2:
The following lines

10
20

LET
FOR

Bm2\C=3
A=C~B TO

NEXT A
PRINT A

ENE

B”C

STEP

will display:

N IMH
When the FOR statement was encountered, it interpreted the initial value as l (3 minus 2), the terminal value as
8 (2 to the third power) and the step value as 3.

Example 3:
The following loop

FOR

I==10

30
99

PRINT
END

STEF‘

“'1

will display:

3
Example 4:
The following loop

FOR

20
30

LET D=D+4
NEXT II

ENE!

D=1

6-23

BASIC

will only be executed once.

Example 5:
The loops in the following program will never be executed:

10
20

FOR D=5
PRINT D

10
20
30

FOR U=1
PRINT U
NEXT D

END

6.7.5.2

STEP

NEXT Statement Format

—

The NEXT statement defines the end of a program loop.

NEXT Statement Form
Line Number

Statement

xxx

Parameters
V

(variable)

where v represents the index value used in the corresponding FOR statement.
A NEXT statement is never used without a FOR statement. The variable value must be identical to the index value

(v) used in the corresponding FOR statement. The variable parameter may not be subscripted.
6.7.6

Control Statements (GOTO, IF-THEN, GOSUB, RETURN)

Normally the statements in a BASIC program are executed in the sequence they appear in a program. Control state—
ments allow this execution sequence to be redirected.

6.7.6.1

Unconditional Branch (GOTO Statement)

The GOTO statement is used to transfer control to another

line statement in a program. BASIC then continues execution at the line number referred to in the GOTO statement.

GOTO Statement Form
Line Number

Statement

Parameter

xxx

GOTO

where n is the line number of the statement to which control should be transferred.
For example, the following program employs two separate GOTO statements to redirect program control.

GOTO

20
30

PRINT
STOP

50
99

GOTO
END

'SECOND'
IFIRSTll
20

When executed, the program displays:

FIRST

SECOND

6-24

BASIC

NOTE
When the program reaches the GOTO statement, the
statements immediately following will not be executed;
instead, execution is transferred to the statement begin-

ning with the line number indicated.
Conditional Branch (IF -THEN Statement)
The IF-THEN statement tests a condition and redirects prothe
control
if
that
condition
is
true.
That
if
condition
is,
gram
specified is true, the IF-THEN statement effectively
executes a GOTO statement for the line number specified. When the condition is false, the next statement is

6.7.6.2

—

executed.
Line Number

Statement

xxx

Parameters
v1 relation v2 THEN n

where

variables, numbers, strings, or expressions to be compared;

v1, v2

are

relation

is the relational operator to be used in comparing v1 and v2 (See Section 6.6.5.4);

THEN

may be replaced with GOTO if desired (either THEN or GOTO is acceptable, but one of
the two must be present); and
is the number of the statement to which the program will jump if the relationship described

is true.
The following examples illustrate the IF -THEN statement.

Example 1:
After executing the first two instructions

20
30

LET

100

ASS
A

333-2

GOTO

100

'NO'

F'RINT

'HER‘E'

control is transferred to line 100.

Example 2:
After executing the following instructions

20
30

LET A=5
IF A=2 SOTO

ENII

100

'NO'

the console will display:

6-25

BASIC

After executing the following instructions

LET

A$==

ENII

A$==

I'YES

GOTO

the program will display:

YES
because the ASCII code for “*” is not equal to the ASCII code ASCII for a “Z”.

Strings may be used in IF-THEN statements, but comparisons are based on the positions of the characters in the
string sequence. The question mark (?) is the highest alphanumeric character and the at sign (@) is the lowest.
The two strings described in the IF-THEN statement are compared one character at a time, from left to right, until
the ends of the strings are reached or until an inequality is found.

If the strings are of unequal length, BASIC lengthens the shorter string by adding spaces to the right until both are of

equal length.

If“AB” is compared to a four-character string, it will be treated as “AB(space) (space”.

When using numerical expressions in the IF-THEN statement, the test for equality may not always work due to the
nature of the arithmetic used by the computer.

One way to get around this is to compare the absolute value of the

difference between the operands to a very small number. For example, instead of
20 IF A=B THEN 50

use

20 IF ABS(B) < 0001 THEN 50

6.7.6.3

Branch to Subroutine (GOSUB Statements)

A subroutine is a group of statements that perform a proc~

essing operation at more than one point in a program.
The GOSUB statement is used to branch to the subroutine and the RETURN statement redirects control back to the
main body of the program.

GOSUB Statement Form
Line Number

Statement

xxx

GOSUB

Parameters
n

where n is the line number of the first line of the subroutine.
When the program encounters a GOSUB statement, the following action occurs:
1.

BASIC internally records the number of the statement following the GOSUB statement.

Control is transferred to statement number n.

GOSUB is always used with the RETURN statement.
The RETURN statement is the last statement in a subroutine. When the program encounters the RETURN state-

ment, control transfers back to the statement following the GOSUB.

6-26

BASIC

A subroutine can call another subroutine. This is called “nesting” (see Example 2).

Programs may be written to

transfer control from one statement to another in the same subroutine, or to a statement in a different subroutine.
When a RETURN is encountered, control returns to the statement following the last GOSUB that was executed.
Subroutines may not be nested more than ten levels deep (Example 2 shows two levels of nesting).
The following examples illustrate the GOSUB statement.

Example 1:
The following program
100

LET

fiSfi'HI'

110

LET

B$V"THERE'

120
130

LET
LET

C$x'
U$3A$

140

GOSUB

150
160

LET U$$A$8C$ZB$
GOSUB 1000

170

STOP

1000

1000
1010

REM PRINT
PRINT U$

1020
9999

RETURN
ENU

will display:

13;.
HI

THERE

Example 2:
The following program, showing subroutine nesting,

100

FOR

110

LET

121

120

UWI
GOSUB 1000

130

LET

140
150

GOSUB 500
NEXT I

200

STOP

500

REM

510

LET

X32*I

CALCULATE

520

U==X+Z2
GOSUB 1000

530

RETURN

1000
1010

'THE

1020

1030

RETURN

9999

END

REM

F'RINT

VALUE
VALUE

IS"?

6-27

BASIC

will display:

THE

VALUE

THE

UALUE

CUNEbl’JbH

IflE_MflLQE_L§_;
THE

VALUE

THE

UALUE
VALUE

IS
IS

THE
6.7.6.4

Return from Subroutine (RETURN Statement)

—

The RETURN statement is used to redirect control

from a subroutine.
RETURN Statement Form

Line Number

Statement

Parameters

xxx

RETURN

(none)

RETURN is always used with the GOSUB statement.
When the RETURN is encountered, it causes control to transfer to the statement following the last GOSUB
executed.
6.7.7

Program Termination Statements (END, STOP)

BASIC is equipped with two statements, the END and STOP statements, that can be used to terminate program
execution

6.7.7.1

END Statement Format

—

The END statement terminates execution of the program. It informs the

BASIC compiler that it has reached the last line of the program.

END Statement Form
Line Number

Statement

Parameters

xxx

END

N one

NOTE

Only one END statement may appear in a program and it
must be the last statement in the program.

6.7.7.2

STOP Statement Format

—

The STOP statement is used to terminate the execution of a program.

STOP Statement Form
Line Number

Statement

Parameters

xxx

STOP

None

Note that there may be several STOP statements in a program.
6.7.8

The INPUT Statement

The INPUT statement permits the operator to specify data during execution of a program.

6-28

BASIC

INPUT Statement Form
Line Number

Statement

Parameters

xxx

INPUT

Variable List

where variables can be a single variable or a list of variables.
The INPUT Statement will cause the program to pause during execution, display a question mark, and wait for you
to enter a value and press the RETURN key. If there are several variables involved, the program will expect you to

type in a value corresponding to each variable. If you press the RETURN without having done this, the system will
display another question mark and wait for the rest of the data. When the values have all been typed in, the program will continue with the variable names now equivalent to the values typed in. The first variable will equal the
first entry, the second will equal the second entry, etc. (See Example 1).

The following values are recognized as acceptable when inputting numeric data:
+ or

sign

digits 0 through 9
the letter B

leading spaces (ignored)
(first decimal point)
All other characters are treated as delimiters for separating numeric data. That is, when the system encounters a
character other than those specified, it will consider that it has come to the end of the entry relating to the variable
it is currently processing and will apply any characters typed in after that to the following variable, if any (see

Example 1).
When inputting numeric data, two delimiters read in succession imply that the data between delimiters is 0 (see

Example 2).
In response to an INPUT statement, you can provide more values than are requested by the INPUT statement. The

remaining or unused values are saved for subsequent use by the next INPUT statement.

The question mark is not

displayed until the program is out of data.
When inputting string data all characters are recognized as part of the string. Quotation marks are not typed in
unless they are deliberately meant to be part of the string.
Each string requested by an INPUT statement must be terminated by a carriage return which acts as the data de-

limiter. This is necessary since all characters except for the carriage return are recognized as part of the data string.

String variables are assumed to be eight characters long unless otherwise described in a DIM statement.
The following examples illustrate the use of the INPUT statement.

Example 1:
If, in response to this statement:
INPUT

100

(«1 y B 9 [3 v [I y E

you type

:....2

3 o 794E325

‘?-=1f+ 1'

6-29

BASIC

the variables will have the following values:
A:

B: 3.7

C: 4000(4E3=4>< 10‘3=4000)
D:9
E: 1(numbers are assumed to be positive unless they are specified to be negative).
The delimiters in the above line are the comma, the letter “A”, the space after the number 3, the left-angle bracket

(<), and the RETURN.
Example 2:
If, in response to the same statement, you type

2""?!

4E3! '1

3.7!

The results will be identical except that variable “D” will have the value 0.

Example 3:
If, in response to the statement:
50

INPUT

you type

'AvC}=+7

the string variable R$ will have the value:

“A,C>=+7

Example 4:
If, in response to:
40

LET

INPUT

A=5

EMA)

you type

?7
B (S) will now have the value of 7.
6.7.9

The REMark Statement

The REM statement is a nonexecutable statement used to insert comments into the source program.
REMARK Statement Form

Line Number

Statement

Parameters

xxx

REM

comments

6-30

BASIC

Ancillary Statements (DIMension, RESTORE, DEFine, RANDOMIZE and CHAIN)

6.7.10

BASIC has five additional statements that are used as helping statements and fall in no particular category.

They are

described in the following paragraphs.
6.7.10.1

DIMension Statement Format

The DIMension statement is used to describe subscripted variables and

to define the length of strings.

DIMension Statement Form

Line Number

Statement

Parameters

xxx

DIM

v(n [,m] )

where v is the name of the subscripted variable.

If the variable name (v) is numeric, and
l.

is omitted, then n+1 is the number of elements in an array or vector (see Example 1).

is specified, then n+1 is the number of rows in a two-dimensional array (see Example 2).

ifv is numeric, then m+l is the number of columns in a two-dimensional array (see Example 2);

If the variable name (v) is alphanumeric, and
l.

is omitted, then n is the length of the string. This describes a single string, not a list, and cannot exceed

80 (see Example 3).
2.
3.

m is specified, then n+1 is the number of strings in the list. This is a one dimensional array (see Example 4).
if v is alphanumeric, then m is the length of each string in a one-dimensional list. It cannot exceed 80 (see

Example 4).
The parameters It and m must be integer constants. They are limited in size only by the amount of available memory.
If a numeric variable is used in the program with a subscript but is not defined in a DIM statement, BASIC assigns it
an

array size of ten.

BASIC assumes a maximum string length of 8 characters unless the variable appears in a DIM statement.

Two-dimensional string variables are not permitted.
When the variable is used in other statements, it is not permitted to have subscripts whose values are higher than
those in the DIM statement.
The first element of every array is automatically assumed to have a subscript of zero. Therefore, the number of
boxes in a one dimensional array is n+1. The number of boxes in a two-dimensional array is (n+l)*(m+l). The
first element in an array is v(O) or v(0,0) (see Example 2). However, the “zero” elements can be disregarded in pro-

gramming unless the user wishes to conserve memory.
More than one array can be defined in a single DIM statement (see Example 7).
In general, wherever you can use a single variable name in a statement, you can use a subscripted one.
are

noted in descriptions of the individual statements (such as the index of the FOR statement).

The following examples illustrate the use of DIMension statement.

6-31

Exceptions

BASIC

Example 1:
The following statement

II I M

A ( 5)

describes six numeric elements as follows:

A(z)

A( 1)

A(O)

A(3)

A(4)

A(5)

To store a 5 in A(3) and then display it, type

A(3)=5

LET

A(3)

Example 2:
The following statement

DIM

ACME)

describes 24 numeric elements (4X 6=24) as follows:
SIX COLUMNS

A(0,0)

A(0,1)

A(0,2)

A(0,3)

A(O,4)

A(O,S)

FOUR

A(1,0)

A(1,1)

A(1,2)

A(1,3)

A(1,4)

A(1,5)

ROWS

A(2,0)

A(2,1)

A(2,2)

A(2,3)

A(2,4)

A(2,5)

A(3,0)

A(3,1)

A(3,2)

A(3,3)

A(3,4)

A(3,5)

Example 3:
The following statement

DIN

C$(12)

describes one string, 12 characters long:

Example 4:
The following statement

DIM

Il$(3720)

describes 4 strings, each 20 characters long:

D$(0)

D$(1)

D$(2)

D$(3)

6-32

BASIC

Example 5:
The following program will fill the array in Example 4 from a DATA list:

DIM

H$<3y20)

FOR

Ymo

30
40

READ
NEXT

FOR

2x0

D$(Y)
T0

D$(Z)

PRINT
NEXT 2
DATA ‘ZERO'v'ONE'v'TUO'y'THREE'

END

60
70

and display each element:

ZERO
3 Z 5‘"!

TN D
THREE

Example 6:
After these statements:

HIM

B(3y5)

FOR

Gal

LET

B(GyO)mG

40
50

NEXT 6
FOR H32

LET

70
9‘?

NEXT
END

B<0rH>3H
H

The area diagrammed in Example 2 would appear as follows:

B(1,0)

B(2,0)

B(3,0)

B(0,2)

B(O,3)

B(0,4)

B(0,5)

Example 7:
The following statement dimensions both the one-dimensional array A and the two-dimensional array B:

10
6.7.10.2

HIM

AGEO) 9

B(4y7)

RESTORE Statement

The RESTORE statement allows the program to go back to the beginning of 3
DATA list (paragraph 6.7.3.1) after using the list in a READ statement (paragraph 6.7.3.2).
,

6-33

BASIC

RESTORE Statement Form
Line Number

Statement

Parameters

xxx

RESTORE

None

Ifit is desired to use the same data more than once in a program, RESTORE makes it possible to recycle through
the DATA list beginning with the first value in the first DATA statement.
The RESTORE statement may be used in programs where DATA statements convey numeric or string data to
READ statements.

The same variable names may be used the second time through the data since the values are being read as though for
the first time.

The following example illustrates the use of the RESTORE statement.
The following lines

READ

20
30

PRINT A9B963n
RESTORE

AyByCyn

REQD

DATA

END

EroGvH
EtFystH
1v;93y4

will cause this display:

6.7.10.3

DEFine Statement

—

This statement allows you to add functions to a program.

DEFine Statement Form
Line Number

Statement

Parameters

xxx

DEF

FNa(x)=expression

where

a is

the capital letter used for identifying the function;

is a dummy variable and must be the same on both sides of the equal (=)

sign; and

expression defines the function by indicating the calculation process (function) involved.
There must be a DEF statement for each function used in the program.
The DEF statement must appear before the first use of the function it defines.

If there is more than one variable involved in the function. BASIC will identify them by their position.

Up to 14 different arguments may be used.

6-34

BA SIC

Up to 26 FN functions may be defined in a single program (FNA, FNB

.FNZ).

The following examples illustrate the use of the DEFine statement.

Example 1:
The statement

DEF

FNB ( S ) ===S"2

will cause the later statement

LET

R===FNII ( 4)

to be evaluated as R=16.

BASIC locates the DEF statement for the function FND, substitutes the 4 for the variable

(S) in the expression (5’2) and calculates the value of FND(4) to be 4A2=l6.
NOTE
A variable that is used as a dummy argument in a DEF
FNa statement can also be used elsewhere in the program.

Example 2:
This program:

FNH(N!F‘)==2*F'+N

IIEF

2-30

LET X==4\LET Y-"-‘-5
PRINT FNH(X!Y)

END

will display:

L4.
BASIC takes the first value in the function (4) as “N”, because “N” appears first in the DEF statement. It takes the
second value (5) as “P”, because “P” is in the second position.
DEF FNH (N,P)
first position

2*P+N

second position

PRINT FNH(X,Y)

6.7.10.4

RANDOMIZE Statement

a different

The RANDOMIZE statement is used with the RND function to generate

set of numbers each time the program is run.

RANDOMIZE Statement Form
Line Number

Statement

Parameters

xxx

RANDOMIZE

None

6-35

BASIC

6.7.10.5

CHAIN Statement

The CHAIN statement allows one program to execute another program. It can be
used to divide large programs into a number of smaller programs that are to be written and stored separately.
—

CHAIN Statement Form
Line Number

Statement

Parameters

xxx

CHAIN

“dev:file.ex”

where “dev:file.ex” is the device and the file name of the program to be executed.
When BASIC encounters a CHAIN statement in a program, it stops execution of that program, retrieves the program
named in the CHAIN statement from the specified device and file, compiles the CHAINed program (if necessary)

and begins execution of that program.
If the program was started in the editor, when execution of all programs in the chain is complete, the workspace will
contain the original program.

All output files that are opened in the original program must be closed before the CHAIN statement is encountered.
A BASIC language program may only CHAIN to another BASIC language program, and the program it CHAINS to
may not have an extension of “.SV” (see the following example).

A compiled program may also execute CHAIN statements, but it can only CHAIN to other compiled programs
which have “.SV” extensions.
In the following example, the program “SECOND” will CHAIN to the program “FIRST”.

NEH

F I HST

READY
21.0 F'R I NT
230

TARGET

END
SYS 2 Ir IRSST

SAVE

Enter FIRST and store it.

READY

NEH

SECOND

READY
10

"ORIGINAL,“

1330

Ell-MIN

“SYSSF‘IRST.B(~1“

liz". NU

SAVE:

SYSS 1 SIEEIIUNII

Enter SECOND and store it.

RISEIQHY
FL'UNNH

Execute the programs.

OR I ['3 I NAL.

Displayed by SECOND.
Displayed by FIRST.

TAR“ 3ET

6.7.1]

File Handling Statements

The file capability provided by 08/78 BASIC allows

6.7.11.1

FILE# Statement

—

writing to or reading from the peripheral devices of the system.

The FILE# statement defines and opens a file.

636

BASIC

FILE# Statement Form
Line Number

Statement

xxx

FILE

Parameters
t

# n: “devzfileex”

where
must be one of the following:

(blank)

—

for an input string file
for an output string file
for an input numeric file

for an output numeric file;

is the number you are assigning to the file (It must be 1, 2, 3, or 4 and can be a numeric

variable); and
“dev: file.ex”

is the standard device, file name and extension; It must be either a string variable or the

string itself in quotation marks.
A file must be opened before it can be used. The only exception is the terminal, which is always available for use.
The n in this statement is the number you must use in all F ILE# statements that refer to the file. The terminal is

always FILE#0.
The following examples illustrate the use of the FILE# statement.

Example 1:
The following statement describes file number 1 to be the string file HPRDATAS on RXAl and opens it for output.

FILEU=I’1 3

RXAI :HF'RIIAT . AS

Example 22
The following statement describes file number 2 to be the numeric file DATA.NU on RXAl and opens it for output.

FILEUNiQ:

R'Xfil :[IATA . NU

Example 3:
The following statement describes file number 3 to be the string file TEST.AB on RXAl and opens it for input.

F'ILEtZ’S:

RXAl :TEST . AB

Example 4:
The following statement describes file number 4 to be the numeric file FILA.CD on RXAl and opens it for input.

10
6.7.11.2

FILEN1¥41

RXAl 3FILA . CD

PRINT# Statement

—

The PRINT# statement writes data into files.

6-37

BASIC

PRINT# Statement Form
Line Number

Statement

Parameters

xxx

PRINT#

nzexpressions

where

is the file number (It may be a numeric lvariable); and

expressions

depends on file type, numeric or string as discussed below.

As long as PRINT# is used for only numbers or numeric variables separated by commas or semicolons (or RETURN
at the end of a PRINT# line), BASIC converts commas to spaces and does not write the carriage return and line feed
to numeric files. The only thing written out will be a “list” of numbers separated by spaces. Each time INPUT# is

used, it will read another number from the list (see Example I).
When dealing with string files, symbols such as RETURNS, Semicolons, and so forth, are used in the PRINT# statement in the same way they are used in the PRINT statement. The PRINT# statement works exactly the same way

that the PRINT statement does, except that the line goes to the file designated instead of to the terminal.
The important difference here is that string files involve lines, while numeric files involve individual numbers. That

is, each INPUT# will read a line (see Example 2).
If PRINT# is used for numerics to a string file, BASIC will convert them to strings. If an attempt is then made to

INPUT# them into numeric variables, BASIC will convert them back to numerics. However, the RETURN and line
feed at the end of a line will be converted to zeros. This can be dealt with by adding two extra variables to the
INPUT# statement.
The following examples illustrate the use of the PRINT# statement.

Example 1:
The following lines

FILEUNtl:'SYS:TST.XX'

20
30

PRINTiitlyE
PRINTi133v4v

PRINT#13596

50
60

FILEN#13'SY83TST.XX'

FOR

80
90

CLOSE#1
X=1 T0
INPUT#1:Z
PRINT Z

100

199

END

will display

lblum
‘I

l'M‘
6-38

BASIC

Example 2:
The following lines

'A'v

'B'

“(3'3

'13";

230

F'RINT
ENE

'E'

will display

r34

CIIE
The same display will also be caused by

DIN

J$(30)

FILEUi23“RXA1:PRUG.XX'

20
30

PRINT
PRINT

#23'A"
#22‘C“$

#2t'E'

50
60

CLOSEiQ

FILE122'RXA13PRDG.XX'

INPUT122J$

INPUT#2:J$

100

199

END

6.7.11.3

'B'
‘3';

INPUT# Statement

The INPUT# statement reads data from a file.

INPUT# Statement Form
Line Number

Statement

Parameters

xxx

INPUT#

nzvariables

where

is the file number of the file being read (it may be a variable); and

variables

is the list of variables into which data will be read. Each variable is separated by a comma.

Normally, data from numeric files is read into numeric variables and data from string files is read into string variables.
It is possible however to write numerics into a string file and then read them into either numeric or string variables,

depending on how it is desired to use them. If numbers are read from a string file into string variables, they will be
in string form and subject to the same rules as other strings.
Numbers read from a string file into numeric variables will be converted to numerics. Line feeds are ignored in this
case.

The following examples illustrate the use of the INPUT# statement.

6-39

BASIC

Example 1:
To read two strings from RXAleILDA, enter

FILE!13'RXA1:FIL.DA'

INPUT#I:A$vB$

A$ will contain the first string, B$ the second string.

Example 2:
To read five numbers from RXAl:TST.XX, enter

FILEN=F32 'RXA12TST.XX'

INF'UTi33fiyB’CyIflyE

Example 3:
The following program writes numerics to a string file and reads them back as numerics:

FILEU#1:'SYS.FILA.ZZ'

FOR

30
40

PRINT#1:I
NEXT I

CLUSE#1

60
70

FILE11:'8YS:FILA.ZZ'
FOR Iml TO 5

1:1

INPUT#1:J;C;L

100

110

END

It will display:

KMkaMH
6.7.11.4

RESTORE# Statement

—

The RESTORE statement resets the file data printer back to the beginning of

the file, that is, RESTORE effectively performs a file close followed immediately by a file open so that the first data

element in the file can be. reread.

RESTORE# Statement Form
Line Number

Statement

Parameters

xxx

RESTORE#

where

is the number of the file to be reset. It may be a number or a numeric variable.

6-40

BASIC

For example, if RXAIIFILBLM is a numeric file containing the numbers 1 through 9, the instructions

100

FILEN132'RXA12FILB.LM'

110

FOR

120

INPUT

130
140

PRINT Z
NEXT I

150

RESTOREtE

160
170

INPUT1322
PRINT Z

199

END

Iml

#332

will display:

lequwuw
NOTE

If n is zero, the DATA list in the program is reset.
6.7.11.5

CLOSE# Statement

The CLOSE# statement finishes the processing of a file and allows it number to be

—

assigned to another file in a FlLE# statement.
FILE CLOSE# Statement Form

Line Number

Statement

Parameters

xxx

CLOSE#

where

is the number of the file to be closed. It may be a variable.

For example, in the following program

FILEV

#1:'SYS:TEST.XX'

#1:'A'»'B';'C'r'fl'
#1

80
9O

FILE #1:“RXGl:FILn.DA“
INPUT #1:J$

END

The CLOSE# statement at line 70 finished the processing of file SYS:TEST.XX and allowed its number, I, to be

assigned to RXAl :FILD.DA in line 80.
NOTE
All output files must be CLOSED before any of the fol-

lowing is executed.
CHAIN

END

STOP

CTRL/C

Failure to do so results in loss of file.

6-41

BASIC

6.7.11.6

IF END# Statement

—

The IF END# statement determines if the End of File (EOF) marker has been

read from a file, and if so, branches to the line specified.
IF END# Statement Form
Line Number

Statement

Parameters

xxx

IF END#

n THEN m

where:

is the file number of the file in question (it may be a variable); and

is the line number to which control passes when the end of the file is detected.

This statement works only for string files.

The IF END# statement should come immediately after the PRINT# or INPUT# statement for that file.

If, as a

result of the IF END# statement, control passes to line m, it means that the last PRINT# or INPUT# was not
successful. That is, nothing was actually read from or written to the file as a result of the last INPUT# or PRINT#
statement.

For Example,
The following lines

10
20

FILEU#1:'SYS:PROGQ.BB'
PRINTtlz'A'

30
50

PRINTilt'B'
CLUSE#1
FILEil:'SYS:PROGA.BB'

INPUT#1:A$

70
80

IF END#1
PRINT A$

GOTO

THEN

100

100
110

PRINT 'END
CLOSEil

199

END

FILE'

will display

9.
B
END
6.8

F I LE

BASIC FUNCTIONS

BASIC functions are standard subroutines incorporated into the BASIC Run Time System (BRTS) to aid computations and text handling.

Function calls consist of a three letter (all capitals) name followed by an argument in parentheses. The argument
may be a number, variable, expression, or another function.
or variable is legal in a mathematical expression.

Generally, functions may be used anywhere a number

Most functions compute a value based on the value of the argument or arguments involved.

this value. For example, SQR(A) “returns” the square root of A.

6-42

They are said to return

BA SIC

Functions may return either strings or numbers. Functions that return strings have names ending in a dollar sign

(STR$, SEG$), while functions returning numbers have names that do not end in a dollar sign (SGN, VAL).
Arithmetic Functions (ABS, INT, EXP, RND, SGN, SQR)

6.8.1

6.8.1.1

BASIC ABS Function

The ABS function returns the absolute value of an expression.

Format

ABS(X)
where

6.8.1.2

is a number, numeric variable, or numeric expression

BASIC EXP Function

—

The EXP function calculates the value of e raised to the X power, where e is

equal to 2.71828. That is, EXP(X) is equivalent to 2.71828 X.
Format

EXP(X)
where
X

6.8.1.3

is a number, numeric variable, expression, or another function.

BASIC INT Function

The INT function returns the value of the largest integer not greater than the

argument.
Format

INT(X)
where
X

is a number, numeric variable, expression, or another function.
NOTE
This function can be used to round numbers to the nearest integer by specifying INT(X+.5).

For example, the function INT(34.67) has the value 34; the functions INT(34.67+.5) and INT(34.37+.5) have
these values of 35 and 34, respectively; and these functions INT(— 23) and INT(— 14.39) have these values of -23 and
-

15, respectively.

6.8.1.4

BASIC RND Function

—

The RND function produces random numbers between (but not including

0 and 1.
Format

RND(X)
where

is a dummy variable in this function.

6-43

BASIC

Every time this function is encountered in a statement, it will produce a different set of decimal numbers. However,
the program is RUN again, the same set of numbers will be produced (see Example I).
If this repetition is undesirable, it can be changed with the RANDOMIZE statement.
For example, the following program is run twice with identical results:

FOR

30
40

NEXT
ENE

RNIKX)
A

RUNNH
0.

361572

332764

0.
0.

633057
50 4a

670166

READY
RUNNH
0.

361572
332764

633057

350342
670166

READY
6.8.1.5

BASIC SGN Function

—

The SGN function creates a value based on the sign of the argument.

Format

SGN(X)
where
X

is a number, numeric variable, numeric expression, or another function.

NOTE
The value of the SGN function will be 1 if the argument
is any positive number, 0 if the argument is zero, and

if the argument is negative.

6.8.1.6

BASIC SQR Function

The SQR function computes the positive square root of an expression.

Format

SQR(X)
where
X

is a number, numeric variable, numeric expression, or another function.

6-44

BASIC

NOTE
If the argument is negative, the absolute value of the

argument is used.

Trigonometric Functions (ATN, COS, LOG, SIN)

6.8.2

BASIC ATN Function

6.8.2.1

—

The ATN function calculates the angle (in radians) whose tangent is given as the

argument of the function.
Format

ATN(X)
where
is a number, numeric variable, expression, or another function, representing the tangent of an angle.

BASIC COS Function

6.8.2.2

The COS function is used to calculate the cosine of an angle specified in radians.

Format

COS(X)
where
is a number, numeric variable, expression, or another function, representing the size of an angle in

radians.
6.8.2.3

BASIC LOG Function

—

The LOG function calculates the natural logarithm of X (to the base e).

Format

LOG(X)
where
X
6.8.2.4

is a number, numeric variable, expression, or another function.

BASIC SIN Function

—

The SIN function is used to calculate the sine of an angle specified in radians.

Format

SIN(X)

whereX

is a number, numeric variable, expression or another function, representing the size of an angle in

radians.
6.8.3

6.8.3.1

String Handling Functions (ASC, CHP$, DAT$, LEN, POS, secs, STR$, TAB, VAL)
BASIC ASC Function

—

The ASC function converts a one character string to its code number

(see CHR$).

6-45

BASIC

Format

ASC(X)
where
X

is a one character string.

To find what will be returned for any character, look for the character in the “CHARACTER” column of Table 6-1.

The number to the left of it is the decimal equivalent.
Table 6-1

Decimal/Character Conversions

Decimal

Character

Decimal

Character

(space)

”

’

(
)

[

;

]

31
_

‘a

6-46

BASIC

For example,
The following program

10
20

LET A$x'*'
PRINT ASCUF“):

ENE

A8C(A$)yASC('9')

will display

6.8.3.2

BASIC CHRS Function

The CHR$ function converts a code number (modulo 64) to its equivalent

character in the 64 character set.
Format

CHR$(X)
where
X

is a number, numeric expression, or numeric variable (0 < X < 63).

To find what will be returned for any number, look for the number in the “DECIMAL” column of the preceding
conversion table. The equivalent character will be to the right of the number.

For example, the following line

CHR$( 1) vCHR$(40)

will display:

L4
6.8.3.3

BASIC DATS Function

—

The DAT$ function returns the current system date.

Format

DAT$(X)
where
X

is a dummy variable in this function.

The date is returned as an eight-character string of the form:

MM/DD/YY
If the date has not been specified with the Monitor DATE command, no characters will be returned.

For example, the following lines

LET

13$

‘~=

DAT$(X)

13$

6-47

BASIC

will display:

07/20/77
if that date was entered with the Monitor DATE command.

6.8.3.4

BASIC LEN Function

The LEN function returns the number of characters in a string.

Format

LEN(X$)
where
is a string or string variable. It may be several concentrated strings and/ or variables.

X$
6.8.3.5

BASIC POS Function

The POS function returns the location of a specified group of characters in a

string.
Format

POS(X$, Y$, Z)
where

is the string to be searched (it may be a string variable or string constant);

is the series of characters you are searching for (it may be a string variable or a string constant); and

is the position in the string at which you want to begin the search.

This function searches X$ for the first occurrence of Y$. The search begins with the Zth character in X$.
Z may not be less than zero or greater than the length of string X$.
If Y$ contains no characters, the function returns a one.
If X$ contains no characters, it returns a zero.
If Y$ is not found, it returns a zero.

For example, the following lines

HIM

8295(12)

B235

ABCDEFGHIDEF

F'OS

(B1256!

'DEF“ 9

will display:

Lg
6.8.3.6

BASIC SEG$ Function

—

The SEG$ function returns the sequence of characters between two positions

in a string.

6-48

BASIC

Format

SEG$(X$,Y,Z)
where

is the string containing the characters to be returned (it may be a string variable or a string constant);

is the position of the first character to be returned; and

is the position of the last character to be returned.

IfY is less than 1, it is set to 1.

If Y is greater than the length of X$, no characters are returned.
If Z is less than 1, no characters are returned.

IfZ is greater than the length of X$, it is set equal to the length of X$.
If Z is smaller then Y, no characters are returned.

For example, the following lines

DIM

B2$(12)

132$

'ABCIIEFGHIIIEF'
SEG$<BE$y3y5)

will display:

CHE
6.8.3.7

BASIC STRS Function

The STR$ function converts numbers to strings.

—

Format

STR$(X)
where
X

is a numeric expression

NOTE

The string that is returned is in the form in which numbers are output in BASIC without leading or trailing
blanks.

6.8.3.8

BASIC TAB Function

—

The TAB function allows you to position characters anywhere on the terminal

line.

6-49

Format

TAB(X)
where
X

is the position (from 1 to 80) in which the next character will be displayed.

This function may only be used in a PRINT or PRINT# statement.
Positions on the line are considered by BASIC to be numbered from 1 to 80 across the screen from left to right.
Each time the TAB function is used, positions are counted from the beginning of the line, not from the current

position of the cursor.
If X is less than the current position of the cursor, the display starts at the current position.
If X is greater than 80, the display will begin at the first position of the next line.
In order to keep track of the cursor, BASIC maintains a “column count”, which represents the position of the cursor
As the cursor moves across the screen, BASIC adds to the column count. When the cursor returns

at any given time.

to the first position, the column count is reset to 0.

The activity of the TAB function is based on this count.

There are circumstances in which the column count does not coincide with the position of the cursor. For example,
the PNT(O7) function will add 1 to the count without moving the cursor.

Also, the PNT(13) function will return

the cursor to the first position on the screen without setting the column count to zero. The user, therefore, must
take this into account when using the TAB function after PNT functions. The column count will be corrected the
next time there is a “normal” return to column one.

For examples, the following lines

LET

B=5
'A'H'ABQDE'C'

will cause this display:

A
6.8.3.9

C
BASIC VAL Function

The VAL function converts a string to numeric data.

Format

VAL(X$)
where

is a string constant or string variable made up of those values that BASIC recognizes as acceptable

when inputting numeric data:
+ or

sign
digits 0 through 9
—

the letter E

leading spaces (ignored)
(first decimal point)
.

6-50

BASIC

NOTE
A

string, even though it is composed of digits, is not

numeric data. It cannot be used in calculations or as
the argument of a mathematical function (SQR, ABS,

EXP, and so forth), without first being converted by
the VAL function.

6.8.4

Display Console Control Function (PNT)

BASIC PNT Function

The PNT function is used to perform special actions on the terminal, such as sounding

the buzzer, erasing the screen, and moving the cursor.

Format

PNT(X)
where
X

is the value of the character to be output.

Special actions that can be performed by the PNT function are as follows:

PNT(07)

sounds buzzer

PNT(08)

moves cursor one space to left

PNT(09)

moves to next tab stop

PNT(10)

moves cursor down one line and

PNT(13)

moves cursor to left

PNT(27);“A”

moves cursor up one line

PNT(27);“C”

moves cursor

PNT(27);“H”

moves cursor to upper lefthand corner of screen

PNT(27);“J”

erases from cursor

PNT(27);“K”

erases line from cursor to

PNT(27);“[”

stops display of any new lines when screen is full. Each time the operator presses the
SCROLL key, another line will be displayed. If he presses SHIFT and SCROLL,

(Tab stops are set every 8 spaces.)
scrolls if required

margin of current line

right one position

(“home” position)

position to end of screen
right margin

twelve new lines will be displayed. This is called Hold Screen Mode.

PNT(27);CHR$(28)

turns off Hold Screen Mode.

NOTE

The PNT function may only be used in a PRINT or

PRINT# Statement.

6-51

BASIC

6.8.5

Trace Function

BASIC TRC Function
program

as it

—

The TRC function causes BASIC to print the line number of each statement in the

is executed.

Format
v =

TRC(X)

where
any letter (it is a dummy argument and has no purpose except to occupy that position on
the line); and

can be

is 1 to turn the function on and O to turn it off.

This function is used to follow the progress of a program and help in tracking down errors.
When BASIC encounters TRC(l) in a program, it displays the line number of each line in the program as it is
executed. The line numbers are displayed between percent signs.
When TRC(O) is encountered, the function is turned off and normal program operation resumes.

Certain types of statements are not recorded by TRC: namely, DATA, DEF, DIM, END, GOTO, NEXT,

RANDOMIZE, REM, and STOP.
For example, the following program

TzTRC(1)

GOSUB

GOTD

140

90 PRINT 'IN OUTER
100 GOSUB 120
110

RETURN

120
130

PRINT llIN
RETURN

140

150

END

TRC(O)

will display:

OUTER

SUB

2—166—2“"“'
z

120

IN INNER
2 130 z
z

110

140

sun

INNER

SUB'

BASIC

6.9

SUMMARY OF BASIC EDITOR COMMANDS
Command

Function

BYE

Exits from the editor and returns control to the monitor

LIst

Displays the program statements in the workspace with a header

LISTNH

Displays the program statements in the work space, without a header

NAme

Renames the program in the workspace

NEW

Clears the workspace and tells the editor the name of the program the user is about to

type

6.10

OLd

Clears the workspace, finds a program on the disk, and puts in into the workspace

RUn

Executes the program in the workspace, after displaying a header

RUNNH

Executes the program in the workspace, without displaying a header

SAve

Puts the program in the workspace on a disk

SCratch

Erases all statements from the workspace

SUMMARY OF BASIC STATEMENTS
Statement

Function

CHAIN

Executes another program

Example:
CLOSE#

Closes 21 file

Example:
DATA

40 CHAIN “SYS:PROG.BA”

100 CLOSE#1

Sets up a list of values to be used by the READ statement

Example: 240 DATA “F IRST”,2 ,3
DEF

Defines functions

Example:
DIM

Describes a string and/or any subscripted variables

Example:
END

50 DIM B(3,5),D$(3,72)

Terminates program compilation and execution

Example:
FILE#

10 DEF FND(S)=S+5

100 END

Defines and opens a file

Example:

20 FILEVN#’2:“RXA1:DATA.NV”

6-5 3

BASIC

Statement

Function

FOR

Describes program loops (used with NEXT)

Example:
GOSUB

Transfers control to a subroutine (used with RETURN)

Example:
GOTO

50 GOSUB 100

Transfers control to another statement

Example:
IF

60 FOR X=1 TO 10 STEP 2

100 GOTO 50

Tests the relationship between two variables, numbers, or expressions

Example: 20 IF A=0 THEN 50
IF END#

Tests for the end of a string file

Example: 60 IF END#3 THEN 100
INPUT

Accepts data from the terminal

Example:
INPUT#

Reads data from a file

Example:
LET

80 INPUT A,B,C

50 INPUT#1:A$

Assigns a value to a variable

Example: 90 LET A$=“XYZ”
NEXT

Indicates the end of a program loop (used with FOR)

Example:
PRINT

Displays data on the screen
Example:

PRINT#

200 PRINT A,“X”;6

Writes data to a file

Example:
RANDOMIZE

140 NEXT I

180 PRINT#1:J

Causes the RND function to produce a different set of numbers each time the program
in run

Example:
READ

Sets variables equal to the values in DATA statements

Example:
REM

50 READ A$,B

Inserts comments into the program

Example:
RESTORE

10 RANDOMIZE

30 REM COMPUTE EARNINGS

Sets program READ statements back to the beginning of the DATA list

Example:

85 RESTORE

6-54

BASIC

Statement

Function

RESTORE#

Resets a file pointer back to the beginning of that file

Example:
RETURN

Returns control from a subroutine (used with GOSUB)

Example:
STOP

130 RESTORE#3

115 RETURN

Terminates program execution

Example: 40 STOP
SUMMARY OF BASIC FUNCTIONS
Command

Function

ABS(X)

Returns the absolute value of an expression

Example:

10 LET X=ABS(— 66)

will assign X a value of 66

ASC(X$)

Converts a one character string to its code number

Example: 20 PRINT ASC(“B”) will display 2

ATN(X)

Calculates the angle (in radians) whose tangent is given as the argument

Example: 30 LET X=ATN(.57735)
will assign X a value of 0.523598

CHR$(X)

Converts a code number to its equivalent character

Example: 40 PRINT CHR$(1) will display A

COS(X)

Returns the cosine of an angle specified in radians

Example: 50 LET Y=COS(45 *3.14159)/180
will assign Y a value of 0.707108

DAT$(X)

Returns the current system date

Example: 60 PRINT DAT$(X)
will display the system date, such as 07/20/77

EXP(X)

Calculates the value of e raised to a power, where e is equal to 2.71828

Example:

30 IF Y>EXP(1.5) GOTO 70

will go to line 70 if Y is greater than 4.48169

INT(X)

Returns the value of the nearest integer not greater than the argument

Example: 60 LET X=INT(34.67)
will assign X the value 34

6-55

BASIC

Command

Function

LEN(X$)

Returns the number of characters in a string

Example:

10 PRINT LEN (“DOG”)

will display 3

LOG(X)

Calculates the natural logarithm of the argument

Example:

10 PRINT LOG(959)

will display 6.86589

PNT(X)

Outputs non-printing characters for terminal control
Example:

50 PRINT PNT(13)

will move the cursor to the left margin of the current line

POS(X$,Y$,Z)

Returns the location of a specified group of characters (Y$) in a string (X$) starting
at a character position

Example:

(Z)

60 LET V= POS(“ABCDBC”, “BC”,4)

will assign V a value of 5

RND(X)

Returns a random number between (but not including) 0 and 1

Example: 70 PRINT RND(X)
will display a decimal number, such as 0.361572

SEG$(X$ ,Y,Z)

Returns the sequence of characters in a string (X$) between two positions in the

string (X,Y)

Example:

30 LET R$=SEG$(“ABCDEF”,2,4)

will assign R$ a value of BCD

SGN(X)

Returns 1 if the argument is positive, 0 if it is zero, and

Example: 200 PRINT 5 *SGN(~ 6)
will display

SIN(X)

Returns the sine of an angle specified in radians

Example:

30 LET B=SIN(30*3.14159/180)

will assign B a value of 0.5

SQR(X)

Returns the positive square root of an expression

Example: 40 PRINT SQR(l6)
will display 4

STR$(X)

Converts a number into a string

Example:

120 PRINT STR$(1.76111124)

will display the string 1.76111

6-56

1 if it is negative

BASIC

Comm and

Function

TAB(X)

Positions characters on a line

Example:

70 PRINT “A”;TAB(5);“B”

will display A B

TRC(1)

Causes BASIC to display the line number of each statement in the program as it is

executed

Example:

10 V=TRC( 1)

will display the line number of each statement executed until a TRC (0) is

encountered

VAL(X$)

Converts a string to a number

Example: 90 PRINT VAL(“2.46111”)*2
will display 4.92222
6.12

BASIC ERROR MESSAGES

6.12.1

Compiler Error Messages

The following error messages are generated by the BASIC compiler:
MISSING LINE NUMBER

ERROR IN CHAIN STATEMENT

ERROR IN DEF STATEMENT

OUTPUT FILE ERROR

ERROR IN DIM STATEMENT

PUSHDOWN STACK OVERFLOW
STRING LITERAL TOO LONG

ERROR IN FILE NUMBER OR NAME

INCORRECT FOR STATEMENT

BAD SUBSCRIPT OR FUNCTION ARG

ERROR IN FUNCTION ARGS

SYMBOL TABLE OVERFLOW

ERROR IN IF STATEMENT

SYSTEM INCOMPLETE

I/C ERROR

PROGRAM TOO BIG

MISSING EQUALS SIGN IN LET

TOO MUCH DATA IN PROGRAM

STATEMENT TOO LONG

TOO MANY CHARS IN STRING

MULTIPLY DEFINED LINE NUMBER

ERROR IN UDEF STATEMENT

MISSING END STATEMENT

FOR STATEMENT WITHOUT NEXT

OPERAND EXPECTED, NOT FOUND

UNDEFINED STATEMENT NUMBER

PARENTHESIS ERROR

USE STATEMENT ERROR

OPERAND OF MIXED TYPE

CHARS AFTER END OF LINE

NEXT STATEMENT WITHOUT FOR

6.12.2

Run-Time System Error Messages

The following error messages are generated by the BASIC run-time system:
BO

NO MORE BUFFERS AVAILABLE

TOO MANY NESTED GOSUBS

IN CHAIN, DEVICE NOT FOUND

ILLEGAL ARG IN UDEF

IN CHAIN, FILE NOT FOUND

ILLEGAL DEVzFILENAME

CHAIN ERROR

INQUIRE FAILURE

READING PAST END OF DATA

TTY INPUT BUFFER OVERFLOW

DEVICE DRIVER ERROR

TAKING LOG OF NEGATIVE NUMBER

NO MORE ROOM FOR DRIVERS

DRIVER ERROR WHILE OVERLAYING

ATTEMPT TO DIVIDE BY ZERO

NUMERIC OR INPUT OVERFLOW

LOGICAL END OF FILE

ILLEGAL ARG IN POS

NEGATIVE NUMBER TO REAL POWER

READING PAST END OF FILE

6-57

BASIC

ENTER ERROR

CONCATENATED STRING TOO LONG

USING FILE ALREADY IN USE

STRING TOO LONG OR UNDEFINED

CLOSE ERROR

READING STRING FROM NUMERIC FILE

FETCH ERROR

STRING TRUNCATION ON INPUT

CLOSING OR USING UNOPENED FILE

SUBSCRIPT OUT OF RANGE

WRITING STRING INTO NUMERIC FILE

FIXING NEGATIVE NUMBER

ILLEGAL FILE NUMBER

READING VARIABLE LENGTH FILE

FIXING NUMBER > 4095

WRITING PAST END OF FILE

RETURN WITHOUT GOSUB

6-58

CHAPTER 7

08/78 FORTRAN IV

7.1

SYSTEM OVERVIEW

08/78 FORTRAN IV provides the full capability of the FORTRAN IV language. In describing 08/78 FORTRAN,
this chapter assumes that you are familiar with the material previously presented on the 08/78 operating system.
FORTRAN IV may be invoked through the commands COMPILE, LOAD, and EXECUTE. (The functions of
these commands are described in detail in Chapter 3.) In addition, your FORTRAN program usually reads and
writes data files under the supervision of the run-time system; FORTRAN [/0 files are treated separately in the
section on the run-time system (Section 7.2.3).

7.1.1

Source Programs

An alphanumeric file containing FORTRAN IV statements is called a source program. Source programs are usually
'

created with the 08/78 Editor (Figure 7-1).

TERMINAL
INPUT

V
.CREATE

EDITOR

POWERFT

Figure 7-1
7.1.2

Creating a FORTRAN IV Program

Using the EXECUTE Command

A FORTRAN source program may be executed

by successively (1) calling the compiler to convert the source

program into a relocatable binary file, (2) calling the loader to link and relocate the binary file, and finally (3)

calling the run-time system to load the program and supervise its execution. OS/ 78 FORTRAN IV provides a
program chaining feature that can usually simplify or eliminate this sequence of explicit program calls. When chaining
is invoked, the first system program to be executed
automatically calls the second program in the compilation/
loader/run-time system sequence and so forth. In this manner, simple FORTRAN programs may be compiled,
loaded, and executed, all as the result of typing the single command EXECUTE (Figure 7-2).

7-1

0S/78 FORTRANIV

POWERFT

.EXECUTE

COMPILES

EXECUTES

Figure 7-2 Executing a FORTRAN IV Program
FORTRAN programs containing subroutines cannot be adapted to the chaining facility because they require
additional input specifications at some point. In general, however, it is usually most convenient to chain from the

loader to the run-time system (combining relocation, loading, and execution).
Errors encountered by the various system programs do not result in the termination of program chaining unless the
error

is such that it is impossible for execution to continue. This permits the system to locate and identify as many

possible before returning control to the Monitor. When chaining is requested, intermediate output files
produced by one system program are automatically deleted after they have been read as input by the next program
errors as

in the chain sequence. This optimizes storage requirements.

7.1.3

Compiling

Once a source program has been prepared, it can be input to the compiler which translates each FORTRAN
statement into one or more relocatable instructions (see Figure 7-3).

Compilation is accomplished in three passes. System program F4.SV begins compilation by building a symbol table
and generating intermediate code. F4 chains to PASSZ.SV automatically, and PASSZ calls PASSZO.SV to complete
the translation into a special assembly language. If a source listing was requested, PASS2O chains to PASS3.SV
automatically, and PASS3 generates the listing. PASS2, PASSZO and PASS3 may not be accessed directly.
The compiler output produced is then automatically assembled by the RALF Post-processor which completes the
translation of FORTRAN IV source statements into relocatable computer instructions.
7.1.4

The relocatable binary file produced by compilation is a machine language version of a single program or subroutine.
This file must be linked with its main program (if it is a subroutine) and with any other subroutines, including

system library subroutines (FORLIB.RL) that it needs in order to execute. The system program LOAD.SV (the
FORTRAN IV loader) accepts a list of relocatable binary file Specifications from the console terminal and builds a
loader image file containing a relocated main program linked to library components required for execution. (See

Figure 7-4.)
The loader image file is executable except for run-time I/O specifications. It may be stored on any mass storage
(directory) device and executed whenever desired. The loader also produces an optional symbol map that indicates

the memory storage requirements of the linked and relocated program.

OS/78FOR TRANIV

POWER.FT

COMPILER
.COMPI LE

————————————

POST-PROCESSOR

POWERRL

Figure 7—3 Compiling a FORTRAN IV Program

POWERiRL

.LOAD

LOADEFI

LOADER

IMAGE

POWERLD

FILE

EXECUTE

RUNvTIME SYSTEM

II
EXECUTION

Figure 7-4 Loading and Executing a FORTRAN IV Program

0S/ 78 FORTRANIV
7.1.5

Run-Time System

The loader image file produced by the FORTRAN IV loader is read and executed by system program FRTS.SV

(the FORTRAN IV run-time system). This program also configures an I/O supervisor to handle any FORTRAN
input or output in accordance with run-time I/O specifications. This arrangement makes the full 1/0 device
independence of the OS/ 78 Operating system available to every FORTRAN IV program, and permits FORTRAN
programs to be written without concern for, or even knowledge of the hardware configuration on which they will
,

be executed.
The run-time system assigns I/O device handlers to the I/O unit numbers referenced by the FORTRAN program,

allocates [/0 buffer space, and also diagnoses certain types of errors that occur when the loader image file is read into
memory. If no load time errors are encountered, the run-time system starts the FORTRAN program and monitors
execution to check for run-time errors such as data I/O, numeric overflow, and hardware malfunctions. Run—time
errors are identified at the

console terminal, and, when a run-time error occurs, the system also provides an error

traceback to identify the full sequence of FORTRAN statements that terminated in the error condition.
The compiler, loader, and run-time system each accept standard OS/78 option specifications. The option specifications are alphanumeric characters that may be thought of as switches whose presence or absence enables or

disables certain program features and conventions. For example, specifying the /N option to the compiler
suppresses compilation of internal statement numbers (described in Section 7.2.1), thereby reducing program
memory requirements (at the cost of disallowing full error traceback during execution).

7.1.6

System Library

The FORTRAN IV system also includes FORLIBRL, a library of FORTRAN functions and subroutines. Almost
every FORTRAN program executes calls to library functions and subroutines that perform such tasks as mathe
matical function evaluation, data I/O and numeric conversion. When the loader recognizes that a program or
subroutine has called a library component, it copies a relocated version of the referenced

library routine into the

loader image file and links it to the calling routine.
7.2

08/78 FORTRAN IV OPERATION

It is important to identify the three processes that must be

performed in the proper sequence to execute a
FORTRAN source program. These processes are compilation, relocation, and execution. A detailed
description of
each process is given in the following sections.

It is also important to identify the types of input that must be supplied to each
process and the types of output,
along with the standard FORTRAN IV extensions, that are produced.

Table 7-1 lists the standard file extensions used to identify various
Table 7-1

types of source and system-generated files.

Standard FORTRAN IV File Extensions

Extension

File Type

.FT

FORTRAN language source file.

.RA

RALF assembly language file.

.RL

Relocatable binary (assembler output).

.LD

Loader image.

.LS

Listing.

.TM

System temporary file. Created by certain multipass
programs and normally deleted automatically after use.

.MP

Load map listing file.

7-4

05/78 FORTRANIV
Table 7-2 briefly illustrates the 08/78 operational process used for FORTRAN IV, showing the input files, the
process and the resulting output.

OS/78 FORTRAN IV Operational Process

Table 7-2
L

Input

Process

Output

FORTRAN Source Programs (FT)

COMPILATION

l. Compiler
Assembly Language File (RA)
2.

Assembler

Relocatable Binary File (RL);

Optional Listing File (LS)
Relocatable Binary File (RL)

RELOCATION

Loader Image File (LD);
Optional Loader Map (MP)

Loader Image File (LD)

EXECUTION

Executed Program

Note that the compilation is a two-step process where the source program, after being compiled, is automaticaliy

assembled by the system. Once a program has been written and debugged, it may be stored as a loader image file
and executed whenever required without the need for further compilation or relocation.
FORTRAN IV Compiler

7.2.1

The FORTRAN IV compiler accepts one FORTRAN source language program or subroutine as input. It then
examines each FORTRAN statement for validity, and produces a list of error diagnostics plus a relocatable binary

version of the source program, along with an optional annotated source listing, as output. A job containing one or

subroutines may be run by compiling and assembling the main program and each subroutine separately, then
combining them with the loader. When compilation is completed, the compiler automatically chains to the
more

postprocessor.
An internal statement number (ISN) is assigned to each FORTRAN IV statement sequentially, in Octal notation,

beginning with ISN 2 at the first FORTRAN statement. When an error is encountered during compilation, the
compiler prints a 2-character error code, followed by the ISN of the offending statement, on the console terminal
during pass 2. When a listing is produced, an extended error message is printed below every erroneous statement in
the listing. Certain errors cause an immediate return to the Monitor, thereby preventing the printing ofa listing file.
Table 74 lists the FORTRAN compiler error messages and describes the error condition indicated by each message.
The RALF postprocessor is called automatically to assemble the output of the compiler. The postprocessor reads
the special assembly language file produced by the compiler as input, generates a binary relocatable file, and routes
it to the first output file specified to the compiler. If this file has a null extension, the default extension .RL is

supplied.
Using the COMPILE Command
Compilation of a source program is done by typing a COMPILE
command. This command has the following format:
7.2.1.1

.COMPILE DEV:RALF .RL,DEV:LIST.LS,MAP.MP<INFIL.FT(Options)
where

DEV:RALF.RL

is the relocatable binary RALF module;

DEV:LIST.LS

is the annotated listing of the compiler output;

MAPMP

is the loader symbol map:

7-5

0S/78 FORTRANIV
<(left-angle bracket)

separates output file specification from input file specification

INFIL

is the input file; and

Options

is any string of alphabetic characters that designates run-time options desired.

The parenthesis may be omitted if each run-time option specification character is preceded with a slash (/). If
you enter the first output file name with a null extension, the compiler appends the default extension .RL. If

the second'output file is a directory device file with a null extension, the assembler appends the default extension
.LS. If no listing or map file is specified, none will be generated. Omitting all output specifications causes the
compiler to generate a relocatable binary file that has the same name as the input file but with an .RL
extension.
The compiler accepts the run-time option specifications (Table 7-3), any combination of which may be requested by

entering the appropriate alphabetic character(s) in the command string. Any Options recognized by the loader or the
run-time system may be entered along with the compiler Options. The options are passed to the loader automatically
unless chaining is suppressed (by an error condition or omission of the /L option specification), in which case they
are ignored.
Table 7-3

FORTRAN IV Compiler Run-Time Options

Operation

Option
/N

Suppress compilation of ISNs. This option reduces program memory requirements
by two words per executable statement; however, it also prevents full error
traceback at run time.

Optimize cross-statement subscripting during compilation. This Option should not
be requested when any variable that appears in a subscript is modified either by

referencing a variable equivalent to it or via a SUBROUTINE or FUNCTION call
(whether as an argument or through COMMON):
/G

Chain to the loader when assembly is complete and chain to the run-time system

following creation of a loader image file (equivalent to the EXECUTE command).
/L

Chain to the loader when assembly is complete. If the /L option is not specified,
the system will return to the Monitor upon completion.

7.2.1.2

Examples of Compilation

The following examples illustrate the use of the various options with the

COMPILE command:

.COMPILE PROG

Compiles PROG.FT and produces RALF module PROG.RL on DSK:

.COMPILE PROG/ G

Compiles PROG.FT into PROG.RL, links it into PROG.LD, then loads
it into memory and executes it. No listing files are produced. This
command is the same as EXECUTE PROG.

.COMPILE PROG,LIST<DSK:PROG
7.2.1.3

Compiler Error Messages

—

Produces PROG.RL and a compiler listing file called LIST.LS.

During compilation pass 2, error messages are displayed at the terminal as a

2—character error message followed by the ISN of the erroneous statement. Typing CTRL/O at the terminal
suppresses the printing of error messages. If a listing was requested, an extended error message is appended to the

listing, immediately following the erroneous statement, during pass 3. Except where indicated in Table 7-4, errors
located by the compiler do not halt processing.

7-6

0S/78 FORTRANIV
Table 7-4

Compiler Error Messages

Error

Code

Meaning

More than six subroutine arguments are arrays.

Bad ASSIGN statement.

Bad dimensions (too big or syntax) in DIMENSION, COMMON or type declaration.

Illegal in BLOCK DATA Program.

Bad COMPLEX literal.

Syntax error in COMMON statement.

Bad syntax in DATA statement.

This type of statement illegal as end of DO 100p (that is, GO TO, another DO).

Bad DEFINE FILE statement.

Hollerith field error in DATA statement.

DATA list and variable list are not same length.

DO—end missing or incorrectly nested. This message is not printed during pass 3.
If it is followed by the statement number of the erroneous statement, rather than the
lSN.

Syntax error in D0 or implied DO.

DO loop parameter not integer or real.

Syntax error in EXTERNAL statement.
Syntax error in GO TO statement.
Assigned or computed GO TO variable must be integer or real.

Hollerith field error.

Error reading input file. Control returns to the Keyboard Monitor.

Logical IF statement cannot be used with DO, DATA, INTEGER, etc.
Argument of logical IF is not type Logical.
Input line too long, too many continuations.
Misspelled keyword.
Multiply defined line number.
Mismatched parenthesis.
Expected operand is missing.
Mixed variable types (other than integer and real)
Error writing output file. Control returns to the Keyboard Monitor.
Illegal Operator.
Type / operator use illegal (for example. A.AND.B where A and/or B not typed
Logical).
Compiler stack overflow; statement too big and/or too many nested loops.

LI
LT
MK

ML
MM
MO

MT
OF
OP
OT
PD
PH

Bad program header line.

QL
QS

Nesting error in EQUIVALENCE statement.
Syntax error in EQUIVALENCE statement.
Attempt to redefine the dimensions of a variable.
Attempt to redefine the type of a variable.
Syntax error in READ/WRITE statement.

Bad arithmetic statement function.

Illegal subroutine name in CALL.
Error in subscript expression, that is, wrong number,
syntax.
Compiler symbol table full, program too big. Causes an immediate return to the

RD
RT

SS
ST

Monitor.
SY
TD

System error, that is, PASSZOSV or PASS2.SV missing, or no room on system for
output file. Causes an immediate return to the Monitor.
Bad syntax in type declaration statement.

Continued on next page

7-7

0S/78 FORTRANIV
Table 7-4 (Cont.)

Compiler Error Messages

Error
Code

Meaning

Undefined statement number. This message is not printed during pass 3. It is
followed by the statement number of the erroneous statement, rather than
the ISN.

Version error. One of the compiler programs is absent from SYS or is present
in the wrong version.

7.2.2

FORTRAN IV Loader

The FORTRAN IV loader accepts up to 128 relocatable binary modules as input and links the modules, along with
any necessary library components, to form a loader image file that may be loaded and executed by the run-time
system. Once all relocatable binary modules and library components have been assigned to some portion of memory

and linked, absolute addresses are assigned to the relocatable binary text.
If the /L or /G option is specified with the COMPILE command, the loader is called automatically to relocate the

output of a successful compilation.
The loader must be called separately to link a set of previously compiled relocatable binary modules by using the

LOAD command.
When several files or subprograms are loaded, the format of the LOAD command is as follows:

.LOAD DEV:OUTFIL.LD,MAP.LS<DSK:INFILl.RL,...,INFIL9.RL (options)
When programs are individually loaded the format is as follows:
.LOAD lNFIL.RL

OUTFI L.LD is the loader image output file and MAP.LS is the loader symbol map output file. The input files
may be either relocatable binary modules or a library file, and“0ptions” is a string of alphabetic characters that

designates any run-time options desired.
If the output file is not specified, the input file is loaded and given the default extension .LD by the system. The

loader symbol map is routed to the second output file provided that a second file is specifically defined. If this is
a

directory device file with no extension, the default extension .LS is assumed.

The-system can only accept nine input file specifications for each command line. If additional files must be input,
the /C (continue) option is used, since it permits the additional files to be put on the following line. Other Loader
run-time options are given in Table 7-5.
A LOAD command is usually terminated by pressing the RETURN key. If a LOAD command is terminated by the

ESCape key, the Command Decoder, a system program, will be called (see Appendix D for a description of the
Command Decoder). The Command Decoder indicates that it is running by prompting with an asterisk (*) on the
terminal screen. The function of this program is to allow the system to accept file l/O specifications.

Pressing

the ESCape key again executes the command line. This feature is discussed in Section 7.2.3.

Any error recognized by the loader during generation of a loader image file results in an error message, immediately
following the input specification line that caused the error condition. Table 7-6 lists the loader error messages
and describes the error condition indicated by each message.

7-8

05/ 78 FOR TRAN IV
Table 7-5

Loader Run-Time Options

Operation

Option

Continue the current line of input on the next line of input. When specifying input files to

the loader, there may be more than nine files in a command line. The C option permits the

additional files to be put on the following line.

Treat the current line as the last line of input, and chain to the FORTRAN IV run-time

system when finished.
Include system symbols in the loader symbol map.

System symbols are identified by an

initial number sign (#). This option is only valid when a symbol map output file was

specifically defined.
7.2.2.1

Examples of Loading

—

The following examples illustrate the use of the various options with the

LOAD command:
Loads POWERRL and produces a loader image file

.LOAD POWERRL

POWER.LD.

.LOAD RXAl :POWER.RL/G

Loads POWERRL, produces a loader image file

POWER.LD, chains to the run—time system, and executes
the program.

.LOAD PROG.LD,RXA1:MAP<MAIN.RL,SUBA.RL

Loads and links MAIN.RL and SUBA.RL to produce

PROG.LD, and produces MAP.LS, a loader symbol map
output file, on diskette RXAl. Using the type command
.TYPE RXAl :MAP will display the MAP file on the
screen.

Typing .EXE PROG.LD will execute the program.

.LOAD PROG.LD<PROG.RL,PROGI .RL -*(more file specifications)
Loads and links two input files to produce PROG.LD, then
calls a special system program for accepting file l/O

specifications (See Section 7.2.3). The second ESCape
executes the entire command.

7.2.2.2

Loader Error Messages

loader image file.

The loader displays error messages on the terminal during generation of a
Except where indicated in Table 7-6, loader errors are fatal. The loader returns to the Monitor
~

when a fatal error condition is encountered.

7-9

0S/78 FORTRANIV
Table 7-6

Loader Error Messages

Error Message

Meaning

BAD INPUT FILE

An input file was not a relocatable binary module.

BAD OUTPUT DEVICE

The loader image file device was not a directory device, or the symbol map

file device was a readbnly device. The entire line is ignored.
The L option was specified on a line that contained some file other than a

MIXED INPUT

library file. The library file (if any) is accepted. Any other input file
specification is ignored.
NO MAIN

No reloeatable binary module contained the section #MAIN.

OVER CORE

The loader image requires more than 16K of memory.

OVER IMAG

Output file overflow in the loader image file.

OVER SYMB

Symbol table overflow. More than 253 (decimal) symbols in one FORTRAN
job.

TOO MANY RALF FILES

More than 128 input files were specified.

The symbol is referenced but not defined.

Multiple Entry.

7.2.3

The symbol is multiply defined.

FORTRAN [V Run-Time System (FRTS)

The run-time system reads, loads, and executes a loader image file produced by the loader. It evaluates arithmetic

and logical operations. It also configures a software [/0 interface between the FORTRAN IV program and the

08/78 Operating system, and then monitors program execution to direct I/O processes and identifies certain types
of run-time errors. The run-time system is automatically called to load and execute the loader image file produced
by the loader whenever the /C option is specified to the loader. When chained to the loader, the run-time system
reacts in one of two ways. If the LOAD command line was terminated by pressing the RETURN key, the program
is executed. If the LOAD command line was terminated with an ESCape, a system program called the Command
Decoder is called and indicates that it is running by printing an asterisk (*).
In response to the asterisk, file specifications can be input to the run-time
system. This allows a source program to
be written that refers to I/O devices as integer costants or variables. Such a
program may be compiled, assembled,
and loaded into an image file. This image file may be run any number of times, each time
specifying different actual
peripheral devices. Thus logical unit 8 may refer in one run to the console terminal and in another run to a diskette

file.
Of the nine I/O unit numbers available under FORTRAN IV, two are initially

assigned to FORTRAN internal device

handlers by the system:

I/O Unit

Internal Handler

Comments

Line printer

LA78 only.

Console terminal

Double buffered output, single character input.

7-10

05/ 78 FOR TRAN IV
The FORTRAN internal handlers listed above are not the same as the 05/78 device handlers. The FORTRAN
internal handlers are designed for ASCII text only and will not transfer noncharacter data.

Additional unit numbers may be assigned, in addition to those listed above, to the FORTRAN internal device
handlers by typing (in response to the asterisk generated by the Command Decoder).

/n=m
where

is a different unit number (I to 9) that is also to be assigned to that internal handler; and

is the I/O unit number (3 or 4) of the internal handlers.

This specification causes all program references to logical unit 11 to perform 1/0 to device m. For example,

/6=3

Assigns the FORTRAN internal line printer handler as 1/0

unit number 6, in addition to unit

number 3.

/3=4

Assigns I/O unit number 3 to the FORTRAN console terminal handler instead of the internal
line printer handler.

I/O unit numbers may be assigned to 03/78 device handlers for nondirectory devices by typing (in response
to the asterisk generated by the Command
Decoder).
DEV: / n

where
DEV:

is the standard or assigned designation for any supported nondirectory device; and

is an I/O unit number (1 to 9).

For example,

LQPz/3

Specifies the 08/78 LQP line printer handler to be used as device #3 instead of the
FORTRAN internal line printer handler.

Existing directory device files may be assigned [/0 unit numbers by typing (in response to the asterisk
generated by the Command Decoder)
DEV: FILEEX/n

Where

DEV:FILE.EX

is the standard OS/78 designation for an existing directory device file; and

is an I/O unit number (1 to 9)‘

For example,
RXAl :FORIO.TM/4

Assigns unit number 4 to Diskette file FORIO.TM rather than to the
FORTRAN internal console terminal handler, where FORIO.TM is an
existing file on Diskette unit 1.

0S/78 FORTRANIV
A directory device file that does not presently exist may be assigned a FORTRAN I/O unit number in the
same manner

by entering it as an output file on the specification line; however, only one such file may be

created on any particular device. For example:

Assigns unit number 9 to file DSK:FORIO.TM, which has not been created

FORIO.TM</9

at load time.

In any case, only one device or file specification is permitted on each line, and no more than six directory

device files may be created by the FORTRAN program. Excess files after the sixth are accepted and written,

but they will not be closed. If a file created by the program has the same file name and extension as a

pre-existing file, the old file is automatically deleted when the new file is closed.
The Command Decoder “[n]

”

specification may be used to optimize storage allocation when assigning files

that do not yet exist, where n is a decimal number that indicates the maximum expected length of the file,
in blocks.

Each time a run—time I/O specification is terminated by pressing the RETURN key, the Command Decoder is

recalled to accept another specification. When a specification is terminated by the ESCape key, the program
is run.

The following examples illustrate the use of device and file specifications.

Exanuflel:
C

PROGRAM

HRITE.FT
DIMENSION
INTEGER

ILT(200)

ILT

SPECIFY LOGICAL
(2910)

WRITE

FORMAT

DU 3 I$19200
ILT(I)KI**2
NRITE

(1X9’URITING

(8920)

FORMAT

NOU

READ (9920)
OUTPUT FILE

RENIND

NUMBER

AND

FOR

READING

TTY

INSTEAD

ASCII

SEQ.

MASS

STORAGE

DATA

FILE’)

(ILT(I)vIflIv200)

(le//v20(10I79/))
8

READ

NRITE

UNIT

DATA

FILE

DATA.DA

FROM

DEVICE

ILT
T0 TTY

(4y20)

ILT

END
The above program raises 200 sequential numbers to the power of two. The following sequence of specifications
are

typed using the COMPILE command as follows:

Lump“ :1: If we 1' rise: .F‘T/G
:1: R'Xri‘. .l. : 1mm DA'CI/E}

yam-4

The /G option in the command line will call the run—time system to execute the program.

Pressing the ESCape
key after the command calls the Command Decoder, allowing device and file specifications to be declared.
In this case, the terminal (4) is assigned to unit number 2 while the load image file is assigned number 8.
A file DATA.DA is created on diskette 1 that contains the results of the
program. The contents of this file
are

then displayed on the terminal screen. The command and the specification strings are executed by the

second ESCape.

7-12

03/78 FORTRANIV
Example 2:
The second example shows an output file RAY.DA being created on the diskette by PROG1.FT, and then

being read from the diskette by PROG2.FT.
(I:

THIS

lr'lii‘tlt’ilitfirfi

INTO

lei

If! IMF "\1 ‘1‘ TON
I N. T l

INSET I HIE}:

WRITEEZES
{ZEGLLED

FILE

r31

I-"CIEICUl-Tilil

(III?

400

VARIABLES

RAY é 1W1

Rt’W II 400.3

R n "i’
I” I LE

1|.

( II. .v 400 5: U r “J?!

...J:I:-‘ Zl.
DU
5

'53

;;;.~

'1
.-

'f It 5' \J )

NF: I TIES
C t-‘i L. L.

4- ()0

F: r1} Y Ii I ) =-'=

If Rrfit Y ( I .3 ., I31]. '1 ~40 f.) )

If: X I I"

END

The above program writes a record into file RAY.DA. The following command and

specification strings

are

typed to accomplish this.
I11‘l13iiilt3].

:CUMF‘ILJEEI

FT/G

It I li'x'lfiil'IL'T

ESC

ERAY

m: C 135 S “x

l“ I LIE:

Ilrft'ii/ 1
Rfll‘)’

RISER-h II

t7,

IIIIEEIU I CE [.7 R 153: {-1 T 133: [1 BY F'RIEEIU I UUS
IIIIIIMlEEINE'” ION RAY ( 4 (1'0?!
I NT E [317:1 Fk‘

R (W

limit” I. NE

FI LE

Illfi

FR OH

H (3185:")

ST ORAL-3153:

F'RtilGli‘mti

2!. C I). y 400 r U y “1)

il.

All!
21.00

C'-

(ll.

IZZ'ORMAT

’

(liiiflr‘t’filll) ,j[::.-:;[ y 400)

( 1H y .f'i’ v 40': 2!. O I 6 v x”) .‘I

IIIUMF'

(SKIIN'HEZV‘H'S

UR I TIES:

( 4 y 100 )

RAY

C A I... l...

Iifir‘l'fv‘t

F I LE

ONTO

TTY

E: X I T

EEINII
The above program then reads out the results of PROG1.FT, and displays the contents of file RAY.DA on

the terminal screen. This is done by typing the following command line.

if: UMP I l... Eff

F5“ F'x' [I] {E} 133 . F T/ (3

*I‘K'AY . 1301/ 1

Although existing files are specified as though they were input files and nonexistent files are specified as though
they were output files, any file that has been assigned a unit number may be used for either input or output.
Run-time system option specifications are described in Table 7-7.
The run-time system recognizes two classes of error conditions. Certain errors are diagnosed while the load image

file is being read from a storage device and loaded into memory. Other errors may occur during execution of the
FORTRAN program. Both classes of run-time errors are identified on the console terminal. Table 7-8 lists the
FRTS error messages and describes the error condition indicated by each message. The run-time system error

traceback feature provides automatic printout of statement numbers lSNs corresponding to the sequence of
executable statements that terminated in an error condition. At least one statement number is always printed.
This number identifies the erroneous statement or, in certain cases, the last correct statement executed prior
to the error.

When a statement was compiled under the /N option, however, the system cannot generate

7-13

03/78 FORTRANIV
Table 7-7

Run-Time System Options

Option
/C

Operation
Carriage control switch. The first character on every output line is

processed as a carriage control character by all FORTRAN internal
handlers and also by the 08/78 handlers TTY and LPT. The first
character on every output line is processed as data, in the same
other character, by all 08/78 handlers except TTY
Entering a C option specification on the command line
that assigns an I/O unit number to a particular handler reverses the
processing of carriage control characters for that device. Thus,
manner as any

and LPT.

TEMP(2C)
assigns file DSKITEMP. as [/0 unit 2. The /C option causes the
first character of every output line to be processed as a carriage
control character. If C were not specified, these characters would
be processed as data.

/C/6=3
assigns the FORTRAN internal line printer handler as 1/0 unit 6,
as well as unit 3.
as a

on unit

The first character of every line will be processed

carriage control character on unit 3, and as a character of data
6.

Ignore the following run-time system errors, any of which indicates
that an error was detected earlier in the compilation loading process:
1.

Illegal subroutine call.

Reference to an undefined symbol.

meaningful statement numbers during traceback. When a statement is reached through any form of GOTO,
the line number for error traceback is not reset.

Thus, an error in such a line will give the number of the

last executed line in the error traceback.
The console terminal serves as FORTRAN l/O unit 4 for both input and output. Terminal input is auto—

matically echoed on the console screen. In addition, the run—time system monitors the keyboard continually
during execution of a FORTRAN program. Typing CTRL/C at any time causes an immediate return to the
Monitor. Typing CTRL/B branches to the system traceback routine, and then exits to the monitor. This
traceback routine generates a printout, similar to the error traceback, including the current subroutine, the
line number in the next higher level subroutine from which it was called, and so forth, to the main program.
This facilitates locating infinite loops when debugging a program. The following additional special
characters are recognized by the console terminal handler and processed as shown:
DELETE

Deletes last character accepted.

CTRL/ U

Deletes current line of input.

CTRL/Z

Signals end-of—file on input.

7-14

03/ 78 FORTRAN IV
Tentative output files (that is, files created by the FORTRAN program) are closed automatically upon successful

completion of program execution provided that one of the following conditions occurs:

An END FILE statement referencing the file was executed. FRTS assigns a file length
equal to the actual length of the file.
The last operation performed on the file was a write operation. FRTS proceeds as

A DEFINE FILE statement referencing the file was executed but an END FILE statement

though an END FILE statement had been executed.
was not executed.

to the length

Upon completion of program execution, FRTS assigns a file length equal

specified in the DEFINE FILE statement.

Execution of a REWIND statement does not close a tentative file, nor does it modify the tentative file length.
7.2.3.1

Run-Time System Error Messages

—

The run—time system generates two classes of error messages.

Messages listed in Table 7-8 identify errors that may occur during execution of a FORTRAN program and
errors that may be encountered when the run-time system is reading a loader image file into memory in
preparation for execution, or accepting [/0 unit specifications. Except where indicated, all run—time system
errors cause full traceback and an immediate return to the monitor. Nonfatal errors cause partial traceback,
sufficient to locate the error, and execution continues.

Table 7-8

Run-Time System Error Messages

Error Message

Meaning

BAD ARG

Illegal argument to library function.

CAN’T READ IT!

I/O error on reading loader image file.

D.F. TOO BIG

Random access file requirements exceed available storage.

DIVIDE BY 0

Attempt to divide by zero.
zero and

The resulting quotient is set to

execution continues.

EOF ERROR

End of file encountered on input.

FILE ERROR

Any of the following conditions occurred:
1.

A file specified as an existing file was not found.

A file specified as a nonexistent file would not

More than one nonexistent file was specified on

fit on the designated device.
a

single device.

The file specification contained an asterisk (*)
as name or

extension.

FILE OVERFLOW

Attempt to write outside file boundaries.

FORMAT ERROR

Illegal syntax in FORMAT statement.

INPUT ERROR

Illegal character received as input.

I/O ERROR

Error reading or writing a file, tried to read from an output device,
or

tried to write on an input device.

Continued on next page

7-15

05/78 FORTRANIV
Table 7-8 (Cont.)

Run-Time System Error Messages

Error Messages

Meaning

MORE CORE REQUIRED

The space required for the program, the I/O device handlers,
the I/O buffers, and the resident Monitor exceeds the available
memory.

NO DEFINE FILE

Direct access I/O attempted without a DEFINE FILE statement.

NO NUMERlC SWITCH

The referenced FORTRAN [/0 unit was not specified to the
run-time system.

The first input file specified to the run-time system was not

NOT A LOADER IMAGE

loader image file.

Result of a computation exceeds upper bound for that class
of variable. The result is set equal to zero and execution

OVERF LOW

continues.

PARENS TOO DEEP

Parentheses nested too deeply in FORMAT statement.

SYSTEM DEVICE ERROR

I/O failure on the system device.

TOO MANY HANDLERS

Too many [/0 device handlers are resident in memory, or files

have been defined on too many devices.

7.2.4

USER ERROR

Illegal subroutine call, or call to undefined subroutine.
Execution continues only if the E option was requested.

UNIT ERROR

[/0 unit not assigned, or incapable of executing the requested
operation.

FORTRAN IV Library:

Library Functions and Subroutines

The 08/78 FORTRAN IV system contains a general purpose FORTRAN library named FORLIBRL.

FORLIBRL contains mathematical functions and miscellaneous subroutines.

Library functions and subroutines are called in the same manner as user written functions and subroutines.

This

section lists the library components that are available to FORTRAN programs and illustrates calling sequences,
where necessary.

Arguments must be of the correct number and type, but need not have the same name as those

shown in the illustrative examples. Certain library routines are used by the FORTRAN system programs and are not

available to a user’s FORTRAN program. These routines may be identified by the initial number sign (#) in the entry

point or section name, and are not included in the following:
ABS (ABSOLUTE VALUE)
ABS calculates the absolute value of a real variable by leaving the variable unchanged if it is positive (or zero)
and negating the variable if it is a negative.

7-16

OS/ 78 FORTRAN IV
ACOS (ARC-COSINE FUNCTION)
ACOS calculates and returns the primary arc-cosine (in radians) or a real argument in the range [-1.1]

according to the following relation:
IF x>0.0, ACOS(x)=ATAN

[SQRT(1— fig]
[SQRT(I—X2)]
X

IF x<0 .0, ACOS(x)=7r+ATAN

IF x=0.0, ACOS(x)=Tr/2.0

AINT (REAL TO INTEGER)
AINT is a truncation function. Given a real argument, it truncates the fractional part of the argument and
returns the integral part as an integer. This is accomplished by taking the absolute value of the argument,

deleting the fractional portion, and then restoring the original sign. AINT, IFIX, and INT perform identical
functions.
ALOG (NATURAL LOGARITHM)

ALOG calculates and returns the natural (Naperian) logarithm of a real argument greater than zero. Any
negative or zero argument returns an error message and a value of 0.0. The algorithm used is an 8-term
Taylor series approximation.
ALOGIO (COMMON LOGARITHM)

ALOGIO calculates and returns the common (base 10) logarithm of a real argument greater than zero.

Any

negative or zero argument returns an error message and a value of 0.0. The calculation is accomplished by
calling ALOG to compute the natural logarithm and executing a change of base.
AMAXO (MAXIMUM VALUE)
AMAXO accepts an arbitrary number of integer arguments and returns a real value equal to the largest of
the arguments.
AMAXl (MAXIMUM VALUE)
AMAXI accepts an arbitrary number of real arguments and returns a real value equal to the largest of the

arguments.
AMINO (MINIMUM VALUE)
AMINO accepts n arbitrary number of integer arguments and returns a real value equal to the smallest of
the arguments.
AMINl (MINIMUM VALUE)
AMINl accepts an arbitrary number of real arguments and returns a real value equal to the smallest of the

arguments.
AMOD (A MODULO B)
AMOD accepts two real arguments and returns a real value equal to the remainder when the first argument
is divided by the second argument. If the second
argument is not large enought to prevent overflow, an
error message

and a value of 0.0 are returned.

7—17

0S/78 FORTRANIV
ASIN (ARC-SINE)

ASIN calculates and returns the primary arc-sine (in radians) of a real argument in the range [- l, 1]

according

to the relation

ASIN(X)=ATAN(X/SQRT(1-X**2))
If the argument falls outside the range [-1, 1]

an error message

results.

ATAN (ARC-TANGENT)
ATAN calculates and returns the primary arc-tangent (in radians) of a real argument. The argument is first
reduced according to the following relations:

(1)
(2)
(3)
(4)

If x<2— 14,
If

atan(x)=x
atan(x)=l/x
atan(x)=1r/2—atan(l/x)
atan(x)=- atan(- x)

x>2_14,

If x>l .0,
If x<0,

and the arc-tangent is then computed by a power series approximation.

ATAN2 (ARC-TANGENT OF TWO ARGUMENTS)
ATAN2 accepts two real arguments, assumed to be an abscissa and an ordinate, and calculates the arc-tangent
of the quotient of the first argument divided by the second argument. This is accomplished by calling ATAN
to find the principal arc-tangent of the quotient and then adjusting the result,

depending on the quadrant in

which a point defined by the arguments falls, according to the following relations:

atan2(y,x)=atan(y/x)
atan2(y,x)=atan(y/x)— 7r
atan2(y,x)=atan(y/x)—7r
atan2(y,x)=atan(y/x)+7r

argument in first quadrant
argument in second quadrant
argument in third quadrant

argument in fourth quadrant
CGET (CHARACTER GET SUBROUTINE)
The calling sequence
CALL CGET (STRING,N,CHAR)
causes the

Nth character to be unpacked from STRING and stored in CHAR as a variable in the range [0,63]

where STRING is a character string in A6 format.

CHKEOF (CHECK FOR END-OF-FILE SUBROUTINE)
CHKEOF accepts one real, integer or logical argument. After the next formatted read operation, this argument will be set to nonzero if the logical end-of-file was encountered, or to 0 if the logical end-of-file was not

encountered. The following example illustrates the use of CHKEOF:

CALI...

CHI"; E OF ( E OF )

'"x'lEECIT-‘nl?

(N y 101 ) DATA

I F"

( ISEZUF 0 NE 0 0 )

T [I]

(99‘?

7-18

OS/ 78 FORTRANIV
CLOCK (INITIALIZE CLOCK SUBROUTINE)
The purpose of the CLOCK subroutine is to initialize the KK8-B real-time clock. The calling sequence for the
subroutine is as follows:
CALL CLOCK (functn,rate)
where function can have the value 0 or 8, specifying multiple or single A/ D channel input, respectively; the
value of rate is preset to 100 (specifying 100 Hz).

Any value of functn other than 0 or 8 causes an error

message; any value of rate other than 100 is ignored.

COS (COSINE FUNCTION)
COS calculates and returns the cosine of a real argument (in radians) by applying the identity:

COS(X)=SIN(X+ /2)
COSH (HYPERBOLIC COSINE FUNCTION)

COSH calculates and returns the hyperbolic cosine of a real argument according to the relations:
If [xl £88029

COSH(X)

1/2

EXP ( x )

If lxl > 88.028 and Ix!

COSH(x)
If lxl

—-

EXPOXI

—

1 0
'

EXP(x)

loge2<88.028
logeZ)

loge2>88.028

COSH(X)

3777377777778

and an error message is returned.

CPUT (CHARACTER PUT SUBROUTINE)
The calling sequence
CALL CPUT (STRING,N,CHAR)
causes CPUT to insert

CHAR as the Nth character in STRING, where STRING is a character

A6 format, and CHAR is a number in the

range

program illustrates the use of CGET and CPUT.

string stored in

[0,63] which is interpreted as a character. The following

7—19

05/78 FORTRANIV
IIIATI'Q

SIR!

’

HEY !

HR 1' TE C «’l a II. 00) STR
FUI'CMIQT (
l-IIEY I
"

100

’

M30 I I

arr-'36:?!

UR I TE ( 4 y 10]. ‘J
101

FIJRMQ'H

110

===-‘

mm...

I'llfi'Y I

1013?.

IllEZClIMm...

’

.‘J

l. v 4

(SIESIEIT ( SIR v I r: I CHAR )

b1 l’x‘ I 'I If}: '1 4 .u 101? ‘1
(3 [I N 'l' I N U E

I (I)

I [ll-lfltli'x'

l3 [fl R H (31 It I <3 )
I! (I1

1320

,.li~

34111 I

(3 fit L I...

I =3: l. 9 r8
CF'UT ( $53 T F: r I ., ...l T:

13 Q

(I: [I] N I I N Ll E

ii. (3'5

FUR Mfil' (

$53 T F:

UR I TE Ii 4 y 103
F.) r"! L l...

NEbJ

SIT-(I Nil-l

:5) c2; 1'

Iii? X .T. 'l'

EN I]

LCUMPILE
HEY!

HEY!

TCHRC/G
ascrr

HEY!

necrmmr

a
’

If?

,2 t.)
3 w
\

N E bl

S T R I N (3

B III E H “J l...

DATE (DATE SUBROUTINE)
DATE accepts three integer arguments, accesses the current system date, and returns an integer from 1 to 12

corresponding to the current month as the first argument, an integer from I to 31 corresponding to the
current day as the second argument, and an integer from 1970 to 1999 corresponding to the current year
as

the third argument.

DIM (POSITIVE REAL DIFFERENCE)
DIM calculates and returns the positive difference of two real arguments. If the first argument is larger than

the second argument, DIM returns the difference between the arguments; if the first argument is less than or

equal to the second argument, DIM returns 0.0.
EXP (EXPONENTIAL FUNCTION)
EXP calculates and returns the exponential function of a real argument. The algorithm uses a numerical
method after Kogbetliantz (IBM Journal of Research and Development, April, 1957, pp 110-5).

EXP3 (BASE RAISED TO AN EXPONENT)
EXP3 accepts two real or integer arguments, that is, a base and an exponent and performs the following

calculation:

a=be
FLOAT (INTEGER-TO-FLOATING-IOINT CONVERSION)
FLOAT accepts an integer argument and returns a real variable equal to the argument.

7-20

OS/ 78 FORTRAN IV
IABS (INTEGER ABSOLUTE VALUE FUNCTION)
IABS calculates and returns the absolute value of an integer variable by leaving the variable unchanged if it is

positive (or zero), and negating the variable if it is negative.
IDIM (INTEGER POSITIVE DIFFERENCE FUNCTION)
IDIM calculates and returns the positive difference of two integer arguments. If the first argument is larger
than the second argument, IDIM returns the difference between the arguments; if the first argument is less

than or equal to the second argument, IDIM returns a value of 0.

IFIX (FLOATING-POINT-TO-INTEGER FUNCTION)
IFIX is a floating-point truncation function. Given a real argument, it truncates the fractional part of the

argument and returns the integral part as an integer. IFIX, AINT, and INT perform the same function.
INT (F LOATING-POINT-TO-lNTEGER)
INT is a floating-point truncation function that performs the same function as AINT and IFIX.
ISIGN (INTEGER TRANSFER OF SIGN FUNCTION)
ISIGN accepts two integer arguments, calculates the absolute value of the first argument, and returns this
value if the second argument is positive (or zero), or the negative ofthis value if the second argument is

negative.
MAXO (MAXIMUM VALUE)
MAXO accepts an arbitrary number ofinteger arguments and returns an integer result equal to the largest of

the arguments.
MAXI (MAXIMUM VALUE)

MAXI accepts an arbitrary number of real arguments and returns an integer result equal to the largest of
the arguments.
MINO (MINIMUM VALUE FUNCTION)
MINO accepts an arbitrary number of integer arguments and returns an integer vaue equal to the smallest of
the arguments.
MINl (MINIMUM VALUE FUNCTION)
MINI accepts an arbitrary number of real arguments and returns an integer value

equal to the smallest of

the arguments.
MOD (INTEGER A MODULO B FUNCTION)
MOD accepts two integer arguments and returns an
ment is divided by the second argument.
error message and a value of zero are

integer value equal to the remainder when the first arguIf the second argument is not large enough to prevent overflow, an

returned.

SIGN (TRANSFER OF SIGN)
SIGN accepts two real arguments, calculates the absolute value ofthe first

argument, and returns this value if

the second argument is positive (or zero), or the negative ofthis value if the second

argument is negative.

SIN (SINE FUNCTION)
SIN calculates and returns the sine ofa real argument (in

radians). The argument is reduced to the first
quadrant, and the sine is then computed from a Taylor series expansion.

7-21

05/78 FORTRANIV
SINH (HYPERBOLIC SIGN)
SINH calculates and returns the hyperbolic sine of a real argument according to the following relations:
1

0.10<|x|<87.929, SINH(x)

|x|<0.10, SINH(x)

lxl>88.028, SINH(x)

l/2[EXP(x) —m

x+x3/6+ x5/120
=

[EXP(lxl

loge2)] '[signum(x)]

SQRT (SQUARE ROOT FUNCTION)

SQRT calculates and returns the (positive) square root of a positive real argument. Any negative argument
results in an error message.
TAN (TANGENT FUNCTION)
TAN calculates and returns the tangent of a real argument (in radians). This is accomplished by computing

the quotient of the sine of the argument divided by the cosine of the argument; thus, if the cosine of the

argument is zero, an error message is returned.
TANH (HYPERBOLIC TANGENT)
TANH calculates and returns the hyperbolic tangent of a real argument by computing the quotient of the
hyperbolic sine of the argument divided by the hyperbolic cosine of the argument.

TIME (READ TIME OF DAY)

TIME may be called as a subroutine with one real or integer argument, or as a function with a dummy argument.

7.3

It returns the elapsed time since the clock was started. This result will be in seconds.

RUNNING OS/78 FORTRAN IV PROGRAMS

The first example that will be used is the power of two program, POWERFT, that was created with the Editor,
and described in Chapter 4. It is repeated here for convenience.

FORTRAN

DEMONSTRATION

COMPUTE

AND

PUwERS

Twfl

DIMENSION

A<1é)
NRITE(4v15)
15

FORMAT(1H

y’POWER

TNO’)

30 leylé
D0
A(N)$2.**N

CONTINUE

FORMQTCIH

wRITE

(4v25)

(Npfi(N)vNW1v16)
’yIEv’$’7F10.1)

v’2**

STOP
END
7.3.1

Executing a Program

This program is executed by typing

LIZ}: X IS:

F' [I] DJ E. F:

F' T

7-22

awZSFORTRANrV
This command compiles, loads, and executes the program with the following results diSplayed on the screen:

POUER OF
in
2*#

7.3.2

Two

2**

4.0

2*#
2**

8.0
1a.o

2mm

32.0

gas

64.0

2**

128.0

2am

25a.o

are
2**

512.0

10m

1024.0

axm

11m

2048.0

2**

12m

4095.0

2**

13x

8192.0

2am

14x

16384.0

2am
2**

15m
16%

327é8.0
55536.0

Compiling a Program

Other commands can be used to compile and load the program in sequential steps, and they are required when

subroutines are used.
The first command is the COMPILE command. This command compiles the source program POWERFT and

generates a relocatable binary file. The format of the command is

LC 0 M F' I l... E:

F' O M IEEI R“ . F T

After execution of this command, the compiled file is listed in the directory as POWER.RL. A source listing of

the compiled file can be output to the terminal by typing

LCOMF‘ILK

F‘OLJEIR . RL , TTY 2 -=ZIF'Ob.lER . FT

This command line will display the following annotated listing on the terminal screen:

FORTRAN

COMPUTE AND PRINT
DIMENSTON 9(16)

0003

NRITE

0006

0007

0012
0013

DEMONSTRATION
POwERS

TwO

(4y15)

FORMAT(1H v’POwER
DO 20 Nfllvlé

TNO’)

fi(N)$2.**N
20

CONTINUE
“RITE (4’25)

FORMAT(1H

0010
0011

17wJUNm77

4Afififi

C
0002

0004
0005

(N9A(N)yNw1y16)

y’2**

’yIQv’m’yF10.1)

STOP
END

This file is useful in tracking any errors that may have been made in the

7-23

program.

PAGE

ONE

0S/78 FORTRANIV
The compiler options discussed previously may be specified when using the COMPILE command. For example, if
the /G option was invoked in the command line

LCUMF' I LIZ-I

F‘UUIEZF.‘ . F'T/ G

the program would be compiled, and then chained to the loader to create an image file, and finally executed by

being chained to the run-time system.
If the /L option is specified with the COMPILE command, the program will be chained to the loader, creating a
POWER.LD file, but will not be executed.
7.3.3

Creating a Loader Image File

After compilation, use the LOAD command to create a loader image file by typing

imam

POWER . RI...

Again, the various options previously described for the loader may be used with this command. After loading, type
_¢ F53 X IE}: (I) U T FBI

IF' CI w E R

I... III

This command will cause the program to be executed and display the results on the terminal screen. This is the
fastest method to use if the program does not have to be recompiled or relinked.
If a loader symbol map is desired, type
_» l...l".lfi'§lil

F'CWER

l... 1",! , T TY i ‘-'ZIF'0|;J|EZFC o R L

The following map for program POWER will be displayed on the terminal screen:

17""JUN"-~77

[ILMIIER

U240

S Y M B (I) L

U r55 L U F

L U l...

filiQGEFL'Fi‘

0011304

EXIT
FMAIN

0021313

10000

00
00

11000

LUL

:::.-

153T

U U L. Y

FWEE

(JULY

LENGTH

10601

LOCATION

The optional loader symbol map lists all symbols defined in the loader image file. The LVL and OVLY entries

apply only to 08/8, a superset of 08/78.
Following the alphabetical list of symbols, the loader prints the address ofthe first free memory location and the
length, in octal words. This information is useful for optimizing memory requirements.
7.3.4

Running Subprograms

The LOAD command is especially useful to link subprograms. Consider the program shown in Figure 7-5, which

computes the volume of a regular polyhedron when given the number of faces and the length of one edge. It
consists ofa main program and a subroutine. The subprogram does the required computation, using a computed

GO TO statement to determine whether the polyhedron is a tetrahedron, cube, octahedron, dodecahedron, or

icosahedron, then transfers control to the proper arithmetic expression for performing the calculation.

7-24

05/78 FORTRANIV
MAIN PROGRAM:

COMPUTE

THE NUMBER OF FACES
COMMON NFfiCESyEDGE

THE

NR1TE<4710)

FORMfiT(lH
REfifl(4y20)

VOLUME

r’TYPE

AND

POLYHEDRON

REGULQR
LENGTH

N0.0F

ONE

FQCES

9ND

AND

GIVEN

EUGE

ONE

LENGTH

EHGE’)

NFQCESyEHGE

FURNQT(IR§FB;S)

BALL SOLUE
GO TO 5
STOP
END
SUBPROGRAM SOLVE:
C

SUBROUTINE
CALLED

SUBROUTINE
COMMON

SOLVE

ANSNER

SOLVE

NFfiCESyEDGEyUOLUME

IF(NFQCE8.GT.RO)OOTO
CUEEUWEDGEXXK
GO

PROBLEM

SOLVE.FT

TO(($969(571y6v21673véy6yéy4véyévév6)96165489596)yNF'flCE-S

VOLUHERCUBEDWO¢117OS
GO TO 20
VOLUMEuCUBEfl
GO

UOLUMEWOUBEDXO.4714O
GO

UOLUMEmCUBEUX7.66312
GO TO 20

UOLUHEflflUBED*201817O
6

GO TO 30
NHITE(4V10)

FORMQT(1HO§'NO

NFQCES
REGULQR

POLYHEURON

HQS’VIKVle’FACES’)

RETURN
20

WRITE(4930)

FORMAT<1H09’UOLUMEm’9F10.2)

VOLUME

RETURN
WRITE(4910)
FORMGT

(1H09"UO

NOT

SPECIFY

THQN

FACES)

RETURN
END

Figure 7-5

Main and Subprogram for Calculating the Volume of a

7-25

Regular Polyhedron

05/78 FORTRANIV
If the number of faces of the solid is other than 4, 6, 8, 12, or 20, or if more than 20 faces are specified the subroutine displays an error message on the terminal. If the correct input parameters are typed in at the terminal

keyboard, the calculation is performed, and the answer is displayed on the terminal screen.
When subprograms are used, the main program and the subprograms must be individually compiled. For the

example program, both the main program and subprogram are compiled as follows:

lCOMF' ILE

MAIN . FT

LGGMF‘ I LE

SOLVE . FT

and

which create relocatable binary files.
The modules must now be loaded to make a single image file that can be executed. This is done as follows:

POLY . LII'iIMA IN . RL y SOLVE: 0 Hi.

LLOAIJ

This command creates a file POLY.LD and returns to the Monitor. Execute the program by typing

LEXECUTE

F'OLY

LII

which results in the program calling for the required input parameters as follows:

TYPE

NO .

FACES

AND

ONE

LENGTH

EDGE

Typing a “6” to represent a cube (preceded by a space since the field specification is 12) and “20” as one length
edge in accordance with the field specification requirement as follows:
b6)

20 o 0

results in the answer being displayed on the terminal screen, and a prompt for the next set of input parameters
as follows:

VOLUME
TYF'E

=-'=:

8000.00
NO.

[3F

F'fiCl-ZS

ANTI

ONE

LENGTH

Do not specify blanks or zeroes for the NFACES and EDGES variables.

7-26

EDGE

Type CTRL/C to return to the Monitor.

0S/78 FORTRANIV
7.3.5

Specifying I/O Devices

A special case of the LOAD command is that in which a system program called the Command Decoder is called up

and signifies it is active by printing an asterisk (*). This program allows the system to accept file I/O specifications.
For example, if the POWER.FT program were written with the output unit designation as a 3, that is,

DJRI TIE?

(’5 v1.5)

the output generated by program execution would be sent to the line printer. However, if your system does not

have a line printer available, the output unit designation is changed by typing

_._|... C] A [I

F' 0 l-JIEEIF.‘ . I... If! *3: F“ 0 DJ E: R

I3: L. 1’6

ESC

Then pressing the ESCape key (which echoes as a dollar sign) calls up the Command Decoder which prompts with an
unit 3 to the console terminal instead of the line printer. Pressing the
which
asterisk. Then

assigns I/O
type /3=4
will
ESCape key again executes the program and program output is sent to the terminal. Thus, the command line
appear as follows:

.LL UAIII

7K/ 13:34

F-“UNEIR ¢ L [I --‘C F‘ OWE: R . R‘ I... / [3

The Command Decoder program also allows you to store the output of an executed program to a file that has
not been created at load time.

For example,

_._l.-Cm I]

PC) I)! ER

I... III {I F‘ 0le R

*RXA 1 2HOL.II

FtL/G

will output the results of the program into a file called HOLD.TM on RXAl.

_._TYF'IEZ

RXA]. :HULII

TM 413/ 4

Typing

will display the contents of this file, that is, the results of the program POWER.F T on the terminal screen.
7.4

FORTRAN IV SOURCE LANGUAGE

The FORTRAN IV language is composed of mathematical-form statements constructed in accordance with precisely

formulated rules. FORTRAN (source) language programs consist of meaningful sequences of FORTRAN statements that direct the computer to perform specified operations and computations.

Each line of a FORTRAN source program contains three fields: statement number field, line continuation field, and
statement field. A fourth field, the identification field consisting of columns 73 to 80, is ignored by the compiler.

It may be used to number statements sequentially or for any other purpose.
A statement number consists of one to five digits entered in columns 1 to 5.

Leading zeros or blanks (leading and
triling) are retained on the listing, but otherwise ignored in this field. Statement numbers may be assigned in any
order, but they must be unique. Any statement referenced by another statement must have a statement number.
Statement numbers on specification statements are ignored.
If a FORTRAN statement is so large that it cannot conveniently fit into one statement field, the statement fields of
up to five additional lines may be used to specify the complete statement. The first line of a statement must

have a blank in column 6. Continuation lines must have some character other than a blank in column 6.

7-27

05/ 78 FOR TRAN IV
FORTRAN statements define arithmetic operations, call for input or output, and alter the sequence of program
execution. Any FORTRAN statement may appear in the statement field (columns 7 to 72). Blanks (spaces) are
ignored and may be used freely for appearance purposes except when they occur as alphanumeric data within a
FORMAT statement, DATA statement, or literal constant. A TAB at the beginning of a line or after the statement
number causes spacing to column 7. The compiler treats the first TAB after column 7 as a blank. It ignores any

input line that does not contain a TAB or at least six other characters in columns 1 to 72.
Comments explaining the program may be written in any format. A line that contains the letter C in column 1 is
interpreted as a line of comments. Comment lines are printed on all listings, but are otherwise ignored by the
compiler. A comment line must not immediately precede a continuation line. The TAB key may be used to locate
the statement field of each line.
7.4.1

Constants

The constants, variables, and expressions described below are basic to expressing data values in the FORTMN

lanugage. Five types of constants are used in 08/78 FORTRAN IV programs: integer, real, octal, logical, and
Hollerith.
7.4.1.1

An integer constant consists of from one to seven decimal digits written without a
Integer Constants
constants
must be preceded by a minus (-) sign; however, the plus (+) sign preceding a
Negative
constant
is
Embedded
commas and blanks are not allowed in integer constants. The following
positive
optional.
are some examples:
_

decimal point.

Examples:
0
+051
-440

6073

Integer constants must fall within the range ~2A23 to 2‘23-1 (-8,388,608 to 8,388,607 decimal). When used as
subscripts, integer constants are taken modulo 2‘12 (4096 decimal). The following are illegal as integer constants:
10.3

5,000
9000000
7.4.1.2

(decimal point)
(comma)
(outside acceptable range)

Real Constants

A real constant is an integer constant followed by a decimal point, a second string
of digits, and an optional exponent. The compiler uses only the leftmost six
digits, aside from leading zeros. A
minus (-) sign must precede a negative constant. The plus
a
real constant is
—

(+) sign preceding

positive

optional.

Real constants may be entered in exponential notation, as illustrated below, by
specifying a positive or negative
decimal value followed by the letter E and a l to 3 digit integer that may be positive,
or zero. The value

negative
preceding the letter E multiplied by that power of
10 indicated by the integer following the letter E. This notation eliminates
leading and trailing zeroes from very
large or very small real constants. The absolute value of any real constant must fall within the approximate range
10‘61 5 to 10615 (or zero). Some examples are as follows:
of the real constant is taken as the value of the decimal number

0.0
.579

—10.794
5.01303
5013+3

5013—3
5.0E0

(i.e.,5000.)
(i.e.,5000.)
(i.e.,0.005)
(i.e.,5.0)

5.50

7-28

0S/78 FORTRANIV
The following are not valid real constants.
6 ,5 1 7 .6

(comma)
(no decimal point or exponent)
(exponent must be integer)

131

20E1.5

An octal constant is a string of octal digits (0 to 7 only) preceded by the letter 0.
Only the 12 low-order digits are used by the compiler. Octal constants are valid only in DATA statements where
they are generally used to set bits for masking purposes. Some examples are as follows:

7.4.1.3

Octal Constants

DATA JOB/Ol/32/
DATA BASE/07777/
The two logical constants .TRUE. and .FALSE. have the internal values 1. and 0.,
Logical Constants
respectively. They may be entered in DATA or input statements as .TRUE. or .FALSE. (or abbreviations .T. or
.F.). The enclosing periods are part of the constant and always appear. Logical quantities are operated upon by
logical operators only.

7.4.1.4

7.4.1.5

Hollerith Constants

—

A Hollerith constant (or literal constant) is a string of ASCII characters. It may

be represented by one of two forms:

Form 1: HH character string
where n is the number of characters following the H.
Some examples are as follows:
SHWORDS
3H123
Form 2:

‘character string’

Some examples are as follows:
‘WORDS’
‘123

’

The apostrophe character that delimits a Hollerith constant in Form 2 may be included in the character
is immediately preceded by a second apostrophe.

string if it

Thus, ‘DON”T’ will be stored as DON’T.

A Hollerith value may be entered in a DATA statement or FORMAT statement as a

string of one to six ASCII

characters per integer or real variable.
7.4.2

Variables

A variable is quantity that is represented by a symbolic name. Arithmetic statements and ASSIGN statements are

used to change the value of a variable, by computation or assignment,

7-29

during program execution. I/O statements

0S/ 78 FOR TRANIV
and subroutine calls can also change the value of a variable. A variable name is a string of one to six alphanumeric

characters, the first of which must be alphabetic:
Valid Names

Invalid Names

1ACT

ALPHA

STANDARD

MAX

FILE

A34

#MAIN

(first character number)
(too long)
(space within name)
(# not alphanumeric)

There are three types of variables: integer, real, and logical. Definitions for these types correspond to definitions of
constants of the same type, that is, integer variables take on a value of from zero to any positive or negative integer

in the range -8,3 88,607 (decimal) to 8,388,607 (decimal); real variables contain a decimal point; and so forth.

Type classification is assigned to a variable explicitly via a type declaration statement or by virtue of the initial letter
ofits name. A first letter of I, J, K, L, M, or N indicates an integer variable. Any other first letter indicates a real
variable. The type declaration statement overrides the type indicated by the initial letter.
Variables can be either scalar (representing a single quantity) or array (representing many
Arrays
quantities with one name). An entire array is identified by its name, while a single element of the array is identified
by a subscript, in parentheses, following the array name.
7.4.2.1

Variable

Refers To

ARRAY (1)

The first element of a one-dimensional array named ARRAY.
The element located in the first row and the third column of a two-dimensional array

B(1 ,3)

named B.
The subscripts of an array variable can be integer constants or expressions. For example,
Subscripts
and
A(I+1 ,2*K+3*J) illustrate valid subscripts. The elements of an array must be of the same
A(l), A(NEW),
that
all
real
or all logical.
is,
type,

7.4.2.2

—

The extent of an array is determined by the dimensions it is assigned. This assignment may be made by means of a

DIMENSION or COMMON statement or as part of a type declaration statement.
7.4.3

Expressions

An expression is a combination of elements (constants, subscripted or nonsubscripted variables, and function refer-

ences), each of which is related to another by operators and parentheses. An expression represents one single value
that is the result of calculations specified by the elements and operators that make up the expression. An expression
may, itself, function as an element in another expression if it is enclosed in parentheses. The FORTRAN language
provides two kinds of expressions: arithmetic and logical.
7.4.3.1 Arithmetic Expressions
An arithmetic expression is a combination of constants, variables, and function
references separated by arithmetic operators and parentheses. In the absence of parentheses, algebraic operations
fl

within arithmetic expressions are performed in the following descending order:
F (x)

functions

exponentiation
unary minus

~—

and /

+ and

multiplication and division
addition and subtraction

equals or replacement sign

7-30

0S/78 FORTRANIV
Parentheses are used to change this order of precedence. An operation enclosed in parentheses is performed before
its result is used in other operations. In the case of operations of equal precedence, the calculations are performed

from left to right. Additional computations (such as sine, cosine, or Square root extraction) may be specified via a

function reference.
An arithmetic expression may consist of a single constant, variable, or function call, referred to as a basic element.
For example,
2 ,7 1828

Z(N)
TAN (THETA)

Any function reference acts as a basic element in an expression, since all functions return a unique value for any
given argument. The reference SQRT(4.), for example, always represents the value 2. in an expression.
Any arithmetic expression may be enclosed in parentheses and considered as a basic element. For example,

IF1X(X+Y)/2
(ZETA)
(COS(SIN(PI*EM)+X))
Compound arithmetic expressions may be formed using numeric operators to combine basic elements. For
example,
x+3

TOTAL/A
PI*EM
A basic element preceded by a plus (+) or minus (-) sign is also an arithmetic expression. For example,
+X

—(ALPHA*BETA)
(SQRT(— GAMMA»
—

With the exception of unary minus, no two arithmetic operators may appear in

sequence. For instance, X*/Y is

illegal.
Parentheses do not imply multiplication, thus (A+B) (C+D) is improper. This
expression must be written as

(A+B)*(C+D)
A typical numeric expression using numeric operators and a function reference, such as the

the roots of the general quadratic equation
-

b+\/b T2 ~4ac
23

might be coded as

(— B+SQRT(B**24.*A*C))/(2.*A)

7-31

expression for one of

03/ 78 FOR TRAN IV
The following examples illustrate conversion of other mathematical expressions into FORTRAN expressions.
a+5

A+5 .0

a.b

A*B

a*(b+2)
L“! 2.5

A**(B+2)
((B+D)/A)**2.5

In general, only real and integer quantities may be mixed in arithmetic expressions. No other type mixing is legal.

Logical variables and constants may only be operated upon by logical operators (.AND., .OR., .NOT., .XOR.,
.EQV.). Hollerith literals in expressions have type integer, with only the first six characters being used.
A logical expression combines logical constants, logical variables, logical function
Logical Expressions
references, and logical expressions, using the logical and relational operators given in Tables 7-9 and 7~lO,
respectively.

7.4.3.2

—

Logical operators can combine only basic elements whose type is logical. Relational operators compare units of type
integer or real. The value of such an expression will be of logical type (that is, .TRUE. or .FALSE.). The relational
operators .EQ. and .NE. may also be used with complex expressions. Complex quantities are equal if the corresponding parts are equal.
Table 7-9

Logical Operators and Their Meanings
Meaning

Logical Operator
.NOT.expr

Has the value .TRUE. only if the expression is .FALSE., and has the value

.FALSE. only if the expression is .TRUE.

exprl .AND.expr2

Has the value .TRUE. only if exprl and expr2 are both .TRUE., and has the
value .FALSE. if either exprl or expr2 or both are .FALSE.

exprl .OR.expr2

(Inclusive OR) Has the value .TRUE. if either exprl or expr2 or both are
.TRUE., and has the value .FALSE. only if both exprl and expr2 are
.FALSE.

expr1.XOR.expr2

(Exclusive .OR.) Has the value .TRUE. if either exprl or expr2, but not both,
is .TRUE., and has the value .FALSE. otherwise.

exprl .EQV.expr2

(Equivalence) Has the value .TRUE. if exprl and expr2 are both .TRUE. or
both .FALSE., and has the value .FALSE. otherwise.

Table 7-10

Relational Operators and Their Meanings

Relational Operator

Meaning

.GT.

greater than

.GE.

greater than or equal to

.LT.

less than

.LE.

less than or equal to

.EQ.

equal to
not equal to

.NE.

The enclosing periods are part of the logical and relational operators, and must be present.

7-32

05/78 FORTRANIV
A logical expression, like an arithmetic expression, may consist of basic elements or a combination of elements, as
in
.TRUE.

X.GE.3.14159

01'

TVAL.AND.INDEX

BOOL(M).OR.K.EQ.IMIT
where BOOL is a logical function with one argument or a singly dimensioned logical array. A logical expression may
also be enclosed in parentheses and function as a basic element.

Thus, the following expressions are evaluated

differently:

A.AND.(B.OR.C)
and

(A.AND.B).OR.C
No two logical operators may appear in sequence, except where .NOT..

Any logical expression may be preceded by

the unary operator .NOT. as in

.NOT.T

.NOT.X+7.GT.Y+Z

BOOL(K).AND..NOT.(TVAL.OR.R)
Logical and relational operations (unless overridden by parentheses) are carried out in the following order:
.GT.,.GE.,.LT.,.LE.,.EQ.,.NE.
.NOT.
.AND.
.OR.

.EQV.,.XOR.
For example, the logical expression

.NOT.ZETA**2+Y*MASS.GT.K-2.0R.PARITY.AND.X.EQ.Y
is interpreted as

(.NOT.(((ZETA**2)+(Y*MASS)).GT.(K-2))).OR.(PARITY.AND.(X.EQ.Y))
There are 16 logical operators theoretically possible between the logical expressions. Two of them are constants
(true and false) and four are unary operators (that is, the value of one of the two expressions is irrelevant to the
value of the operation). These six are marked by asterisks in Table 7-11. The
remaining ten operators can be
most conveniently represented as shown at the right of the table, with A and B
representing the two logical
expressions involved.

7-33

03/ 78 FOR TRAN IV
Table 7-11

Truth Table for Logical Expressions

Expressions involved:
Operands
A

7.4.4

FORTRAN IV Expression

Result

Function Name

FALSE

.FALSE.

AND

A .AND. B

A .AND. .NOT. B

.NOT. A .AND. B

XOR

A .XOR. B

A .OR. B

NOR

.NOT. (A .OR. B)

EQV

A .EQV. B

NOT B

.NOT. B

A .OR. .NOT. B

.NOT. A

.NOT. A .OR. B

NAND

.NOT. (A .AND. B)

TRUE

.TRUE.

NOT A

Assignment Statements

A variable may be assigned a value by an assignment statement anywhere in the source program.

During program

execution, the most recent assignment determines the variable’s value in subsequent statements. The statements
that may be used to assign a value to a variable are the arithmetic and logical statements, which assign a numeric
or

logical value, and the ASSIGN statement, which assigns a statement number.

7.4.4.1

Arithmetic Statements

—

Arithmetic statements indicate computations to be performed by 03/78

FORTRAN IV.

734

05/ 78 FOR TRAN IV
Form

v=e

where

is a variable name;

is an expression; and

is the replacement operator.

The variable v is assigned the value of expression e.

Effect

The arithmetic statement associates a variable name with a value. This name may then be used in subsequent
B=A+1
expressions to represent the value. Thus, if the arithmetic statement A=2 is executed first, the statement
is equivalent to the statement B=2+1, or B=3.

Since the equal sign in the arithmetic statement does not indicate equality, but rather a replacement, statements of
the form

I=I+1

perfectly legal. The arithmetic statement is, in fact, the only means in FORTRAN by which the results of
computations represented by expressions may be stored.

are

In the following examples, the expression to the right of the equal sign is evaluated and converted when necessary
to conform to the type of the variable to the left of the equal sign. The converted value is stored in the storage

location associated with the variable name to the left of the equal sign. That is, if a real expression is assigned to
an integer variable, the value of the expression is converted to an integer before assignment.

Examples:
ANS=Y*(X**2+Z)
I=1*N

X(J)=A(J)—B(J)
P=.TRUE.
S=D.LT.5
The expression to be assigned must be capable of yielding a value that conforms to the type attribute of the variable
to which it is being assigned. The compiler performs conversions in accordance with Table 7-12.

Table 7-12

Conversion Rules for Assignment Statements

Logical

Literal

Real

Integer

Constant

Real

D ,6

Integer

D,6

Logical

N,6

To:
From:

Conversion between integer and real.

N= Convert non-zero to 1.0 (logical truth).

31‘?

Direct replacement.
Use the first character in the literal and five characters following.

7-35

0S/78 FORTRANIV
The GO TO Assignment Statement

7.4.4.2

—

The ASSIGN statement is used in conjunction with an assigned

GO TO statement to permit symbolic referencing of statements.
Form

ASSIGN n to var

where

is a statement number; and

var is an

Effect

integer or real variable.

The variable represents the assigned statement number and may be used in an assigned GO TO
statement.

The statement number assigned must be that of an executable statement. If more than one ASSIGN statement
refers to the same integer variable name, the value assigned by the last executed statement is the current value.
A variable that has obtained its value via an ASSIGN statement must be redefined via an arithmetic statement

before it can be used in any context other than the GO TO statement. For example, the statement
F4238 III GN

T (I)

1".) [i] UNT

associates the variable name COUNT with statement number 10. The following statement is then invalid:

COUNT =-‘=={I:(I]l.lNT-t~ 2|.
The statement becomes valid, however, if preceded by an arithmetic assignment statement such as

[IDLJNTm ll. 0
which assigns COUNT the integer value of 10. The use of an arithmetic assignment, however, invalidates any
future use of the variable COUNT in an assigned GO TO.
Never use an assigned GO TO to transfer program control outside of the program or subprogram in which it
appears.

7.4.5

Control Statements

Statements are normally executed in the sequence in which they appear in the source program. This sequence
may be altered by the use of one of the following FORTRAN control statements: GO TO, IF, DO, CONTINUE,

PAUSE, STOP, CALL, and RETURN. The CALL and RETURN statements, which transfer control to and
from subroutines, are described in this section.
7.4.5.1
are

GO TO Statements

—

The GO TO statement transfers control directly to a specified statement. There

three forms of the GO TO statement: unconditional, computed, and assigned. A GO TO statement may appear

anywhere in the executable portion of the source program except as the terminal statement in a DO loop.
Unconditional GO TO Statement
Form

GO TO 11

where

n= the

Effect

Control is transferred to statement 11

statement number of an executable statement.

7-36

0S/78 FORTRANIV
When control is transferred by a statement of the form GO to n, the usual sequential processing continues
at the statement whose number is n.

NOTE

Explicit statement numbers should not be confused
with internal statement numbers (ISN’s) which are

sequential octal numbers assigned by the FORTRAN
compiler.

Examples:
(30

1020

Computed GO TO Statement
Form

GO TO (n1,n2,...,nK),e
A comma must follow the right parenthesis.

where

n1 ,n2,...,nK are statement numbers; and
e is a

positive (nonzero) integer expression whose value is less than or equal to the number

of statement numbers within the parentheses.
Effect

Control is transferred to the statement whose number is eth in the list of statement

numbers.
The integer expression in a computed GO TO statement acts as a switch, as in the following example:

(30

(2091096) 9

If K=l, control will be transferred to statement 20; if K=2, to statement 10; or if K=3, to statement 6. If
K has a value less than 1 or greater than the number of statements within the parentheses, unpredictable

results occur.

Assigned GO TO Statement
Form

GO TO v

GO TO v,(n1,n2,...,nk)
where

v is an

integer variable; and

n1 ,n2,...,nk are statement numbers whose values may have been assigned to v.

Effect

Control is transferred to the statement whose number is currently associated with the
variable v via an ASSIGN statement.

7-37

0S/78 FORTRANIV
An ASSIGN statement defines an integer or real variable as a statement number.

881' (SN

Thus, when the statement

LUOF'

has been executed, control is transferred to statement 10 by the assigned GO TO statements:

(3 f.)

l... C] 0 F'

[30

I... (I) 0 F" 9 ( 10 a: 20 v 1 00 )

either of which may be used to transfer control to whichever statement number is currently associated with

LOOP.
Never use an assigned GO TO statement to transfer program control outside of the program or subprogram
in which it appears.

7.4.5.2

IF Statements

—

An IF statement causes control to be transferred on the basis of the value of a

specified expression. There are two forms of the IF statement:

arithmetic and logical.

Arithmetic IF Statement

Form

IF (arithmetic expression) n1,n2,n3

where

n1,n2,n3 are statement numbers.

Effect

Control is transferred to
n1 if expression <0;

n2 if expression =0; and
n3 if expression >0.
An IF statement transfers control to one of three statements, as shown in the model, according to the value of
the expression given. For example, the following statements transfer control to statement number 30:

Q I... F‘ H (31 =2: "(5
(m...F'|-~m )

3|.0y1230r130

If fewer than three statement numbers are present, control passes to the next sequential statement for each of
the missing conditions. Thus,

caLPHn)

sTnP
transfers control to 10 if ALPHA is negative; otherwise, it executes the STOP statement.

Logical expressions may be used in an arithmetic IF statement. In such statements, the logical expression is
first converted to an integer.

Logical IF Statement
Form

IF (logical expression) statement

where

statement may be any executable statement except another logical IF or a DO statement.

7-38

05/ 78 FOR TRAN IV
The statement given is executed if the expression has the value .TRUE.; otherwise, the next

Effect

statement in sequence is executed.

Examples:
LOGICAL T,F

1F (T.OR.F)X=Y+1
IF (Z.GY.X) CALL SWITCH (S,Y)

IF (K.EQ.INDEX) GO TO 15

7.4.5.3

D0 Statement

—

DO statements provide for the repeated execution of a statement or series of

statements.

Form

DO n i=ml ,m2,m3

where

n is a

statement number;

i is a nonsubscripted integer or real variable; and
ml ,m2,m3 are integer or real constants or expressions.
i is set to ml and statements following the DO statement up to and including statement 11 are

Effect

executed repeatedly increasing i by m3 at the end of each iteration, until i is greater than m2.
The statements that are executed as a result of a DO statement are called the range. The variable i is called the
index. The values ml, m2, and m3 are, respectively, the initial, limit, and increment values of the index. Note
that the range of a DO need not be merely a section of straight line code following the DO statement. A control
statement that causes instructions elsewhere in the program to be executed is permissible, as long as control

eventually comes back to the terminal statement. When the range contains such control statements, it is called
an

extended range.

If m3 is omitted, an increment of l is assumed. A zero or negative increment is not permitted. The range of a
D0 is always executed at least once, regardless of the values of the limit and increment. After each execution

of the range, the increment value is added to the value of the index and the result is compared with the limit
value. If the value of the index is not greater than the limit value, the range is executed again using the new
value of the index.

Examples:
[1U

'35

1133]. v 100

(I=100 during last iteration of DO loop)
DE)

13510

1' 31-": y 100 v 12

(I=99 during last iteration of DO loop)
DU

100

1330.9 100 y 13.?

(I=100 during last iteration of DO loop)
After the last execution of the range, control passes to the statement immediately following it. This exit from
the range is called the normal exit. Exit may also be accomplished by the execution of a control statement
within the range.

7-39

0S/78 FORTRANIV
The values of the initial, limit, and increment variables or expressions of the DO loop may be altered within
the range of the DO statement. Such alteration will not affect the operation of the 100p, since the values of

ml, m2, and m3 are remembered by the program. Altering the index will affect the number of iterations
of the loop, however. This value is available for program use as a variable. For example,

IIIU

I==Il v 10

TEMF‘=== I
.40

'--

ANUMBRK-TTEMF‘DKO . II.
R00'I'===8QRT(ANUMBR)

In this example, the value of the index I is used as the minuend in determining the value of TEMP. Also, when
a

statement transfers control outside the range of a DO loop, (for example, by a GO T0 or IF), the index

retains its current value and is available for use as a variable. A transfer into a DO loop from outside its range
may cause improper partial execution of the loop unless the transfer into the range is a return from the extended

range.

The terminal statement ofa DO range must not be a GO TO, DO, RETURN, STOP, PAUSE, or an arithmetic
IF statement. A logical IF statement is allowed as the last statement of the range, provided that it does not
contain any of the statements mentioned above. For example,

K33 II. 7 4

1110

5333

‘5}

(X(l‘1()o(3T.L.)

\’
L)

0-96

Y(l‘i)t=:X(l"()

In this example, the range is considered ended when, and if, control would normally pass to the statement

following the entire logical IF statement. Statement 5 is executed four times whether or not the statement
Y(K)=X(K) is executed. Statement 6 is not executed until statement 5 has been executed four times. Note
that statement 5 would be an error ifit were

1‘55

(X ( K) 961‘s» ‘1’ ( L. 1‘! )

(30

Any statement that serves as the range limit of a DO loop must not be used as the transfer point for IF or GO
TO statements which are outside the DO loop. The range of a DO statement may also include other DO
statements. This is referred to as nesting. The range of any nested DO statement must fall entirely within
the range of the next outermost DO statement; that is, every statement in the range of an inner loop must
be within the range of its enclosing outer loop. It is possible for a terminal statement to be the terminal
statement for more than one DO loop, however. Figure 7-6 illustrates the order in which nested DO’s
are

executed.

Figure 7-6

Nested DO Loops

740

08/78 FORTRAN IV
DO loops may be nested to a depth of ten levels maximum. In calculating this depth, one implied D0 in an
I/O statement counts as one level whose range is the single statement, and n implied DO’s within one [/0
statement count as n levels, the ranges of which are all within the single statement.
7.4.5.4

CONTINUE Statement

—

The CONTINUE statement consists of the text

CONTINUE
and causes continuation of the normal sequence of program execution. CONTINUE is principally used as

the range limit of DO loops in which the last statement would otherwise be a GO TO, IF, PAUSE, STOP,
or

RETURN statement. The CONTINUE statement is also used as a transfer point for IF and GO TO

statements within the DO loop that are intended to begin another repetition of the loop.

133 ’2':

113:

Ill-{Ti .,~ 0
(A B310y30925

I :2: II. 2: 1-30

Awn+tto

BmbfitO
CON‘KNUE

:51 (3'

[:1

53‘

For example,

r-‘a 't‘ 1‘}

A CONTINUE statement may be used as the range limit of any number of DO loops, as in this example:

DE]

U331? :2 (z': y 3

3.7353

“:3 '33

if. 0 N T I N U E

A CONTINUE statement serving as the range limit ofa DO loop must not be used as the transfer point for
IF or GO TO statements that are outside the DO loop. Ifit serves as the range limit of several DO
loops,
as

above, it must not be used as the transfer point for IF or GO TO statements that are outside the innermost

loop.

For example,
[I0

1320

I I77

[IO

(N
10

(N.EG.4)

WRITE

123:]. 9 1’30
a

EU . 4 )

(30

( :i. rum: 1‘ react; )

£90

(crcn‘r‘ecrt)

Is: 1 v {30
(1‘ij 10071ny1<

It. 0

(I: 0 N T I N U IE".

2-3 0
1 00

(I: 0 N T 1‘ N U E
15-” [1] IR‘ rm T ( 1'5 I (a )

E N 1]

7-41

05/78 FORTRANIV
PAUSE Statement

7.4.5.5

The PAUSE statement interrupts program execution.

—

Form

PAUSE

PAUSE number

where

number is an integer variable or expression.

Effect

number, if any, is displayed
is typed on the console.
The

STOP Statement

7.4.5.6

—

the terminal. Execution is

suspended

until

The STOP statement is placed at the logical end of a program.

Form

STOP

Effect

terminates the program and returns control to the Monitor.

The STOP statement terminates program execution. No continuation is possible. If no STOP statement is

present in a program, a STOP occurs when control passes to the END statement in the MAIN program.
A CALL EXIT statement is equivalent to a STOP and closes tentative files at the last block written on the file.

Control returns to the Monitor.
7.4.5.7

END Statement

—

The END statement consists of the text

END

and is placed at the physical end of a program or subprogram. In the main program, the END statement is

equivalent to STOP; in a subroutine, END is equivalent to RETURN. The compiler assumes the presence of
an END statement if it fails to
an

find one before the end of the source input file. A program can not reference

END statement.

7.4.6

Data Transmission Statements

Three distinct types of data transmission statements control the transfer of data between computer memory

and I/O devices: data description (FORMAT) statement, input/ output (READ and WRITE) statements, and
device control (BACKSPACE, REWIND, and END FILE) statements.
7.4.6.1

FORMAT Statement

—

The FORMAT statement describes the form and arrangement of data on

a record.

Form

FORMAT (specl ,spec2,.../...)

where

specl,spec2 define consecutive series of characters within a record;
/ is the end of the record description; and

) is the end of a statement.
Effect

Specifies either the type of conversion to be performed between the internal and external
representation of data or the format of fixed data.

A FORMAT statement must have an explicit statement number that is used in READ or WRITE statements

for reference.

742

character

08/78 FORTRAN IV
The field specification (spec) is one of the following:
nAw

nEw.d
an.d

nGw.d
nH
an
an

-sP
Tw
nX

‘string’
“string”
where

is an unsigned

nonzero

integer stating the number of times the field specification is to be

repeated;
3

is scale factor;

A,E,F,G,H,I,L,P,T,X are types of conversion;
w

is nonzero, unsigned integer constant specifying width of field; field width must be large enough
to provide for all characters (including decimal point, sign, and

exponent) necessary to constitute

the data value plus any blank characters needed to separate it from other data values; the data

value within a field is right-justified; if the value is too large for the field, the field is filled with

asterisks; and
.d

is unsigned integer constant (may be zero) specifying number of digits t0 the right of the decimal

point or, for G conversion, the number of significant digits.
Field specifications must be wirtten in the same sequence as the data record being described, except when the
T specification is used.
A FORMAT statement may describe one or more records, each of which can consist of one or more field

specifications.

The slash character (/) indicates that a new record is being described. For example, the

statement

l" (33F: M r353 T

II (310 e 33/ 1'15 y EFB 0 4 )

is equivalent to

F'UFQMfiiT

( (3310 e 2’. ‘1

for the first record, and

lii'OF'a‘Mtfi'l' ‘1 I '53 so BF 8

4 II

for the second record. Field specifications are separated by commas as shown above. The separating comma
may be omitted when a slash is used. When n slashes appear at the end or beginning of a format, n blank

records may be written on output or n records skipped on input. When n slashes appear in the middle of a

format, n-l blank records are written or n-l records skipped.

7-43

0S/78 FORTRANIV
For example,

I’II‘JI‘lIMA’T'

( III (S v ff,”

IBF‘LJ‘ . 1 3

where / / / indicates that two records are to be skipped.
A group of field specifications may be repeated by enclosing the group in parentheses and preceding the enclosed
group with a repetition number as shown in the following example:

Iii'tlllitfiril'l‘

II 3 C I '53 y F II. (.‘J o '3 ‘J J

Both the slash and the closing parenthesis at the end of the format indicate the termination of a record. If the
list of an input/output statement dictates that t'ansmission of data is to continue after the closing parentheses
of the format is reached, the format is repeated starting with that group repeat specification terminated by the last

preceding right parenthesis, or, if none exists, then to the first left parenthesis of the format specification. Thus,
the statement:

Iti‘tiitl‘i'tr‘irfh'l’

( I377

$3.? :: 1'5 ( 1 1-3 s- £13.? '1 I 3 y IEI‘?

3 .‘J I I? 1' I)

Terminator

Group Repeat
specification

Last preceding

right parenthesis
causes

(F7 .2,3(12,2(I3,E9.3)I7)
to be used on the first record, and the format

3(12,2(I3,E(.3)I7)
to be used on succeeding records.

As a further example, consider the statement

i3"l‘llll3ifi"’i{‘»i'l

'7. I371?"

1?}? ( .133 '1' IEE: Cl. '55} t '55 y 1331113"; s 4 2'! y I 7 3 T!

The first record has the format

(F7 .2)
and successive records have the format

(2(E15.5,E15.4),I7)
FORMAT statements may be placed anywhere within the executable portion of the source program. Unless the

FORMAT statement contains only Hollerith data for direct input/output transmission, it will be used in conjunction
with the list of a data transmission statement. Because FORMAT statements are referenced by READ or WRITE

statements, each FORMAT statement must have an explicit statement number.
The ASCII character string comprising a format specification may be stored as an array.

Input/output statements

may then refer to the format by giving the array name, rather than the statement number of a FORMAT statement.

744

05/78 FORTRANIV
The stored format has the same form as a FORMAT statement excluding the word FORMAT. The enclosing

parentheses are required.
Field specifications in a FORMAT statement should be of the same type as the corresponding items in the I/O list;
that is, integer quantities require integer (1) conversion, etc. There are three types of field specifications: numeric,

logical, and alphanumeric (including Hollerith). In addition, a blank field description may be given to skip portions
of an input record or to embed blanks within an output record.
Numeric Fields

Numeric fields are specified by one-letter codes (E, F, G, or I) that designate the type of

conversion to be performed. Two parameters may appear in a numeric field description, depending on the field type;

(1) an integer (w) specifying the field width (which may be greater than required to provide for blank columns
between numbers); and (2) an integer (d) specifying the number of decimal places to the right of the decimal point,
or, for G conversion, the number of significant digits.

Decimal points are not permitted in I conversion.

(For E, F,

and G input, the position of the decimal point, if present in the external field, takes precedence over the value of
d in the format.) Conversion codes and the corresponding internal and external forms of the numbers are listed in

Table 7-13. Numeric fields are right-justified with the addition of leading spaces and, if necessary, trailing zeroes.

Single precision l/O specifications will transfer a maximum of six decimal digits of accurate data.
Table 7-13

Numeric Field Codes

Conversion

Internal

External

Code

Form

Input Form

Output Form

Decimal number

Real

with or without a

and an E exponent

decimal point or

with a decimal point

exponent field.

field.

Decimal number

with or without a

with a decimal point.

decimal point or

exponent field.
G

Real

Integer

Decimal number

with or without a

with a decimal point

decimal point or

and with or without

exponent field.

Decimal number

without a decimal

point or exponent.

Four numeric field specification forms are allowed:
l.
2.

Ew.d,
Ew.d,

Iw, and

Gw.d.

For example,

F'DIIQMI‘N'T'

E exponent field.

( I 5:5 y l5" 1 0 . 2'3 )

7-45

05/78 FORTRAN IV
could be used to output the following line on the output listing
bb32bbbb-l7 .60
where b represents a blank. Since there is no carriage control character in the statement, the first character is

interpreted as a carriage control character and is not printed.
The G format is the general format code used to transmit data having a specific number of significant figures, no
matter what the magnitude of the number. This format is intended to allow use of the simplest output format that
can express

the desired value in the space allowed. The rules for input are the same as those for E format.

The form of the output conversion (E or F) is a function of the magnitude of the data being converted. Table 7-14
shows the magnitude of the internal data, M, and the resulting method of conversion.

Table 7-14

Conversion Under G Format

Magnitude of Data

Resulting Conversion

0.1<=M<l

F (w-4).d ,4X

1<=M=<10

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

10(d-2)<=M<lO(d-l)

F(w-4).1,4X

lO(d-l)<=M<lO(d)

F(w-4).0,4X

All others

Ew.d

Scale factors may be specified for E, F, and G conversion. A scale factor is of the form
sP

where s is a signed or unsigned integer that specifies the scale factor; and P is the identifying character.
For F type conversions, the scale factor specifies a power of ten, so that

external number

(internal number)*10(s‘3‘ille faCtOY)

For E conversions, the scale factor multiplies the fraction by a power of ten, but the exponent is decreased
accordingly leaving the number unchanged except in form. For example, if the statement

FORMAT

(F78 . 3 v E 1 6 . 5)

corresponds to the line
bb26,451bbbb-0.41321E-01
then the statement

FORMQT

(- IlI‘i‘F-"B 9 '3 9 EF'Elé . 5)

7-46

05/ 78 FOR TRAN IV
would correspond to the line

bbb2.645bbb41.3215E-03
For C type output conversions, the scale factor is not used unless the magnitude of the number is such that E format
is used. In input operations, the scale factor is not used if there is an exponent in the external field.

When no scale factor is specified, a scale factor of zero is assumed. Once a scale factor has been specified, however,
it holds for all subsequent E, F, and G type conversions within the same format unless another scale factor is

encountered. A zero scale factor may be resumed via an explicit specification. The scale factor is initially 0. Scale
factors have no effect on I type conversions.

Logical Fields

Logical data can be described in a manner similar to numeric data. A logical field description has

the form:
Lw

where L is the conversion code character and w is an integer specifying the field width. The data is transmitted as
the value of a logical variable in the input/output list. On input, the first non-blank character in the data field must
be .T., T, .F. or F; the value of the logical variable will be stored as .TRUE., .TRUE., .FALSE., or .FALSE.,

respectively. Leading blanks and the period preceding the T or F are ignored. If the data field is blank, a value of
false will be stored. On output, w-l blanks followed by the letter T or F, according to the variable’s value, will be
transmitted. For example, if the specification were L10, the output for the value .TRUE. would be
bbbbbbbbbT
Hollerith Data A Conversion

—

Hollerith data that is to be stored by the program is specified by

where A is the conversion code character and w is the number of characters in the field. The alphanumeric

characters are transmitted as the value of a variable in an input/output list. The variable may be of any type. The

following sequence causes four characters to be read and placed in memory as the value of the variable V:
liiiliifi it]
'i

C 2?. y '5 3 ‘v‘

lii'OFEMA'l'

( A «’l )

The value of w is limited to the maximum number of characters that can be stored in the space allotted for a single
variable. If w exceeds this amount, the extraneous rightmost characters are lost on input, and on output the
characters are lost on input, and on output the characters after the sixth are not significant. If w is less than the
number of characters that can be stored in the space allotted to the variable, on input the characters are left-justified
and blank—filled on the right of each list item. On output the leftmost w characters in the variable are transmitted to

the output field.
Hollerith Data H Conversion

—

Hollerith data that is not changed by the program is specified by one of two

forms. One, called H conversion, is
nH

where H is the control character; and n is the number of characters in the string (including

blanks). For example, the

format in the statement below can be used to print PROGRAM COMPLETE on the output listing.
FORMQT

( 1 7H

F'RIJGI‘R’AM

[BUMP I... l” T IEEZ )

7-47

OS/ 78 FOR TRAN I V
Referring to this format in a READ statement causes the 17 characters to be replaced with a new string of characters
from the input file.
In the second form, the Hollerith data is simply enclosed in single quotes. The output result is the same as in H conversion; on output the characters between the quotes (including blanks) are written as part of the output data, as

with Hollerith constants. For example,

FOFEMQ'I'

(

F'FL'CJG RAM

’

CUMHuff-TIE.

A single quote within the data is represented by two successive quotes.
A Hollerith format field may be placed among the other fields of the format. For example, the statement:

l‘i'Ul‘ifM («T
can

I; I '53 , ".7 H

Flffilli'x'tillfimil: 10

'53)

be used to output the line

bbb22bFORCE=bb 17.68901
Note that the separating comma may be omitted after a Hollerith format field. The legal characters in a Hollerith

field are the 64~character 6-bit subset of ASCII, with the exception of @, which must not be used.

Carriage Control

—

The first character of each ASCII record controls the spacing of the line printer or terminal.

This character may be established by beginning a FORMAT statement for an ASCII record with ch, where c is the
desired control character. The line spacing actions, listed below, occur before any printing:

Effect

Character

blank

Advance carriage to next line.

0 (zero)

Skip a line (double space).

1 (one)

Form feed

go to top of next page.

Suppress skipping

plus

—

overprint previous line (used only for hard copy output).

For example, the following program moves the line printer paper to the top of the next sheet (1H1) and prints b3.5
on

the first line.:

55‘}

DJ I31“ I T IEE.‘
10"?)

F’CIRMIIXT

i.‘ 1‘3 .‘I ./ 2.33

'53}

( 4 9 21.00 I) X
( 1H]. 7 I74 0 ii. .3

533 T [I] F"

ISSNIII
If any unexpected character appears first in the FORMAT statement, it is processed as a blank.
It is often desirable to print a prompting sequence, such as a question, to which a response is to be entered on the
same line.

To cause such prompting, a dollar sign ($) is included at the end of a FORMAT statement that is

7-48

03/ 78 FOR TRAN IV
associated with a WRITE statement that has been satisfied. This will inhibit the carriage return and line feed before
the next input. For example,

NR I TE t’. 4 v ll. '1) O 3' A
R IE: :31 I) ( it y EGO )1?!

100
1230!)

l15'(315:i'iifi'l"(

$30MF'I...!E§

I :3 a»

’

IS?

’

it)

F-"URMQT '1 $36)
NRI‘T'IEZ (4 ,, 1:30-17.- )1-3
IE: N17!

The output is

SiifiiMF'l...l-I~I

1253:

RED

REID
Record Layout Specification

Input and output can be made to begin at any position within a FORTRAN

record by use of a field description of the form
Tw

where T is the spacing control character; and w is an unsigned integer constant specifying the character position in a

FORTRAN record where the transfer of data is to begin.
For printed output, w corresponds to the (w-1)th print position, since the first character of the output buffer is a
carriage control character and is not printed. (A blank carriage control indicator is assumed.) For example,

FORMAT

(T130 7

’

BLACK "T501

’

UHITE ’)

BLACK to appear in columns 29-33 and WHITE to appear in columns 49—53. The statement

causes

FORMAT

(T50 y

’

EILhIH‘J

’

T130

’

NHITEI ’)

BLACK to appear in columns 49-53 of one line and WHITE to appear in columns 29-33 of the following line.
The two lines will constitute two separate records iflater read as input. On input, the statement:
causes

FORMAT (‘TBENF'ELO7
READ (3’1)X

causes the

first 34 characters of the input data to be skipped, and the next five characters to be used as the value

of X. If an input record containing
ABCbbeYZ
is read with the format specification

FORMAT

(T7 7 A3 ! T1 7 (#3)

then the characters XYZ and ABC are read, in that Order. The T-code can be used in a FORMAT statement with any
other type of format code.

‘

Blanks can be introduced into an output record or characters skipped on an input record by use of the specification:

7-49

05/ 78 FOR TRAN [V
where the spacing control character is X and n is the number of blanks or characters skipped and must be greater
than zero. For example, the statement

FORMAT
can be

ESS‘I'EIF‘IS y 1 0X2HY===F7 . 3)

(15H

used to output the line

STEPbbeSbbbbbbbbbbY=b-3.872
The FORTRAN program may read or write chosen records from a direct-

7.4.6.2

DEFINE FILE Statement

access file

without reading intermediate records. In the more common sequential access, the FORTRAN program

——

must read or write each record in turn until the correct record is found. No DEFINE FILE statement is required for

sequential access to mass storage files.
The DEFINE FILE statement is required so that mass storage files may be referenced as direct access files by input/

output statements:
Form

DEFINE FILE a1(bl,cl ,U,vl),a2(b2,c2,U,V2),

where

a is an

integer constant or integer variable name that is the symbolic designation for this file

specified to the run~time system;
b is an integer expression that defines the number of records in the file;
c is an

integer expression that defines the length of each file record; each integer or real variable or

constant is stored in floating point notation that requires three storage locations per value;

U is a fixed argument designating that the file is unformatted; and

integer variable name, called the associated variable, which is set at the conclusion of and
input/output operation on the file to point to the next record.

v is an

Effect

Describes a mass storage file for use with input/output statements.

The associated variable (v) in a DEFINE FILE statement is used to maintain an index of records processed. It is set

automatically after an input/output statement is executed. The statement
DEFINE

FILE

1(1000 y 100 v U VII-U1)

specifies a IOOO-record file, each record of which is 100 floating point variables long. The variable IVl will maintain
of records processed, providing a pointer to the next record.

an index

The symbolic file designation (a) cannot be passed as a dummy argument to a DEFINE FILE statement in a
subroutine.

The input/output statements, READ and WRITE, govern transfer of data
records between internal storage and peripheral devices. Each statement can contain an input/output list naming the

7.4.6.3

Input/Output Statements

variables and array elements to be given values on input or whose values are to be transmitted on output. Both

formatted and unformatted records can be transmitted. A formatted record requires the use of a format

specification.

7-50

0S/78 FORTRANIV
Input/Output Lists
assigned
can be

input

—

An input/output list contains variable names and array elements whose values will be
output. Constants are not allowed as list items. During input, the new values listed

or written on

used in subscript or control expressions for variables appearing later in the list. For example,

READ

if 6 , 100 7 I... v A ( I...) y B ( I...+2I.)

reads a new value for L and uses this value in subscripts of A and B.

The transmission of array variables may be controlled by indexing similar to that used in the DO statement. This is
called an implied DO and includes as a list element a parenthesized list of control variables followed by the index
control. For example,

RESIN]

('7 y 10)

(X ( K) , K31 y 4)

is equivalent to:

REEF-11f!

('7»Ei())>((1l. ) MHZ?) yX(3) vXUi)

The indexing may be compounded by nesting the implied DO’s as in the following:

READ

((MQSSS ( l\‘( p I...) , Kflfiil y 4) y L513]. 9 "5)

'1‘? 9 70)

The above statement reads in the elements of array MASS in the following order:

MASS(I,1),MASS(2,1),

,MASS(4,1),MASS(1,2),

,MASS(4,5)

If an entire array is to be transmitted, the indexing may be omitted and only the array name written. The array is
transmitted in order of increasing subscripts with the first subscript varying most rapidly. Thus, the example above
could have been written

lii‘lfiifliIII

('.?y’.75'j)

lamest;

assuming that the array MASS is dimensioned MASS(4,5). The same statement can transmit integer and real
quantities. If the data to be transmitted exceeds the items in the list, the extra data is ignored.

Input/Output Records

All data is transmitted by an input/output statement in terms of records. The maximum

amount of information in one record and the manner of separation between records depends on the medium.

On a

terminal, a record is one line; for ASCII records, the amount of information is specified by the FORMAT reference
and the I/O list.
Each execution of an input or output statement initiates the transmission of a new data record. If an input/ output
statement requests less than a full record of information, the unrequested part of the record is lost and cannot be

recovered by another input/output statement without repositioning the record.

Repositioning is not possible on all

devices, however. If an input/output statement requires more than one ASCII record of information, successive
records are read in sequence.
READ Statement
Form

READ (u,f) list

formatted READ

READ (u,f)
READ (11) list

unformatted READ

READ (u)
READ (a’r) list

direct access mass storage READ

7-51

05/78 FORTRANIV
where

u is an

input unit designation;

f is a format statement reference number;
list is an [/0 list of variable names;
a is a

symbolic mass storage file number (unsigned integer constant or integer variable);

designates direct access; and
is the record number in which transfer begins (integer expression).

Input is performed according to the arguments of the READ statement.

Effect

The unit designation (u) referred to in READ statements must be an integer in the range 1 to 9. If device specifications are not given, the system assumes the standard device designations mentioned in Section 7.2.3 describing the
run-time system. Thus, READ (4,1 1) could refer to input from the terminal. Formatted and unformatted records

should not be mixed on a single unit.
A formatted READ statement causes information to be read from the specified unit and put in memory. The data
converted from external to internal form as sepcified by the referenced FORMAT statement. If an I/O list is
provided, the data are stored as the values of the listed variables. The second form of the formatted READ
statement is used if the data are transmitted directly into the specified format. For example,
are

F: If: {-1 I]
.

.f, (:11)

C 1’53 v 1’30 .3 (it y I?- 5! [I3
'1 ,5), z: 100)

Detection of end-of—file during input causes a fatal run-time system error unless the library subroutine CHKEOF is

called. CHKEOF should only be used with formatted [/0 involving a single record.
A comma encountered on input signals the end of the current field and causes the next input character to be read as
the first character of the next input field.

An unformatted READ statement causes binary information to be read from the unit designated and stored in
memory as values of the variables in the I/O list, if any. If the record contains more words than the list requires,
that part of the record is lost. If more items are in the list than are in one record, additional records are read until

the list is completed.
A direct access READ statement provides random access to fixed-length records in a mass storage file. The file whose

records are to be read must be defined by a DEFINE FILE statement. For example,

IIIEZIF INIEC

READ

I57 ILIEI

(ATE:

’

{itTl'Zlf 100 y 100 9 U y F'T )

'53 ) QRIMY

WRITE Statement

Form

WRITE (u,f) list

formatted WRITE

WRITE (u.f)
WRITE (u) list
WRITE (u)

unformatted WRITE

WRITE (a’r) list

Direct access mass storage WRITE

7-52

0S/78 FORTRAN [V
where

u is an

output unit designation (unsigned integer constant or variable);

f is a format statement number;
list is an I/O list of variable names;
a is a

symbolic mass storage file number;

designates direct access; and

r is an

Effect

associated variable (record pointer).

Output is performed as specified by the arguments of the WRITE statement.

The unit designation (u) referred to in WRITE statements may be an integer in the range I to 9. Unit
designations are assigned via a device specification command to the run»time system.
A formatted WRITE statement may appear with or without an I/O list. If a list is provided, the values

of the variables in the list are read from memory and written on the unit designated in ASCII format. The
data is converted to external form as specified by the designated FORMAT statement. If no list is supplied,

information is read directly from the specified format statement and written on the designated unit in
ASCII format. For example,

wl-"(ITE
100

FORMAT

(4! 100)
(

OUTF'UT

RECORD" )

will produce the following record on unit 4:

OUTPUT

RECORD

For an unformatted or direct-access WRIT E, the values of the variables in the list are read from memory
and written on the unit designated in binary format. A record holds 85 real or integer variables. If the
list elements fill more than one record, successive records are written until the list is completed. If the

list elements do not fill the record, the remaining part of the record contains undefined data.
two records are used if the list contains 100 variables:

one

Thus,

record contains 85 variables and the second

record contains 15 variables and undefined data. For example,

[IIMENfEEIlIlN
URITE

(6)

X(200)
X

will produce three records on unit 6, the first will contain X(l) to X(85); the second will contain X(86) to X(l70),
and the third will contain X(l7l) to X(200).
A direct access WRITE statement outputs a fixed-length record directly into a mass storage file. The file must
have been defined previously via the execution of an appropriate DEFINE FILE statement. This
means minimum
record size is one block. For example,

7.53

0S/ 78 FOR TRAN IV
DIHENSION
DEFINE
.Jwi
HO

1(12y109UyJ)

1341712

READ (4,100) RAY
FORMAT (10FB.0)

100

HRITE

(1 ’J)

RAY

CONTINUE
STOP
IiEINII

‘5

7.4.6.4

RAYCIO)

FILE

Device Control Statements

—

There are three device control statements:

BACKSPACE, END

FILE, and REWIND, that apply to diskettes. Their forms and effects are listed in Table 7-15.
Table 7-15
Statement

Device Control Statements

Effect

BACKSPACE u

If the next record, which would be read or written on unit (u), is n,
BACKSPACE repositions the unit so that the next record to be
read or written will be n-l. If the first record is the next record to
be read or written, the BACKSPACE statement has no effect.

REWIND u

Repositions the designated unit (u) to the beginning of the file.
If the unit is at the beginning of the file, the REWIND statement
has no effect.

END FILE u

Writes an END—OF-FILE character in the specified file (u),

provided that the file has been written on by a formatted WRITE.
END FILE does not execute a REWIND.

7.4.7

Specification Statements
Specification statements may be divided into three categories:
1.

7.4.7.1

Storage specification statements (DIMENSION, COMMON, and EQUIVALENCE) that give the
compiler storage allocation instructions;
Data specification statements (DATA and BLOCK DATA) that are used to enter values; and
Type declaration statements (INTEGER, REAL, and LOGICAL) that specify the type of variable.

Storage Specification Statements Dimension Statement

—

Form

DIMENSION namel (ul,...,u7), name2 (v1,...,v7),...

where

u1,...,u7 and v1 ,...,v7 are the maximum values of the subscripts they represent, up to
a

Effect

maximum of seven subscripts.

The array name assigns the type to the array.

Storage is allocated according to the

dimensions given.
Each array specification gives the array name and the maximum size that each of its subscripts may assume.

Array size is limited to 4095 elements. Each size specification must be a nonzero positive integer constant.
For example,
[IflIi‘ifiNSIlitJN

A( 10) 9 17¢ ( 4 9 (S) y X ( 53 y '5 .v '55)

7-54

03/78 FORTRAN IV
defines A as a onedimensional array variable with storage locations for 10 words. Array B is defined as a two
dimensional array with storage for 24 words (4x6). Array X is a three-dimensional array with 125 words (5x5x5).
In certain cases involving subroutines, a dimension may be an unsubscripted integer parameter.

Any number of arrays may be declared in a single DIMENSION statement. Each array variable appearing in
the program must represent an element of an array declared in a DIMENSION statement, unless the dimension

information is given in a COMMON or type statement.
Dimension information may appear only once for a given variable. The DIMENSION statement must

precede any reference to the variable including reference in a DATA or EQUIVALENCE statement.
Type declarations or COMMON statements may appear anywhere in a program unless they include
dimension information.
A subprogram can establish adjustable arrays at execution time if both the array name and the subscript
size are expressed as dummy arguments in the subroutine, as in

DJHAT ( fir y X y 'Y 3 Z)

SUBROUT I NIEEI

IlZlIMEINSION

(a ( X 9 Y r 7 ‘I

To do this, the programmer must establish A, X, Y and Z as required arguments. The dummy array must
not exceed the dimensions of the main program array but may be smaller if the call provides lower subscript

sizes than those of the main program dimensioning or if the initial array element referenced is not the

beginning of the main program array.
COMMON Statements
Form

COMMON/block l/a,b,c/block2/d,e,f/...

where

blockl, block2,..., are the block names; and a,b,c,d,e,f are the variables to be assigned
to each block.

Effect

Specified variables or arrays are stored in an area available to other programs.

By means of COMMON statements, the data of a main program and/or the data of its subprograms can share
storage area. The common area can be divided into separate blocks identified by block names

a common

that may not be the same as any program variable names. A block is specified as follows:

/block name/varl var2,...
,

The variables following the block name indicate scalar or array variables assigned to the block, They are
placed in the block in the order in which they appear in the block specification. For example, the
statement:

[IDMl‘llilNl-‘lziif'kt' 9 Y y TD-"C/U v 'v' v L-J a! Z
indicates that the elements X, Y, and T are to be placed in block R in that order, and that U, V, W, and Z

placed in block C. Variables whose names appear in the formal parameter list must not also
in
COMMON declarations within the subroutine. A COMMON block may not have the same name
appear

are to be

as a variable

in the same program. In addition, a COMMON block may not have the same name as any
used
at the same time as the COMMON block.
subprogram

7-55

05/ 78 FOR TRAN IV
Block entries are linked sequentially throughout the program, beginning with the first COMMON statement.
For example, the statements

[3 (I) NW) N. x’l] ./ 9‘1 l... F“ H 15% 2/ R/ (-‘I 9 13 x’C ,-" ‘3

COMM UN ,I'EI/ K .v "r’I’R/ [J .v V 2: Id
have the same effect as the statement

l..:l:ll"ll"l[:ll‘~l-'I].-"ft!LPNf3.3.r‘"‘l:'\'./It{3 y 17.4 5: U v V ,, l:J.="l.'.T;-"'$3 .u X y Y
One block of COMMON storage, referred to as blank COMMON, can be left unlabeled. Blank COMMON is
indicated by two consecutive slashes. For example,

C U i"! M [:1 N .r" F:

‘1! v '1’ .2’ ,,-’ I3 :- C r [I

indicates that B, C, and D are placed in blank COMMON. The slashes may be omitted when blank COMMON
is the first block of the statement, as in

C (I) H i"! [2) hr

lit 9 if. 9 1]

Storage allocation for blocks of the same name begins at the same location for all programs executed together.
For example, if a program contains
COMMON

as its first

A 5! H/R/X v Y ., K

COMMON statement, and a subprogram has

{IZOMMClt-J/R/U , U y w/x’U y E: ,, F
as its

first COMMON statement, the quantities represented by X and U are stored in the same location, that

is, X and U both refer to the third variable. A similar correspondence holds for A and D in blank COMMON.
COMMON blocks may be of any length, subject to the limitations on available memory.

Array names appearing in COMMON statements may have dimension information appended if the arrays have
For example:

not been declared via a DIMENSION statement or a type declaration.

COMMON

{Mufi‘l’ifi y'l'(1ll‘5y10 v '5) y GAMMA

specifies the dimensions of the array T while entering T in blank COMMON. If array dimensions are not
defined in :1 COMMON statement, they must be defined in some other type statement.
EQUIVALENCE Statement
Form

EQUIVALENCE (v1,v2,...,),(vk,vk+l,...),...

where

vl,v2,..., vk are the variable names.

Effect

The variables within the parentheses identify the same storage location.

The following example

ECHJIL’UHLENCE ( REII v BLUE)

7-56

0S/78 FORTRANIV
specifies that the values of the variables RED and BLUE are stored in the same location. If used at different
times, multiple variables with different values can occupy the same storage location or if used at the same
time, multiple variables can be assigned the same value through the use of EQUIVALENCE.
The master variable in an equivalence group is either the variable in the group that is in COMMON (only one

such variable per group is legal) or the first variable in the group. All the other variables of an equivalence
group are considered subordinate and can only appear in one group. The master of an equivalence group
should be large enough to encompass all the subordinate variables equivalenced to it.
The subscripts of array variables in an EQUIVALENCE statement must be integer constants as shown in the
following example:

l3§Gl.l.l:U(Al.-ENC[—I ( X y A ( 3) I Y ( '2. v 1 v 4) ) 7 (BETA ( 2 y 2) v ALPHA)
The formal parameters of a subroutine must not appear in EQUIVALENCE statements within that subprogram.
The variables assigned by an EQUIVALENCE statement must be within the same main program or within the
same

subprogram.

Variables EQUIVALENCE and COMMON Statements

Variables may appear in both COMMON and
but
no
two
in
COMMON
EQUIVALENCE statements,
may be set equivalent to one another.
quantities
7

Quantities placed in a COMMON block by means of EQUIVALENCE statements can cause the end of the
COMMON block to be extended. For example, the statements

COMMON/R/X , Y , Z

IIIIIMENSION

A ( 4)

EQUIVALENCE ( IQ 9 Y )
cause

the COMMON block R to extend from X to A(4), arranged as follows:
x
Y A(1)
2 A(2)

A(3)
A(4)
EQUIVALENCE statements that would require extension of the start of a COMMON block are not allowed.
For example, the sequence
COMMON/FUR 9 Y a: Z
litIMEINSIlIUN {it ( 4)
IZZGLJIUN...IESIN.CE ( X ,. r-‘-i ( 3))
is not permitted, since it would require A(1) and A0) to extend the starting location of block R.

Care must be exercised when using EQUIVALENCE and COMMON statements.

7-57

05/78 FORTRANIV
Data Statement
Form

DATA var listl/val listl / ,var list2/val list2/,...

where

var list contains a

string of variables separated by commas; and

/val list/ contains a string of data items separated by commas.
Effect

A value from val list is assigned to the corresponding variable in var list.

The DATA statement is used to supply initial or constant values for variables. The specified values are

compiled into the object program, and become the values assumed by the variables when program execution
begins. Such values may also be provided via a BLOCK DATA subprogram. It is recommended that variables
in COMMON be initialized only by means of a BLOCK DATA subprogram.
Elements in the variable list may be either single subscripted or unsubscripted variables, or the name of an
entire array.

When an entire array is given, data values must be specified for every element of the array. Data elements
are

stored in the array in the same order as that used for the data transmission and storage arrays, that is, in

order of increasing subscripts with the first subscript varying most rapidly.

Allocation to memory locations in the array stops when
1.

the data item list is exhausted, or

data items have been allocated to the entire array; if so, additional data items will be allocated
to succeeding variables listed.

When Hollerith or literal constants are encountered in the values list, they are assigned to the associated
variables in the same manner as in assignment statements.
A Hollerith value in a data statement will occupy ( n+5)/ 6 words of storage, with any partial words filled to

the right with blanks.

The data items following each list of variables must have a one-to-one correspondence with the variables of
the list, since each item of the data specifies the value given to its corresponding variable.
Data items assigned may be numeric, Hollerith, Octal, or logical constants. For example,

IMTA

Al... F‘ HA ,

BETA/{:3 9 16 9 E3-133/

specifies the value 5 for ALPHA and the value 0.16 for BETA. Any item of data may be preceded by an
unsigned nonzero integer constant followed by an asterisk.
repeated. For example,
IIIATA

This notation indicates that the item is to be

r-“M ll. ) ant-H.133) 9fi(3)/3)I<0./

specifies the value zero for array elements A(1)—A(3). As another example,
III I i‘ilEZNEE-l I UN

Illrfll'lr’v‘r

(3) ( 333 3:11.? 3

175 ( 3 )1

[TIMI-31/2232iill. 90 y 336.33. » 0 y 35 o 0 r 4 y /

7-58

08/78 FORTRAN 1V
will initialize

A(l,l) and A(2,l) to l
A(1 ,2), A(2 ,2) and B(l) to 2
B(2) to 3, and B(3) to 4
7.4.7.2

Type Declaration Statements

Form

type v1,v2, V3,...

where

type may be INTEGER, REAL, LOGICAL; and

v1, v2, V3,... represent variables.
Effect

All variables in the list are assigned the given type.

A variable can appear in only one type declaration statement.

Type declaration statements override any

implicit type specifications determined by the first letter of a variable. Type statements can be used to give
dimension specifications for arrays. Adjustable arrays in subprograms can also be defined via type declaration
statements. Each variable or function name in a type declaration statement is defined to be of that specific
type throughout the program; the type cannot change.
Examples :
it 1-?- Cl 5' Il'. Jl‘i v KY .2'

I 'r‘l'l'liiit'I-EIEZI'E
RE (11!...
7.4.8

l-"l 4: 1.'.’. v 4 ) y I 9 .J r K

Subprogram Statements

The use of subprograms allow a statement or group of statements to be written once and then referenced
whenever the implemented operation is to be executed. The use of subprograms saves programming time
and computer memory. There are three categories of subprograms in FORTRAN: FUNCTION subprograms,

SUBROUTINE subprograms, and BLOCK DATA subprograms.
Functions and subroutines consist of one or more FORTRAN statements, which may be invoked by name,

and, as appropriate, with values upon which they are to operate. A function differs from a subroutine in that
it is always called with at least one parameter, and always returns at least one value as the value to be used for

the function reference in the expression in which it occurs. A subroutine may be called with or without

parameters and may return values only through its parameters or via COMMON. BLOCK DATA subprograms
contain specification statements only and are used to specify initial values for variables in COMMON.

The transmission of arguments between a subprogram reference and the subprogram itself is accomplished

by the use of dummy variables within the subprogram definition. The dummy variables in the subprogram
are listed in the subprogram definition statement. References to the subprogram may then supply values
for these arguments in the same order and be substituted for them whenever they appear in the subprogram.
7.4.8.1

Functions

—

An internal function is defined via a form of the arithmetic assignment statement

and may be referenced only by the program in which it is defined.

Form

name

where

name is the

(arg1,...)

expression

function name;

arg1,... are dummy arguments; these variables will be altered whenever the function is
used and should not be referenced elsewhere in the program; and

expression is the function definition.
Effect

Defines an internal function.

7-59

03/ 78 FOR TRAN IV
An arithmetic statement function definition is a single statement. The expression which defines the function
may include dummy arguments, ordinary variables, external functions and previously defined internal functions.

In the following definition,

ACOSH(X)

(EXP(X/ A) + EXP(— X/ A))/ 2.

X is a format argument and A an ordinary variable. When the function is referenced, the current value of A and
the supplied value of X will be used to evaluate it. All function definitions of this type must precede the first

executable statement of the program in which they appear, and follow the last specification statement appearing
in the program.

7.4.8.2

FUNCTION Statements

An external function, one which may be referenced by other programs,

—

is defined via the FUNCTION statement. A function reference may only appear within an expression and must,

like other elements of expressions, have a specified type.

Type may be specified in the definition itself or via

any other FORTRAN type specification facility.

Form

t FUNCTION name

where

t is an optional type specification (for example, REAL)
name is the

argl
Effect

are

(argl ,...)

function name; and

dummy arguments.

Defines an external function subprogram.

The function name must be a legal symbol that is assigned a value within the subprogram definition. The last
value assigned to this name is the function’s value. There must be at least one argument and arguments must
agree in number, order, and type with actual arguments given by the calling program.
dummy arguments may be given in a FUNCTION statement. For example,

FUNCTION
I I35"

( X

M3310

GE:

A maximum of six

MOO
9

11?.

2’.

'33

0 )

GE]

1‘3 0

RETURN
50

Mm.5

5.0

PZTURN

Dummy arguments may represent the following elements in the function definition: expressions,
alphanumeric strings, array names, or elements and subprogram names. Dummy arguments that represent
array names must appear within the subprogram either in a DIMENSION statement or in one of the type
statements that provide dimension information. Dimensions given as constants must not exceed the
dimensions of the corresponding arrays in the calling program. Dimensions given as dummy variables may
be used to specify adjustable dimensions for array name arguments. For example, in the statement sequence
FUNCTION

TABLE

(AyMyNythrY)

DIMENSION

A(M9N) vB(10) “C(50)

7-60

0S/78 FORTRANIV
the dimensions of array A are specified by the dummy arguments M and N, while the dimension of array B is given
as a constant.

The various values given for M and N by calling the program must be within the limits of the actual

arrays that the dummy array A represents. Various arrays may be substituted for A. These arrays may each be of
different size. Dummy dimensions may only be given for dummy arrays. Note in the example above that array C,

which is not a dummy argument, must be given absolute dimensions. A dummy argument can not appear in an

EQUIVALENCE statement in the function subprogram.
A function may modify any arguments that appear in the FORTRAN arithmetic expression calling the function.
The only FORTRAN statements not allowed in a function subprogram are SUBROUTINE, BLOCK DATA, and

another FUNCTION statement.
A function is called by placing the function name in the program. Control then passes to the function subroutine.
The quantity resulting from the function subroutine replaces the function name in the calling expression.

7.4.8.3

Subroutine Subprograms

A subroutine subprogram is defined external to the program that references it.

Subroutine definition is initiated by a SUBROUTINE statement. A subroutine is invoked by a CALL statement and
returns control to the calling program by means of a RETURN statement.

SUBROUTINE Statement
From

SUBROUTINE name

SUBROUTINE name (argl ,...)
where

name

is subroutine name; and

arg1,... are optional dummy arguments.
Effect

The program that follows is declared a subroutine program.

The arguments in the parenthesized list are dummy arguments representing the arguments of the subprogram. The

dummy arguments must agree in number, order, and type with the actual arguments used by the calling program.
A subroutine subprogram need not have any arguments at all; a maximum of six dummy array arguments are
allowed. When supplied, they may be expressions, alphanumeric strings, array names, array elements, scalar variables, and subprogram names.

Dummy variables representing array names must be dimensioned within the subprogram by a DIMENSION or type
declaration statement. As in a function subprogram, either constants or dummy identifiers may be used to specify
dimensions in a DIMENSION statement. The dummy arguments may not appear in an EQUIVALENCE or COMMON statement in the subroutine program.
A subroutine or function subprogram may use one or more ofits dummy arguments to represent results. For

example,
SUBROUTINE COMPUT (A,B,ANS)

might require the user to supply numeric values for A and B to be computed, and a variable for ANS in which to
store the results. The only FORTRAN statements not allowed in a subroutine
subprogram are FUNCTION,
BLOCK DATA, and another SUBROUTINE statement.

7-61

05/78 FORTRANIV
Constants in call lists of subroutines and function subprograms are not protected. Therefore, a function such as the
following will result in erroneous values:
FNCl

FUNCTION

(Y)

FNClm100.0
YmY+2.0

RETURN
[2".er
CALL Statement
Form

CALL name

CALL name (argl,
name identifies a

where

argl,

subprogram; and

.are actual

arguments.

Control is transferred to the subroutine subprogram.

Effect

The arguments of a CALL statement can be expressions, array names, array elements, scalar variables, alphanumeric

strings, or subprogram names; arguments may be of any type, but must agree in number, order, type, and array
size (except for adjustable arrays, as discussed under the DIMENSION statement) with the corresponding arguments in the SUBROUTINE statement of the called subroutine. A subroutine cannot be referred to as a basic

element in an expression.
7.4.8.4

RETURN Statement

—

The RETURN statement consists of the text

RETURN
This statement returns control from a subprogram (subroutine or function) to the calling program.
last statement executed in a subprogram is a RETURN statement.

Normally, the

Any number of RETURN statements can appear

in a subprogram.

7.4.8.5

BLOCK DATA Statement

—

The BLOCK DATA statement is used to establish a block data subprogram,

which is a data specification subprogram used to enter initial values for variables in common blocks. No executable
statements can appear in a block data subprogram. A block data subprogram is established by a BLOCK DATA
statement consisting of the text

BLOCK DATA
This statement declares that the program that follows is a data specification subprogram and mustbe the first statement of the subprogram.

For example,

DATA

BLOCK

fivaC
COMMON/XKARRAY(100)
COMMON

INTEGER

fiyC

REAL

DQTA

fivaCXSrltSyO/

LUGICQL

fiRRfiY/100*O/

END
This subprogram causes blank common to be initialized with the integer 5, a real variable 1.5, an integer O as its first

three variables and an array with 100 zeros.

7-62

05/ 78 FOR TRAN IV
The subprogram contains only type statements, EQUIVALENCE, DATA, DIMENSION, and COMMON arguments.
A complete set of specifications must be given for an entire common block. A single block data subprogram can
initialize any number of named COMMON blocks.

EXTERNAL Statement

7.4.8.6

—

Function and subroutine subprogram names can be used as the actual argu-

ments of subprograms. When they are, their names must be distinguished from ordinary variables by their

appearance in an EXTERNAL statement.

identifier

Form

EXTERNAL identifier, identifier,

where

identifier is the name of a subprogram.

Effect

Theidentifier is declared a subprogram name and may be used as the argument of other sub-

programs.

Any subprogram name given as an argument to another subprogram must have previously appeared in an external
declaration in the calling program (that is, as an identifier in an EXTERNAL). Example:
EXTERNAL

ERIIGMABZTHETA

Emit

TRI GF' ( SI {mm ! 11. ‘5; y fiNSNEFx‘)

C ALI...

TR I [3F ( TH E. Tfit y 187 7 fiNSbJISEFt‘ )

IESZNL‘I
Tl‘i' III (3F ( FUNK; , ARE! 9 ANSWER )

S?) U I?! F: (I) U T I NEE

fiNSiNEIF:===l'-’lJNC ( fit I13: (3 ‘1

R ETUFIN

EINIII
7.4.9

OS/78 FORTRAN IV Statement Summary

A summary of all OS/78 FORTRAN IV statements is given in Table 7-16.
Table 7-16

FORTRAN IV Statement Summary

‘

Statement
Arithmetic

Effect

Form

The value of expression b is assigned to the

variable a.
Arithmetic statement

The value of expression x is assigned to f(a1...)

t name(a1...)=x

function definition
ASSIGN

after parameter substitution.
ASSIGN n TO v

Statement number n is assigned as the value of

integer variable v for use in assigned GO TO
statement.

Continued on next page

7-63

anFDRnMNIV
Table 7-16 (Cont.)

FORTRAN IV Statement Summary

Form

Statement

Effect

BACKSPACE

BACKSPACE 11

Peripheral device u is backspaced one record.

BLOCK DATA

Identifies a block data subprogram.

CALL

CALL prog
CALL prog(al ...)

arguments when required.

COMMON/blockl/a,b,c/

Variables (a,b,c) are assigned to a common

COMMON

Invokes subroutine named prog, supplying

block.

CONTINUE

No processing, target for transfers.

DATA

DATA var listi/val listl /

Assigns initial or constant values to variables.

DEFINE FILE

Declares a mass storage file for direct access I/O.

DEFINE FILE

a1(b1,cl ,U,vl)...
DIMENSION

(v1..., v7)...

DIMENSION array

Storage allocated according to dimensions
specified for the array.

DO 11 i=ml ,m2,m3

Statements following the DO up to statement
n are

repeated for values of integer variable

1, starting at i=mi, incrementing by m3, ter-

minating when i>m2.
END

END

Cease program compilation; equivalent to
STOP in main program or RETURN in

subprogram.
END FILE

END FILE u

Writes END-OF-FILE character in file u.

EQUIVALENCE

EQUIVALENCE(V1 ,v2,...,),

Declares same storage location for variables
within parentheses.

EXTERNAL

EXTERNAL subprog,

Declares subprograms for use by other subprograms.

FORMAT

FORMAT(specl ,spec2,.../ ...)

Specifies conversions between internal and
external representation of data.

FUNCTION

FUNCTION name(a1...)

Indicates an external function definition.

GO TO

Transfers control to:

U)GOT0n

(1) statement n

(2) GO TO (n1,...,nk),e

(2) to statement n1 if e=1 : to statement nk
if e=k.

Continued on next page

7-64

05/ 78 FOR TRAN IV
Table 7—16 (Cont.)

Effect

Form

Statement

FORTRAN IV Statement Summary

(3) GO TO v
GO TO v(ni,...,nk)
GO TO v,(ni,...,nk)

(3) Transfers control to statement number
assigned to v optionally checking that v
is assigned one of the labels n1...nk.

IF (arith expr)nl ,n2 ,n3

Transfers control to n1 if expr <0, n2 if

expr=0, n3 if expr >0.
IF

IF (logical expr) statement

Executes statement if expression has value

.TRUE., otherwise executes next statement
in sequence.

Logical

V=E

Value of expression E is assigned to variable V.

PAUSE

PAUSE
PAUSE number

Program execution interrupted and number
displayed, if given.

READ (u,f) list

Reads a record from a peripheral device

READ (u,f)

according to specifications given in the

READ (u) list

arguments of the statement.

READ

READ (a’r) list

RETURN

Returns control from a subprogram to the

RETURN

calling program.
REWIND

Repositions designated unit to the beginning of

REWIND u

the file.

STOP

Terminate program execution.

SUBROUTINE

SUBROUTINE name(a1,...)

Declares name to be a subroutine and al

,...,

if supplied, as dummy arguments.

Type Declaration

Where the variables vn are assigned the

type v1,v2,v3,...

indicated type, that is, REAL, INTEGER
or

WRITE

LOGICAL.

WRITE (u,f) list

Writes a record to a peripheral device accord-

WRITE (u,f)

ing to specifications given in the arguments

WRITE (u) list

of the statement.

WRITE (u)
WRITE (a’r) list

7-65

CHAPTER 8
BATCH

OS/ 78 BATCH processing is ideally suited to lengthy or after-hour jobs consisting of fixed sequences of OS/ 78
system commands not requiring operator intervention. BATCH permits the preparation of a job as a file or
part of a file that can be left for an operator to start and run. Output is returned usually in the form of line
printer listings and/or diskette files that include program output as well as a comprehensive summary (log)
of all action taken by the program, 03/78 BATCH, and the operator.

BATCH provides optional spooling of output files. This allows output to be diverted to fast devices such as
diskettes instead of slower devices such as line printers for later transferral to the slower device. This feature
serves to increase

perceived throughput on a system. A line printer, although optional, makes the use of

BATCH more effective. BATCH will support only input/output devices that are present in the hardware

configuration.
With a few exceptions, BATCH executes standard 08/78 commands.

8.1

BATCH PROCESSING UNDER 05/78

08/78 BATCH maintains an input file and an output log. The BATCH input file is a diskette file consisting
of a series of BATCH commands. The input file must reside on the system device. Its default extension is

.31. Each command in the BATCH input file generally occupies one line.
The BATCH output file is a line printer listing (log) on which BATCH prints job headers, certain messages
that result from conditions within the input file, an image of each line in the input file, and certain types of
user

output. BATCH supports only the LA78 line printer for the output of the BATCH log. Listings, however,
output on the LQP78 line printer. If a line printer is not present in the system, the output file is

can be

displayed on the terminal.
BATCH accepts program and data files from any input device in the system. User output files may be directed
to any output device in the system.

BATCH also permits optional spooling of output files. When spooling is requested, every output request to
a non-file-structured device

output file is assigned a file name from a list of names maintained by BATCH and
Spooling of output files increases

directed to a file-structured spool device instead of the user specified device.

BATCH throughput, permitting slow output operations to be postponed until a more favorable time. For

example, a batch processing run that generates many output listings may be initialized to reroute all listings
to a specified diskette. The listings on the diskette may then be dumped onto the appropriate hard copy device
after the run, when more time is available. The spool device may be either RXAO, RXAl SYS: or DSK:.
,

BATCH is called via the SUBMIT command. The format for a BATCH command string is

.SUBMIT SPDV: <DEV:INPUT/0ption/option
where SPDV: is the device on which to spool output. If “SPDV: <” is not specified, no spooling is performed.

Note that spooling applies only to non-file-structured output. DEVlePUT is the input device and file. The
default extension for BATCH input files is .B1. The default input device is DSK. The Run-Time Options and
their meanings are listed in Table 8-1.

8-1

BA TCH

Table 8-1

Option

BATCH Run-Time Options

Meaning

/E
(Error option)

Treat Monitor errors as non-fatal errors. If /E is not specified,

/H
(Hush option)

Do not output a BATCH log.

/Q
(Quiet option)

Output an abbreviated BATCH log, consisting of $JOB and

Output the BATCH log to the terminal. This option need be
specified only if a line printer is available. If a line printer is
not available, the BATCH log is automatically output to the

(Terminal option)

Monitor errors cause the current BATCH job to be terminated.

$MSG lines.

terminal.

(Unattended mode)

BATCH will not pause for operator response to $MSG lines.

Any attempt to use TTY: as an input device to an unattended
BATCH stream will cause the current job to be aborted.

/V
(Version option)
8.2

Print the version number of 05/78 BATCH on the terminal.

BATCH MONITOR COMMANDS

A BATCH command is a character or string of characters that begins with the first character of a line in the

BATCH input file. Each BATCH command must be followed by a RETURN. The files may contain form
feed characters, but form feed characters are ignored by BATCH on input.
BATCH recognizes four monitor level commands. These commands allowroutine housekeeping operations
in a multi-job, batch processing environment and provide communication between the BATCH user and the

operator. Table 8-2 lists the BATCH monitor commands, which may be considered as an extension of the
03/78 Monitor command set. Note that the first character of the $J OB, $MSG and $END commands is a
dollar sign (SHIFT/4).

Any record that begins with a dollar sign character but is not one of the BATCH monitor commands listed
above is copied onto the output file and ignored by BATCH.
A BATCH processing job consists of a $J OB command line and all of the commands that follow it up to the
next $J OB or $END command.

Normally, all the commands submitted by one user are processed as a single
job, and all output from these commands appears under one job header.
After BATCH encounters a $J OB command, it scans the input file until the next OS/78 command is read.

Any lines that follow the $1 OB command and precede the first command are written onto the log and
ignored by BATCH.
The first character of every command is a dot (.). An exception to this is the use of an asterisk (*) whose
function is to accept a command string that indicates input/output files. It is further described in the
example program given in Section 8.3. The rest of the line contains a command, which should appear in

8-2

BA TCH

Table 8-2

BATCH Monitor Commands

Command

Meaning
Initialize for a new job and display a job header on the output file.

$1 OB

The remainder of the $JOB record is included in the job header but

ignored by BATCH. It should be used for job identification, to
provide correlation between terminal output, line printer output
and spool device output.
Sounds the terminal buzzer and prints an image of the message at

$MSG

the terminal. If the /U option was not specified, implying that an

operator is present, BATCH will pause until any key is struck at the
keyboard. If the /U option was specified, processing continues

uninterrupted.
Terminate batch processing and exit to the Monitor. A $END

$END

command should be the last line of every BATCH input file.

Copy the line onto the output log, then ignore it. BATCH assumes
that every line beginning with a slash is a comment.

standard OS/78 format. However, commands that would be terminated with an ESCape under interactive
OS/78 should be terminated with a dollar sign under BATCH. Every standard command is legal input to
BATCH except the MEMORY command.

However, the CDT command will go to the terminal for input

instead of the BATCH file. It is not usually meaningful to invoke ODT, EDIT, or other interactive programs
under BATCH.

BATCH executes an OS/78 command by stripping off the initial dot character and passing the remainder
of the line to the Monitor. BATCH then passes control to the Monitor, which executes the command as

though it had been typed at the keyboard. Monitor commands that return control to the Monitor level
should be followed by a BATCH monitor command or another Monitor command. The SUBMIT command
can be

used to chain from one BATCH stream to another, if desired.

The general rules and conventions associated with BATCH processing are as follows:
1.

The dollar sign ($) is always in the first character position of the BATCH command
lines $J OB, $MSG and $END.

3.
4.

Each job must have a $JOB and $END command.
08/78 commands can be spelled out entirely or in accordance with the accepted
abbreviations. Also, a dot (.) must precede every command.
Wild cards can be specified only for the 08/78 commands COPY, DELETE,

DIRECT, LIST, RENAME and TYPE.
Comments may only be included as separate comment lines.

GUI Only
7.

80 characters per control statement are allowed.

Continuation file specification lines (where necessary) are specified by an initial

asterisk. For example,

L. {I} If] III

F'Iitlfllft‘: 9 S3 U I} (51! y .33 L] 1'3 175$

2*553LIBC3 v SUI-RI} v 533 U 1-3 E52 ‘lS

BA TCH

8.3

THE BATCH INPUT FILE

The following file is an example of a Batch input file.

$JOB
.DATE

3*JUN"-77

/LIST

ANY

.DIR
$MSG

XhHL/B
INSERT

/LIST

HELP

FILES

DISKETTE
OF