⟦727955478⟧

TextFile

T_I_M_E_F_O_R_B_R_U_G_

80.03.24 SB 3.5 timer skrivning
80.03.25 SB 2.0 timer skrivning
80.04.17 SB 3.5 timer skrivning
80.04.18 SB 7.5 timer skrivning
80.04.22 SB 4.5 timer skrivning
80.05.05 SB 6.5 timer rettelser og skrivning
80.05.06 SB 3.0 timer rettelser og udskrivning
80.05.08 SB 5,0 timer rettelser og udskrivning
80.05.16 SB 875-, \f

ALGOL Begin

A/S Regnecentralen af 1979 January 1981
Information Department RCSL No 42-i1330 \f

AUTHOR: Kjeld Christensen

KEYWORDS: Textbook for education, ALGOL, BOSS, FP, administrative
data processing.

ABSTRACT: The book is a part of a series of textbooks covering
fundamental education in computer science. This one is
an introduction to the programming language ALGOL. The
textbooks are used for teaching purposes at RC Com-
puter, but are also well suited for self-tuition.

P_R_E_F_A_C_E_

After publishing an RC8000 basic software textbook it is quite
natural that the next publication in Regnecentralens textbook
series should deal with ALGOL, - because Algol is the main pro-
gramming language at Regnecentralen. The language is based on the
ALGOL60 report but is much extended in relation to this.

This means, that where in general ALGOL has a reputation for
being a programming language specially suited to arithmetical
problems, and is also a language widely known in connection with
articles in periodicals where it is generally used in the publi-
cation of algorithms, the version of ALGOL used at Regnecentralen
has reached the stage of development where it is a language very
suitable for the solution of administrative problems also.

This book describes the basic use of ALGOL and it will be fol-
lowed by another textbook dealing with input/output handling
using ALGOL amongst other things.
Programmer Kjeld Christensen from Handelshøjskolens Datacenter in
Århus is the author of the book.

The good and constructive co-coperation shown by Kjeld Christen-
sen in the work on this manual is much appreciated, and thanks
must be given to Handelshøjskolens Datacenter for the positive
attitude to the project.

The book is meant to be used as part of a Regnecentralen course
but at the same time is written with the aim that it can be used
for self function. Thus suggested solutions are given to the
examples in the textbook.

It must be emphasized that this book is a textbook and not a ma-
mual. That is, the book does not describe all the possibilities
of the ALGOL language but on the contrary only a limited and ma-
nageable selection. It is thus recommended that the student, as
soon as possible after this book is studied, uses the relevant\f

manuals in daily work so that one does not become too dependent
on the textbook. If there are cases where one is in doubt the re-
levant chapter in the textbook may be referred to.

We hope that the users of this book will be as happy with it as
we are at Regnecentralen.

April 1979
A/S Regnecentralen \f

iii

T_A_B_L_E_ _O_F_ _C_O_N_T_E_N_T_S_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _P_A_G_E_

1. INTRODUCTION .......................................... 1

2. PROBLEM DESCRIPTION ................................... 4
2.1 A File ........................................... 7
2.1.1 Construction .............................. 7
2.1.2 File Organisation ......................... 9
2.2 File Description ................................. 10
2.2.1 Record Description ........................ 11
2.2.2 Data List ................................. 13
2.3 Form Description ................................. 15
2.3.1 Format Sheet .............................. 15
2.3.2 Form Contents ............................. 16
2.4 A Programming Problem ............................ 17
2.4.1 Problem Description ....................... 17

3. CONSTRUCTION OF THE COMPUTER .......................... 24
3.1 Hardware ......................................... 24
3.1.1 Components ................................ 24
3.1.2 Working Store ............................. 26
3.1.3 Backing Store ............................. 29
3.1.4 Terminal .................................. 30
3.1.5 Reader .................................... 30
3.1.6 Punch ..................................... 30
3.1.7 Line Printer .............................. 31
3.2 Software ......................................... 31
3.2.1 The Operating System BOSS ................. 31

4. PROGRAMMING ........................................... 35
4.1 Primary Store Usage .............................. 36
4.2 Description of the Course of Action .............. 43

5. SOFTWARE .............................................. 52
5.1 LOGIN Routine .................................... 52
5.2 JOB .............................................. 53
5.3 Running a Job .................................... 55 \f

T_A_B_L_E_ _O_F_ _C_O_N_T_E_N_T_S_ _(_c_o_n_t_i_n_u_e_d_)_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _P_A_G_E_

5.4 Run Analysis ..................................... 57
5.5 BOSS Commands .................................... 59
5.5.1 JOB Interrupt ............................. 59
5.5.2 End Use of Terminal ....................... 60
5.5.3 Assignment of Line Numbers ................ 61
5.5.4 Editing ................................... 62
5.5.5 Printing .................................. 65
5.5.6 Job File Handling ......................... 66
5.5.7 Messages from BOSS ........................ 68
5.6 FP Commands ...................................... 69
5.6.1 Message from FP ........................... 70
5.7 ALGOL ............................................ 70
5.7.1 Calling ALGOL ............................. 73
5.8 Illustrative Run Examples ........................ 74

6. ALGOL ................................................. 78
6.1 Character Set .................................... 79
6.2 Identifiers ...................................... 80
6.3 Numbers .......................................... 82
6.4 Basic Symbols .................................... 83
6.4.1 Compound Symbols .......................... 91
6.5 Program Structure ................................ 92
6.5.1 Declarations .............................. 93
6.5.2 Statements ................................ 93

7. DECLARATIONS .......................................... 99
7.1 Simple Variable .................................. 99
7.2 Array ............................................ 101
7.3 Field Variable ................................... 104
7.4 Zone ............................................. 106
7.5 Procedure ........................................ 107
7.6 Uniqueness ....................................... 107
\f

T_A_B_L_E_ _O_F_ _C_O_N_T_E_N_T_S_ _(_c_o_n_t_i_n_u_e_d_)_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _P_A_G_E_

8. EXPRESSIONS ........................................... 110
8.1 Arithmetic Expression ............................ 110
8.2 Logical Expression ............................... 113
8.3 String Expression ................................ 116
8.4 Multiple Choice Expression ....................... 118
8.5 Two-Way Choice Expression ........................ 119

9. STATEMENTS ............................................ 120
9.1 Unconditional Statement .......................... 122
9.1.1 Simple Statement .......................... 122
9.1.2 Compound Statement ........................ 125
9.1.3 Block ..................................... 126
9.2 Selection Statement .............................. 127
9.2.1 Conditional Statement ..................... 127
9.2.2 CASE Statement ............................ 132
9.3 Loop Statement ................................... 139
9.3.1 FOR STEP Statement ........................ 139
9.3.2 WHILE Statement ........................... 143
9.3.3 REPEAT Statement .......................... 144
9.3.4 FOR Statements ............................ 145

10. CHARACTER INPUT AND OUTPUT ............................ 147
10.1 Integer Procedure READCHAR ....................... 147
10.2 Character String ................................. 148
10.3 Integer Procedure READ ........................... 149
10.4 Integer Procedure WRITE .......................... 151

11. USE OF BACKING STORE .................................. 155
11.1 The Procedure OPEN ............................... 155
11.2 The Procedure CLOSE .............................. 156
11.3 Record Input and Output .......................... 156
11.3.1 Integer Procedure INVAR ................... 157
11.3.2 Integer Procedure OUTVAR .................. 158
\f

T_A_B_L_E_ _O_F_ _C_O_N_T_E_N_T_S_ _(_c_o_n_t_i_n_u_e_d_)_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _P_A_G_E_

12. COMMENTS .............................................. 161
12.1 COMMENT .......................................... 161
12.2 END .............................................. 161
12.3 Comment Brackets ................................. 162

13. PROCEDURES ............................................ 163
13.1 Construction ..................................... 165
13.2 Procedure Declaration ............................ 166
13.3 Procedure Call ................................... 166
13.4 Use .............................................. 167

14. PROGRAM DESIGN ........................................ 172
14.1 Problem Definition ............................... 172
14.2 First Stage ...................................... 173
14.3 Second Stage ..................................... 174
14.4 Third Stage ...................................... 176
14.5 Fourth Stage ..................................... 177
14.5.1 Calculate Groups .......................... 178
14.5.2 Calculate pct ............................. 179
14.5.3 Output Group .............................. 180
14.5.4 Terminate Group ........................... 180
14.5.5 Initialisation ............................ 181
14.5.6 Processing ................................ 181
14.5.7 Termination ............................... 183
14.6 The Last Stage ................................... 183

15. RECOMMENDED SOLUTIONS ................................. 187

A_P_P_E_N_D_I_C_E_S_:

A. CHARACTER VALUES AND CHARACTER CLASSES ................ 201

B. SELECTED MESSAGES FROM BOSS ........................... 205

C. SELECTED MESSAGES FROM FP ............................. 209
\f

vii

T_A_B_L_E_ _O_F_ _C_O_N_T_E_N_T_S_ _(_c_o_n_t_i_n_u_e_d_)_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _P_A_G_E_

D. SELECTED MESSAGES FROM ALGOL .......................... 210

E. SELECTED MESSAGES FROM THE RUNNING PROGRAM ............ 213

F. INDEX ................................................. 215

viii
\f

1_._ _ _ _ _ _ _ _ _I_N_T_R_O_D_U_C_T_I_O_N_ 1.

The object of this book is to give a description of the program-
ming language ALGOL, and to show its use particularly with the
view of solving administrative problems. As the title ALGOL BEGIN
infers the book is mainly for the use of readers without ALGOL
programming experience.

In the book the necessary concepts are presented in such a way
that readers without any EDP experience can understand them.
Naturally this means that those with programming experience must
be prepared for a little revision, as the basic terms and the
fundamental description tools must necessarily be introduced be-
fore the actual description and use of the language.

The book is based on a typical but somewhat simplified admini-
strative problem and using this as a frame of reference various
fundamental terms and some elementary definition methods are in-
troduced. The concepts thus dealt with are closely related to the
chosen example as it is not the intention to define a complete
EDP terminology, but solely to introduce a few concepts which
must be familiar before the work of programming can begin.

Knowledge of the main principles in the contruction and operation
of the computer is also necessary before a start on programming
can be made. This subject is dealt with in chapter 3, also very
concisely, the basis for treating the subject of the book.

In chapter 4 the experience gained from chapters 2 and 3 is used
to compose the first ALGOL program in the book. The basic
structure of the language and of programs is thus introduced.

All programs have a running-in phase, where the completed program
is debugged, the syntactical and logical errors are removed, and
the program functions as designed. This requires the use of a
computer, thus chapter 6 contains guidance in the operation of
the computer, solely concerned with the running-in of examples
given in the book. It is assumed that the computer is equipped
with the BOSS operating system, as is usual for RC4000 & RC8000.
\f

The loose ends are gathered in the following chapters. Chapter 7
contains a more rigorous description of the components of ALGOL.
This description is not complete, dealing only with the most
important elements of the language.

The principle of treating only the most important concepts holds
for the following chapters. These chapters contain syntax des-
criptions for permitted constructions together with semantic
descriptions which explain their meaning. The chapters deal with
declarations, expressions, statements, input, output, and serial
use of the backing store. The accompanying examples give the op-
portunity to become familiar with the new concepts.

Chapter 13 gives a very short introduction to procedures and the
book is completed in chapter 14 by a short introduction to pro-
gram design illustrated by an example.

NOTATION The notation in the book is adopted from the ALGOL ma-
nual ref. 1. The following symbols should be known:

::= the symbol is called a defining equality sign, and is
used in definitions. The concept to be defined is
placed on the left of the symbol, and the definition is
placed on the right

E_x_. <digit' ::= 0 1 2 3 4 5 6 7 8 9

A digit is defined as being a number in the range 0 to
9.

The symbol is used as a separating character between
several possibilities. See the previous example.
\f

< ' The symbol surrounds an explanatory text.

E_x_.
<number' ::= <integer' <real'

The symbol contains a list of several possibilities.

M_m_m_ E_x_. <integer'
<number'::=
P_p_p_ <real'

Brackets may be supplied with a number which relates the number
of times the element may occur

E_x_.
M_m_m_ + 1 14
<integer'::= - <digit'
P_p_p_ 01

An integer may be signed. There must follow at least one digit
and not more than 14 digits.
\f

2_._ _ _ _ _ _ _ _ _P_R_O_B_L_E_M_ _D_E_S_C_R_I_P_T_I_O_N_ 2.

For all problem solving, the problem must be described in such a
way as to give all the necessary information. If the true rate of
interest on a mortgage is to be calculated, the basic rate of
interest and the exchange rate must be known as these are used in
the calculation:

true rate = basic rate x 100/exchange rate.

The calculations may be carried out manually or on a computer as
desired. If the latter, a program must be written which instructs
the computer to execute the simple calculation. In this case the
problem description must be more detailed as the program to solve
the problem must naturally consist of the following three parts:

1. Reading in of the current and basic rates of interest.
2. Calculation of the true rate of interest.
3. Output of the result.

To ensure the correct input of the rates of interest it must be
previously defined in which order these numbers appear, how they
are to be given - for example as reals or integers.
Such details are defined in an input description, so-called. The
user of the program also requires the input description to define
which numbers are valid for the calculation. The calculation
gives no speciel problems as the formula is known.

To be of use the result of the calculation must be output in a
readable form. The method to be used must also be defined as a
number can be output in many ways. E.g. the number of decimal
places should be given.

A detailed description of the problem concerning the calculation
of the true rate of interest can be completed without the use of
special description tools (programs). It is quite another matter
with most administrative problems where normally the input data
as well as the output data is very elaborate.
\f

Even such a simple problem as the printing of a list of all the
items in an item file makes big demands on the input and output
description. The amount of detail necessary is dependent on the
solution of the problem being manual or run on a computer.

If the item file is kept on a card index, the list of records can
be made manually and an experienced employee can solve the pro-
blems despite the vague formulation of the problem. On the other
hand if a less experienced employee is to carry out the work a
more precise description of the problem is required, as problems
will arise which an inexperienced employee cannot solve alone.
Such as:

Which data is to be put on the list?
Where is this data found?

A rough outline of the layout of the list answers the first
question.
E.g. the outline may be thus given:

F_i_g_._ _2_._1_
ITEM REPORT

I_T_E_M_N_O_N_A_M_E_I_T_E_M_N_O_N_A_M_E_

1CPU7Core Store
17Line Printer1293Tape Punch
2047Tape Reader3011Terminal
3001Disc Store3500Print Board

A copy of a card index card with the relevant data solves the
second problem:

F_i_g_._ _2_._2_

ITEMNO = 2047
NAME = TAPEREADER
STOCK = 3000 ITEMS
SALES
JANUARY 200 MAY 375 SEPTEMBER 423
FEBUARY 10 JUNE 3000 OCTOBER
MARCH 119 JULY 25 NOVEMBER
APRIL 517 AUGUST 7 DECEMBER

The problem description is still not detailed enough for a new
employee, and problems such as the following will arise:

How many lines to be written on each page.
How much space to be allowed for an item no.
How much space does an item name use.
Where is the item card index.
How are the cards ordered in the card index.

These questions can be answered with an improved problem descrip-
tion. An outline of the item report can be made which defines how
much space the individual pieces of data may take, and the maxi-
mum number of lines per page can be stated. A more detailed des-
cription of the card index may be given stating position, con-
tents and sequence.

If the item file is stored using a computer, a computer must also
be used to print the item report, and thus more demands are made
on the program description. The computer which is to do the work
is totally without intelligence and all problems must be solved
before the task is handed over to it. Only by instructing (pro-
gramming) it on what to do in all possible situations can the
computer carry out the work.
\f

The programming work is treated in chapter 4. The basis for the
programming work is a detailed problem description in the form of
an input and an output data description. The input data for the
print problem is the item file, and the next section deals with
the concept of files in general, and an appropriate mode of des-
cription is shown. Output data on paper, as shown in section 2.3,
can be described using a format sheet.

2.1
2_._1_ _ _ _ _ _ _ _A_ _F_i_l_e_

2_._1_._1_ _ _ _ _ _C_o_n_s_t_r_u_c_t_i_o_n_ 2.1.1

The item file discussed contains information about different
items and for it to be of any use it is necessary that the infor-
mation is always up to date, so that the file gives a correct
picture of the situation. Each time there is a movement in the
stock, either outflow or inflow, it must be noted in the item
file. Data in the file must thus be stored so that it is easy to
make amendments. This holds true for card index files as well as
files only accessible via the computer. \f

DATA Information found in a state where it can be dealt with
is called data.

FIELD Thus the item file is a collection of data concerning
items. Each piece of information on a card in a card
index has a fixed position. A position is usually cal-
led a field. It is quite common to use the words field
and information interchangeably so that there is no
difference between the concept of a field and the in-
formation it contains.

RECORD A card index card exists for every item containing all
the relevant information about the item. The informa-
tion about the items is thus organised in such a way
that logically coherent data is collected. Such a
logical collection of data is called a record.

KEY A special field exists to establish a unique correspo-
FIELD dence between a record and the object it describes.
This field is called key field or record number, as
liked. In the item file the item number connects the
record with the item.

Ex. 2.1. An identification number connects uniquely a
record in the identification file with a person.

The special meaning of the key field makes it senseless
to change just this field.

FILE The collection of card index cards which constitute the
item file can also be regarded as a collection of
similar records. A reasonable file definition, also
valid in connection with a file accessible via computer
can now be formulated.
\f

A file is a collection of records.

The connection between the different terms is shown more clearly
by fig. 2.3.

This shows that a file is made up of a collection of similar
records. Each record contains a number of fields, and each field
holds information.

F_i_g_._ _2_._3_

F_I_L_E_ R_E_C_O_R_D_ F_I_E_L_D_

RECORD FIELD 1 INFORMATION
1
FIELD 2

FIELD 3
RECORD
2

RECORD
3

2_._1_._2_ _ _ _ _ _F_i_l_e_ _O_r_g_a_n_i_s_a_t_i_o_n_ 2.1.2

The cards which make up the item file may be sorted in different
order, and that chosen defines the organisation of the file. The
following possibilities seem equally suitable:

Sorting after ascending item number order.
Sorting in alphabetical order.
\f

In choosing the type of organisation it is important to take into
account the most common use of the file, as the organisation de-
fines the seeking methods which can be used when the file is to
be utilised. Ex. 2.2 gives an illustration.

E_x_ _2_._2_

Assume that the cards in the item file are ordered
alphabetically using the item name, and the item report
is to be printed in ascending item number order.
Clearly the sorting order of the cards complicates the
job of printing the item report.

Many different storage methods, each catering for a different
utilisation, have been developed for files accessible via a com-
puter. It lies beyond the scope of this book to deal with these
various methods, and in the book only serial files are used

SERIAL In a serial file the records are stored according to
the record number, usually in ascending order.
Searching in the file can only be carried out by
reading the records, starting at the first record and
ending with the required record or the last record in
the file.

END It is difficult to determine when the last record in
RECORD a computer stored file is found, as the records are not
visible as in a card index file. Thus a special end
record is inserted which serves to relate that there
are no more records to be found.

2_._2_ _ _ _ _ _ _ _F_i_l_e_ _D_e_s_c_r_i_p_t_i_o_n_ 2.2

To use a file it must be known how the file is organised and how
the records are constructed. If one is building a new system the
file structure must be described as one quickly forgets details
and the users need to know the system also. Incomplete knowledge
of a system always results in many expensive misunderstandings. \f

The necessity of a good description is made clear in the compre-
hensive technical literature found on the subject. Instructions
as to how a good file description is prepared may be found in the
technical literature. To ensure that all relevant information is
described a number of forms are designed. Thus a fairly good de-
scription may be achieved by completing a number of forms. If it
is a big system lots of forms have to be completed, and if it is
a small system fewer forms are necessary.

For small systems an adequate file description can be made using
two different forms. Records are described using a record de-
scription form, and all the data is described by a so-called
data list. Both of these forms begin with a heading which should
be completed to give the user, date of completion and in what
connection the form was used.

2_._2_._1_ _ _ _ _ _R_e_c_o_r_d_ _D_e_s_c_r_i_p_t_i_o_n_ 2.2.1

This form is used to describe records. The fields in a record are
numbered, and for each field is stated

. Field no.
. Field address
. Field name
. Reference to the data list
. Comment, if required

The fields in a record are numbered 1,2,3, etc.

The field address gives the position of the field in the record.
In a computer stored file a field address can be uniquely defined
by an integer. This is more fully explained in chapter 4.

A field is always named with a descriptive name so that it is
immediately obvious what the field is used for.
\f

A reference to the data list is stated for each field, here the
field is described in detail.

If there is anything special to note a comment is made.

F_i_g_._ _2_._4_

2_._2_._2_ _ _ _ _ _D_a_t_a_ _L_i_s_t_ 2.2.2

All data found in the system is described on this form.
For each item of data is stated:

. Data name
. Abbreviated name
. Data type
. Range of values
. Data definition

An item of data is given a name, and a description name should
always be used, so that the data name makes it immediately
obvious what the information is about.

A descriptive name can be quite long, so for those people
acquainted with the system an abbreviated name which they can
use, is stated.

There are two types of data, namely numerical information (num-
bers) and alphanumerical information (letters and characters). It
is important to distinquish between these two types of data as
they are handled differently in the computer.

Information regarding the range of values is also important, as
this defines the space needed for storage in the computer and
when the data is output.
\f

A complete description of the data is given in the data
definition.

F_i_g_u_r_e_ _2_._5_

2_._3_ _ _ _ _ _ _ _F_o_r_m_ _D_e_s_c_r_i_p_t_i_o_n_ 2.3

To describe the output data a graphic description and a descrip-
tion of the contents of the form are required.

2.3.1
2_._3_._1_ _ _ _ _ _F_o_r_m_a_t_ _S_h_e_e_t_

The form desired is entered on a format sheet, and thus it is
possible to see where each item of information is to be written
and how much space the information may use.

F_i_g_u_r_e_ _2_._6_

2_._3_._2_ _ _ _ _ _F_o_r_m_ _C_o_n_t_e_n_t_s_ 2.3.2

This form is used to describe forms. Each item of information is
numbered, named and a reference to the data list is given. An
additional comment may be given, if required.

F_i_g_u_r_e_ _2_._7_

2_._4_ _ _ _ _ _ _ _A_ _P_r_o_g_a_m_m_i_n_g_ _P_r_o_b_l_e_m_ 2.4

Assume that the item files previously referred to, exist on a
RC8000 computer and an item report is required. To this end an
ALGOL program is to be written. The remainder of this chapter
contains a detailed problem description.

The solution of the problem demands knowledge of the construction
and action of the computer. This is dealt with in chapter 3, be-
fore the problem can be actually tackled in chapter 4.

2_._4_._1_ _ _ _ _ _P_r_o_b_l_e_m_ _D_e_s_c_r_i_p_t_i_o_n_ 2.4.1

On the following pages the problem is described in detail using
the four forms previously named. The item file, a serially stored
file, is described using a record description form and a data
list. The item report is entered on a format sheet, and each
field is described individually on the appropriate form.
\f

RECORD DESCRIPTION\f

DATA LIST\f

FORMAT SHEET\f

FORM CONTENTS\f

3_._ _ _ _ _ _ _ _ _C_O_N_S_T_R_U_C_T_I_O_N_ _O_F_ _T_H_E_ _C_O_M_P_U_T_E_R_ 3.

To obtain full use of the computer a little knowledge of its
function and components in necessary, both for programming and
implementation of programs.

3_._1_ _ _ _ _ _ _ _H_a_r_d_w_a_r_e_ 3.1.

The physical components such as wires and transistors are called
hardware.

3_._1_._1_ _ _ _ _ _C_o_m_p_o_n_e_n_t_s_ 3.1.1

F_i_g_u_r_e_ _3_._1_

BACKING STORE

HIGH

READER WORKING STORE PRINTER

MEDIUM TERMINAL

TERMINAL PUNCH

LOW

COLUMN 1 COLUMN 2 COLUMN 3
\f

The gereral outline of a computer in figure 3.1 shows that the
working store is connected to several units from which or to
which data may be transferred. The arrows show the possible di-
rection. Thus data may be transferred from the reader in Column 1
to the working store whith data may be transferred in both
directions between the working store and the units in Column 2.
For the units in Column 3 data may only be transferred in one
direction, namely from the working store.
Regarding the horizontal division of the diagram it is seen that
the units are grouped corresponding to the speed of which the
data may be transferred. Distinction is made between low, medium,
and high speed.

CENTRAL The control of the activity of the computer is carried
PROCESSOR out by the C_entral P_rocessor U_nit.
The CPU is not shown in figure 3.1.

CONFIGU- The size of the following store and the number of
RATION peripheral units connected determine the configuration
of the computer. The configuration may be altered as
necessary, and figure 3.1 thus shows a possible confi-
guration.

3_._1_._2_ _ _ _ _ _W_o_r_k_i_n_g_ _S_t_o_r_e_ 3.1.2

In the working store, often called the primary store, data to be
processed as well as programs which control the processing, are
kept. By a program is understood the instructions which state how
the processing is to be executed. Chapter 4 describes how a
program may be written. A program must be present in the working
store in order that it may be executed. Data being processed must
also be present in working store.

SOFTWARE When the computer is working, one machine instruction
at a time is read from the working store and the
instruction is executed. There may be several programs
with associated data in the primary storage simultane-
ously (figure 3.2). Then, it is said, the computer is
multiprogrammed.

F_i_g_u_r_e_ _3_._2_

PRIMARY STORE

PROGRAM A

DATA A

PROGRAM B

DATA B

CONTROL The CPU is not alone in controlling the program as a
PROGRAM control program exists and shares this function. Such a
control program is called an operating system.
\f

F_i_g_u_r_e_ _3_._3_

PRIMARY STORE

CONTROL PROGRAM

PROGRAM A

DATA A

PROGRAM B

DATA B

As the control function is a continous process, the operating
system must be permanently active in the primary store, as ap-
posed to other programs, which are only active in the primary
store in the limited time taken to execute the individual
program.

DATA When data is to be transferred between primary store
TRANSFER and a peripheral unit a so-called zone buffer is used.
On transfer from a peripheral unit to primary store
data is placed in the zone buffer where it is available
for processing. A zone buffer is thus a part of the
primary store which contains data for a program. On
transfer from primary store to a peripheral unit the
data to be transferred is placed in a zone-buffer, from
which the transfer may be made.

F_i_g_u_r_e_ _3_._4_

PRIMARY STORE

PROGRAM

DATA

ZONE \f

All this activity is carried out partly because the primary store
is usually very limited, allowing only room for the data being
processed, and partly because all information existing in the
primary store disappears when the computer is turned of, for
example by an electricity cut.

WORD The primary store is divided into a number of equally
large parts, called words. A word is
exactly the amount of primary store which can hold a
machine instruction.

MACHINE A machine instruction is an order which the computer
INSTRUC- can interpret and execute. Therefore a program consists
TIONS of a sequence of machine instructions and when the
program is run the individual instructions are executed
in order.

Data may also be placed in the primary store. All data does not
take up the same amount of space, and depending on the type of
data to be stored it may be held in a half word, a word, or a
double word.
\f

The following table shows the space required:

WORD USE RANGE

Half Integer min -2048
max 2047

Whole Integer min -8388608
max 8388607

Machine
Instruction

Double Integer min -140737488355328
max 140737488355327

Alpha numeric 6 characters

M_m_m_ 616
P_p_p_ Real number min -10
M_m_m_ 616
P_p_p_ max 10

A character may be a letter, a digit, or one of the other symbols
found on a typewriter keyboard.

3_._1_._3_ _ _ _ _ _B_a_c_k_i_n_g_ _S_t_o_r_e_ 3.1.3

Data as well as programs are stored on backing store. There are
different types of backing store. A common characteristic for all
types is that the data is written on a magnetic material, so that
information is kept even though the computer is switched off.
Thus information is only deleted when it is overwritten by new
data. As the processing of data is carried out in the primary
store there are many data transfers between the primary and
backing store. Connection is made by a high speed cable so that
least time possible is used on data transfers.
\f

3_._1_._4_ _ _ _ _ _T_e_r_m_i_n_a_l_3.1.4

Terminals may be used for both input and output. Input is made
using a normal typewriter keyboard. Output appears on paper or on
a screen (display) depending on the terminal type. A terminal may
be connected with a fixed cable or via the telephone network.
Transfer speed is defined by terminal type and mode of connec-
tion.

3_._1_._5_ _ _ _ _ _R_e_a_d_e_r_3.1.5

The computer may be equipped with both a paper tape and a card
reader. Input via one of these units is usually made where large
amounts of data are concerned. For example it could be data
amendments punched on normal registering apparatus.

F_i_g_u_r_e_ _3_._5_

There are several punch codes for both cards and paper tape. The
tape shown in figure 3.5, is punched in flexowriter code.

3_._1_._6_ _ _ _ _ _P_u_n_c_h_3.1.6

Data and programs stored in the computer may be produced on
paper tape using a paper tape punch. One of the various punch
codes is used as desired.

3_._1_._7_ _ _ _ _ _L_i_n_e_ _P_r_i_n_t_e_r_ 3.1.7

A large amount of administrative data processing consists of
printing out file information in various ways. The material to be
output is usually extensive thus output to a terminal would be
too time-consuming.
A line at a time is output on a line printer and thus data is
output at quite a good speed.

3_._2_ _ _ _ _ _ _ _S_o_f_t_w_a_r_e_ 3.2

Programs are called software. Software consists of both control
program and the other utility programs delivered with the com-
puter.
Programs written by the user (application programs) are also
called software.

Section 3.2.1 deals with the basic principles involved in using
the control program and thus in using the computer. A more de-
tailed description is given in chapter 5 with the sole aim of
showing how to use the computer to run the demonstration pro-
grams.

3_._2_._1_ _ _ _ _ _T_h_e_ _O_p_e_r_a_t_i_n_g_ _S_y_s_t_e_m_ _B_O_S_S_3.2.1

The computer to be used to run the demonstration problems, is
expected to be supplied with the BOSS operating system. BOSS at-
tends to tasks in the computer which may be compared to those
tasks, attended to by a librarian in a library.

A librarian lends out books to borrowers with a library card. If
there are several borrowers who wish to borrow the same book, the
librarian decides in which order the book is to be lent out. When
a loan has been made, and the book is not back on time, the li-
brarian must take action and see that the book is returned. Apart
from attending to these administrative functions, the librarian
must also act in an advisory capacity where necessary. The led-
ning of the computer resources is administered by the operating
system, BOSS.
\f

The resources which may be borrowed are the various units shown
in figure 3.1. If a program is to be run the various units must
be borrowed when output or such like is to be carried out.

BOSS has at its disposal a little file containing the names of
the legitimate users of the computer, as well as a list of code-
words which must be declared before permission is granted. BOSS
decides in which order the jobs at the computer are to be execu-
ted. If a user exceeds his run time BOSS takes action and imme-
diately recalls all the resources lent out, which means effecti-
vely that the run is terminated. Queries may also be made to BOSS
concerning the activity of the computer.

E_x_e_r_c_i_s_e_ _3_._1_

State some resources which are administered by the
operating system.

DATA The operating system also carries out many service
TRANSFER functions. A very important function is handling all
the data transfers between the primary store and the
other units. As stated previously, the data to be
output to a peripheral unit should be put in a zone
buffer, thereafter the operating system deals with the
transfer to the correct position. The various positions
to which the data may be transferred are defined by
name. When the position is to be found on backing store
AREA the name refers to an area.

BACKING A user may borrow part of the backing store. The users
STORE share may consist of one or more areas which are
individually named.
\f

F_i_g_u_r_e_ _3_._6_
USER A
AREA 1AREA 2

FREE AREA

AREA 3 AREA 4
USER B

Figure 3.6 illustrates the utilisation of the backing store by
two separate users, and it can be seen that there is free area
also.

AREA NAME A name can have a maximum of 11 characters. The
characters may be letters or numbers. The first
character must always be a letter.

E_x_e_r_c_i_s_e_ _3_._2_

Which of the following names are valid?

storagefile31
storagefile
itemfile
a2
7b
a162:
monkey face

SEGMENT Areas can have different sizes. The size is given in
segments, a segment contains 128 double words.

BLOCK Transfer between the primary store and the backing
store is made in blocks, these contain an integral
number of segments. Block size is determined by the
size of the zone buffer.
\f

As an integral number of whole words is transferred simultaneous-
ly to an area, an area contains an interrupted string of double
words. In certain cases the words are solely made up of charac-
ters, thus the area contains a character string.

DOCUMENT A string of double words is called a document. Documents
are supplied with names, the same rules must be followed
as for area names.

E_x_a_m_p_l_e_ _3_._1_

A string of characters is output to the paper on the
printer, and this is called a document.

A paper tape consists of a character string, and this
is called a document.

Thus an area is also a document.

E_x_e_r_c_i_s_e_ _3_._3_

Data is to be transferred to an area on backing store.
Which of the following block sizes may be used? 128,
256, 24, 368, 512, 917.

PROGRAM Areas may contain many different types of information.
NAME Perhaps an area may contain an employee file or a pro-
gram. Area reference is always made with the help of
the area name. When a program is stored in the area,
the area name is called the program name.
\f

4_._ _ _ _ _ _ _ _ _P_R_O_G_R_A_M_M_I_N_G_ 4.

The foundation is now laid for the solution of the printing pro-
blem stated in chapter 2. It remains to state the problem solu-
tion so that it may be executed by the computer. This can be done
using the programming language ALGOL, and it would be natural at
this point in the book to give a detailed description of the
language. Such a language description is in the meantime rather
complicated, and here the decision has been made to introduce
those elements of the language as they occur in the solution of
the problem.

Using this method the book is less formal and thus less detailed,
but it is done in the hope that the subject matter may be more
easily understood.

There are many details in the printout, shown in figure 2.1,
which complicate the problem solving, so in the first case let us
simplify the result required so that the item report should only
give the item number and name as shown in fig. 4.1

F_i_g_u_r_e_ _4_._1_

1 CPU
7 Core Store
17 Line Printer
1293 Paper tape Punch
2047 Paper tape Reader
3011 Terminal
3101 Disc Store
3300 Print Board

PROGRAM The problem is now to make a set of instructions which
can tell the computer how to act. Such a set of instruc-
tions is called a program and it must, down to the
smallest detail, describe the following:
\f

. The use of the primary store by the program
. Course of action

4_._1_ _ _ _ _ _ _ _P_r_i_m_a_r_y_ _S_t_o_r_e_ _U_s_a_g_e_ 4.1

The data to be processed must be present in the primary store
during the processing, thus the item file records must be read
into the primary store. For this purpose a zonebuffer is required
as stated in section 3.1.2. Space in the primary store must also
be reserved for various working variables, such as field addres-
ses (see section 2.2.1).

Halfwords, words, double words or a consecutive string of any of
these may be reserved for data storage. A reserved entity is
localised by the name, and this name must be stated on reserva-
tion. Fig. 4.2 shows part of the primary store where a halfword,
a word, and a double word are reserved under the names A, B and
C.

F_i_g_u_r_e_ _4_._2_

Data may be held in a reserved element of the primary store. The
word, B, for example, is used to hold the number seven. The num-
ber is put in B using the ALGOL statement:

b:= 7;

DECLARATION Reservation of primary store is stated in ALGOL by
means of a declaration.
\f

E_x_._ _4_._1_

integer b;

The name B is declared to be of type INTEGER, which
means that a word is reserved under the name B.

In a declaration it must be stated which element type is to be
reserved, and under which name the element is to be identified.

E_x_._ _4_._2_

boolean a;

A half word is reserved under the name A.

TYPE The special word which defines the element type is cal-
DECLARA- led a type declarator. INTEGER and BOOLEAN are examples
TOR of this.

IDENTI- The name which identifies the reserved element is cal-
FIER led an identifier. A and B are examples of valid names.

VARIABLE A reserved element is called a variable as it may be
assigned various values during the run of a program.
The name is called the variable name.

Several variables may be reserved in the same declara-
tion.

E_x_._ _4_._3_

long a,b;

Two double words are reserved, and the two variables
are given the names A and B respectively.
\f

It is worth noting that there is no assignment of values in con-
nection with declarations, and before a variable is used it is
normally necessary to give it an initial value, e.g. 0. This can
be done for the variables declared in ex. 4.3 using the following
ALGOL statement:

a:= 0;
b:= 0;

SIMPLE Variables which may be held by a half word, a word, or
VARIABLE a double word are called simple variables. BOOLEAN
variables may be assigned logical values (see section
8.2), INTEGER and LONG variables may be assigned
various integer values as shown in the following
table.

TYPE RANGE

BOOLEAN TRUE
FALSE

INTEGER Min. -8388608
Max. 8388607

LONG Min. -140737488355328
Max. 140737488355327

FIELD As stated when describing records, the individual
VARIABLE fields could only be addressed using a word address.
Addresses of this type are held in field variables.

E_x_._ _4_._4_

integer field nr;

A field variable with the name NR is declared.
\f

A field variable declaration is similar to an ordinary declara-
tion except that the word FIELD is inserted after the type de-
clarator. The type declarator determines whether the field de-
fined by the field variable consists of a halfword, a word, or a
double word. The assignment of a value to a field variable is
expressed in a way corresponding to that for a simple variable.

E_x_._ _4_._5_

integer field no;
no:= b;

ZONE As previously mentioned a zone buffer is required
for the input of records from the item file. A
zone buffer is declared using the declarator ZONE.
It must naturally be named and various parameters
must be stated.

E_x_._ _4_._6_
zone item (128,1,stderror);

Declarator

Name

Specification

Using the zone specification it is possible to tailor the zone
buffer to its current use. In the specification, enclosed by
brackets, 3 items of data, separated by commas, are given. The
first parameter defines the size of the zone buffer given in
double words. The second parameter states whether the zone buffer
is to be devided into shares. The third parameter is the name of
the error routine to be used if something unexpected happens. The
error routine will then give an explanatory message. The zone
declaration shown in ex. 4.6 will be the most suitable for use in
connection with the problems in this book.
\f

FIELD A field in a record read into a zone buffer, is acces-
REFERENCE sed by referring to both the zone and that field vari-
able which defines the required field. The two names
are separated by a full stop.

E_x_._ _4_._7_
Item.no

FIELD A record fills an integral number of words in the zone
ADDRESS buffer. These words contain the record fields, and to
facilitate reference to the individual fields all the
half words in the record are numbered in ascending
order starting with 1. The field address is now defined
by the number of the last half word in the field.

E_x_._ _4_._8_

RECORD IN ZONE ITEM

1 2
3 4
5 6
7 8
9 10
11 12
13 14

zone item (128,1,stderror);
integer field no;

no:= 6;

ITEM.NO will then define the field consisting of half
word 5 and half word 6.

VARIABLE The item name cannot be contained by one double word
thus several must be used. A consecutive set of double
words is called a variable set, and to state that the
field consists of several double words the word ARRAY
is added in the declaration of the field variable which
defines the field. \f

E_x_ _4_._9_

long array field name;

The individual elements in the variable set are numbered 1, 2, 3,
etc. and reference may be made to the variable set as a whole or
to a chosen element.

E_x_._ _4_._1_0_

zone item (128,1, stderror);
long array field name;

ITEM.NAME refers to the array as a whole.
ITEM.NAME (1) refers to the first element.
ITEM.NAME (2) refers to the second element.
ITEM.NAME (3) refers to the third element.

The field address of an array is determined by the number of the
half word which is placed before the first half word of the first
element.

E_x_._ _4_._1_1_

RECORD IN ZONE ITEM

1 2
3 4
5 6
7 8
9 10
11 12
13 14
15 16
17 18

zone item (128,1,stderror);
long array field name;
name:= 6;
\f

The half words 7, 8, 9, and 10 make up the element
ITEM.NAME(1).

In the printout program it is necessary to have a zone buffer for
record input and a field variable as a pointer for the two fields
to be printed out. The field types are described in the data list
section 2.2.2, and the declarations necessary can be made as
shown below; note that declarations are separated by a semicolon.

zone item (128,1,stderror);
integer field no;
long array field name;

Three areas are thus reserved in the working store. A zone buf-
fer, consisting of 128 double words, and two field variables each
consisting of a word. Figure 4.3 illustrates this.

F_i_g_u_r_e_ _4_._3_

WORKING STORE

NAME

ITEM

STANDARD The item report is to be output to a terminal, and a
ZONE zone buffer is to be used in this connection. Usually
both input and output are required, therefore a zone-
buffer for each of these functions is declared automa-
tically. A zone thus declared is called a standard
IN zone. The standard zone, IN, is used for input, and the
OUT standard zone, OUT, is for output. OUT is used for the
output of the item report.
\f

STANDARD Most programs need various working variables and
just as standard zones may be declared so may
standard variables, these may be used without
further difficulty. The working store now appears
as shown in fig. 4.4.

F_i_g_u_r_e_ _4_._4_

WORKING STORE

PROGRAM-
DECLARED
VARIABLES

NO
NAME
ITEM

STANDARD
VARIABLES

STANDARD
ZONES

IN
OUT

4_._2_ _ _ _ _ _ _ _D_e_s_c_r_i_p_t_i_o_n_ _o_f_ _t_h_e_ _C_o_u_r_s_e_ _o_f_ _A_c_t_i_o_n_ 4.2

An initial analysis of the problem clarifies the course of action.
The following must be executed:

. Read a record from the item file.
. Output the item number and item name fields.
\f

The above actions must be executed for each record in the item
file, the end record excepted. The process may be further
illustrated by a graph (fig. 4.5).

F_i_g_u_r_e_ _4_._5_

READ

1 INITIALISATION 2 PRINTOUT 3 4

TERMINATION

GRAPH A graph consists of several nodes, which are connected
by edges.

LINE

NODE

The edges are finished with arrows which show the direction to be
taken in the graph. E.g. in the graph, fig. 4.5, the edge may be
followed from node 1 to node 2 but not vice versa.

Nodes in a graph symbolise states, while edges symbolise actions
(activities).
\f

NODE The nodes in the graph shown in fig. 4.5 symbolise the
following states:

1. start
2. Ready for output
3. Ready for input
4. Stop

EDGE Each edge in fig. 4.5 is supplied with an explanatory
text which describes the symbolised action.

The edges in fig. 4.5 symbolise the following actions:

. Initialisation
. Input of a record
. Output of item number and item name.
. Termination

The course of action described by the graph may be further
illustrated by a run through of the graph.

GRAPH The initial state is the START state. From this state
RUN an edge leads to a new node. The action symbolised by
THROUGH this edge is called INITIALISATION, and this consists
of accessing the item file and reading the first record
If the first record is an end record, which is theore-
tically possible, an edge leads from node (2) directly
to the last node (4), and the course of action is
terminated.

A more interesting path is followed when the first record is an
actual item record. An edge symbolising the output of the
required data then leads to node (3). There is only one exit from
node (3) the edge symbolising the input of a record, this returns
to node (2). Depending on the record input, one of two possible
edges exitting from node (2) is chosen. The loop is repeated as
long as item records are input and output and input are executed
for each time round the loop. Exit is made from the loop only
when the end record is read.
\f

INPUT A record may be read from the item file on the backing
store using the following ALGOL statement.

invar (item);

INVAR INVAR is the name of a subroutine which carries out the
input, and ITEM is the name of the zone which is to be
used for input. A subroutine is activated by stating
the name and an associated parameter list enclosed in
brackets. In the parameter list for a procedure, as a
subroutine is usually called, information may be trans-
ferred to the procedure, just as information may be
returned by the procedure, using the parameters.

The records which may be input using INVAR must have a defined
construction, the first four halfwords are reserved for internal
data to be used by INVAR. Fig. 4.6 shows this:

F_i_g_u_r_e_ _4_._6_

INTERNAL
PART

USER
PART

The internal part consists of two word-length data items. The
first word contains the length of the record as a number of
halfwords. The second word contains internal control data called
a CHECKSUM.

Actual data information makes up the user part, and this can vary
in size from record to record. The item file is constructed so
that all records are of fixed length. The presence of the
internal part explains why the first user field has the field
address 6, - cf. the record description.
\f

The output of the required data must be supplemented with printed
information, and every time a record is output to the printer the
following must be carried out:

1. Make a line change.
2. Print the item number
3. Print space
4. Print the item name

The following activation of the output procedure WRITE executes
all that is required:

write (out,"nl",1,item.no,"sp",2,item.name);

Procedure name
zone
line-feed
item number
space
item name

WRITE The procedure WRITE is entered by stating the procedure
name followed by a list of parameters enclosed by brac-
kets. The parameter list specifies the zone to be used
for output and the data which is to be output. The in-
dividual data items in the parameter list are separated
by comma, and they are output in the order in which
they occur.

OUT The first parameter must always be the name of the zone
which is to be used. In this case it is the standard
zone, OUT.

"NL",1 The next parameter begins with a boolean constant.
Boolean constant indicate that a certain character is
to be output, and following parameter determines how
many times that character is to be repeated.
\f

The output of the constant "NL" causes a new-line char-
acter to be output, and as the following parameter is
the digit one, only one new-line character is output.

ITEM.NO This parameter refers to the field which contains the
item number, and this is output.

"SP",2 "SP" is a boolean constant, and the following parameter
determines the number of times that character is to be
output. The output of the constant "SP" results in the
output of a space. Thus two space characters are out-
put.

ITEM.NAME This parameter refers to the field which contains the
item name, and the name is output.

Program statements are executed in the order in which they are
written, and the writing of the two statements

invar (item);
write (out, "nl",1, item.no, "sp",2, item.name);

just as many times as there are records in the item file would
give one solution to the printing out problem.

LOOP A more realistic solution is obtained by expressing the
loop in the graph, fig. 4.5, in a WHILE construction,
and the graph is thus converted to the following
program sequence:

invar(item);
while item.no < 4000 do
begin
write (out, "nl",1, item.no, "sp",2, item.name);
invar (item);
end;
\f

In the first statement in this program sequence, the first record
in the item file is read in.
The next statement is a WHILE statement. A WHILE statement begins
with the word WHILE, which is followed by a condition. The value
of this condition may be TRUE or FALSE, dependent on the item
number in the record last input being less than 4000 or not. When
the condition is true the sequential operator DO indicates the
next statement to be executed. In this case the next statement is
a compound statement consisting of the two well-known statements
enclosed by the statement bracket BEGIN END. After the compound
statement is executed the condition is reexamined, and the cur-
rent value determines whether the compound statement is to be
executed again. When the condition is false exit is made from the
WHILE statement and thus from the loop, and the follwing state-
ment in the program is executed.

As the end record has item number, 4000, it is clear that the
program sequence is a direct transformation of the loop in the
graph, fig 4.5, to ALGOL statements.

E_x_e_r_c_i_s_e_ _4_._1_

Assume that the following records exist in the item file.

10 transistors
17 diodes
209 print cards
4000 end record

Use the outlined program sequence for this item file
and state in detail what occurs.

A few details are still required to complete the output program.
The zone ITEM has to be connected the area on the backing store
where the item file is kept, and it must be released after use.
The required initiation of field variables can be easily managed
as shown in examples 4.8 and 4.10.
\f

OPEN A zone is connected to a stated area of the backing
store by using the procedure OPEN. The zone ITEM may be
connected to the area ITEMFILE, where the item file is
assumed to exist, by the following activation of OPEN:

open (item,4,<:itemfile:',0);
Procedure name
Zone
Backing store type
Area name
Error statement

The backing store type determines to which unit the zone is to be
attached. For disc storage, the only type used for the problems
in the book, 4 is stated.

The error statement is used when error routine, defined in the
zone declaration, is activated. If the standard error is to be
used, a zero is given as parameter, this will be sufficient for
the solution of the problems given in the book.

CLOSE The procedure CLOSE terminates the used of a zone,
and the connected area is released. The use of the
zone ITEM may be terminated with the following
statement:

close (item, true);

Procedure
Zone
Release

The last parameter in the procedure call may be used in various
ways, amongst others when a zone is connected to a magnetic tape,
TRUE must always be used in the solution of problems in this book.

The output program may now be put together using the completed
parts, it must be noted that the individual declarations and
statements must be separated by semicolons, and the whole program
should be contained in the brackets BEGIN END.
\f

begin
integer field no;
long array field name;
zone item (128,1,stderror);

open (item,4,<:itemfile:',0);
no: = 6;
name: = 6;

invar (item);

while item.no <4000 do
begin
write (out,"nl",1,item.no,
"sp",2,item.name);
invar(item);
end;

close (item,true);
end
\f

5_._ _ _ _ _ _ _ _ _S_O_F_T_W_A_R_E_ 5.

The last part of the programming effect, where the program is to
be debugged using test runs, demands knowledge of the software.
This chapter describes the use of the software with debugging in
mind. The following must be done:

1. Get permission to use a terminal.
2. Type in the problem the computer is to execute.
3. Make the computer execute the program.

NOTATION In the illustrating examples a special notation is
used, amongst other things to show whether a line
has been typed in by the terminal user or has been
output by the computer.

X shows that an ATTENTION signal is to be sent.

-' indicates that the line is typed in.

Note that a line is not processed before the complete line is
typed in and terminated with a new line character; this cannot be
seen in the examples.

5_._1_ _ _ _ _ _ _ _L_O_G_I_N_ _R_o_u_t_i_n_e_ 5.1

Permission to use a terminal is obtained after a conversation
with BOSS. This dialogue is called the LOGIN routine, and this
may run as shown in ex. 5.1

E_x_ _5_._1_

X att -' boss
type user name and project number

-' kc 1234
in: 1978.09.03 11.04
\f

From ex 5.1 it is seen that the dialogue with the computer is
initiated by sending an ATTENTION message. The message is answer-
ed by the output ATT, the user then types BOSS and a new line
character, the BOSS operating system then asks for identification;
the user replies with a codeword consisting of a name and a num-
ber. These data items are separated by a space character. BOSS
shows acceptance of the codeword by outputting the time of day,
and the terminal is at the disposal of the user.

A so called JOBFILE is also reserved for the user where the fol-
lowing lines, which are to describe the project, are stored.

5_._2_ _ _ _ _ _ _ _J_O_B_ 5.2

A defined project which the computer is asked to execute is cal-
led a job. The job, which can manage the running of the output
program, may be typed in to the JOBFILE as shown in ex. 5.2.

E_x_._ _5_._2_

-' 10 u=algol
-' 20 begin
-' 30 integer field no;
-' 40 long array field name;
-' 50 zone item(128,1,stderror);
-' 60 no:= 6;
-' 70 name:= 6;
-' 80 open(item,4,<:itemfile:',0);
-' 90 invar(item);
-' 100 while item.no < 4000 do
-' 110 begin
-' 120 write(out,"nl",1,item.no,"sp",2,item.name);
-' 130 invar(item);
-' 140 end;
-' 150 close(item,true);
-' 160 end
-' 170 u
-' 180 finis

LINE In ex. 5.2 each line begins with a line number which is
NUMBER separated from the contents by a space, lines typed in
in this fashion are stored in the JOB FILE in ascending
line number order. If any typing errors occur it is not
a catastrophe, as by using the line number it is possi-
ble to access the lines which are to be amended, or
perhaps completely altered. The editing possibilities
are described in section 5.5.4.

FP Apart from the output program (lines 20-160), the
COMMAND typed-in job contains various directions (control
cards), which tell the computer which programs are to
be activated during the job run. \f

A directive is called an FP Command as a special pro-
gram FP undertakes the interpretation and execution of
these commands.

There are 3 FP commands in the job.

The first (line 10) states that the ALGOL compiler is to be acti-
vated and is to compile the typed program to machine instruc-
tions. It is further stated that the compiled program is to be
stored under name U.

The second FP command (line 170) is the name of the compiled
program, and by giving this name it is stated that the compiled
program is to be run.

The third FP command is FINIS, and with this command the FINIS
program is activated, this informs BOSS that the job is to be
terminated.

The three FP commands are executed in the order they are stored
in the JOB FILE, and the job thus consists of compiling the out-
put program, running the compiled program and stating that the
job has no more requirements.

FP commands are further described in section 5.6

5_._3_ _ _ _ _ _ _ _R_u_n_n_i_n_g_ _a_ _J_o_b_ 5.3

A job, described in the JOB FILE, is brought to execution by
asking BOSS to run the job. Requests to BOSS are made by typing
in various commands.

GO The GO command request BOSS to execute a job.
\f

E_x_._ _5_._3_

-' go
finis kc0 at 13.02

The GO command is acknowledged with a message stating the
likely time for the run of the job.

SYNTAX go

FUNCTION When the GO command is used, the job described in the JOB-
FILE is entered to be run. As there may be several jobs
entered, the new job is just inserted in the queue of jobs
waiting to be run. Following a well-defined strategy, BOSS
chooses from the job queue those jobs which are to be run
first. Thus the small jobs with short run times are usually
run first.

On job entry, an expected time for the job to be terminated is calcu-
lated, and output to the terminal user.
The terminal keyboard is then locked, and the terminal may not be used
to run other jobs before the entered job is run.

JOB NAME A job name is run under a job name made up of a user name
and an index. The job name is output in the reply to the GO
command. In ex. 5.4 the job name is KC0.

Ex. 5.4 shows the complete terminal conversation.
\f

E_x_._ _5_._4_

X att -' boss
type user name and project number
-' kc 1234
in: 1978.09.03 11.04

-' go
finis kc0 at 13.02

5_._4_ _ _ _ _ _ _ _R_u_n_ _A_n_a_l_y_s_i_s_ 5.4

The entered job (ex. 5.4) produces various messages during its
run, these are in the following given explanatory comments.

1 : begin output from
15: end the ALGOL compiler
algol end 25 \f

9 transistors Output from the item file.
11 diodes
17 terminals
111 curcuits
900 power supply
1234 disc store
2345 tape station
3456 magnetic tape

end 18 Final output from the out-
put program.
end 2 sec job kc0 Output from BOSS
log opdate 1978.09.03 12.59
''

The output from the ALGOL compiler relates that the compilation
is completed without error, and the compiled program fills 25
segments of backing store. The ALGOL compiler is described in
section 5.7.

The output from the item file is poor as regards layout, but this
will be improved later.

The run of an ALGOL program is always terminated with the message
END XX. The number after END relates how many times a program
segment has been transferred from backing store to main store.
This number may be less than the total number of segments in the
program, as it is only those segments required which are read in.

A job is always terminated with a message from BOSS stating the
run time and termination time.

'' informs that the terminal is free for further use.
\f

5_._5_ _ _ _ _ _ _ _B_O_S_S_ _C_o_m_m_a_n_d_s_ 5.5

A BOSS command is a directive to BOSS concerning the execution of
a certain action. Such a directive must be typed in from a termi-
nal. The GO command is an example of a BOSS command.

There are many BOSS commands which a terminal user may type in,
and in the following section only a limited number are dealt with.
After a job is entered from a terminal the keyboard may not nor-
mally be used until the job is completed. Permision to do so may
be obtained by sending an ATTENTION signal.

Ex. 5.5
X
''
-' kill
The ATTENTION signal receives the reply '', this states
that a BOSS command may be typed. The KILL command is
then typed.

Typing of BOSS commands can cause various error messages to be
output. There are many reasons for these, but the most common are
elementary spelling mistakes and the invalid use of a command.
Appendix B gives a list of the error messages which may arise
from the use of the BOSS commands given in this book.

5_._5_._1_ _ _ _ _ _J_O_B_ _I_n_t_e_r_r_u_p_t_i_o_n_ 5.5.1

KILL A job may be stopped using the KILL command.

E_x_._ _5_._6_

X
''
-' kill
Permission to type a KILL
command is obtained \f

by sending an ATTENTION message.

SYNTAX kill

FUNCTION When the KILL command is typed at the terminal the job
that is entered from that terminal to the job queue
administered by BOSS, is stopped.

5_._5_._2_ _ _ _ _ _E_n_d_ _U_s_e_ _o_f_ _T_e_r_m_i_n_a_l_ 5.5.2

LOGOUT The LOGOUT command states that the terminal is no
longer required.

E_x_._ _5_._7_

-' logout
logged in 3 min operations 5

The LOGOUT command is typed and receives the following
line in reply.

SYNTAX logout

FUNCTION The use of the LOGOUT command gives the following
result:

1. The removal of an entered job, if any
2. All LOGIN areas are removed. (see section 5.5.6.3)
3. An explanatory message concerning the time and the
scope of the use of the terminal is output.

When this is completed the use of the terminal is terminated, and
a new session at the terminal must be initiated with the LOGIN
routine.
\f

5_._5_._3_ _ _ _ _ _A_s_s_i_g_n_m_e_n_t_ _o_f_ _L_i_n_e_ _N_u_m_b_e_r_s_ 5.5.3

AUTOLINE If the AUTOLINE command is used BOSS creates and
outputs line numbers. This saves typing line numbers.

E_x_._ _5_._8_
-' autoline
10 -' u=algol
20 -' begin
30 -' integer field no;
40 -' long array field name;
50 -' zone item(128,1,stderror);
60 -' no:= 6;
70 -' name:= 6;
80 -' open(item,4,<:itemfile:',0);
90 -' invar(item);
100 -' while item.no < 4000 do
110 -' begin
120 -' write(out,"nl",1,item.no,"sp",2,item.name);
130 -' invar(item);
140 -' end;
150 -' close(item,true);
160 -' end
170 -' u
180 -' finis
190 -' X

SYNTAX autoline

FUNCTION After the AUTOLINE command is typed each of the follo-
wing lines is assigned with a line number.
The lines are numbered 10, 20, 30..., etc. The line num-
ber is output concurrently with the typing in, and the
terminal user thus does not have to type the numbers.
Assignment of line numbers is concluded when an ATTEN-
TION message is sent.
\f

5_._5_._4_ _ _ _ _ _E_d_i_t_i_n_g_ 5.5.4

It is nearly impossible to work at a terminal without making
errors. Perhaps a word may be misspelt or a line forgotten.
Various amendment facilities exist to correct such errors.

5_._5_._4_._1_ _ _ _D_e_l_e_t_i_o_n_ 5.5.4.1

A delete character may be used when an error is discovered while
typing a line.

BACKSPACE or RUBOUT deletes the last character
(RUBOUT is eccoed as , BACKSPACE deletes the last character).
A press on CTRL + E at the same time delets the whole line
(CTRL + E is eccoed as ).

E_x_._ _5_._9_

u=alf gol
f is deleted.

u=alf u=algol
the line is deleted and rewritten.

The delete symbol may be used several times in succession, and
the characters in the line are successively deleted from right to
left.

E_x_._ _5_._1_0_
u=fortran algol
fortran is deleted.

5_._5_._4_._2_ _ _ _A_m_e_n_d_m_e_n_t_ _C_o_m_m_a_n_d_s_ 5.5.4.2

It is possible

. to insert
. to delete
. to modify

lines in the JOB FILE. \f

Each line in the JOB FILE is supplied with a line number. When
typing, the line numbering may be chosen freely by the terminal
user. The lines are stored in ascending line number order regard-
less of the order in which they were typed.

INSERT The typing of the job in example 5.2 consisted of
inserting a set of lines into an empty JOB FILE. Thus
new lines are inserted by typing the line number and
the contents of the line, separated by a space.

E_x_._ _5_._1_1_

35 invar(item);

If a line with the given line number already exists, the newest
version of the line will replace the original line.

E_x_e_r_c_i_s_e_ _5_._1_

Define the contents of the JOB FILE after the following
input has been made:

10 p=algol
20 integer sum,a,b;
30 write(out,sum);
15 begin
40 end
10 u=algol list.yes
25 read(in,a,b);
26 sum:=a+b;

Insert an empty line immediately after line 10.

DELETE The typing of a line number alone causes that line to
be deleted in the JOB FILE.
\f

E_x_._ _5_._1_2_

MODIFY A line may be modified, and the following must then be
stated:

. the line number,
. the characters (the graphical picture) which
are to be replaced,
. the new graphical picture to be inserted.

The data must be separated by the same delimiter.
Naturally the delimiter must not be found in
either graphical picture.

E_x_._ _5_._1_3_
50 name: =6;
50/: =/:=/
The graphical picture >: => is replaced by >:=> in line
50. / is used as delimiter.

The graphical picture first stated is searched for, from left to
right. The first time it is met the correction is made, and the
modification executed.

DELIMITER A delimiter may be chosen freely with the following
restrictions:

1. A space following immediatly after the line number
causes the whole line to be inserted, thus a space
may not be used as a delimiter.

2. The delimiter must not occur in the graphical
pictures.

/ > " are given as examples of separators.
\f

E_x_e_r_c_i_s_e_ _5_._2_

The JOB FILE used in exercise 5.1 is to be altered so
that the variable name A, whereever it occurs, is to be
replaced with SALE.

5_._5_._5_ _ _ _ _ _P_r_i_n_t_i_n_g_ 5.5.5

The whole or part of the JOB FILE may be printed using various
commands.

5_._5_._5_._1_ _ _ _V_e_r_i_f_y_ 5.5.5.1

The line last typed or modified is printed using the VERIFY com-
mand.

E_x_._ _5_._1_4_

verify

A chosen line may be printed by adding the line number.

E_x_._ _5_._1_5_

verify 10

A limited group of lines may be printed, by adding the number of
lines to be output.

E_x_._ _5_._1_6_

verify 10 4

line 10 plus the 3 following lines are output.

5_._5_._5_._2_ _ _ _L_i_s_t_ 5.5.5.2

The whole of the JOB FILE may be output using the LIST command. \f

E_x_._ _5_._1_7_

list

If a line number is added then the specified line plus all the
lines with a greater line number will be output.

E_x_._ _5_._1_8_

list 30

A group of lines may be printed out by stating the lower and
upper limits for the group.

E_x_._ _5_._1_9_

list 30 97

5_._5_._6_ _ _ _ _ _J_o_b_ _F_i_l_e_ _H_a_n_d_l_i_n_g_ 5.5.6

An area on the backing store may be reserved for the JOB FILE by
using the GET command. The area is then called a BASIS FILE.

The CLEAR command frees a BASIS FILE, the area then being
returned to normal status. The contents of the JOB FILE may be
copied to an area using the SAVE command.

5_._5_._6_._1_ _ _ _G_E_T_ 5.5.6.1

E_x_._ _5_._2_0_

get text
The area TEXT is transferred to the JOB FILE.

SYNTAX get <area name'
\f

GET An existing basis file, if any, is released, and the
specified area is made a BASIS FILE, thus causing the
contents to be transferred to the JOB FILE.
The lines are numbered, 10, 20, 30 ..., etc.

5_._5_._6_._2_ _ _ _C_L_E_A_R_ 5.5.6.2

E_x_._ _5_._2_1_

clear

The contents of the JOB FILE are erased, and a BASIS
FILE, if it exists, is released.

SYNTAX clear

CLEAR The contents of the JOB FILE are erased and a BASIS
FILE, if it exists, is released.

5_._5_._6_._3_ _ _ _S_A_V_E_ 5.5.6.3

E_x_._ _5_._2_2_

save rctext

The JOB FILE is transferred to the area RCTEXT.

SYNTAX save <area name'

SAVE The contents of the JOB FILE are transferred to the
backing store to an area with the specified name. If a
LOGIN area with this name exists, the contents are
transferred to this area. If not, a new LOGIN area with
the name given is created, and the contents are trans-
ferred. A LOGIN area is saved until the terminal user
removes it, perhaps by using the LOGOUT command. When a
LOGIN area is a BASIS FILE, the contents of the JOB\f

FILE will be erased after transfer back to the BASIS
FILE with the SAVE command.

The last command in the series
get rctext
save rctext
save rctext
thus transfers an empty JOB FILE back to RCTEXT, and
the original contents have disappeared.

5_._5_._7_ _ _ _ _ _M_e_s_s_a_g_e_s_ _f_r_o_m_ _B_O_S_S_ 5.5.7

The conversation with BOSS concerning terminal use, editing, JOB
FILE handling and job control can give rise to various unexpected
messages from BOSS. The reasons are many, but the most common are
probably spelling mistakes and wrong use of BOSS commands.

A selection of the messages output by BOSS masy be found in
appendix B.

5_._5_._7_._1_ _ _ _N_O_ _R_O_O_M_ 5.5.7.1

The messages NO ROOM occurs when there is no capacity to make
further amendments to a JOB FILE. To proceed the JOB FILE must be
transferred to the backing store, and the area used converted to
BASIS FILE. This capacity problem may be solved by using SAVE and
GET commands, and further corrections may be made. The command
which gave rise to the message has not been executed, and must
thus be repeated.

When the VERIFY command is used immediately after a GET command,
the line number of the last line in the JOB FILE is output.

If NO ROOM appears when typing, the VERIFY command thus ensures,
in a simple way, the further input, as shown in the following:
\f

. . .
-' 136 begin
-' 137 integer
no room
-' save rcjob
-' get rcjob
-' verify
510
-' 510 integer
. . .

5_._6_ _ _ _ _ _ _ _F_P_ _C_o_m_m_a_n_d_s_ 5.6

JOB requirements are expressed in JOB FILE with the aid of FP
commands. A program is activated by an FP command, and at the
same time parameters may be transferred to the program.

E_x_._ _5_._2_3_

u=algol list.yes

Call of the ALGOL compiler, accompanied by control
parameters.

E_x_._ _5_._2_4_

u
u data
p=algol txt list.yes bossline.yes

Various FP commands.

M_m_m_ 1 1
SYNTAX <name' = <programname' <parameter list'
P_p_p_ 0 0

Names may consist of at most 11 characters (letters/digits). The
first character must be a letter. \f

FUNCTION The specified program is activated, and any control
parameters are made available to the program.

5_._6_._1_ _ _ _ _ _M_e_s_s_a_g_e_ _f_r_o_m_ _F_P_ 5.6.1

FP commands are read and interpreted by the FILE PROCESSOR pro-
gram, FP. When FP reads an erroneous FP command, one or more
error messages are output, these are often accompained by an
explanatory text, informing where the command in question was
input.

E_x_._ _5_._2_5_

*** FP SYNTAX TESTPROGRAM1
* READ FROM PRIMARY INPUT

The first message states that there is a syntax error
in the name TESTPROGRAM1, and the second states that
the command was read on PRIMARY INPUT, i.e. JOB FILE.

Appendix C contains some of the messages output by FP.

5_._7_ _ _ _ _ _ _ _A_L_G_O_L_ 5.7

COMPILER is the name of both the programming language and the
ALGOL program which can translate an ALGOL text and compile a
program in machine instructions. This section deals
with the program ALGOL or the ALGOL compiler, as it is
called.

A program written in ALGOL cannot be run as it stands, as the
computer can only execute machine code. Before it is run the
program must be transformed into a series of machine instruc-
tions.
\f

SYNTAX When the ALGOL compiler is working, a source text
ERRORS composed of a series of letters and numbers, e.g. the
output program, is read. Compilation can only be made
correctly if the source text is written according to
the defined rules. Syntax errors occur when these rules
are broken, and under compilation any error is noted
and a message output. An error message states when and
where the error occurs as well the type of error.

E_x_._ _5_._2_6_

Assume that line 70 in the output program is, by
mistake, written thus:

70 name = 6

Compilation of the program will give the following
error message:

6. LINE 6.1 DELIMITER

When

Where

Error type

When The compilation is executed with several passes, and
the first number in the error message relates in which
pass the error was discovered.

Where The information consists of a line number followed by
an operand number, which gives the position in the
line. Under compilation each line is given a number,
and it is this number which is used in the message.
\f

Error There are various error types. DELIMITER in ex. 5.26
type means that the delimiter = is placed incorrectly. It
should have been :=.

Appendix D contains a selection of the messages from the ALGOL
compiler.

LOGICAL Even if the compilation is completed successfully,
ERRORS there may still be errors. These are the logical errors
caused by a badly designed program. An incorrectly de-
signed program can come to the stage where the run must
be abandoned. An error message will be output stating
why, and where in the program, the program was aban-
doned.

E_x_._ _5_._2_7_

10 u=algol
20 begin
30 integer field no;
40 long array field name;
50 zone item(128,1,stderror);
60 no:= 6;
70 name:= 6;

90 invar(item);
100 while item.no < 4000 do
110 begin
120 write(out,"nl",1,item.no,"sp",2,item.name);
130 invar(item);
140 end;
150 close(item, true);
160 end
170 u
180 finis

A run of the familiar job where line 80 is absentminded-
ly omitted, will give the following error message.
\f

ZONE STATE 4 INVAR
CALLED FROM LINE 6-7

It is obvious that the error is caused by the zone not
being connected to the item file.

Appendix E contains some of the messages output while the program
is being run.

5_._7_._1_ _ _ _ _ _C_a_l_l_i_n_g_ _A_L_G_O_L_ 5.7.1

The ALGOL compiler may be activated in several ways, and thus it
is possible to control the input as well as the output.

E_x_._ _5_._2_8_

u=algol
u=algol source
object=algol source list.yes bossline.yes

M_m_m_ 1
SYNTAX <object name'=algol <source name'
P_p_p_ 0
M_m_m_ 1 1
list.yes bossline.yes
P_p_p_ 0 0

The syntax given shows just a few of the possibilities for trans-
ferring parameters which can influence execution.

<source name'
When this data is given, the ALGOL text is read from\f

the area with the stated name. Where no name is given
the ALGOL text must occur immediately after the ALGOL
activation.

LIST.YES When this parameter occurs the source text is output
during the compilation. Each line starts with an as-
signed line number thus facilitating the location of
errors.

BOSSLINE.YES
This parameter causes the lines to be output with a
supplementory line number which may be used when edi-
ting with BOSS.

5_._8_ _ _ _ _ _ _ _I_l_l_u_s_t_r_a_t_i_v_e_ _R_u_n_ _E_x_a_m_p_l_e_s_ 5.8

Runs of the exercises in the book may be executed as shown in
example 5.4. Though sometimes it may be easier to put the runs
through in a different way. Many program tools exist, of which
the following will be discussed:

O Using the O program, the output data may be redirected
to an area on the backing store.

CONVERT Using the CONVERT program, an area may be output to the
printer.

SET An area is created using the SET program.

In the example BEGIN ..... END defines an arbitrary ALGOL text.
\f

E_x_._ _5_._2_9_

Separation of the program text and FP commands.

10 begin
20 end
save text
clear
10 u=algol text line.yes bossline.yes
20 u
30 finis

The ALGOL text is typed in and stored in the area TEXT. Then the
contents of the JOB FILE are deleted, and various FP commands
typed in. The ALGOL call states that the sourcetext is to be
found in the area TEXT, and that output is to be made during the
compilation so that each line is supplied with two line numbers.
The first line number may be used for a subsegment editing using
BOSS, while the second is used in connection with error messages
from ALGOL.

The separation of the ALGOL text and FP commands results in the
first set of line numbers corresponding exactly with the line
numbers assigned when the ALGOL text is transferred to the JOB
FILE by the GET command.

E_x_._ _5_._3_0_

It can be an advantage to save the FP commands on the
backing store if several runs are to be made.

10 begih Type in the program.
20 end
save text Save program.
clear Delete JOB FILE.
\f

10 u=algol text Type FP commands.
20 u
30 finis
save command Save FP commands.

go Book the job.
. . .
. . . Wait run results.

get text Fetch ALGOL text.
10/begih/begin Correct ALGOL text.
save text save edited text.

get command Fetch FP commands.
go Book job.

E_x_._ _5_._3_1_

Using SET, O, and CONVERT it is possible to redirect
output data from a job so that it is printed out on the
printer.

10 outfile=set 200 An area of 200 segment is
reserved for output.
20 o outfile Output is redirected to OUT-
FILE
30 u=algol text list.yes
40 u
50 o c Output directed back again,
thus terminating the use of
OUTFILE
60 convert outfile OUTFILE output on the printer
70 finis

The data redirected consists of output from ALGOL, any
error messages from the running program as well as data
output to the standard zone OUT.
\f

E_x_._ _5_._3_2_

Input which is read using the standard zone IN, is
normally read from the JOB FILE.

10 u=algol text
20 u
30 1 2
40 3 4
50 finis

The program u requires 4 numbers, and these are stated
immediately after the call of the program.

E_x_._ _5_._3_3_

The input of data using the standard zone IN can be re-
directed by altering the call of the program. This is
done by stating the name of the data area immediately
after the program name.

10 1 2 Type input
20 3 4
save data Save in area DATA
clear

10 u=algol text
20 u data Call of program so that input taken
from DATA
30 finis
\f

6_._ _ _ _ _ _ _ _ _A_L_G_O_L_ 6.

The output program was written using the familiar characters
found on a typewriter keyboard. Those characters which may appear
in an ALGOL program are dealt with in this chapter, also the com-
bination of characters into symbols, number constants, and names.
The construction of declarations, expressions, and statements
using the symbols, numbers, and names is also dealt with, but not
exhaustively as a detailed explanation is given in the following
chapters. This chapter ends with the last step in the language
description, namely the construction of programs using statements
and declarations.

NOTATION The language description is simplified by using a
special notation.

Definitions are described with the aid of a so called
defining equality sign ::=.

<concept'::= <definition'

In definitions the various possibilities are separated
by vertical stroke.

E_x_._ _6_._1_

<digit'::= 0 1 2 3 4 5 6 7 8 9

A digit is defined to be one of the symbols which ap-
pear to the right of the defining equality sign.

A choise between several possibilities may also be shown by sta-
ting the possibilities in a column enclosed by brackets.

E_x_._ _6_._2_
<integer'
<number'::= <real number'

A number is defined to be either integer or real. \f

When an element may be repeated, it is defined by repetition
statement on the brackets.

E_x_._ _6_._3_
M_m_m_ 1 14
<integer'::= + <digit'
P_p_p_ -
U_U_U_ 0 1

The definition shows that an integer may be signed, and
must consists of at least one digit and at most 14
digits.

6_._1_ _ _ _ _ _ _ _C_h_a_r_a_c_t_e_r_ _S_e_t_ 6.1

The following characters may be included in an
ALGOL-program:

a b c d e f g h i j
k l m n o p q r s t
u v w x y z æ ø å
A B C D E F G H I J
K L M N O P Q R S T
U V W X Y Z Æ Ø Å

0 1 2 3 4 5 6 7 8 9

= - * /
< ' =
. , > : ; ( )
<SP' <NL' <FF'

-
& !
" % ?
\f

Some symbols cannot be represented graphically, and a
special notation is used to designate these characters.

<SP' space.
<NL' new line.
<FF' form feed.

The individual characters are separated by a vertical stroke and
grouped as letters, digits, basic characters and graphic charac-
ters. The characters are used to create indentifiers, numbers,
basic symbols, strings, and to separate the various elements.

6_._2_ _ _ _ _ _ _ _I_d_e_n_t_i_f_i_e_r_s_ 6.2

In connection with the use of working storage it is necessary to
declare indentifiers to define the working storage areas (see
4.1). Procedures (see 4.2) are also defined by names.

Names in ALGOL must obey the following syntax:

M_m_m_ N
<letter'
<name'::= <letter'
<digit'
P_p_p_ 0

The syntax definition shows that a name must begin with a letter,
which may be followed by an arbitrary number of letters and
digits in no definite order.
\f

E_x_._ _6_._4_

The first column consists of various valid names while
the second column contains only invalid names.

a 9
b2 2b
abekat G=7
a99k vat-month
paymentofwages payment of wages

E_x_e_r_c_i_s_e_ _6_._1_

Find the errors in the invalid names in 6.4.

The last name in the first column in ex. 6.4 is rather difficult
to read unhappily so, as the advantage of having explanatory and
easily read names cannot be disconnected. A name may not contain
a space character but an underline character, _ , may be used.
This character may be placed between any two characters in the
name.

E_x_._ _6_._5_

payment _of _wages

_ serves only to help the legibility of a name, and has
no significance from a programming viewpoint.

When choosing a name it must be remembered that various character
combinations are reserved for special purposes.

A name is terminated by <SP', <NL', <FF', or one of the delimi-
ters described in section 6.4

6_._3_ _ _ _ _ _ _ _N_u_m_b_e_r_s_

A distinction is made between integers and reals.

<integer'
<number'::= <real'

An integer must obey the following syntax.

M_m_m_ 1 14
<integer'::= + <digit'
P_p_p_ - 0 1

E_x_._ _6_._6_

+1 -1 +01 0987 88
17 9 -8 000 000 97

Ex. 6.6 shows that it is valid to separate digits with
a space for better logibility. 6.3

A real number consists of a decimal number which may be followed
by an exponent part.

M_m_m_ 1
<real number'::= <decimal number' <exponent part'
P_p_p_ 0

A decimal number must obey the following syntax:
M_m_m_ 1 k 14-k
<decimal number'::= + <digit' .<digit'
P_p_p_ - 0 1 1

The exponent part states a power of 10, and the following syntax
must be obeyed:

M_m_m_ 1 3
<exponent part'::= > + <digit'
P_p_p_ - 0 1

E_x_._ _6_._8_
M_M_m_ 2
P_P_p_ 5.7>2 represents the value 5.7* 10 = 570.0 \f

6_._4_ _ _ _ _ _ _ _B_a_s_i_c_ _S_y_m_b_o_l_s_ 6.4

A basic symbol may consist of a single character or of a group of
characters. The basic symbols are divided into 4 groups:

<letter'
M_m_m_ <digit'
<basic symbol'::=
P_p_p_ <logical value'
<delimiter'

LETTER A letter is defined in exact agreement with the
grouping of the character set in section 6.1.

<letter'::= a b c d e f g h i j
k l m n o p q r s t
u v w x y z æ ø å
A B C D E F G H I J
K L M N O P Q R S T
U V W X Y Z Æ Ø Å

By definition is meant that a letter may be one of the characters
in the list on the right of the defining equality sign.

DIGIT A digit is similarly defined in exact agreement with
the character set grouping, thus:

<digit'::= 0 1 2 3 4 5 6 7 8 9

Letters and digits are included in variables, amongst other
things, as shown in section 6.2. As individual characters they do
not normally have any significance.

Single digit numbers and single letter indentifiers are the
exception.

LOGICAL The third group of basic symbols is the logical values.
VALUE Logically a statement may have the value true or the
value false. \f

The relation >item.no < 4000> in the printout program
could thus be true or false. A logical value is defined
thus:

<logical value'::= true false

TRUE and FALSE represent the values true and false.

A number of boolean constants may be used by writing individual
characters.

<boolean constant'::= "<character symbol'"

DELMITER The fourth group of basic symbols is a motley selection
which may be subdivided using the following definition:

<operator'
M_m_m_ <separator'
<delimiter'::=
P_p_p_ <parentheses'
<declarator'

OPERATOR In a common arithmetic expression such as

10 + 5*7

two operators exist, + and *.
There is no doubt as to what the expression means, and
that the value of the expression is 45.

To be able to calculate the expression it is necessary to know
the operations which the operators symbolise, as well as to know
the order of operation. It is not insignificant if the addition
is executed then the multiplication or vice versa.

Operators for various uses exist in ALGOL.
\f

<arithmetic operator'
<relational operator'
<operator'::= <logical operator'
<sequetial operator'

ATITHMETIC Arithmetic expressions may be included in ALGOL
OPERATOR programs, and the usual arithmetic operators may
naturally be used. The denifition below shows the
most common.

<arithmetic operator'::= + -
* **
/ // mod

+, - and / have their usual meaning.
* multiplication sign
** to the power of
// integer division
mod the calcutation of the remainder on
integer devision.

E_x_._ _6_._9_

7*5 = 35
10 ** 2 = 100
101 // 10 = 10
101 mod 10 = 1

The operators // and MOD may not be used in connection with real
numbers.

Arithmetic expressions in ALGOL programs are calculated using the
same rules as normal arithmetic expressions. The operators have
the following precedence.

1. **
2. * / // mod
3. + -
\f

E_x_e_r_c_i_s_e_ _6_._2_

Calculate the value of the following expression:

7-8*10/2+3

RELATIO- The relational operator < has already been used in the
NAL printout program. This operator was used to compare two
OPERATOR numbers, namely the value of the field, ITEM.NO, and
the constant 4000. The result of the comparison may be
true or false. That is, a logical value, represented in
ALGOL by TRUE or FALSE.

<relational operators' ::= < <= = '= ' <'

The operators have the following significance:

< less than
<= less than or equal to
= equal to
'= greater than or equal to
' greater than
<' not equal to

E_x_e_r_c_i_s_e_ _6_._3_

Put a cross in the correct column.

RELATION TRUE FALSE

5 < 7

6 ' 80

7 <= 2

5 <' 5

9 '= 5 + 4

2 = 7 - 5 \f

LOGICAL A logical operator may be used in a logical expression,
OPERATOR that is an expression which takes the value TRUE or
FALSE. A relation is an example of a logical expression.
The definition below includes the 3 most common logical
operators.

<logical operator'::= -, and or

E_x_._ _6_._1_0_

item.no ' 2000 and item.no < 3000

The logical expression has the value TRUE when the item
number is greater than 2000 and less than 3000.

The significance of the three operators is shown in the following
diagram.

p true true false false
q true false true false

NEGATION -, q false true false true

CONJUNCTION p and q true false false false

DISJUNCTION p or q true true true false

Logical operators and expressions are often called boolean
operators and expressions after the English mathematican, George
Boole, who founded symbolic logic.

SEQUEN- In the printout program DO was used to select the next
TIAL statement:
OPERATOR

while item.no < 4000 do
begin
write(out,"nl",1,item.no,"sp",2,item.name);
invar(item);
end;
\f

An operator which selects the next statement or the next
expression, is called s sequential operator.

<sequential operator'::= if then else
case of
for do
repeat
goto

All these operators are explained in detail in the following
chapters.

SEPARATOR In the printout program the individual declarations and
statements are separated by semicolons, and the
parameters in procedure calls are separated by commas.

There are many different separators.

<separator'::= , . > : = := ;
while
step until
comment
<FF' <NL' <SP'

The last separators, those which do not have a graphic picture,
differ from the others in that an arbitrary number of these se-
parators may be used in those positions where such separators are
allowed. This facility was used in the printout program to make
the program legible by beginning several lines with spaces, and
by inserting blank lines (see section 4.2).

It is also valid to surround the separators

, : ; :=

with an arbitrary number of spaces and line feeds.

The usage of the various separators will be clarified in the
order in which the necessity of using them occurs in the \f

following chapters.

BRACKETS In normal arithmetic expressions brackets may be used
to alter the order of calculation. The same is possible
in ALGOL programs.

There are several different kinds of brackets in ALGOL, these are
defined below:

<brackets'::= begin end
( )
<: :'
<< '
<* *'

BEGIN END is called a statement bracket and is used to bracket
statements in a program. The statement brackets was used in the
output program to create a compound statement, similarly the
whole program must be enclosed in a statement bracket.

() is used in expressions, declarations, procedure calls, and on
reference to elements in an array.

E_x_._ _6_._1_1_

(7 + 9)/3
zone item(128,1,stderror);
invar.name(1);

<: :' is solely used to enclose a text string. E.g. this could be
the name of an area:

open(item,4,<:itemfile:',0).

<< ' is used to enclose a layout. When printing, a layout
string states the graphical picture required. This facility
should be used in the output program. \f

E_x_._ _6_._1_2_

<<dddd'

In the layout string a d is given for each digit
position required to be output.

<* *' is used to enclose a comment which is directed only to
someone reading the program. The characters thus enclosed have
therefore no significance in relation to the program.

E_x_._ _6_._1_3_

<*this is a meaningless comment*'.

DECLARATOR
Declarators are used in declarations.

<declarator'::= boolean integer long real
array
field
zone
procedure

Declarators are discussed in chapter 7.

6_._4_._1_ _ _ _ _ _C_o_m_p_o_u_n_d_ _S_y_m_b_o_l_s_ 6.4.1

The basic symbols in the following list are called compound. In
an ALGOL program a compound symbol must always be separated from
the previous and the following letters and digits. This can be
done using <SP' <NL' or another delimiter which is not it self a
compound symbol. The compound symbols may not be used as identi-
fiers. Several of the symbols in the list will be discussed in
more detail in the book, but all the symbols shown should be
known to avoid invalid use of the names.

abs external own
add extract procedure
algol false real
and field repeat
array for round
begin goto shift
boolean if step
case integer string
comment label switch
do long then
else message true
end mod until
entier of value
extend or while
zone
\f

E_x_e_r_c_i_s_e_ _6_._4_

Analyse the printout program and find all the brackets
delimiters, operators and compound symbols.

E_x_e_r_c_i_s_e_ _6_._5_

Find all the errors in the following version of the
printout program.

begin
integer field no;
long array field name:
zone item(128,1,stderror),
no:= 6;
name:= 6;
open(item,4<:itemfile:'0);
invar(item);
while item,no < 4000 do
begin
write (ont,"nl",2,item.no,"sp",2,item.name)
invare (item);
end;
close(item,true):
end;

6_._5_ _ _ _ _ _ _ _P_r_o_g_r_a_m_ _S_t_r_u_c_t_u_r_e_ 6.5

The rules as to how the characters from the character set may be
combined to form indentifiers, numbers, and basic symbols were
stated in the previous section. It remains to combine these en-
tities into declarations and statements which can form a program.

6_._5_._1_ _ _ _ _ _D_e_c_l_a_r_a_t_i_o_n_s_ 6.5.1

There are various types of declarations.

<type declaration'
<array declaration'
<declaration'::= <field declaration'
<zone declaration'
<procedure declaration'

Declarations are separated from each other and from statements by
using the delimiter semicolon. The syntax and use of declarations
are discussed in detail in chapter 7.

6_._5_._2_ _ _ _ _ _S_t_a_t_e_m_e_n_t_s_ 6.5.2

Statements may be sub-divided thus:

<unconditional statement'
<statement'::= <conditional statement'
<loop statement'

The various statement types are covered in detail in the follow-
ing chapters, the remainder of this chapters deals only with a
few.

<simple statement'
<unconditional statement'::= <compound statement'
<block'

6_._5_._2_._1_ _ _ _S_i_m_p_l_e_ _S_t_a_t_e_m_e_n_t_ 6.5.2.1

ASSIGN- An assignment statement is an example of a simple
MENT statement.
\f

E_x_._ _6_._1_4_

no:= 6;

A simple statement followed by the delimiter, semicolon.

<assigment'::= <identifier':= <expression'

FUNCTION To the right of the dynamic equality sign, as the
delimiter := is called, an expression must be stated,
and the value of this expression is assigned to the
identifier stated on the left.

PROCEDURE A procedure call is also an example of a simple state-
CALL ment.

E_x_._ _6_._1_5_

invar(item);

A simple statement followed by delimiter, semicolon.

<procedure call'::=
M_m_m_ 1
<procedure identifier' (<parameter list')
P_p_p_ 0

FUNCTION The procedure specified is activated, and using the
parameter list it is possible to transfer data to and
from the procedure.
\f

6_._5_._2_._2_ _ _ _C_o_m_p_o_u_n_d_ _S_t_a_t_e_m_e_n_t_ 6.5.2.2

A Compound statement means one or more statements which are
enclosed in statement brackets.

E_x_._ _6_._1_6_

begin
write(out, "nl", 1, item.no, "sp", 2, item.name);
invar(item);
end

<compound statement'::=
M_m_m_ N
begin <statement'; <statement'; end
P_p_p_ 0

The individual statements are separated by a semicolon.

USE When several statements are to be executed, in connec-
tion with a loop for example, it is necessary to define
how many, and this is done by using a compound state-
ment.

RECURSIVE In the definition of a compound statement the concept
STATEMENT appers on the right hand side, this may seen
strange when definition of a statement (see 6.5.2.2) is
remembered. A reduction of the definition gives the
following:

<statement' = begin <statement'; <statement'; end
\f

T_A_B_L_E_ _O_F_ _C_O_N_T_E_N_T_S_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _P_A_G_E_

1. INTRODUCTION ........................................... 1

2. INSTALLATION ........................................... 2

A_P_P_E_N_D_I_C_E_S_:

A. REFERENCES ............................................. 3

B. FUNCTION KEYS .......................................... 4

ii
\f

1_._ _ _ _ _ _ _ _ _I_N_T_R_O_D_U_C_T_I_O_N_ 1.

CalcStar runs under the control of CP/M, which is assumed to be
installed on your RC700/RC855 Microcomputer System.

The CalcStar package you have received contains the following:

- this manual "CALCSTAR for the RC700/RC855 Microcomputer System,
Installation Guide".

- MicroPro documentation of CalcStar:
"CalcStar User's Guide"

- One 8" diskette or one 5 1/4" diskette containing the CALCSTAR
system.

Do not write on your distribution diskette as it is your master
copy, and serves as a last resort backup in case of grave errors.

2_._ _ _ _ _ _ _ _ _I_N_S_T_A_L_L_A_T_I_O_N_ 2.

Start by making a copy of the original CalcStar distribution
diskette, see ref. 1, backup, and continue in the following
way:

1) Transfer all CalcStar files to your CP/M diskette by means of
either PIP or TRANSFER

2) Call the "INSTCS" program which will carry out the
installation of CalcStar

3) Answer to the installation questions as shown:
First time installation of CalcStar (Y/N): N_
CALSTAR installation menu please enter
selection (A/B) : A_

Installation menu A

Please enter selection: <_n_o_>_ _<_c_r_>_
enter 33 for RC855 Work Station
or 34 for RC855 Work Station with\f

A_._ _ _ _ _ _ _ _ _R_E_F_E_R_E_N_C_E_S_ A.

1 RCSL No 42-i1995:
CP/M for the RC700 Microcomputer System, User's Guide

RCSL No 42-i1687:
RC855 Work Station, User's Guide.
\f

B_._ _ _ _ _ _ _ _ _F_U_N_C_T_I_O_N_ _K_E_Y_S_ B.

K_E_Y_: F_U_N_C_T_I_O_N_: C_O_M_M_A_N_D_:

-> GOTO @G

<- NEXT ROW, FIRST COL @Z

RUBOUT (RC702) DELETE CHAR LEFT DEL

DEL CHAR (RC855)

@ CURSOR UP @E

CURSOR DOWN @X

<- CURSOR LEFT @S

-> CURSOR RIGHT @D
\f

T_A_B_L_E_ _O_F_ _C_O_N_T_E_N_T_S_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _P_A_G_E_

1. INTRODUCTION ........................................... 1

2. INSTALLATION ........................................... 2

3. KEYBOARD ............................................... 3

4. SCREEN CONTROL ......................................... 4

5. PRINTERS ............................................... 5

A_P_P_E_N_D_I_C_E_S_:

A. REFERENCES ............................................. 7

B. INSTALLATION PROGRAM ................................... 8

ii
\f

F_ 1_._ _ _ _ _ _ _ _ _I_N_T_R_O_D_U_C_T_I_O_N_ 1.

The present version of the WordStar 3.0 text processing system
has been adapted specially for use on your ITT3290 Intelligent
Terminal Work Station.

WordStar runs under the control of the CP/M operating system. It
is assumed that CP/M has been installed on your ITT3290
Intelligent Terminal Work Station and that you are familiar with
the basic CP/M procedures.

The WordStar package you have received contains the following ma-
terials:

- This manual: "WordStar for the ITT3290 Intelligent
Terminal Work Station, Installation Guide".
- MicroPro's documentation of WordStar including:
"WordStar, General Information Manual"
"WordStar, Reference Manual"
"WordStar, Installation Manual".
- A diskette containing the WordStar system.

Do not write on the original distribution diskette as it is your
master copy and last resort in case of errors.

F_ 2_._ _ _ _ _ _ _ _ _I_N_S_T_A_L_L_A_T_I_O_N_ 2.

Start by making a copy of the original WordStar distribution
diskette - for details on how to do this, see ref. 1, BACKUP.
After the copy has been made, proceed as follows:

Transfer all WordStar files to a CP/M system diskette by means of
CP/M command PIP or TRANSFER (see ref. 1).

The file WS.COM is now ready for you to use with a standard
matrix printer.

In case you are not satisfied with the standard ITT 3290 WordStar
adaption (WS.COM) please follow the instructions in appendix B in
order to install a different adaption.

!_W_A_R_N_I_N_G_!_

Due to a peculiarity in WordStar 3.0 the following sequence will
bring the system down:

R ; "run a program" command
@U ; cancel command
--- ; any normal command
X ; exit command

The remedy for this problem is to avoid the cancel command and to
run a harmless program in place of the cancel command (e.g. STAT).
\f

F_ 3_._ _ _ _ _ _ _ _ _K_E_Y_B_O_A_R_D_ 3.

The standard ITT 3290 WordStar 3.0 enables you to use the special
function keys included on the ITT3290 keyboard.

M_ K_e_y_: =_ _F_u_n_c_t_i_o_n_ =_ _C_o_m_m_a_n_d_

CLEAR = NOT USED =
ERASE INP = DEL CHAR = @G
ERASE EOF = NOT USED =
PF1 = LINE UP = @Z
PF2 = LINE DOWN = @W
PF3 = SCREEN UP = @C
PF4 = SCREEN DOWN = @R
PF5 = INSERT LINE = @N
PF6 = DELETE LINE = @Y
PF7 = DELETE WORD RIGHT = @T
PF8 = HELP MENU = @J
PF9 = BLOCK MENU = @K
PF10 = ON SCREEN MENU = @O
PF11 = PRINT MENU = @P
PF12 = QUICK MENU = @Q
TEST REQ = NOT USED =
PROC MODE = NOT USED =
USM = NOT USED =
CURSR SELECT = NOT USED =
PRINT = PRINT = P
RESET CURSR = POSITION BEFORE PREVIOUS COMMAND = @QP_
<- = DELETE CHAR LEFT = RUB
ESC = ESC = ESC
-> = WORD RIGHT = @F
<- = WORD LEFT = @A
<- = RETURN = RETURN
RESET = STOP COMMAND = @U
ENTER = RETURN = RETURN
PA1/3 = FORMAT (ON SCREEN MENU) = @O
PA2/4 = BLOCK MOVE = @KV_
-> = TAB = @I
INS MODE = INSERT ON/OFF = @V
DEL = BLOCK DELETE = @KY_
DUP = BLOCK COPY = @KC_
FIELD MARK = BLOCK MARKER DISPLAY/HIDE = @KH_
@ = CURSOR UP = @E
= CURSOR DOWN = @X
<- = CURSOR RIGHT = @S
-> = CURSOR LEFT = @D
P_ \f

F_ 4_._ _ _ _ _ _ _ _ _S_C_R_E_E_N_ _C_O_N_T_R_O_L_ 4.

The headings and menues of the standard system are displayed in
the form of inverted and underlined text. For further information
about attributes please see ref. 1, appendix B.

The standard screen options can be changed by patching the sys-
tem. For further information about patching please see ref. 2.

F_ 5_._ _ _ _ _ _ _ _ _P_R_I_N_T_E_R_S_ 5.

The WordStar system is provided with a set of printer options
which make it possible to use almost any type of printer.

You may choose the standard option (this is already done for you
in WS.COM) by typing U in the printer menu during installation
(as described in appendix B of this manual) If, on the other
hand, you wish to choose an alternative printer option, then
please note the following points:

- The printer chosen may make special demands on the communi-
cations protocol and/or the printer driver. Please check
ref. 1 appendix F, ref. 2 chapter 7 and the technical
specifications of the printer.

- The following printers may, however, be used without chang-
ing the printer driver or the communications protocol:

1) OKI 83A, Matrix Printer (RC867)
2) Qume Sprint 9 (RC868)
3) NEC Spinwriter 7710/772 Thimble Printer

To ensure full utilization of the printer facilities, choose
option U for OKI, F for Qume and G for NEC in the printer
menu. If you choose options F or G, INSTALL will advise you
to select a special communications protocol. Answer by typ-
ing U.

F_
\f

F_ A_._ _ _ _ _ _ _ _ _R_E_F_E_R_E_N_C_E_S_ A.

1 RCSL No 42-i2134:
ITT3290 Intelligent Terminal Work Station
CP/M Operating System, User's Guide

2 WordStar, Installation Manual (by MicroPro)

F_ B_._ _ _ _ _ _ _ _ _I_N_S_T_A_L_L_A_T_I_O_N_ _P_R_O_G_R_A_M_ B.

Invoke the Installation Program by typing:
A> I_N_S_T_A_L_L_ _<_c_r_>_

Once you have started INSTALL you should see the following
question at the bottom of the display:

DO YOU WANT A NORMAL FIRST-TIME INSTALLATION OF WORDSTAR?

Answer by pressing the key marked Y. The display will change and
at the bottom of the display you will see the text
m_
OK (Y/N):
p_
This is a question. Answer by pressing the key marked Y.

At the bottom of the display you should now see the text:
m_
PLEASE ENTER SELECTION (1 LETTER):
p_
This is a menu. Respond by pressing the key marked U or the
key for your required selection.

There should now be a new question of type OK(Y/N) at the bottom
of the display. This question is one of a series of questions and
menues which will be displayed on the screen in succession, a new
one appering each time you respond a question or menu. Each time
you see a question of the type OK(Y/N) at the bottom of the
display, press the key marked Y, and each time you see a menu at
the bottom of the display press the key marked U or the key for
your required selection, - just as you did above.

After you have answered the last question there will be a number
of display changes and then the first WordStar menu will appear.
At the top of the menu you should see the text:
m_
EDITING NO FILE
p_ \f

To make the function keys active, you have to exit the WordStar
by pressing the key marked X.

When the CP/M prompt appear, extend the keyboard function keys by
entering the command as shown below.

A> E_X_T_E_N_D_ _W_S_ _<_c_r_>_

where keyboard input is underlined and <cr> means pressing the
return key.

After the following print out:

ITT KEYBOARD EXTENTION VERS.1.0 - 82.09.22
FUNCTION COMPLETE - KEYBOARD TABLES EXTENDED

you have a full function key implementation as shown in chapter
3, Keyboard.

F_
\f

«eof»

DataMuseum.dk

CP/M

⟦727955478⟧ TextFile

Derivation

TextFile