AZ98 02_wordpro Ref Man_Jul83 02 Wordpro Man Jul83
AZ98-02_wordproRefMan_Jul83 AZ98-02_wordproRefMan_Jul83
User Manual: AZ98-02_wordproRefMan_Jul83
Open the PDF directly: View PDF 
.
Page Count: 248
| Download | |
| Open PDF In Browser | View PDF |
HONEYWELL
MULTICS
WORDPRO
REFERENCE
MANUAL
SOFTWARE
MULTICS WORDPRO REFERENCE MANUAL
SUBJECT
Detailed Description of the Multics Word Processing System (WORDPRO),
Including Comprehensive Discussions of the Various Facilities and Related
Commands
SPECIAL INSTRUCTIONS
This manual supersedes AZ98, Revision 1, dated November 1978, and its
addendum AZ98-01A, dated August 1979. Refer to the PREFACE for "Significant Changes;'
The manual has been extensively revised and reorganized. Throughout the
manual, change bars in the margins indicate technical additions and asterisks
denote deletions. Sections 1, 2, 3, A, B, C, and D are either new or completely
rewritten and do not contain change bars.
SOFTWARE SUPPORTED
Multics Software Release 10.1
ORDER NUMBER
AZ98-02
July 1983
Honey",ell
PREFACE
This manual describes the capabilities of WORDPRO, the Multics word processing
system.
WORDPRO is comprised of a set of standard Multics commands that perform
various word processing tasks. Readers of this manual should have a working knowledge
of the Multics command environment. New Users' Introduction to Multics - Part I
and Part II. (Order No. CH24 and CH2S), respectively, are very useful as they provide
programmers and other users with a basic introduction to Multics.
This manual contains references to the Multics Commands and Active Functions.
(Order No. AG92), referred to as the Multics Commands, Multics Programmer's
Reference Manual, (Order No. AG91), referred to as the Multics Reference Manual,
and the Multics Subroutines and I nputlOutput Module, (Order No. AG93), referred
to as the Multics Subroutines.
Often, user-typed lines and lines displayed by Multics are shown together in the
same examples. To differentiate between these lines, an exclamation point 0) precedes
user-typed text. This is done only to distinguish user text from system-generated text,
it is not to be included as part of the input line. Also, a "carriage return" is implied
at the end of every user-typed line.
Note:
Because of page constraints in this document, certain character strings of
data used in examples may not match exactly the information as seen on a
user's terminal. That is, the character strings in examples may be folded
(contained on several lines), whereas the actual interactive (live) session may
display the same information on a single line or multiple lines with
different line breaks than shown.
'e information and specifications in this document are subject to change without notice. This
. cument contains information about Honeywell products or services that may not be available
!."'ide the Fpit.ed Statii'B Ct:>m:mlt. Y01Jr HOnl,!ywp.l1 Marketing Representative.
11
C
9
Honeywell Information Systems Inc .. 1983
File
No.:
1L13
AZ98-02
Significant Changes in AZ98-02
The following commands and subroutine have either changed. been deleted. or are
new in this version.
REVISED
NEW
DELETED
add_symbols
change_sym boIs
compose
process_list
sort_list
append_list
compdv
compose_index
convert_runoff
describe_list
display _comp_dsm
display _list
expand_device_wri ter
expand_list
format_document
hyphenate_word_
modify _list
process_com pout
qedx
Many of the formatting controls in Section 2 were revised and are not identified in the
following list. Only deleted and new controls are included. The deleted controls were
undocumented in this release, but will remain fully supported for an indefinite period.
NE¥l
DELETED
'"
.bb
.bbe
.bbe
.bbi
.be
. bee
. bee
.br
.cbn
.fl
.fIa
. fIe
.fIo
.hI
.hIa
.hIe
.hIo
.ps
.sp
.spc
.tb
.tbb
.tbe
.tIc
.tlh
.tre
.trf
.unn
.wi
+
*
.bbt
.bcf
.bch
.bet
.bpf
.bph
.btc
.btt
.efl
.chl
.ecf
.ech
.else
.eIseif
.endif
.epf
.eph
.eqc
.etc
.ett
.fnt
.frf
.fth
.ftr
.gl
. hit
.if
.ift
.indetl
.pfl
.phl
.spt
.tae
. tel
. then
.trn
. ttl
.unb
.unh
The section on "Electronic Mail" was deleted; refer to Mail System Users' Guide,
(Order No. CH23).
iii
AZ98-02
CONTENTS
I
I
Section 1
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
II
I
I
j
I
I
I
I
I
i
I
I
I
I
I
j
I
I
I,
I
I
!
i
I
I
I
I
I
Section 2
General Information ......
Text Editing ........
Wordpro Formatter
.....
Wordpro Text Formatter
General Syntax . . .
Section Organization
Formatting Terminology
Formatting Features · ..
Change Bars . . . . . . . .
.....
Character Translation
......
Comments · .....
Default Conditions ..
· ....
Delimiter, Symbol
.....
Delimiter, Title · ..
Document Indexing .. .....
Hi t Strings . . .
Delimiters .
Hit Types
......
Error Messages · .....
Escaping Characters .....
Escaping from the Formatter · .... .....
Expression Evaluation .....
File Insertion ..........
Files, Auxiliary Output . . . . .
Fon ts and Type Sizes. Changing
Footnotes · ........
Hyphenation . . . . . . . .
Indention · .......
Input Line Continuation
Linespacing ........
Page Definition · ....
Page Headers and Footers
· ....
Page Numbers . . . . . . .
Page Numbers. Structured
Picture Blocks . . . . . . .
.....
Printwheel Changing . . .
Programming Features
Punctuation Space
To::I"hlAC'
.I.u.I.'J ..... .:t,
,J.
V.1J..I..lu. ... "''''''''''''
~.
iv
2-1
2-1
2-2
2-2
2-10
2-10
2-10
2-10
2-10
2-11
2-11
2-11
2-11
2-11
2-12
2-13
2-14
2-14
2-14
2-15
2-15
2-15
2-15
2-16
2-16
2-16
2-16
2-17
2-17
2-17
2-18
2-18
2-18
2-19
2-19
2-19
P,.,.... n"'\o::I1"To~
Tabulation · ....
Text Alignment · .
Text Biocks, Secondary
Text Breaks
Text Filling . . . . .
1-1
1-1
1-1
.....
2-20
2-20
2-20
2-21
2-21
AZ98-02
Text Headers and Captions .
Title Lines
Undention . . . . . . . . .
Variables. Built-in . . . .
Variables, Substitution of
Variables, User . . . . . .
White Space (Extra Lead)
Built-in Variables . . . . . . .
Creating Artwork . . . . . . .
Artwork Conventions .
Artwor k Syntax . . . . . . . .
Artwork Constructs . . . . . .
Formatting Controls . . . . . . . .
Comprehensive Control Summary
2-21
2-21
2-21
2-21
2-22
2-22
2-22
2-22
2-26
2-27
2-28
2-29
2-31
2-58
Section 3
Word pro Commands .. .
compdv . . . . . . . .
compose (comp) .. .
compose_index (cndx)
convert_runoff (cv_rO ..
display _comp_dsm (ddsm)
expand_device_writer (xdw)
format_document (fdoc)
process_compout (pco)
3-1
3-2
3-3
3-8
3-11
3-12
3-18
3-20
3-28
Section 4
WORDPRO Dictionaries . .
Dictionary Use . . . . .
Standard WORDPRO Dictionary
User-Supplied Dictionaries
Dictionary Files . . . . . . . . . . .
Hyphenation . . . . . . . . . . . . . .
When Hyphenation Is Needed
Hyphenation Problems Solved by WORDPRO ..
WORDPRO Hyphenation Technique
Spelling Errors . . . . . . . .
Spelling Error Detection .
Unwant.ed Words . . . . .
Wordlist Segments . . . .
Spelling Error Correction
add_dict._words (adw) ..
count_dict_words (cdw) . .
create_wordlist (cwI)
delete_dict_words (ddw)
find_dict_words (fdw)
hyphenate_word_ ..
list_diet_words (ldw)
locate_words Ow) ..
print_wordlist (pwl)
revise_words (rw) .
trim_word list (twI)
4-1
4-1
4-1
4-1
4-2
4-2
4-2
4-2
4-2
4-4
4-4
4-4
4-5
4-5
4-6
4-8
4-9
4-11
4-13
4-14
Speedt.ype . . . . . . . . .
Speedtyping
Speedtype Features
5-1
5-1
5-1
Section 5
v
.1_1"
--,- .Lv
4-17
4-19
4-21
4-23
AZ98-02
5-1
5-2
Text Segments . . . .
Symbol Dictionaries
Expansion Process
Escapes
.... .
Suffixes .. .
Prefixes . . . .
add_symbols (asb)
change_symbols (csb)
delete_symbols (dsb)
expand_symbols (esb)
find svmbols (fsb) . . .
list_symbols (Isb)
option_symbols (osb) ..
print_symbols_path (psbp)
retain_symbols (rsb)
show_symbols (ssb)
use_sym boIs (usb) .
Section 6
5-3
5-4
5-5
5-6
5-7
5-10
5-11
5-12
5-13
5-14
5-15
5-17
5-18
5-19
5-20
List Processing . . . . .
List Processing Functions
List Processing Files .
Listin File . . .
Lister File . . . .
Listform File . . . . . . . .
Field Insertion
Angle Bracket Escapes
Sorting . . . . . . . . . . . .
Selection . . . . . . . . . . .
Sample List Processing Files
Sample Letter .
append_list (als) .
copy _list (cpls) . . .
create_list (cIs) . .
describe_list (dIs)
display _list (dils) .
expand_list (e1s) .
merge_list (mls) .
modify_list (md1s)
process_list (pIs)
sort_list (sIs) . . .
trim_list (tls) . . . .
6-1
6-1
6-1
6-2
6-3
6-3
6-4
6-5
6-5
6-6
6-8
6-11
6-13
6-14
6-15
6-16
6-18
6-19
6-20
6-23
6-24
6-27
6-29
Appendix A Compose Metacharacter Table
A-I
Appendix B Reference to Commands/Subroutines by Function
Word pro Commands . . . . . . . . . .
Dictionary Commands/Subroutines . .
Speed type Commands . . .
B-1
B-1
B-1
B-2
B-3
T
~"'tL.ll..,"
D~I"\~o",,,,~,,,~
1. 1
V~\o.<"''''lUe
1"""'
............ " ... 14'"
'-'VllJ.J.UU.J.H•• ..,
•••
Appendix C Device Support Tools . . . . . . . . .
Device Writer Source Expander
Expansion Constructs
Expansion Definitions . . . .
vi
C-l
C-1
C-1
C-2
AZ98-02
Variables and Arrays
Scalar Variables .
Array Variables . .
Fixed Arrays
Varying Arrays
List Arrays . .
Stack Arrays .. -. . . . . . . .
Value Assignment . . . . . . . .
Expression Evaluation
Accessing Variables
Scalar Accesses
Subscripted Accesses
Array Accesses . . . . .
Accessing Arguments . . . .
Single Argument Accesses
Multiple Argument Accesses
Argument Count.
Protected Strings . . . .
Arithmetic Expressions
Iteration . . . . . . . .
Condi tional Execution . .
Expansion Calling . . . .
Active Function Calling .
Miscellaneous Features .
Built-in functions
Length Function
Substr Function
Usage Function
Comments . . . . .
Emptying Arrays ..
Error Reporting
General Terminator Token
Null Separator Tokens
Quote Processing
Rescanning
Return . . . . . .
White Space Control
Expansion Tokens ..
Self -terminating Constructs
Matching Character Terminator Constructs
General Terminator Token Constructs
Specific Terminator Token Constructs
Sorted Token List
Reserved Words . . . . .
Annotated Example
Device Writer . . . . .
. . . . .
Variables and Code Fragments . . . . .
Device Table Compiler . . . . . . .
The Device Description Language . .
General Syntax
Literals ..
Comments .
Names
Fonts . . .
vii
C-3
C-4
C-4
C-5
C-5
C-5
C-6
C-6
C-7
C-7
C-7
C-7
C-8
C-9
C-9
C-9
C-IO
C-IO
C-IO
C-12
C-12
C-13
C-14
C-14
C-15
C-15
C-15
C-16
C-17
C-17
,....
1 '"1
\..,..- .11
C-18
C-18
C-19
C-19
C-20
C-20
C-20
C-21
C-21
C-21
C-21
C-22
C-22
C-23
C-24
C-24
C-26
C-27
C-27
C-27
C-27
C-27
C-27
AZ98-02
Braces, Ellipses, and Vertical Lines
Input . . . . . . . . . . . . . . . . .
Range . . . . . . , . , , , , . , , , ,
Output . . . . . . . . . . . . . . . .
Media Characters . . . . . . . . . .
Media Character List . . . . . . . .
Media . . . . . . . . . . . . . . . . . . . ..
Switch . . . . . . . . . . . . . . . . .
Numbers . . . . . . . . . . . . . . .
Syntax of the Sections . . . . . . . . . .
Global Values . . . . . . . . . . . .
Symbol Declarations . . . . . . . . .
Media Character Table . . . . . . .
Media Tables . . . . . . . . . . . . .
View Tables
. . . . . . . . . .
Definitions . . . . . . . . . . . . . .
Font Table . . . . . . . . . . . . . .
Size Table
. . . . . . . . . .
Device Table . . . . . . . . . . . . .
Global/Local Device Values .. . . . . .
Unique Local Device Values . . . . . . .
Artwork Part Descriptions . . . . . . . .
Appendix D Glossary
C-28
C-28
c-28
C-28
C-29
C-29
C-29
C-29
C-29
C-30
C-30
C-34
C-34
C-35
C-36
C-36
C-38
C-39
C-39
C-40
C-41
C-42
D-1
viii
AZ98-o2
SECTION 1
GENERAL INFORMATION
WORDPRO is the Multics word processing system. It consists of a set of related
Multics commands that assist Multics users in the input. update, and maintenance of
high-quality documents. Applications of WORDPRO range from simple form letters to
complex technical manuals. Because the WORDPRO system is integrated into the
Multics operating software, users can develop and maintain all types of documents at
the same time they are performing other data processing activities. Document
preparation is accomplished rapidly, increasing productivity and producing characteristically
superior results. Additionally, WORDPRO provides Multics security, ease of use, and
document management tools that are not available in other systems.
TEXT EDITING
The ted text editor is the suggested editor to be used for WORDPRO application
because of its many powerful features that can simplify the task of maintaining large,
complex documents. (Refer to Multics Text Editor (TED) Reference Manual. (Order
No. CP50) for information about this command.) The ted text editor is but one of
several editors available on Multics (others are edm, emacs, q~x, and teco).
WORDPRO FORMATIER
The WORDPRO system off ers two methods of text formatting by utilizing the
format_document command (the elementary text formatter) or the compose command
(the full-blown WORDPRO text formatter). To decide which command to use, you
must first decide what you want the text formatter to do. The format_document
command is easy to learn and performs quite quickly when compared to either compose
or runoff (see note below). but it performs only a small number of special actions.
(Refer to Section 4, format_document command, for a complete description of
elementary text formatting. including the supported format controls. examples, etc.) The
compose command, on the other hand, is more difficult for a new user to understand
because it offers many more features (refer to Section 2 for a description of the
formatting controls, and to Section 4 for a description of the compose command). In
other words, if you are preparing a simple memo or business letter, you may choose to
use format_document, but if you require more complex items such as footnotes, change
bars, or multicolumn text, you must use compose and its extended group of formatting
controls. Keep in mind that what you learn about format_document also applies to
compose (Le., fO!'!!1c.t_document utilizes a small, compatible subset of compese controls),
so you may wish to first learn about format_document and then switch to compose.
You can use compose to process a document with format_document controls.
Note:
The runoff command. described in the MULTICS COMMANDS, is a
forerunner to compose and is capable of performing only a limited number
of compose functions.
In addition to the formatting choices noted above, WORDPRO provides the
following outstanding features:
1-1
AZ98-02
•
A graphic function for generating simple artwork.
•
An online dictionary f or detecting spelling errors.
•
A Speedtype function for improving productivity.
•
A list-processing function for creating and maintaining data records.
1-2
AZ98-02
SECTION 2
WORDPRO TEXT FORMATTER
The WORDPRO system uses the compose Text Formatter (referred to here simply
as the Formatter) to format text. The input to the Formatter is a file (or group of
files) containing plain text and embedded formatting directives (or controls).
Control arguments supported by the compose command (described
allow the user to "tailor" an invocation of the Formatter by modifying
of default values and initial parameter values. Most importantly, the
device is given with a control argument, thus allowing input files to
device-independen t.
in Section 3)
a wide variety
desired output
be completely
Formatted output may be directed back to the user's terminal or to a file for
eventual transcription to another online device (e.g., a lineprinter or typesetter) or
medium (such as magnetic tape) to be transported to an offline device. If the output is
directed back to the user's terminal, it may be printed page by page to allow
positioning of forms. Pages damaged during printing (e.g., by a paper jam or ribbon
failure) may be reprinted without having to restart the entire document.
GENERAL SYNTAX
A formatting control is a delimited character string having the form .XXX variable-field,
where XXX is a one- to six-character keyword token and variable-field contains
parameter values andlor keywords that are interpreted during processing of the control.
A single blank (ASCI! SP) is required between the keyword token and the variable
field. The delimiter may be either of the delimiters described under input line or
symbol delimiter (both discussed under "Formatting Terminology" below). The interpretation
of the variable field for each control is discussed later in the descriptions of the
controls. If .XXX is the first non-blank string in an input line, then the line is
processed as a control line; otherwise, it is processed as a line of text (also refer to
"Notes" under Formatting Controls later in this section).
In the control descriptions, if a space is shown in a symbolic variable field, then at
least one blank must appear in that position. Any other space given (outside quoted
strings) is ignored.
The symbols that appear repeatedly in variables fields are:
#
an unsigned number. If an integer is needed, the given value is
truncated to next smaller integer.
n
a signed or unsigned number. If an integer is needed, the given
value is truncated to next smaller integer.
expr
a numeric or string expression.
a
any single character.
ab
any character pair.
2-1
AZ98-02
name
a name string up to 32 characters beginning with an alphabetic and
containing only alphanumerics and the underscore L).
string
an arbitrary character string.
title
a three-part title of the form
IpartIl part21 part31.
SECTION ORGANIZATION
The remainder of this section is subdivided as follows:
•
Formatting Terminology (description of technical terms)
•
Formatting Features (description of features that make up the Formatter)
•
Built-in Variables (description of variables)
•
Creating Artwork (description of the artwork features)
•
Formatting Controls (description of Formatter controls. alphabetically organized)
•
Comprehensive Control Summary (grouping of all controls with associated page
reference to the control description)
The Formatter controls can be divided into three groups:
basic
features with which users can process most prose documents having
straightforward formatting requirements.
intermediate
features with which users can process sophisticated technical documents
having complex formatting requirements.
advanced
features with which users can create "macro" procedures that are
context- or device-dependent, or provide for automatic generation
of Tables of Contents. Glossaries, Bibliographies, and Cross Reference
Indexes.
Refer to the "Comprehensive Control Summary" located at the end of this section
for a GROUPING of all controls. Using this summary, a working unit can tailor
responsibilities for individuals or work groups based upon the complexity required for
the job or the individual. That is, the Formatter can be learned in steps of complexity:
basic. intermediate, and advanced.
FORMAITING TERMINOLOGY
The terminology needed for understanding of the formatting features and controls
are discussed here. In addition, any terms appearing below in italics are discussed
separately.
back page
the page to the left when a book is open for reading. It normally has an even
page number"
blank space
some amount of empty space intentionally left in the output line. When
fill i ng text. it is preserved when it appears at the left margin otherwise. it is
2-2
AZ98-02
compressed to a single blank (ASCII space). When not fi /I i ng, it appears in
the output as given.
bottom margin
the amount of white space appearing between the page footer and the bottom
of the page.
built-in variables
a collection of data items maintained by the Formatter whose values are
available to the user by means of substitution of variables.
center
the midpoint between the left margin and the right margin.
change bar
a mark in the page margin space showing that text has been added, deleted, or
modified.
column
a vertical area of text on the page, located by its left and right margins which
determine its width. By convention, "column 0" refers to the entire page and
is always defined.
control
a formatting control as discussed in "General Syntax" above.
control line
a control and all its parameters.
counter variable
a numeric variable whose val-ue is incremented by some specified· amount
every time it is referenced during substitution of variables.
display mode
the number system in which numeric values are shown in substitution of
variables. The available modes are Arabic, binary, octal, hexadecimal, uppercase
and lowercase Roman, and uppercase and lowercase alphabetic.
equation block
an inline text block containing title lines.
escape
a technique for ensuring the literal appearance of certain characters that have
special meaning to the Formatter or whose occurrence causes some special
action.
expression
a construction made up of symbolic references, special references, literal
constants, arithmetic, relational, or logical (Boolean) operators, and conforming
to the normal rules of algebra. An expression may be a numeric expression,
a string expression, a relational expression, or a logical expression.
expression evaiuation
the procedure of reducing an expression by substitution of variables and
perf orming the actions of the operators. The order of precedence in evaluating
mixed expressions is:
numeric expressions
string expressions
relational expressions
logical expressions
2-3
AZ98-02
filling
the procedure of accumulating text from input lines so as to fit as much as
possible into the output line. When not filling, each input text line generates
an output line; over length
margin.
flush both (justified)
text aligned to
both
margins
by
adjustment of
the
wordspace
and/or
I etterspace.
flush left
text aligned to the left margin.
flush right
text aligned to the right margin.
font
the design of a typeface supported by the output device. Examples are "ascii",
"Helvetica", "Times Roman", and "Italian Gothic". Fonts may have several
intensities such as light, medium, bold, and ultra bold, and may be available in
both italic and upright form. A font may also be the name of a collection of
type having no filial association such as "News Commercial Pi" or "Universal
Greek and Math". The fonts supported by an output device are given in its
device description table.
rooter margin
the minimum amount of white space appearing between the last text block (or
footnote) and the page footer.
footnote
a specially formatted text block contaImng additional or descriptive information
that normally appears at the bottom of a page (see footnote description under
"Formatting Features" later in this section).
formatted tables
the formatting style in which the columns are further subdivided into "table
columns". The table columns are independent of each other and each has its
own set of formatting parameters. In this style, the text block is a "table
entry" that may contain text for each of the table columns. The text block is
not finished until processing reverts back to the containing column. Text from
one table column does not spill over into the next. When producing output in
this style, the Formatter is said to be in "table mode".
front page
the page to the right when a book is open for reading. It normally has an odd
page number.
gutter
when more than one column appears on a page, the amount of space between
the right margin of one column and the left margin of the next. If the first
col umn does not He at the page left margin. the page is said to have a "left
gutter".
header margin
the amount of white space appearing between the page header and the first
text block.
2-4
AZ98-02
indention
a local adjustment to the left margin or right margin value. It does not
change the margin value, but does control the position of the first and last
characters in an output line.
input line
the unit of information from the input file processed by the Formatter.
Control lines are delimited by unprotected semicolon (;) characters or NL
(ASCII code 012) characters; text blocks are delimited by NL characters only.
It may contain any combination of text and/or controls. After all control s in
the. line are processed, any remaining text is accumulated into a text block for
eventual placement on the output page.
lead
the amount of space appearing between output lines. For example. ali nespace
value of 1 produces output with zero lead while ali nespace value of 1.5
produces output with a half -line of lead.
leader
a character or short character string that is replicated to fill the blank space
between any tabulation stop position or formatted table column and the end
of text in the preceding one (or the preceding margin). The leader string is
formatted flush right in the blank space.
left margin
the left-most position on the page or column in which text is allowed to
appear.
letterspace
the amount of space appearing between individual characters in the output line.
line art
a simple graphic "picture" constructed from rules and other special symbols
supported internally by the Formatter (see "Creating Artwork" below),
linespace
the amount of automatic vertical advance when proceeding from one output line
to the next.
logical expression
an expression consisting of any number of relational expressions and logical
(Boolean) operators. The order of precedence of the logical operators is:
&
A
I
I
Boolean AND
Boolean EXCLUSIVE OR
Boolean OR
m'ath symbol
one of a subset of the symbols of mathematics that the Formatter can fabricate
to span multiple lines. The subset is limited" to those symbols normally needed
for the syntax definitions of various meta-languages (particularly COBOL) and
the dyadic (or infix) operators (see "Creating Artwork" later in this section).
numeric expression
an algebraic expression consisting of symbolic references to numeric variables
and counter variables, literal numeric constants. other numeric expressions.
arithmetic operators, and parentheses. The order of precedence of the arithmetic
operators is:
arithmetic negative sign
2-5
AZ98-02
+
~':
/
\
+
arithmetic positive sign
multiplication
division
modulus (remainder)
addition
subtraction
numeric variable
a variable whose value may only be a decimal number.
page footer
an optional special text block that appears at the bottom of every page. It may
contain both title lines and normally-formatted text.
page header
an optional special block that appears at the top of every page. It may contain
both title lines and normally-formatted text.
page mode
the state of the Formatter when it is processing text according to page
formatting parameters, not constructing a special text block, and not in any of
the intermediate or advanced special processing modes.
picture block
a form of keep (see text block (intermediate) later in this section) that allows
following text to be inserted ahead of it (also known as a "float").
plain text
an ordinary text paragraph.
pointsize
the size of type in the current font given in typographic points (1 point
1/72 inch).
protected character
a character whose literal presence is ensured. One of the design philosophies of
the Formatter is that its control syntax does not require "blind" (or nonprinting)
characters in the input files. Therefore. there must be a mechanism for
signalling the literal occurrence of the several graphic characters that have
special syntactic meaning. The mechanism chosen is flagging the literal
occurrence by preceding it with the asterisk (*) (also called "escaping" the
literal occurrence).
relational expression
an expression consisting of two numeric expressions or string expressions
and a relational operator. The order of precedence of the relational operators
is:
=
<
<=
>
>=
equal
not equal
less than
less than or equal
greater than
greater than or equal
righ t margin
the right-most position on the page or column in which text is allowed to
appear.
2-6
AZ98-02
rule
a horizontal. vertical. or slant line of some given thickness and length.
running multicolumn
the formatting style
placed side by side
over to the top of
Formatter is said to
in which parallel. identically formatted columns of text are
on the page. Text from the bottom of one column spills
the next column. When producing output in this style, the
be in "multicolumn mode".
special reference
a form of symbolic reference in which the referred object (that is, the string
contained within the symbol delimiters) is not a variable name but has one of
the following forms:
•
•
any Formatter control
an active function (see Multics Commands) also contained within brackets
•
an expression evaluation also contained within braces (n)
([])
string expression
an expression consisting of symbolic references to string variables, literal
string constants, and substring expressions. There are no "string operators";
the only operation is concatenation, that being implied by the positions of the
given strings. A string expression must begin with a double quote (n).
string variable
a variable whose value may be any character string.
substi tution o-f - variables
the procedure by which symbolic references to
variables are replaced with their current values.
built-in variables and user
substring expression
an expression consisting of a symbolic reference to a string variable or a
literal string constant, followed by one or two numeric expressions given in
parentheses, and having the form:
"string" ({i-expr} {,k_expr J})
where i_expr has the value i and k_expr has the value k. Note that k_expr
may not be given unless i_expr is also given. The substring extracted depends
on the values of the numeric expressions as follows:
the substring begins with the ith character of string and
continues through its end.
-i
the substring begins with the i th character f rom the end of
stri ng and continues through its end.
i. k
the substring begins with the ith character of string and
consists of the next k characters.
- i. k
the substring begins with the ith character from the end of
string and consists of the next k characters.
i, -k
the substring begins with the ith character of string and
continues through the kth character from its end.
-i, -k
the substring begins with the i th character from the end of
string and continues through the kth character from its end.
2-7
AZ98-02
In all cases, the length of the substring must be posItIVe (or 0) and neither i
nor k may reference a character outside the range of string.
symbol delimiter
the graphic character used to delimit a reference to a symbolic variable. (see
"delimiter, symbol" under "Formatting Features" later in this section).
symbolic reference
a reference to the current value of a variable, made by enclosing its name in
symbol delimiter characters.
symbolic variable
the reference name of a built-in or user-created variable.
tabulation
the technique of presetting specific horizontal positions on the page and then
signalling an advance to the next such position with a single keystroke. This
technique is directly analogous to "tab stops" on c.t typewriter; however, the
concept is extended by the ability to provide a leader.
text block (basic)
(also see "intermediate" description below) the unit of output produced by the
Formatter. A text block is constructed by accumulating output text from the
input file until a text break is encountered. A text block may contain up to
1000 output lines.
The text blocks supported by the Formatter are plain text paragraphs, white
space, page headers, page footers, footnotes, text headers, and text captions.
Text blocks are classed according to the way they are placed on the page
(primary type) and their content or usage (secondary type). When a text block
is piaced on the page, it may be subjected to widow processing:
Primary blocks are physically separate entities and may not intersect or overlap.
Two primary block types are defined:
inline
text blocks that are placed
on the page immediately upon
completion.
special
text blocks that are handled according to special considerations
determined by their secondary type.
Secondary blocks may overlap and may be contained within each other and
within primary blocks. Several secondary block types are defined; however, for
basic formatting, the only one of interest is:
title
blocks that may contain title lines.
Under certain conditions. the processing of a text block may be suspended.
When a block is suspended, certain parameters and variable values associated
with the block are set aside (or pushed) for resumption of processing in that
block. When the block is resumed. those items set aside are restored to an
active state (or popped) and processing of the block continues as though it had
never been interrupted.
text block (intermediate)
(also see "basic" description above) the remaining secondary block types are:
art
blocks that may contain conventional artwork constructs which
are converted to symbols or line art.
keep
blocks that are not subject to widowing.
2-8
AZ98-02
literal
blocks that may contain lines that appear to be control lines.
text break (basic)
(also see "intermediate" description below) an interruption of normal processing
in the current text block. Three different basic breaks are defined:
format
finish processing the current output line (i.e., finish formatting
any pending text as' a short line), then begin a new output line
in the current text block.
block
finish processing the current text block by finishing the current
output line as above, adding any pending text caption, then
begin a new text block.
page
finish processing the current page by finishing
as above, then finish the page by adding any
and page footer, and eject the page. Any
widowing is picked up at the top of the next
the current block
pending footnotes
text deferred by
page.
text break (intermediate)
(also see "basic" description above) one additional text break is defined:
column
finish the current column by finishing the current text block
and filling out the column with nontrimmable white space, then
begin processing in the next column (or the next page, if the
col umn is the last on the page).
text caption
an optional special block that is i~separably attached to the end of an output
text block. It may contain both title lines and normally formatted text.
text header
an optional special block that is inseparably attached to the beginning of an
output text block. It may contain both title lines and normally formatted text.
title
an output line that is formatted in
part, and a right margin part.
distinction is made between "blank"
in the output and "nUll" titles which
three parts: a I ef t margin part, a center
In page header s and page footer s, a
titles which appear (as white space lines)
do not appear in the output.
title delimiter
the graphic character used to delimit the parts of a title line.
top margin
the amount of white space appearing between the top of the page and the
page header.
user variables
a collection of data items defined and maintained by the user whose values may
be retrieved by means of substitution of variables.
whIte space
some amount of empty space intentionally left on the page; e.g., interparagraph
space appearing at the top of a page or column (discarded), or space left for
later "paste up" of hand drawn artwork (left as-is). It may be trimmed,
discarded, or left as-is.
2-9
AZ98-02
widow
some (usually small) fragment of text that is the least amount that may be split
away from its containing block in the event of a page or column overflow.
The default fragment size is two lines but may be changed by the user.
width
the amount of space in a column or the page available for text.
wordspace
the amount of space appearing between words of justified text in the output
line.
FORMATIING FEATURES
The formatting features are described below.
Change Bars
Text revision marks may be used on a line-by-line basis to show addition, deletion,
or modification of text material. The marks may be associated with a change level
character and the Formatter allows generation of marks based on these characters. By
default, the marks are a vertical line (I) for additions and modifications, and an
asterisk (*) for deletions, and they appear in the outside page margin. See the .cba,
.cbd, .cbf, and .cbm controls below. and the -change_bars and -change_bars_art control
arguments in the description of the compose command in Section 3.
Character Translation
In rare instances it is necessary to translate certain characters in the input file into
other characters in the output, due primarily to unavoidable conflicts among various
characters that have special meaning. For example. it is not possible (because of the
nature of the Multics command processor) to evaluate an active function expression that
contains parentheses or brackets. Such characters must be processed as other characters
without special meaning and translated to the desired characters when the processing is
complete. See the . trn control.
Note:
Due to the complexity of this translation feature and the number of times
it must be performed during file processing, it is quite expensive and users
are advised not to use it indiscriminately. It should be disabled as soon as
the need for it has passed.
Comments
Comments not affecting the formatted output may be placed as desired in the input
files. See the .* control.
Default Conditions
The Formatter initializes the values of the formatting parameters at the beginning
of each input file given in the command line so as to produce the following default
output:
•
8.5 x 11 inch page (85 columns x 66 lines)
2-10
AZ98-02
•
•
•
•
•
1 inch margins; top. bottom, left, and right (6 lines top and bottom; 10
columns left and right)
6.5 inch single-column text; filled, justified, and unhyphenated (65 columns)
no page headers or footers
initial font and pointsize as given in the device description table for the output
device
output written to the user_output I/O switch (normally connected to the user's
terminal)
Certain of these default values may be changed with control arguments given in the
command line that invokes the Formatter (see the description of the compose command
later in this section). All formatting parameter values may be changed locally with
formatting controls in the input file(s).
Delimiter, Symbol
The default symbol delimiter character is percent (%). Although isolated literal
appearances of the character may be protected with the escape convention, certain input
files may require extensive use of the character in lines that are subject to evaluation.
To relieve the user from the necessity of escaping all the literal appearances, the
symbol delimiter character may be changed to any other character whenever desired.
See the .csd control.
Delimiter~
Title
The default title delimiter character is the vertical line (i). Although isolated literal
appearances of the -character in title lines may be protected with the escape convention,
certain input files may require extensive use of the character. To relieve the user from
the necessity of escaping all the literal appearances, the title delimiter character may be
changed to any other character whenever desired. See the .ctd control.
Document Indexing
Many technical documents require a cross reference index showing those pages
whereon certain keywords and phrases are mentioned. The formatter provides the ability
to gather data for such indexes during the processing of the input files. The feature
may also be used to gather data for bibliographies and glossaries. Data for as many as
ten indexes may be gathered simultaneously. See the .hit control.
HIT STRINGS
A hit string is a line of text made up of keywords or phrases and various
delimiters. There are no restrictions on the number or location of hit strings in a
document. When the formatter encounters a .hit control, it prepends the given hit
string(s} with the input line number. a constant string (used for synchronization), the
delimiter change string (if any), and the hit type character, then appends the end
delimiter and page number, and emits it to one of ten raw data files. The receiving
raw data file is specified by an index number given with the .hit control.
2-11
AZ98-02
Del imiters
Three characters are reserved for use as delimiters in the hit line. Their definitions
and default values are shown below. If any of them are needed in the text of a hit
string, the delimiters must be changed to allow their use as text See the .hit control.
hit "'"
the hit delimiter is used to signal the beginning of a hit string. It
appears only once in a hit string. but a hit line may contain multiple
hit strings, each with its own hit delimiter.
key""""
the key delimiter is used to separate individual keys in the hit string.
I t may appear as often as needed.
end ";"
the end delimiter separates the keys from the page number. This
delimiter is not given in the hit line but is appended (with the page
number) by the formatter.
Hit Types
The following type of hits are available. The desired type is specified by giving its
hit type character.
N
a null key. The hit string does not cause an entry in the index and
may be of any ar bi trary form, only the hi t delimiter is required.
This hit type is useful for annotating the raw data file as desired
(e.g., with the names of the sections of the document).
K
specific keys. A
more subordinate
given (depending
program; see the
they are treated
For example:
primary keyword or phrase followed by one or
keys. Any number of subordinate keys may be
on the number allowed the hit data processing
description of the compose_index command) and
as primary, secondary, etc., in the order given.
K I commands-command lines
K i commands-argumen ts
K Icommands. . . control arguments
s
a "see" reference. As for the K type above except that there must
be at least one subordinate key and the last key must make
reference to some other primary key. No page number is shown in
the index for a hit of this type. For example:
S 'control arguments"'See commands
u
permuted uppercase keys. This type may have a primary key only.
The primary key is expanded into the equivalent of a set of
specific keys by extracting each word (or protected phrase; see the
description of the compose_index command), translating it to all
uppercase, and appending the given primary key as a secondary key.
For example:
U Icommand control arguments
expands into the equivalent of:
2-12
AZ98-02
K ICOMMAND"'command control arguments
K ICONTROL"'command control arguments
K IARGUMENTS"-'command control arguments
permuted lowercase keys. As for the U type above except that the
new primary keys are forced to all lowercase.
L
permuted initial caps keys. As for the U type above except that
the new primary keys are forced to ini tial caps only.
permuted as-is keys. As for the U type above except that the new
primary keys are left as-is.
A
Error Messages
The Formatter issues error messages for all situations it determines to be errors,
whether internal program errors, implementation restrictions, errors in the user's input
file, or explicitly requested messages (see the .err contron. The form of these error
messages is:
filename; linenumber: {filename; linenumber: ... }
Error message(s)
Source line
The first line shows the history of inserted files at the time of the error beginning
with the command line input file. The second line shows any system error message that
may have been returned to the Formatter and a message from the Formatter. The third
line shows the offending input line (this line is not shown for messages requested with
the ,err control).
If formatted output is being written to an output file or the -check control
argument (see the description of the compose command) was given with the invocation
of the Formatter, then error messages are written directly to the error_output I/O
switch.
If formatted output is being written back to the user's terminal, then error
messages are accumulated in a list in the process directory. If the invocation of the
Formatter is allowed to complete normally, the list is printed at termination. If the
user QUITs the Formatter invocation, then the error list is printed in response to the
program_interrupt command (see Multics Commands); otherwise, it is discarded when
the Formatter invocation is released.
The Formatter also supports the severity active function (see Multics Commands)
with the following severity schedule:
o-
No errors
2 - User errors (undefined variables, misspellings. invalid control parameters,
etc.) that prevent specific actions.
3 - Missing/inaccessible insert files.
4 - Program errors/limitations and/or internal inconsistencies that may
cause a formatter abort.
5 - Command line errors that prevent any execution.
Note, however, that error severity is NOT shown in the error messages.
2-13
AZ98-02
Escaping Characters
Certain applications require the literal use of characters that have special meaning
to the Formatter or are treated in some special way. Among such characters are the
symbol delimiter, the title delimiter, and the ASCII motion characters (HT, SP. LF. FF.
etc.). To subvert any special meaning or action. the characters may be protected by
preceding them with an asterisk (*). Escaping of characters is part of expression
evaluation (see "Expression Evaluation" below), therefore. lines containing escaped
characters must be evaluated.
In general, any character may be escaped; however. there is a special set that is
replaced with something other than the escaped character. This set is:
*'
*'
*-
left double quote (")
righ t double quote (n)
an EN dash (useful in phototypeset documents where proportional spacing
tends to nearly obliterate the hyphen -- use for the unary negation operator
in text or the first character of control arguments)
an EN space (single SP for ascii device)
an EM space (double SP for ascii device)
backspace (ASCII 010)
newline (ASCII 012)
wordspace (ASCII 040)
*s
horizontal tab (ASCII 011)
*t
form feed (ASCII 014)
*f
*cddd character whose decimal rank in the ASCII collating sequence is ddd.
*Cddd
*c#ooo character whose octal rank in the ASCII collating sequence is
*C#ooo
000.
Escaping from the Formatter
In some applications it is necessary to perform actions that are not supported by
the Formatter but are supported by the Multics command processor. To perform such
actions, it is possible to escape from the Formatter to the command processsor. See the
.exc control.
Expression Evaluation
Expression evaluation takes place for:
•
all specified expressions in the variable fields of controls.
•
the substring expression of string expressions (see "Formatting Terminology"
above).
•
explicit evaluation constructs of the form n% {expr} %".
•
all title line parts when they are inserted into the output.
The action is performed by scanning the expression once from left to right for
symbol delimiters. When a delimiter is found, the action proceeds according to the
character following the delimiter.
•
If it is a symbol delimiter. then the pair is reduced to a single delimiter and
scanning continues with the character following the pair.
2-14
AZ98-02
•
If it is an opening bracket (D, then the expression is scanned for a matching
closing bracket (]) and the string thus contained is processed as an active
function.
•
If it is an opening brace (n, then the expression is scanned for a matching
closing brace (}) and the string thus contained is processed (recursively) as an
expression.
•
If it is a dot (.), then the expression is scanned for the matching closing
symbol delimiter and the string thus contained is processed as a formatting
control.
•
Otherwise. the expression is scanned for the next symbol delimiter and the
string thus contained is processed as a symbolic reference.
File Insertion
Any Formatter input file may be inserted into the text at any point. This inserted
file is a normal input file in all respects and may contain any valid combination of
input text and formatting controls. Inserted files are located by means of the
Formatter's search list (see "Search List" in the compose command description). See the
.ifi and .rt controls.
Note:
Once inserted, a file should remain in existence throughout the invocation
of the Formatter. It is an error to attempt to reinsert a file that has
ceased to exist after the first insertion.
Files, . Auxiliary Output
In many applications (particularly in the preparation of Tables of Contents and
cross reference Indexes), it is necessary to write text-dependent information to a file
other than the normal output file. The Formatter supports the ability to write
information to up to ten such auxiliary files. See the .wrt control.
Fonts and Type Sizes, Changing
The typeface or pointsize of the output may be changed at any time. See the .fnt
control.
Footnotes
Footnotes are text blocks contalmng additional or descriptive material pertammg to
some item in the material on the page and possibly referenced at one or more points
on the page. Footnotes are placed at the bottom of the page, just ahead of the page
footer or bottom margin. They may be formatted differently than main body text and
are separated from the main body by a special footnote header line. When the
Formatter is processing a footnote. it it is said to be in footnote mode.
Footnotes may be referenced or unreferenced, either globally throughout the
document or locally for individual notes. Unreferenced footnotes may be defined
anywhere on the page: referenced footnotes are defined at the point of their first
references. Referenced footnotes are assigned sequential reference numbers that advance
automatically as they are used on a page and resets to "1" at the top of each new
page.
2-15
AZ98-02
The amount of page space needed for un referenced footnotes is deducted from the
available text space.
The number of, and i:>pal.-C for, referenced footnotes are attached to the text blocks
containing their first references on a line-by-line basis, and are accounted for during
widow processing. When output lines are spilled into the next column or page, any
attached footnotes accompany them. Any footnote references in output lines spilled
onto the next page are resequenced beginning with "1" at the top of the new page. If
an output line and some of its footnotes (but not all) fit, then the line and those
footnotes that fit are retained and the remaining footnotes are spilled. In this case,
reference numbers for footnotes spilled onto the next page are not resequenced and
output lines continue to be placed on the page until a true overflow condition occurs.
See the .bbf, .bef, .frf, .ftp, .ftu, and .hlf controls.
Hyphenation
Hyphenation may be enabled and disabled either globally for the entire document
or locally on a line-by-line basis.
Note:
The Formatter does not support local discretionary hyphenation, so there is
no mechanism to resolve the hyphenation of homographs such as "record"
(rec-ord or re-cord) whose hyphenation is context-dependent.
The algorithm for hyphenation is not internal to the Formatter but is supplied as a
free-standing subroutine in the Multics WORDPRO system. Because of this, the user is
free to replace the Honeywell-supplied hyphenation routine with any other routine
implementing a different algorithm as long as the interface requirements are met.
See the .hy, .hyf, and .hyn controls, the -hyphenate control argument in the
description of the compose command in Section 3, and the description of the
hyphenate_word_ subroutine in Section 4.
Indention
The position of the first and last characters in an output line may be adjusted with
respect to the margins by giving left and right indention values, respectively. Positive
indention values move the affected characters away from the margin (toward the center
of the page) and negative values move them toward the margin. See the .in, .inl, .inr,
and .inb controls.
Input Line Continuation
In some applications it is necessary to give very long input control lines.
particularly lines having logical expressions containing other mixed expressions. Since
such lines quickly exceed the line length for some terminals, an input line continuation
feature is provided to improve input file readability. See the .+ control.
The amount of automatic vertical advance given with each output text line can be
set to any desired value. See the .ls control.
2-16
AZ98-02
Page Definition
The user may specify the dimensions of the output page for:
left and right page margins
top margin
header margin
footer margin
bottom margin
and the number of, widths, depth offsets, and gutters for columns on the page.
Values are given in terms of 10-pitch terminal characters and lines, and may have
up to three decimal places (for example. "1.333" or "2.75"). It is an error to give any
values that do not conform to the maxima or minima for the output device as given in
the device description table. See the .pd, .pdl. .pdw, .vm, .vmt, .vmh, .vm£. and .vmb
controls and the -indent control argument in the description of the compose command
in Section 3.
In addition to the basic page definition parameters. the user may also specify the
number of, widths, gutters, and initial page depths for, columns on the page. See the
.pdc control.
Page Headers and Footers
A page header is a block placed at the top of each page; a page footer is a block
placed at the bottom of each page. Both may contain title lines and normally formatted
text. Any title lines may be assigned an index number so that they may be individually
managed without affecting the rest of the lines in the block~ Indexed lines in a page
header are numbered from the top down; indexed lines in a page footer are numbered
from the bottom up. Page headers and page footers may be specified the same for all
pages or separately for front and back pages. The margin space is the same for all
pages but the header and footer text may be specified the same or separately for odd
and even pages. See the . phI, .bph, .eph, . pfl, .bpf, and .epf controls.
Running column header/footer capability is also available. This feature can be used
to generate header lines BETWEEN the page header and first text (e.g., "Section XXX
(cont.)") and BETWEEN the last text and the page footer (e.g., "Continued on page
xxx").
The column header/footer lines are independent of any page header/footers and
any local text titles/captions. The syntax for their use is (almost) the same as that for
page headers and footers. See .chI, .cf!, .bch, .ech, .bcL and .ecf controls.
Page Numbers
The Formatter counts pages as they are produced. The value of the page counter
may be inserted into any title line part by enclosing its symbolic name in symbol
delimiter characters as:
%PageNo%
The page counter is initialized to "1" at the beginning of each input file, but it can
be reset to any desired value. Its value may be displayed as Roman or Arabic numerals,
or as an alphabetic; the default is Arabic. See the . brp control. (Also, see "Page
Numbers, Structured" below).
2-17
AZ98-02
Page Numbers, Structured
The internal page counter maintained by the Formatter is not a simple integer
counter as implied in "Page Numbers" above, but is an array of 20 counters in which
counting is done in the last element. The counters hold the numeric values to be used
in a "structured" page number of the form, for example, "2-6.3(b)14-V". The counters
are initialized by giving the desired initial page number structure as the parameter of
the .brp control.
Note:
Presently, the entire structure must be repeated with the desired values for
every structure change.
The page number structure is given as an ordered set of counter values and field
separators. The counter values are given as they are to appear in the display of the
structure in the formatted output and the separators may be chosen from ".", "-", "(",
and ")".
Note that the "appearance" of a counter value may be ambiguous. For example, the
desired values of the first five counters in the structure cited above are obvious but the
value of the sixth is not. It could be either the uppercase roman representation of "5"
or the alphabetic representation of "22". In order to resolve these ambiguities, a second
parameter is supported by the .brp control. Ambiguities not explicitly resolved by the
second parameter are resolved to Roman, hexadecimal, or alphabetic, in that order. See
the .brp control.
Picture Blocks
A picture block is a special block that is placed on the page on a space-available
basis. If a picture block does not fit, it is deferred to the next column or page and
following text continues to be placed on the page until an overflow occurs. Picture
blocks are stacked and placed on the page in the order they are given. See the .bbp,
and . bep controls.
Printwheel Changing
For incremental plotting terminals that employ a "daisywheel" (or similar interchangeable
typeface element), the Formatter supports the production of pages that require the use
of more than one such prin twheel.
This feature is accomplished as follows:
1.
In the device description table for the terminal, the printwheels are assigned
index n urn bers.
Note: . In the device support modules available from Honeywell as Priced
Software Products, the printwheel index numbers are:
1
2
3
10-pitch PICA
10-pitch APL
12-pitch ELITE
These assignments are merely suggested; authors of device description
tables for other devices are free to make any convenient assignments.
2.
Each font defined for the terminal gives the required printwheel index number.
3.
As a page is formatted, the index numbers of all required printwheels are
recorded for later use.
2-18
AZ98-02
4.
When the page is printed (either directly to the terminal by the Formatter or
from an output file by the process_compout command), the printwheel index
number for the first unprinted text is compared to the index number of the
mounted printwheel. (If the page is the first printed page for the invocation of
the Formatter (or process_compou!), the mounted printwheel is assumed to be
#1; otherwise, it is the printwheel left mounted from the previous printed
page, which may well be this same page with a different printwheel.)
5.
If the printwheel index numbers differ and printwheel n is needed, then n
BEL/HT sequences are transmitted, leaving the print head at column 10n.
The user then changes the printwheel and types a single NL (ASCII code 012).
Then, one more BEL/HT sequence is transmitted and the print head returns
to the left margin. The user then repositions the page as necessary and types
another NL.
6.
If the printwheel index numbers are the same, then all text on the page for
the mounted printwheel is printed in its required position with empty space left
for text requiring other printwheels.
7.
At the end of the page, if any text for other printwheels remains unprinted,
the next printwheel index is taken from the recorded list and processing returns
to step 4 above.
Note:
Terminals that allow the simultaneous mounting of multiple printwheels
are not supported.
Programming Features
The essence of programmability is the control of processing -according to dynamic
conditions and communication with the user during processing. The Formatter supports
these abilities but the experienced user should note that the level of sophistication is
about that of the BASIC programming language, except that conditional groups may be
nested to any convenient depth. See the .!S, .go, .la, .if, .elseif, . then, .else, .endif, .ty,
.rd, and .wt controls.
Note that it is an error to leave a group with the .go or .rt controls.
Punctuation Space
The Formatter recognizes sentence termination punctuation normally used in written
English. When such punctuation is encountered in fil1e-d text, extra blank space may be
inserted to separate the sentences. The amount of space inserted is given in the device
description table for the output device.
Tables, Formatted
A formatted table is a form of output in which the text colum.!! is subdivided into
vertical areas known as table columns, and each table column is given its own limited
set of formatting parameters. The formatting parameters of each table ~olumn are
independent of all the others and independent of the parameters of the containing text
column.
When the Formatter is producing output in this form. it is in table mode. In table
mode, the unit of output is not a simple text block, but is a table entry that may (and
usually does) contain text from all the table columns. The completion of a table entry
2-19
AZ98-02
is signalled by an event that returns the Formatter to the formatting parameters of the
containing text column.
The Table of Contents of this manual is a formatted table with the Section
numbers in the first table column, the titles in the second, and the page numbers in
the last.
Additional capabilities provided by the table mode feature are:
1.
Alignment of a given text string to the specified column. This can be used for
applications such as aligning a column of prices at the decimal point. The TOe
feature also uses this feature to align page numbers at the ,,_u separator. See
ASTR as described in the .tab control.
2.
Positioning of the text in a table column at the top, center, or bottom of the
table entry. This can be used for applications such as aligning the price column
(in a catalog) at the bottom line of the item description. The TOe feature also
uses this feature to align the page number at the last line of the title. See V
as described in the .tab control.
See the . tab, .taf, and .tan controls.
Tabulation
Up to 20 typewriter-like horizontal tabulation stop patterns may be defined. Each
pattern may have up to 20 stop positions and an optional leader string may be given
for each stop position. All defined tabulation patterns may be active at the same time,
each with its own assigned "tab character". When tabulation is active. text filling is
temporarily disabled and text alignment is forced to align-left. See the .htd, .htn, and
.htf controls.
Text Alignment
Text may be formatted as flush left, flush right, flush both (justified), centered. or
flush to the inside (binding edge) or outside. See the .all, .alr, .alb.. ale, .ali, and .alo
controls.
Text Blocks, Secondary
Secondary text blocJ<.s are subblocks that may overlap and intersect and may be
contained in primary text blocks. The secondary block types defined are:
art
blocks that may contain conventional artwork. See the .bba and
.bea controls.
keep
blocks that may not be split between columns and/or pages. See
the .bbk and .bek controls.
literal
blocks that may contain lines that appear to be control lines or
have certain special characters in them. See the . bbl and . bel
controls.
title
blocks that may contain title lines.
controls.
2-20
See the . bbt and .bet
AZ98-02
Text Breaks
Text breaks are used to control the appearance of paragraphs and their placement
on the page. See the .brb, .brf, .brn, and .brp controls.
Text Filling
Filling of text may be enabled or disabled at any point. See the .fi. .fif. and .fin
controls.
Text Headers and Captions
A text header is a specially formatted subblock attached to the beginning of a text
block: a text caption is a specially formatted subblock attached to the end of the text
block. Both may contain title lines and normally formatted text. The attachment is
inseparable; that is, the subblock is not split away from the text block if the text block
must be split between pages. See the .tcl, .ttl, .btt, and .ett controls.
Title Lines
A title line is an output line that has three parts: a left margin part, a centered
part, and a right margin part. The parts are given (where allowed) by enclosing them
in title delimiter characters as follows:
Ileft part Icenter part Iright part I
The default title delimiter character is the vertical line
any other character. (See the .ctd controD.
(I>
but may be changed to
Delimiters for empty parts to the right of the last desired part may be omitted. If
two successive delimiters are given, the corresponding part is set to an empty
(zero-length) string. If the title consists only of two or more occurrences of the
delimiter, then all parts are empty and the affected line becomes a white space line.
The parts may contain symbolic references
however, in basic formatting, usually only the
are retained in symbolic form and substitution
variables when the title line is inserted into the
to program built-in and user variables;
page number is wanted. The references
is done with the current values of the
document.
Undention
Reverses the sense of direction of indention and is effective for only the next
output line. See the .un, .unl, .unh, .unr, and .unb controls.
Variables, Built - in
The Formatter supports an extensive set of built-in variables that allow the user to
retrieve and use the values of various formatting parameters, device description
parameters. formatting mode switches, and command line values. See "Built-in Variables"
described below.
2-21
AZ98-02
Variables, Substitution of
During expression evaluation, all symbolic references are replaced with the current
value of the referenced variable. ~tnng values are used as they exist. Numeric values
are converted to strings according to the· current display mode for the referenced
variable. See the .srm and .ur controls.
Variables, User
User variables may be defined and their values assigned or changed whenever
desired. The names of variables are constructed with alphanumeric characters and
underscores L) with a maximum length of 32 characters. The first character of a
variable name must be alphabetic.
Variables may be assigned numeric values or string values. Given variable values are
evaluated before they are assigned (see "Expression Evaluation" above).
If a variable value is given as a quoted string, the evaluated string value is assigned
to the variable. The maximum string length is 1092 characters.
If a variable value is not given as a quoted string. it may contain only numeric
digits and a decimal point (.) with a possible leading sign character. The evaluated
numeric value is assigned to the variable. The maximum absolute value of a numeric
variable is 2,097,151 and it may have up to three decimal places. See the .srv and .src
controls.
Note:
When % [] % and % {} % contain variables, the user must double the 0/05. For
example:
.ur .if %% [active_fnc_name %Variable%
~X]
%%
. ur . ur xxxx %% {%variable%-l} %% xxxx
White Space (Extra Lead)
Any arbitrary amount of white space may be added to improve readability, either
as an independent text block or within a text block. See the .spb and .spf controls.
In some instances, the user may want to advance to some specific page depth (such
as the extra space usually found at the top of the first page of a chapter or section),
or to ensure that some amount of "protected" (nontrimmable) white space appears at
some place on the page. or to ensure than no less than some given amount appears
without having sure knowledge of how much has already been given (this case occurs
frequently in "macro" formatting packages). See the .spd and .spt controls.
BUILT- IN VARIABLES
Descriptions of the built-in variables of the Formatter are described here. Each
description begins with a title line of the form:
Name:: type:: default value:: controls
where:
Name
the name of the variable as it is used in symbolic references.
2-22
AZ98-02
type
the storage type of the variable.
numeric, counter, string, or logical.
def aul t value
the initial value of the variable and the value assigned if no
value is given with a control or control argument that affects
its value.
controls
any controls and control arguments that affect the value.
The possible values are
The double colon (::) has no meaning other than a field separator in the title
line.
AlignMode :: string:: "both" :: .alb .alc .ali .all .alo .alr
the current text alignment mode. The possible values are "both", "left", "right",
"inside", "outside". and "center".
ArgCount :: numeric :: 0:: .ifi.. .rt
the number of arguments passed to an inserted file.
Argj :: string :: "" :: .ifi .. .rt
the ith argument passed to an inserted file. i may have the values 1, 2, ...
ArgCount.
ArtMode:: logical :: 0 :: . bba . bea
true when in an artwork subblock; otherwise, false.
CallingFileName:: string:: "" :: .ifi.. .rt
the "current reference name" (no suffix) of the caller of an inserted file.
CaliingLineNo :: numeric :: 0 :: .ifi .. .rt
the cutrent input line number in the caller of an inserted file.
CommandArgCount :: numeric :: 0 :: -arguments
the number of values given with the -arguments control argument in the
command line.
CommandArgi :: string :: "" :: -arguments
the ith argument given with the -arguments control argument. i may have the
columns 1.2.... CommandArgCount.
Date :: string :: dateO :: none
the current date in the form mm/dd/yy (month/day/year).
Device :: string :: "ascii" :: -device
the "current reference name" of the output device as given with the -device
control argument in the command line. This "current reference name" usually
implies some specific configuration or operating mode. There must be a device
description table with this "current reference name".
DeviceClass :: string:: "typewriter" :: -device
the class of the output device as given in the device description table. The
class of a device generally implies the technology used in its design and typical
values are "typewriter", "printer". and "typesetter".
DeviceName :: string:: "ascii" :: -device
the generic "current reference name" of the output device as given
device description table. The generic "current reference name" of a
usually implies some set of major features or the existence of multiple
description tables for different configurations and typical values are
"diablo", "Dymo". and "VIP".
2-23
in the
device
device
"ascii",
AZ98-02
Eqcnt:: counter:: 1 :: .eqc
the equation reference counter.
the amount of extra left margin space to be added.
FileName :: string :: entry in input file list :: none
the "current reference name" of the current command line input file.
FillMode :: logical :: -1 :: .fi .fif .fin -nofill
true when text filling is enabled; otherwise. false.
FontName :: string :: device dependent:: .fnt -device
the name of the current font. The initial value for the variable is the initial
font as given in the device description table.
Footcnt :: counter:: 1:: .bef .ftp .ftu
the footnote counter.
FootnoteMode :: logical :: 0:: .bbf .bef
true when processing a footnote; otherwise. false.
FootReset :: string :: "paged" :: .ftp .ftu
the footnote numbering mode. Its value is "paged" when footnotes are being
inserted and numbered for each page, and "unref" when unreferenced footnotes
are being used.
From :: numeric :: 1 :: -from
the number of the first output page to print as given with the -from control
argument.
FrontPage:: logical :: -1 :: .brn .brp .brs
true when the current page is a front or facing page, that is, the left edge is
the binding edge; otherwise, false.
Galley:: logical :: 0 :: -galley .gl
true when the -galley control argument or the .gl control are given; otherwise,
false.
HeadSpace :: numeric :: 0 :: .spb .spd .spt .tcl . ttl . vmh . vmt
the amount of white space on the page immediately preceding the current page
depth given in terms of 10-pitch lines.
Hyphenating :: logical :: 0 :: .hy .hyn .hyf -hyphenate
true when hyphenation is enabled; otherwise, false.
Indent:: numeric :: 0:: .inb .inl
the current value of left margin
characters.
indention
given
in
IndentRight :: numeric :: 0 :: .inb .inr
the current value of right margin indention given in
characters.
terms of
10-pitch
terms of lO-pitch
InputDirName :: string:: '''' :: .ifi .. .rt
the "current reference pathname' of the directory containing !nputFileName.
InputFileName :: string :: '''' :: .ifi .. .rt
the "current reference name" of the current input file.
InputLineNo :: numeric :: 0 :: none
the current line number in InputFileName.
2-24
AZ98-02
KeepMode:: logical :: 0 :: .bbk .bektrue when processing a keep block; otherwise, false.
Linelnput :: string :: "', :: none
the contents of the next line in the caller of an inserted file, that is, the line
whose input line number is one greater than the value of CallingLineNo. Use
of this built-in also advances the value of CallingLineNo.
LinesLeft :: numeric:: 54 :: .bbf .hef .bpf .efp .pfl .pel .pdl .vm .vmf .vmb
the amount of space left on the page available-for text. given in terms of
IO-pitch lines.
LineSpace :: numeric:: 1 :: .Is -linespace
the current linespacing value given in terms of IO-pitch lines.
Measure(string) :: numeric :: 0 :: none
the visual width of the string as measured in the current font and pointsize,
and given in terms of IO-pitch characters.
NextPageNo :: string:: "2" :: . brn .brp
the page number of the next page to be printed.
OutputFileOpt :: logical :: 0:: -output_file
true when the -output_file control argument is given; otherwise, false.
PageLength :: numeric :: 66:: .pd .pdl -device
the current page length given in terms of IO-pitch lines. The default may be
less than the value shown here due to device restrictions in the device
description table.
PageLine :: numeric:: 0 :: none
the depth of the current output line on the page given in terms of IO-pitch
lines.
PageNo :: string:: "1" :: .brn .brp
the page number of the current page.
PageWidth :: numeric:: 65 :: .pd .pdw -device
the current page width given in terms of lO-pitch characters. The default may
be less than the value shown here due to device restrictions in the device
description table.
Parameter :: string :: "" :: .ifi .. -parameter
the entire list of arguments passed to an inserted file given as a single string,
or (before any file is inserted) the value given with the =parameter control
argument.
ParamPresent:: logical :: 0:: .ifi ..
true if any arguments are passed to an inserted file; otherwise, false.
Pass :: numeric:: I :: -passes
the number of processing passes remaining to be performed (including the
current pass). Output is produced only when the value is 1.
PointSize :: numeric:: 7.2 :: .fnt -device
the current type size given in typographic points (72 points
default value is the size of IO-pitch characters.
I inch).
The
Print :: logical :: -1 :: -galley -from -to -pages -passes
true when the current line is to be printed; otherwise, false.
2-25
AZ98-02
StopOpt:: logical :: 0:: -stop
true when the -stop control argument is given; otherwise, false.
SymbolDelimiter :: string:: "%" :: .csd
the character being used to delimit symboiic references.
TableMode :: logical :: 0 :: . tan . taf
true when in table mode; otherwise, false.
Time:: string:: timeO :: none
the time of day at command invocation in the form hhmm.m (hours minutes
tenths of minutes).
TitleDelimiter :: string :: "I" :: .ctd
the character being used to delimit title parts.
To :: numeric :: -1 :: -to
the page number of the last page to be printed as given with the -to control
argument. The default value implies the end of the input file.
Undent :: numeric :: 0 :: .unl .unh
the value of any pending left undention.
UndentRight :: numeric :: 0 :: .unr
the value of any pending right undention.
User Input :: string:: "" :: none
one line as typed in by the user on the user_input I/O switch.
VMargBottom :: numeric: 4: .vm .vmb
the bottom margin, that is, the amount of white space between the page footer
and the bottom of the page given in terms of lO-pitch lines.
VMargFooter :: numeric: 2: .vm .vrnf
the fooler margin, that is, the amount of white space between the last text line
and the page footer given in terms of 10-pitch lines.
VMargHeader :: numeric: 2 : .vm .vmh
the header margin, that is, the amount of white space between the page header
and the first text line given in terms of lO-pitch lines.
VMargTop:: numeric : 4: . vm . vmt
the top margin, that is, the amount of white space between the top of the page
and the page header given in terms of IO-pitch lines.
WaitOpt:: logical :: 0 : -wait
true when the -wait control argument is given; otherwise, false.
Widow -:: numeric :: 2 : .wit
the current text widow size given in terms of 10-pitch lines.
CREA TING ARTWORK
The artwork feature of the Formatter permits the user to insert certain conventional
overstruck character patterns into an input file and have them displayed as fabricated
symbols and line art diagrams. The detection and expansion of the artwork constructs
described here is controlled with the block-begin-art and block-end-art formatting
controls described below.
Note:
Overstriking of characters is inappropriate for video terminals.
2-26
AZ98-02
For the purposes of discussion here, artwork constructs are shown with
dollar signs representing the backspace character as in "X$X".
The replacement strings for the artwork symbols are contained in the device
description table for the output device (see Section 4) and are the closest approximation
to the desired output symbol possible for the device.
The occurrence of an artwork construct in an
(or symbol fragment) with some definite position,
either given in the device description table or are
and are used by the Formatter when placing output
input line implies an output symbol
width, and height. The values are
part of the input artwork construct
symbols and text on the page.
In this section, the word "rule" refers to a typographic rule, that is, a line of given
length. thickness, and orientation.
Artwork Conventions
Two subsets of the 95-character ASCII graphic set are defined: the "line art" set
and the "math symbol" set. Members of the line art set are syntactically significant if
they are overstruck with another character from the set. Members of the math symbol
set are syntactically significant if they are overstruck with a valid size character (see
the discussion of size characters following). The characters in a construct may be given
in any order.
Line
art
Math
Symbol
Meaning
element of a horizontal rule
elemen t of a vertical rule or a vertical bar, depending on overstrike
pattern
/
/
\
element of a +45 degree slant rule or a division sign, depending on
overstrike pattern
element of a -45 degree slant rule
(
(
left semicircle or left parenthesis, depending on overstrike pattern
)
)
right semicircle or right parenthesis, depending on overstrike pattern
up arrow, diamond top vertex, or upward movement, depending on
overstrike pattern
v
v
dov:n arrov:, diamond bottom
depending on overstrike pattern
<
<
left arrow, diamond left vertex, or leftward movement, depending on
overstrike pattern
>
>
right arrow, diamond right vertex, or rightward movement, depending
on overstrike pattern
[
left bracket
]
right bracket
{
left brace
vertex,
or
downward
movement,
right brace
X
multiplication sign (one-high math symbol only)
2-27
AZ98-02
vertical or slant rule terminator
i:
"
-;'c
I
horizontal rule terminator or text deletion symbol, depending on
overstrike pattern
replicator
meaning
character
showing
overstrike
but
having
no
pictorial
H h
half-line control, up or down, depending on overstrike pattern
S s
superscript/subscript control, depending on overstrike pattern
=
double vertical bar concatenate symbol
c
text modification change bar (one-high math symbol only)
0
bullet (one-high math symbol only)
If any of the characters in the math symbol set is overstruck with a numeric or
alphabetic character it is considered a math symbol and the overstrike character is
interpreted as the symbol size as follows:
1 - 0
a - z
A - Z
1 through 10
11 through 36
31 through 56
Examination of the two subsets reveals that there are ambiguous cases (the most
obvious of which is "v") that may be either line art, a math symbol, or a size
character depending on usage. These ambiguities may be resolved in favor of line art
by adding a replica tor character Clor I) to the input artwork construct.
The four movement constructs perform "micropositioning" and the size character
represent the count of increment to be moved. The amount of space for an increment
is given in the device description table.
Artwork Syntax
The syntax for artwork construction is as follows:
1.
Text filling should be off to preserve element position in an artwork diagram. This
does not prevent the use of artwork in filled text, however.
2.
Line art may be contained within math symbols and vice versa.
3.
All rules continue through intersections unless they are specifically terminated by an
appropriate terminator character.
4.
Unterminated horizontal rules generate a reported syntax error at the right margin.
5.
Unterminated slant rules generate a reported syntax error at either margin or
continue to the end of the artwork block.
6.
Un terminated vertical rules continue to the end of the artwork block.
I.
Positioning of plain text is the responsibility of the user. The movement constructs
are provided for this purpose.
2-28
AZ98-02
Artwork Constructs
Boxes:
Boxes are created by defining their corners with the constructs below. The
corners appear on the page at the position of the first character of the
construct.
upper
upper
lower
lower
left corner
right corner
left corner
right corner
Artwork constructs are shown with dollar signs representing the backspace
character.
Superscripts and Subscripts:
Superscripts and subscripts are created by placing the letter usn (either
uppercase or lowercase) and one of the two vertical motion characters at the
point where the baseline change is to occur. Alternatively, "S" for superscript
or "Si! for subscript and a replicator character may be used. When using
superscripts and subscripts, you must remember to return to the normal text
baseline. For example, the chemical symbol for water could be given as:
Hs$v2"'$sO
The baseline offset for superscripts and subscripts is one third of the current
linespace value but the feature is effective only for those devices capable of
fractional linespacing. The superscript and subscript constructs are:
S$'"
s $'"
S$II
S$ I
S$v
s$v
s $11
s$ I
superscript
superscript
superscript
superscript
subscript
subscript
subscript
subscript
Half -Lines:
Half-lines are created by placing the letter "h" (either uppercase or lowercase)
and one of the two vertical motion characters at the point where the baseline
change is to occur. Alternatively, "H" for half-line up or "h" for half-line
down and a replica tor character may be used. When using half lines, you must
remember to return to the normal text baseline.
The baseline offset for half -lines is one half of the current linespace value but
the feature is effective only for those devices capable of fractional linespacing.
The half -line constructs are:
H$'"
h$'"
H$II
H$ I
H$v
h$v
h $ II
h$ I
half-line
half-line
half -line
half-line
half-line
half-line
half -line
half -line
up
up
up
up
down
down
down
down
2-29
AZ98-02
Math Symbols:
Multi-line math symbols are created by "stacking" the appropriate math symbol
construct in the same input column for the desired number of lines. The
constructs are given as one of the math symbol selector characters below and
the alphanumeric character representing the desired size. The maximum symbol
size is 56. The size character and the number of input lines given should be
the same.
Note:
It is critical that text filling be off and that alignment be flus
left for these constructs so that the positioning of the symbol
elemen ts are maintained.
The math symbol selector characters are:
[
]
{
}
(
)
I
=
opening bracket
closing bracket
opening brace
closing brace
opening parenthesis
closing parenthesis
single line (Boolean OR)
double line (concatenate)
In addition, the following symbols are available as "one-high" only (these
symbols may also be used in filled text):
0$ 1
/$ 1
X$ 1
$1
c $1
bullet
division sign
multiplication sign
asterisk
heavy vertical line (change bar mark)
Diamonds:
Diamonds are created by defining their vertices with the constructs below. The
vertices appear on the page at the position of the first character of the
construct. The left and right vertices must be in the mid line of the diamond
and the top and bottom vertices must be in the center column of the diamond.
Diamonds must always be an odd number of lines high and an odd number of
columns wide. The number of lines and columns must be the same. If only
the left and right vertices are given, the resulting diamond is one line high. If
only the top and bottom vertices are given, the resulting diamond is a half-line
high and is centered between the two lines giving the vertices (if the output
device is capable of fractional linespacing; otherwise, the vertices are separated
by the value of linespacing). Symbols for diamond vertices may not appear in
column 1. The diamond vertex constructs are:
A$"
v$ II
top vertex
bottom vertex
<$"
left vertex
>$ II
right vertex
Lozenges:
Lozenges (flattened diamonds) are created by defining their corners and vertices
with the constructs below. The corners and vertices appear on the page at the
position of the first character of the construct. The left and right vertices
2-30
AZ98-02
must be in the mid line of the lozenge. The left pair of corners must be in
the same column and the right pair of corners must be in the same column.
Lozenges must always be an odd number of lines high but may be any number
of columns wide. The minimum lozenge height is three' lines. If all four
corners are given in the same column. the top and bottom lines of the resulting
diamond are one column wide. Symbols for lozenges may not appear in column
1. The lozenge constructs are:
- $/
upper left corner
upper right corner
lower left corner
lower right corner
left or right vertex
~'~$\
- S'"
~'~$ '"
\$ '" $/
Rules:
Typographical rules are created by giving their starting and ending points with
the constructs below. Rules are started by giving a rule selector character and
any other line art character in an artwork construct and are ended by giving
the appropriate terminator character with any other line art character in an
artwork construct.
Un terminated rules that attempt to cross either page margin result in error
messages. Unterminated rules that reach the end of the artwork block are
terminated gracefully. The rule selector and terminator characters are:
{
/
horizon tal rule
vertical rule
left slant rule
righ t sIan t rule
horizontal terminator
vertical and slant terminator
Circles and Rounded Boxes:
A circle may be created by glvmg a left and a right parenthesis, each with a
replicator character, in artwork constructs separated by exactly one column. The
separating column may be used for a single character. The resulting circle is
three lines high; no other circle size is provided.
A rounded box may be created by opening up the space between the
parenthesis constructs and giving horizontal rules in the line before and the line
after the one containing the parentheses. The extra space may be used for any
text.
The circle constructs are:
($ II
) $ II
left semicircle
right semicircle
FORMAITING CONTROLS
Each of the formatting controls will be described in detail. The presentation is in
alphabetical order of the control tokens. An index style summary is included
immediately following these descriptions.
Each description begins with a line showing the control token and its variable field,
ending with the optional input line delimiter (;), then gives the name of the control
2-31
AZ98-02
and which break type it implies, if any.
executed before the action of the control.
Note:
When a break is indicated. the break is
The use of the semicolon in these descriptions does not imply that its use
in actual practice is required. It is used here merely for clarity of
presentation. The NL character (ASCII code 012) is the usual input line
delimiter.
Formatting controls need not be entered individually in a control line of an
input file. They may be entered as:
.pdl 66
.pdw 71
.vm 3,1,1,3
.sr fw "325"
or as:
.pdl 66:.pdw 7t.vm 3,1,1,3;.sr fw "325"
using the semicolon (;) as the control delimiter and the newline (NL) as the
input line delimiter. One exception to this type of construction (i.e..
multiple formatting controls on a single entry line) is with the controls: .if,
. then, .else, and .elseif. These unique controls accept either a semicolon or
space character as the delimiter. For example:
.if
. then;.srv
or.
.if
. then .srv
Any parameters in the variable fields shown in braces ({}) are optional and their
default values are used if they are not given. The left slash (\) is used to show that
exactly one of the values so separated may be given. An ellipsis C.. ) indicates
continuation of a parameter string to the extent given in the explanation .
.*
any text; comment, no break
a comment line having no effect on any output.
.+string; continue. no break
stri ng is appended directly onto the previous input line. That is, thc two lines
are processed together as a smgle line as though there had been no intervening
input line delimiter character. Note that. in this single exceptional control, no
SP character following the control token is needed. If one is given. it becomes
part of the input line.
2-32
AZ98-02
.. path - {arg 7/ arg·2, ... }; insert-file, no break
suspend processing of the current input file and begin reading input from the
given file. (This form is supported as being less "programmatic" than the ".ifi"
form.)
If any arguments are given, the built-in variable "ParamPresent" is set to "-1"
and the entire argument list is copied into the built-in variable "Parameter",
destroying any existing value. If no arguments are given, "ParamPresent" is set
to "0" and the contents of "Parameter" are not changed. (See "Built-in
Variables" above.)
The individual arguments (if any) are copied into the built-in variables "ArgI
"Arg2", ... and the number of arguments given is assigned to the built-in
variable "ArgCount". Any existing values of these built-in variables are saved in
a push-down/pop-up stack and are restored when processing returns to the
suspended input file. If any of the arguments contain blank space, they must
be given as quoted strings.
IT,
The depth to which files may be inserted is not limited.
.alb; align-both, format break
align the text at both the left and right margins as adjusted by the indention
values. Filling must be enabled for this alignment to operate. If filling is
disabled, .alb has the effect of .all. This is the default alignment mode.
.ale: align-center, format break
center the text between the left and right margins as adjusted by the indention
values. Filling has no effect on this alignment.
.ali; align-inside; format- break
align the text at the inside margin (binding edge) as adjusted by the appropriate
indention value. Filling has no effect on this alignment.
.all; align-left, format break
align the text on the left margin as adjusted by the left indention value.
Filling has no effect on this alignment.
.alo: align-outside, format break
align the text at the outside margin (away from binding edge) as adjusted by
the appropriate indention value. Filling has no effect on this alignment.
.alr; align-right, format break
align the text at the right margin as adjusted by the right indention value.
Filling has no effect on this -alignment.
.bba {#}; block_begin_art, no break
begin flagging output text lines as artwork lines to be processed by the artwork
expander. The parameter is given as the number of input lines following whose
generated text should be flagged in the output. (See "Creating Artwork" above.)
If the parameter is not given, then the flagging of output continues until the
occurrence of a .bea control and may span blocks, columns. and pages .
.bbf {ul Lc\p}: block-begin-footnote, no break
suspend processing of the current text block and begin processing a footnote.
The first parameter may only be "u" to indicate that the footnote is to be
unreferenced. If the footnote is to be referenced, then a footnote reference
string is constructed according to the current value of the footnote counter and
the style and procedure for the output device (see Appendix C) and is placed
as a hanging undent on the first line· of the footnote.
2-33
AZ98-02
If the second parameter is not given, then the footnote formatting parameters
are carried forward from the previous footnote. or set from the default
footnote formatting parameters if this is the first footnote. If the second
parameter is given; then the footnote formatting parameters are set according to
its value. Any formatting parameters changed while processing the footnote are
carried forward to subsequent footnotes but do not affect main body text.
If the second parameter is given and the first is omitted, the separating comma
(,) must still be given. The allowed parameter values are:
c
format the footnote according to the current column formatting
parameters. This is the default when the Formatter is in
multicolumn or table modes.
p
format the footnote according to the current page formatting
parameters. This is the default when the Formatter is in page
mode.
.bbk {#}; block-begin-keep, format break
finish the current line and begin flagging output text lines as exempt from
being split across column or page boundaries. The parameter is given as the
number of following input lines whose generated text should be flagged in the
output. If the parameter is not given. then the flagging of output continues
until the occurrence of a .bek control. When this flagging is active, all block,
column, and page breaks are inhibited. If the size of a keep block exceeds the
maximum page space available it is treated as a normal text block and split
across pages.
· bbl {#}; block-begin-literal, no break
process input lines as text lines even though they may appear to be control
lines or contain apparent escape sequences. The parameter is given as the
number of following input lines to process. If it is not given, then continue
until the occurrence of a .bel control.
·bbp {#}; block-begin-picture, no break
if # is given, then define an unbreakable picture block of exactly # lines of
vertical white space: the parameter is given as an unsigned number. If # is not
given, then accumulate output lines into an unbreakable picture block until the
occurrence of a block-end-all or block-end-picture control. Text headings
and/or captions given while in picture mode (# not given) pertain to the
picture and not to a possible containing text block. A picture block is vertical
white space or a formatted block that is inserted on a space-available basis. If,
at the completion of a picture block, sufficient space remains on the current
page, it is inserted immediately. If the picture block does not fit on the
current page, inline text is inserted from behind to ahead of the picture and
the picture is placed at the top of the next page. If the size of a picture
block exceeds the maximum text space available on a page as determined by the
vertical margins and any headers and footers, an error diagnostic message is
produced and the block is broken into full and partial pages. Multiple picture
blocks are queued, not merged into a single block. Up to ten picture blocks
may be queued. Queued picture blocks are inserted in the order in which they
were defined.
·bbt {#}; block-begin-title, format break
begin accepting three-part title lines as input. The parameter is given as the
number of input lines to accept. If it is not given. the title lines are accepted
until the occurrence of a .bet control.
f
2-34
AZ98-02
.bcf In}; begin-column-footer, no break
cancel the column footer block and begin a new formatted column footer
block. The formatting of the block proceeds as for any in line text block except
that none of the "special block" features (e.g., page header lines, footnotes) may
be used and title lines are allowed. The variable field parameter is:
n
the initial indention value for this footer block.
If it is omitted or given as "0", then the block is aligned at
the left column margin. If it is given as any other value, then
the block is aligned at the given position in the column .
. bch {n}; begin-column-header, no break
cancel the column header block and begin a new formatted column header
block. The formatting of the block proceeds as for any inline text block except
that none of the "special block" features (e.g., page footer lines, footnotes) may
be used and title lines are allowed. The variable field parameter is:
n
the initial indention value for this header block.
If it is omitted or given as "0", then the block is aligned at
the left column margin. If it is given as any other value, then
the block is aligned at the given positiin in the column.
· bea; block-end-art. no break
stop flagging output lines for artwork conversion .
. bef; block-end-footnote, no break
if the Formatter is in fo()tnote mode,. then save the cllrrent formatting
parameters for use in the ri-ext· {ootnote, then --finish- the footnote and resume
processing the suspended text block (if any), leaving footnote mode.
If there is no suspended block, then attach the footnote to the page header as
an "orphan;" otherwise, attach the footnote to the current output line of the
suspended block.
If the footnote is referenced, then insert the footnote reference string (see the
.bbf control above) into the output line, and advance the footnote reference
counter.
· bek; block-end-keep, no break
stop flagging output lines as exempt from block splitting and reactivate the
block, column. and page breaks.
· bel; block-end-literal, no break
stop ignoring control lines in the input.
· bep; block-end-picture, no parameters, no break, no substitution
Stop accumulating output lines into the current picture block and revert to
in1ine block processing. If the picture will fit in the space remaining on the
current page, then insert it immediately: otherwise. queue the picture block for
insertion on a space-ava1lable basis. If the picture mode is not in effect.
ignore the control.
· bet; block-end-title, no break
stop accepting title lines as input.
.bpf {n} {e\o\a}: begin-page-footer, no break
cancel the page footer block of the type specified by the second parameter and
begin a new formatted page footer block of the same type. The formatting of
2-35
AZ98-02
the block proceeds as for any inline text block except that none of the tlspecial
block" features (e.g., page header lines, footnotes) may be used and title lines
are allowed. The variable field parameters are:
n
the initial indention value for this footer block.
If it is omitted or given as "0", then the block is aligned at
the left page margin. If it is given as any other value, then
the block is aligned at the given position on the page.
the footer block to be created.
e\o\a
The meaning of the three allowed parameter values are:
e
a
a
even page footer block only
odd page foot block only
all page footer blocks (Default)
.bph {n} {e\o\a}; begin-page-header, no break
cancel the page header block of the type specified by the second parameter and
begin a new formatted page header block of the same type. The formatting of
the block proceeds as for any inline text block except that none of the "special
block" features (e.g., page footer lines, foot) may be used and title lines are
allowed. The variable field parameters are:
n
the initial indention value for this header block.
If it is omitted or given as "0", then the block is aligned at
the left page margin. If it is given as any other value, then
the block is aligned at the given position on the page.
the header block to be created.
e\o\a
The meaning of the three allowed parameter values are:
e
a
a
even page header block only
odd page head block only
all page header blocks (Default)
. brb; break-block, block break (see Note)
finish the current output line as for the . brf control following and then finish
processing the current block as appropriate for the block type.
Note:
For certain special blocks and formatting modes (e.g., page headers,
and keeps) that require some other control to signal completion,
this control causes only a format break, that is, completion of the
block is inhibited until the expected control is encountered .
. brc {#}; break-column, column break
finish the current block (if any), fill the remainder of the current column with
white space, and then change to the column indicated by the parameter. The
parameter is given as an unsigned number.
If the parameter is not given, then advance to the next column on the current
page or, if the next column is not defined, finish the current page and begin
with the first column on the next page. The appearance of this control on a
n')O"o
Y"o'"
f'')l1COC
~u. .............,
tho
"u,,",
n'lO".:>
YLlb....
tA
LV
"ho
v'-'
f A .. ..,.,'ltto~
> V> lUULL'-'U
" .........
U.:I
UU
11 ...
UlJ
l-. .... 1 ........... "rl
va1a1!'",'--""
... " .... '"
pa.S'--'
If the parameter is given as zero, finish the current block (if any), balance all
the columns at the current page depth, and revert to the formatting parameters
of the full page until the occurrence of a . brc control that selects a defined
column.
2-36
AZ98-02
If the parameter is given any nonzero value, finish the current block (if any).
fill the remainder of the current column with white space, and then change to
the given column, filling all intervening columns with white space. If the given
column is less than the current, then the current page is also finished (without
balancing) and processing begins in the given column on the next page. It is an
error to give a column number that is not defined for the page.
It is an error to give this control when not running multicolumn mode.
.brf~ break-format, format break
finish the current output line by formatting any pending text as a short line .
.brn {#}; break-need, no break (see Note)
if the parameter is greater than the amount of text space available, advance
immediately to the next page or column, as appropriate. The default value is l.
Note:
This control performs all the actions of the page or column breaks
except that the current output line is not finished, that is, any
pending text is carried forward to the next page or column.
Therefore, it is considered as not causing a break.
.brp {e\o\n}; break-page, page break (see Note)
(basic)
finish the current page and set the page counter according to the parameter
given. "Finishing" the page involves finishing the current block (and possibly
balancing the columns, if in multicolumn mode) and ejecting the page, inserting
any footnotes and/or the page footer. If the parameter is omitted, the page
counter is advanced by 1. If the page is empty when this control is given, it is
not ejected and the automatic page counter advance does not take place,
however the page counter is still set from a given parameter. (Also refer to
the "intermediate description" of the .brp control below.)
Note:
For certain special blocks and formatting modes (e.g., page headers
and keeps) that require some other control to signal completion,this
control is ignored.
The allowed values of the parameter are:
e
set the page counter to the next even value.
o
set the page counter to the next odd value.
n
new value for the page counter, given as an integer. If given
as an unsigned number, set the page counter to the value
given. If given as a signed number, change the page counter
by the amount given .
.brp {e\o\page-number} {mode-string}; break-page. page break (see Note)
(intermediate)
finish the current page and set the page number structure according to the
parameters given. "Finishing" the page involves finishing the current biock (and
possibly balancing the columns, if in multicolumn mode) and ejecting the page.
inserting any footnotes and/or the page footer. If no parameters are given, the
last page number counter in the existing structure is advanced by 1. If the
page is empty when this control is given, it is not ejected and the automatic
page counter advance does not take place, however the page number structure is
still set from any given parameters. (Also see the "basic description" of the
. brp control above.)
2-37
AZ98-02
Note:
For certain special blocks and formatting modes (e.g., page headers
and keeps) that require some other control to signal completion,
this control is ignored.
The variable field parameters are:
e
set the last page number counter to the next even value.
o
set the last page number counter to the next odd value.
page-number
if given as a signed number, change the last page number
counter by the amount given.
Otherwise, a new value for the page number structure is given
as a character string consisting of counter values and separators.
Counter values are given as they are to appear in the output
(with ambiguities resolved by mode-string below). Separators
may be chosen from ".", "-", "(n, ")", and "I" where n I"
represents a null separator. For example, the page number A2
would be given as "A 12".
mode-string
a string made up of keywords and commas where the keywords
are in one-to-one correspondence with the counter values in
the given page- number and the commas are in one-to-one
correspondence with the separators. Any of the keywords may
be omitted or chosen from the list below. All commas except
those following the last desired keyword must be given.
Key
ar
bi
hx
oc
al
au
rl
ru
Value displayed as
Arabic
binary
hexadecimal
octal
lowercase alphabetic
uppercase alphabetic
lowercase Roman
uppercase Roman
A keyword given for a counter value overrides the apparent
value given in the page- number. For example:
Page
i
I-i
.brp parameters
"i rl" or "1 rln
"I-i au" or "9-1 aU,rl"
(Also see the .srm control.)
.brs {#} {"text"} {"header"} f'footer"}; break-skip, page break
finish the current page, and then create blank pages according to the given
parameters. The pages created are not assigned page numbers and the page
counter is not advanced. Any parameters given apply only to the blank pages
created by the current use on the control; they do not carry forward. Note
that all parameters must be given as quoted strings regardless of whether they
contain blanks or not The variable field parameters are:
#
the number of blank pages desired. The default value is 1.
text
a line of text to appear centered on the blank page(s) created.
header
a single page header line for the blank pages created. It must
be given as a title line. If a header is wanted, then a text line
2-38
AZ98-02
must also be given, but the text may be given as a blank line
(nn).
footer
a single page footer line for the blank pages created. It must
be given as a title line. If a footer is wanted, then a header
must also be given, but the header may be given as a null
header ("") .
.btc {n}; begin-text-caption, no break
suspend processing of the current block (if any) and
formatted text caption block. The formatting of the block
inline text block except that none of the "special block"
footer lines and footnotes) may be used and that title lines
begin processing a
proceeds as for any
features (e.g.. page
are allowed.
If a pending text caption already exists. then add the output generated to it;
otherwise, use the output generated as the caption.
The variable field parameter is:
n
the initial indention value for this caption.
If it is omitted, then the caption is aligned according to the
text left indention.
If it is given as an unsigned number, then the caption is
aligned at the value given relative to current column or page
left margin.
If it is given as a signed number. then it is used as a local
adjustment to the text left indention .
. btt {n}; begin-text-title. no break
suspend processing of the current biock
formatted text title block. The formatting
inline text block except that none of the
footer lines and footnotes) may be used and
(if any) and begin processing a
of the block proceeds as for any
"special block" features (e.g., page
that title lines are allowed.
If a pending text title already exists, then add the output generated to it;
otherwise, use the output generated as the title.
The variable field parameters are:
n
the initial indention value for this title.
If it is omitted, then the title is aligned according the text left
indention.
If it is given as an unsigned number, then the title is aligned
at the value given relative to current column or page left
margin.
If it is given as a signed number, then it is used as a local
adjustment to the text left indention .
.cba {e}: change-bar-add, no break
if change bars are enabled by giving the -change_bars or -change_bars_art
control argument in the compose command line. and the parameter is less than
or equal to the change level character (in the ASCII collating sequence sense;
O.1. ... 9,A,B. ... Z,a,b, ... z) given with the control argument, then flag output lines
2-39
AZ98-02
contaInIng the text and white space following with text modification marks. If
neither control argument is given. then ignore the control.
The. parameter is given as a single character and is the change level (Revision
number or Addendum letter) for the text change. . The default character is a
blank (ASCII SP).
By default, the mark is a vertical line (I) and it is placed in the outside page
margin. separated from the text by one column. The mark and placement may
be changed with parameters given with the control arguments.
.cbd {c} ~ change-bar-delete. no break
if change bars are enabled by giving the -change_bars or -change_bars_art
control argument in the compose command line, and the parameter is less than
or equal to the change level character (in the ASCII collating sequence sense;
0,1. ... 9.A.B .... Z.a.b, ... z) given with the control argument. then flag the next
output line only with a text deletion mark. If neither control argument is
given, then ignore the control.
The parameter is given as a single character and is the change level (Revision
number or Addendum letter) for the text change. The default character is a
blank (ASCII SP).
By default, the mark is an asterisk (*) and it is placed in the outside page
margin, separated from the text by one column. The mark and placement may
be changed with parameters given with the control arguments.
. cbf {c}; change-bar-off, no break
if change bars are activated by a preceding .cba or .cbm control and the
parameter is equal to the currently active change level. then stop flagging
output lines with text modification marks. If change bars are not active for the
given change level then ignore the control.
The parameter is given as a single character and is the change level (Revision
number or Addendum letter) for the text change. The default character is a
blank (ASCII SP) .
. cbm {c}; change-bar-modify, no break
if change bars are enabled by giving the -change_bars or -change_bars_art
control argument in the compose command line, and the parameter is less than
or equal t.o the change level character (in the ASCII collating sequence sense;
0.1. ... 9,A.B. ... Z,a,b, ... z) given with the control argument. then flag output lines
containing the text only following with text modification marks. If neither
control argument is given, then ignore the control.
The parameter is given as a single character and is the change level (Revision
number or Addendum letter) for the text change. The default character is a
blank (ASCII SP).
By default, the mark is a vertical line (I> and it is placed in the outside page
margin, separated from the text by one column. The mark and placement may
be changed with parameters given with the control arguments .
.cn {#l {n} {title}; column-footer-line, no-break
define column footer lines according to the values given in the variable field.
If no parameters are given. then all column footers are cancelled. The variable
field parameters are:
2-40
AZ98-02
#
the index value for the line.
If it is omitted or given as "0", then the current column
. footer block is cancelled and this line becomes line 1 of a
new column footer block~
It it is less than or equal to the highest index number in the·
block. then title replaces that line in the block. If no title is
given. then the line is replaced with a null line.
If it is greater than the highest index line number in the
block, then title becomes the indexed footer line with the
given number in the block and any intervening indexed lines
become null lines. If no title is given. then the control is
ignored since all lines involved would be null lines.
n
the indention value for this footer line. Note the n may not
be given unless # is also given since they both appear as
simple, unsigned numbers.
If it is omitted or given as "0", then title is aligned at the left
page margin. If it is given as any other value, then title is
aligned at the given position on the page.
title
the three-part title used as the footer line. Any references to
symbolic variables in the title are evaluated when the line is
placed on the page.
.chl {#} in} {title]; column-header-line, no-break
define column header lines according to the values given in the variable field.
If no parameters are given, then all column headers are cancelled. The variable
field parameters are:
#
the index value for the line.
If it is omitted or given as "0", then the current column
header block is cancelled and this line becomes line 1 of a
new column header block.
It it is less than or equal to the highest index number in the
block. then title replaces that line in the block. If no title is
given, then the line is replaced with a null line.
If it is greater than the highest index line number in the
block, then title becomes the indexed header line with the
given number in the block and any intervening indexed lines
become null lmes. If no title is given. then the control is
ignored since all lines involved would be null lines.
n
the indention value for this header line. Note the n may not
be given unless # is also given since they both appear as
simple, unsigned numbers.
2-41
AZ98-02
If it is omitted or given as "0", then title is aligned at the left
column margin. If it is given as any other value, then title is
aligned at the given position on the page.
title
the three-part title used as the header line. Any references to
symbolic variables in the title are evaluated when the line is
placed on the page.
.csd {e}; change-symbol-delimiter, no break
change the symbol delimiter to the given character. The default value for the
parameter is percent (%) .
.ctd {c}; change-title-delimiter, no break
change the title delimiter to the given character. The default value for the
parameter is the vertical line (I) .
.ecf; end-column-footer, no break
finish the column footer block begun with the preceding .bcf control and use
the block as the appropriate column footer .
.ech; end-column-header, no break
finish the column header block begun with the preceding .bch control and use
the block as the appropriate column header .
.else; conditional-else, no-break
begin the conditional execution group clause that is executed when the expr of
the preceding .elseif or .if control is false. The end of the clause is marked by
a .endif control and the eventual occurrence of that control is required. (See
"Notes" under Formatting Controls above,)
.elseif expr; conditional-elseif, no break
begin the conditional execution group clause that is executed when the expr of
the preceding .elseif or .if control is false, evaluate expr and proceed according
to the result. The end of the clause is marked by a .elseif, .else, or a .endif
control and the eventual occurrence of one of them is required. (See "Notes"
under Formatting Controls above.)
If expr is given as a logical expression, then evaluate it as given. If it is given
as a string expression, then evaluate it as though it were expr "'= "". If it is
given as a numeric expression, then evaluate it as though it were expr "'= O.
If the result of evaluation is true or expr is not given, then execute the clause;
otherwise, skip the clause.
.endif; conditional-end, no break
marks the end of a conditional execution group.
.epf; end-page-footer, no break
finish the page footer block(s} begun with the preceding .bpf control and use
the block as the appropriate page footers .
.eph; end-page-header, no break
finish the page header block(s) begun with the preceding .bph control and use
tho
~u""
hl",.l,r
U~VwA
... co tho ................. "' ............... .0
0." ~u .... Uptn Vp~ 10. .......
l-.o ... ~o .. t"
paO.... u ....au""" h) •
1""'n".o
.eqc {n}; equation-count, no break
if the parameter is given. it must be an unsigned number and is the value to
be assigned to the internal equation counter. If the parameter is not given,
then the internal equation counter is advanced by 1.
2-42
AZ98-02
·err string; error, no break
.
generate a Formatter error message using string as the text of the message but
do not show this control line.
.etc: end-text-caption, no break
finish the caption block begun with the preceding .btc control.
.ett; end-text-title, no break
finish the title block begun with the preceding .btt control, resume the
suspended block (if any).
If there is no suspended block. then use the title block as the title for the
next text block.
If there is a suspended block and it is untitled, then prepend the title block
onto the suspended block as a text title.
If there is a suspended block and it is already titled, then insert the title block
between the existing text title and the first line in the suspended block .
.exc string; execute-command, no break
string is passed to the Multics command processor for execution as a command
line .
.fi; fill-default, format break
enable or disable filling according to the default given for the invocation of
the Formatter. See the -nofill control argument in the description of the
compose command. If the -no fill control argument is not given, filling is
enabled by default.
.fif; fill-off, format break
disable filling .
.fin; fill-on, format break
enable filling .
. fnt {name} {/member J {size J; font. no break
change to the given font.
If any parameters are given, then push the current font and pointsize onto a
20-element push-down/pop-up stack. If no parameters are given, then use the
top element of the stack as the given font and "pop" the stack. It is an error
to attempt to "pop" an empty stack. The variable field parameters are:
name
the name (or alias) of the desired font. It must be supported
by the output device and registered in the device description
table. If no name is given, then the name of the current font
is used. I t is an error to give the name of an unsupported
font.
I member
the name (or alias) of the desired family member of the given
font. It must be supported by the output device and registered
in the device description table. If no name is given, then the
current member name is used. It is an error to give a member
name not in the given font family. (Refer to "Device Table
Compiler" in Appendix C for a discussion of font naming.)
size
the desired type size, given in typographic points. It must be
within the range of sizes supported for the given font as
registered in the device description table. If no size is given,
then the current pointsize value is used. It is an error to give
2-43
AZ98-02
a size value outside the allowed range; however. if the device
supports only one size (as for an ASCII terrninal or lineprinter),
then this apparent error is ignored .
. frf {#}; footnote-reference, nO break
prepare a footnote reference string (see the .bbf control above) for the #'th
previous footnote according to the style and procedure for the output device
(see Appendix C) and insert it into the current output line. The default value
for the pararneter is "1", that is, the imrnediately prior footnote. It is an error
to give this control when not in a text block or to refer to a footnote that has
already been placed on the current or some earlier page.
.fth; footnote-hold, no pararneters, no break, no substitution
do not insert footnotes on the page of their reference, but hold them aside for
insertion by the user with the insert-footnote control or, by default, at the end
of the document. The footnote counter runs continuously until reset by another
footnote control.
.ftp; footnote-paged; no break
format all following footnotes as "paged" and referenced .
.ftr {#}; footnote-running, no break, no substitution
same as paged in .fth above except runs continuously until reset.
.ftu; footnote-unreferenced, no break
format all following footnotes as unreferenced .
.gl; galley; no break
format the file in galley mode. This control rnust be the first control line read
by the formatter; it is an error for it to appear elsewhere.
.go label name; go-to, no break
repoSItion the input file to the first occurrence of a .la control defining
label name and continue processing with that line. label name must be unique
in the file for correct operation. however uniqueness is not required. It is an
error to make ref erence to an undefined I abel name. If I abel name is
undefined, processing continues with the input line following the .go control.
.hit {n} {=ABC} KSTR {STR}; hit-line, no break
ernits a line of data to one of several auxiliary files collecting data to be
processed as cross-reference indexes, bibliographies, etc. (see "Document Indexing"
above).
The variable field parameters are:
n
an unsigned number showing which data file is to be used.
The allowable values are 0 through 9 with 0 being the default.
=ABC
the three hit string delirniters to be used for this hit line only.
The default delimiters are the triplet I""';, however, if any of
those characters are needed in the text of a hit string. then
the delimiters must be changed to allow their use as text. For
example:
=!"';S!segno Ioffset""See addressing
= I"-'+K ILine Terminators"-;+
KSTR
the hit type character.
2-44
AZ98-02
STR
a hit string.
.hlf {title}; header-line-foonote, no break
use the given title as the footnote header line. The default footnote header is a
full page or column width horizontal rule.
.htd {name} {#s,#s,#s, ... }; horizontal-tab-define, no break
define a horizontal tab stop pattern according to the ordered set of parameters.
If no parameters are given, then all tab stop patterns are cancelled. No more
than 20 tab stop patterns may be defined but all may be simultaneously active.
The variable field parameters are:
name
the name of the pattern.
#s,#s,#s, ...
the position/leader string pairs for the 1st, 2nd, 3rd,
stop positions. Up to 20 positions may be given.
positions are given, the pattern is cancelled.
tab
If no
The position values are given as unsigned, nonzero numbers in
increasing order from left to right across the page.
The leader strings may be any character strings, however, if
any contains a blank (ASCII SP), it must be given as a quoted
string.
. htf {aa ... }; horizontal-tab-off, format break
disable horizontal tabulation for the given set of characters. If no tab stop
pattern is associated with any of the characters, then the control is ignored. If
no characters are given. then horizontal tabulation is disabled for all patterns,
.htn a name\ #s,#s,#s, ... ; horizontal-tab-on, format break
enable horizontal tabulation according to the parameters given. Text filling is
suspended if enabled, and text alignment is forced to flush left. Both
parameters are required and are:
a
the character to be assigned a temporary role of horizontal
tabulation character and associated with the given tab stop
pattern. The character is not available for use in text during
this use.
name
the name of the tab stop pattern to be enabled.
#s,#s,#s, ...
position/leader string pairs as for .htd above. Patterns defined
by this mechanism are not retained as they are for named
patterns with the .htd control, but are discarded upon the next
occurrence of .htf or .htn .
. hy; hyphenate-default, no break
enable hyphenation according to the initial default and syllable size as set with
the -hyphenate control argument for this invocation of the Formatter. If no
syllable size is given with the control argument, then the defauit size is used.
If the control argument is not given. then the initial default is off. The
default syllable size is 2.
. hyf; hyphenate-off, no break
disable hyphenation .
. hyn {#}; hyphenate-on, no break
enable hyphenation and set the syllable size according to the given parameter.
The parameter is given as an unsigned integer and specifies the number of
2-45
AZ98-02
characters in the smallest allowed hyphenated syllable. If the parameter is not
given, then the default syllable size is used.
Thp
rlpf~l11t
.... .a. ............. ,.....1. ................
~"lhhlp
tJ.}.L.I.""',,",.I.-'"
~17P
...
.a.",....
1~
.I.~
thp
.... .1..1'-"
~;7P
...,..I.LI ......
o-1"pn ur;th
thp
_h17nhpn<:ltp
,.nnt'l"nl
<:t'l"ft"t11"nP.... t
" • .1."'.&..1
,.,.1..1. ......
.I.J.J.t-".J.J.,"".I..1.LL",-""
'-"'V.ll .... .lVj.
U.iOU..lJ.l"-'.l.lL..
0.1.1''-''.1..1
for this invocation of the Formatter. If no size is given with the control
argument or the control argument is not given, then the default syllable size is
2.
.if expr; conditional-if, no break
begin a conditional execution group, evaluate expr and proceed according to the
result. (See "Notes" under Formatting Controls above.)
If expr is given as a logical expression, then evaluate it as given. If it is
given as a string expression. then evaluate it as though it were expr /\= "". If
it is given as a numeric expression. then evaluate it as though it were expr /\=
o.
If the result of evaluation is true or expr is not given, then execute the "then"
group (if any) following; if it is false, then execute "else" group (if any)
following. (See .then and .else controls.)
The end of the conditional execution group created by this control is marked
with the .endif control and the eventual occurrence of that control is required.
.ifi path {arg 1, arg2, ... }; insert-file, no break
refer to ".. " control above for a description of this format control.
.ift; insert-footnotes, no parameters, no break, no substitution
Insert all pending footnotes and reset the footnote counter to 1.
.in {nJ; indent-left, format break
refer to .inl control below for a description of this format control.
.inb {n}; indent-both, format break
set the indention for both left and right margins according to the value of the
parameter. The default value is O.
If it is given as an unsigned number. set the indention to the value given.
If It IS given as a signed number, change the indention by the amount given.
jndctl {state}; indent-controls, no break
where state is "on" or "off". The existing state is remembered in a circular (20
entries) queue and the state is set as specified. If the state is not given, the
queue is popped.
The program no longer requires that a control line have the control starting in
column 1, but merely that the control (.XXX) be the first non-blank string in
the line (refer to "General Syntax" above). This feature allows indention of
control lines and makes macros more readable. Note however, that this is an
incompatible change since certain applications may rely on the now-ignored
leading white space to "protect" certain formatting controls. Such protection
will now have to be done with a .bbl control.
.inl {nJ; indent-left, format break
set the indention for the left margin according to the value of the parameter.
The default value is O.
If it is given as an unsigned number. set the indention to the value given.
2-46
AZ98-02
If it is given as a signed number, change the indention by the amount given .
.inr In}; indent-right, format break
set the indention for the right margin according to the value of the parameter.
The default value is O.
If it is given as an unsigned number, set the indention to the value given.
If it is given as a signed number, change the indention by the amount given .
.la I abel name; label, no break
record label name as a target for .go controls. The Formatter supports up to
2000 labels for each input file .
.ls {nl; linespace, no break
set the linespace value according to the parameter given.
If n is omitted, then set the linespace to "1"; however, if -linespace was given
in the invocation of the Formatter, then use that value.
If n is given as an unsigned number, then set the linespace to the value given.
If n is given as a signed number, then change the linespace by the amount
given .
.pd {I} {,w} {,C { (d}}} ~g,c { (d) }... } {,b\u};
page-define-all, column break (see
Note)
define the page according to the ordered set of parameters. If a value is not
given for a parameter (i.e., its field is blank or null), then its default value is
used; however, if-following values-are -g-i-ven, -tbecomma for the skipped value
must still be given.
Note:
I
This control causes a column break only if the column structure
changes: otherwise, no break occurs.
the length of the page given as a number of 10-pitch lines.
The default value is 66.
If it is given as an unsigned number, then the length is set to
the value given.
If it is given as a signed _number, then the length is changed
by the amount given.
w
the width of the page given as a number
characters. The default value is 65.
of
IO-pitch
If it is given as an unsigned number, then the width is set to
the value given.
If it is given as a signed number, then the width is changed
by the amount given.
c( d },g,c(d J...
See the description of the .pdc control.
b\u
See the description of the .pdc control.
.pdc {cUd)} 1 {,g,c{(dj} ... } {,b\u}: page-define-column, column break (see Note) •
define text columns according to the ordered set of parameters. If a value is
not given for a parameter (i.e., its field is blank or null), then its default value
2-47
AZ98-02
is used; however, if following values are given, the comma for the skipped
value must still be given.
The width and depth offset values for column 1 are given first. then gutter,
width, and depth offset values for the remaining columns are given. All
nonzero width values given must be the same; however, gutter and depth offset
values may vary from column to column. If no parameters are given or all
values are 0, then the Formatter returns to page mode.
Note:
This control causes a column break only if the column structure
changes: otherwise, no break occurs.
The variable field parameters are:
c(d},g,e(d}...
the widths (e) of text columns, their depth offsets (d), and
their separating gutters (g). The widths and gutters are given
in terms of IO-pitch characters and the depth offsets are given
in terms of IO-pitch lines. The columns are numbered 1, 2, 3,
with column 1 placed at the left page margin. The
maximum number of columns allowed is 20.
If any width value is
appear on the page and
is, there are no breaks
default value for widths
skipped or given as "0", it does not
is not assigned a column number; that
in the column number sequence. The
is O.
The depth offsets are relative to the first text position on the
page, that is, the first available text line position following the
page header (including the header margin). If a value is given
as an unsigned or positive number, the top of the column is
moved downward by the given amount; if it is given as a
negative number, the top of the column is moved upward (into
the header margin) by the given amount. The default value for
depth offsets is O.
If any gutter value is given as "0", then the two adjoining
columns have no separating space. The default value for
gu tters is 3.
b\u
the column balancing action to be used.
The two allowed parameter values are:
b
at a page break event, balance the columns so that their
bottoms are at the same page depth level. The success of
this balancing is limited by the capabilities of the output
device and is affected by widowing constraints. This is the
default.
u
do not balance the columns, leaving their bottoms ragged .
.pdl {n}: page-define-Iength, no break
define page length only, as described for the .pd control.
.pdw {n}; page-define-width, no break
define page width only, as described for the .pd control.
.pfl {e\o\a} # {n}} {title} :page-footer-line, no break
define page footer lines according to the values given in the variable field. If
no parameters are given, then all page footers are cancelled. The variable field
parameters are:
2-48
AZ98-02
e\o\a
the page posItIOn key indicating even, odd, or all pages. If
this field is omitted, then all pages are assumed.
#
the index value for the line.
If it is omitted or given as "0", then the current page footer
block is cancelled and this line becomes line 1 of new footer
block.
If it is less than or equal to the highest index number in the
block, then title replaces that line in the biock. If no iitle is
given, then the line is replaced with a null line.
If it is greater than the highest index line number in the
block, then title becomes the indexed footer line with the
given number in the block and any intervening indexed lines
become null lines. If no title is given, then the control is
ignored since all lines involved would be null lines.
n
the indention value for this footer line. Note that n may not
be given unless # is also given since they both appear as
simple, unsigned numbers.
If it is omitted or given as "0", then title is aligned at the
left page margin. If it is given as any other value, then title
is aligned at the given position on the page.
title
the three-part title used as the footer line. Any references to
symbolic variables in the title are evaluated when the line is
pI aced- on -the page•
.phl {e\o\a} {# {n}} {title}; page-header-line, no break
define page header lines according to the values given in the variable field. If
no parameters are given, then all page headers are cancelled. The variable field
parameters are:
{e\o\a}
the page position key indicating even, odd, or all pages. If
this field is omitted, then all pages are assumed.
#
the index value for the line.
If it is omitted or given as "0", then the current page header
block is cancelled and this line becomes line 1 of the new
page header block.
If it is less than or equal to the highest index number in the
block, then title replaces that line in the block. If no title is
given, then the line is replaced with a null line.
If it is greater than the highest index line number in the
block, then title becomes the indexed header line with the
given number in the block and any intervening indexed lines
become null lines. If no title is given, then the control is
ignored since all lines involved would be null lines.
n
the indention value for this header line. Note that n may not
be given unless # is also given since they both appear as
simple, unsigned numbers.
2-49
AZ98-02
If it is omitted or given as "0", then title is aligned at the
left page margin. If it is given as any other value, then title
is aligned at the given position on the page.
title
the three-part ti tIe used as the header line. Any references to
symbolic variables in the title are evaluated when the line is
placed on the page.
.rd; read, no break
read one line from the user_input I/O switch and process it as an input line.
Processing continues with the line following the .rd control unless the line read
from user_input is a .go control that sends processing elsewhere.
.rt; return, no break
stop reading input from the current file and return to reading input from the
suspended input file, if any.
If there is no suspended file (i.e., the current input file is given in the
command line invoking the Formatter). then the remainder of the file is
ignored and processing proceeds according to the command line parameters.
If there is a suspended file, then any arguments saved in the push-down/pop-up
stack are restored to their prior values.
.spb {#}; space-block, block break (see Note)
finish the current output line, formatting any pending text as a short line, and
then, if sufficient space remains in the column or page, add the given amount
of space, then finish the current text block. If there is not sufficient space,
then finish the current column or page and do not add white space. Any space
created with this control is discarded if it appears at the top of a column or
page. The default value for the parameter is 1. A blank or null line in the
text has the effect of a .spb 1 control.
Note:
For certain special blocks and formatting modes (e.g., page headers
and keeps.) that require some other control to signal completion,
this control causes only a format break, that is, it has the action
of the .spf control following .
.spd n; space-to-depth, block break
finish the current block and then advance the page depth by adding white
space as determined by the given parameter. The parameter is required; there is
no default.
If the parameter is given as an unsigned number, then add a nontrimmable
white space block such that the next output line appears at the page depth
given. It is an error to give a value less than the current page depth.
If the parameter is given as a positive number, then add a nontrimmable white
space block such that the page depth advances by the amount given. It is an
error to give a negative number .
.spf {#}; space-format, format break
finish the current output line and then add the given amount of white space to·
the current block. The default value for the parameter is 1. If the white space
is at the beginning of the text block and the text block is placed at the top of
the page or column. then the space is trimmed .
.spt {#}; space-total, no break (see Note)
ensure that at least the given amount of white space appears on the page. The
parameter is given as an unsigned number and its default value is 1.
2-50
AZ98-02
This is accomplished as follows:
•
calculate the space required as the difference between the given parameter
and the amount of white space in the output immediately preceding
occurrence of the control.
•
if the space required is zero, then ignore the control.
•
if the space required is greater than zero, then execute a .spb or .spf
control (as appropriate for the current block) with the amount of required
space.
Note:
This control does not cause a break. However, if a .spb or
.spf control is executed in the course of processing, then the
executed control causes a break as described above .
.src name {value-expr {by incr-expr}}; set-reference-counter. no break
define name as a user counter variable. If it does not exist, then create and
initialize it; otherwise. convert the existing variable to a counter variable. Both
given expressions must be numeric expressions.
If value-expr is given. then evaluate it and assign it as the value of name;
otherwise. do not change any existing value. If i ncr-expr is given. then
evaluate it and assign it as the increment of name; otherwise. do not change
any existing increment.
For newly created counters, the default value is 0 and the default increment is
l.
.srm mode name{ name ... }; set-reference-mode. no break
set the numeric -display--mode-for the named variables according to the given
mode keyword. mode is required and at least one name must be given. It is
an error to give the name of the page counter. PageNo. in the list of names.
The valid mode keywords are:
Mode
ar
bi
hx
oc
al
au
rl
ru
Dispiay
Arabic numerals (0,1.2, .. .)
binary numerals (0,1.10.11.100, ... )
hexadecimal numerals (O.1.2, ... ,D,E,F.10,11 ... J
octal numerals (0,1,2, ...• 7.10,11, ... )
lowercase alphabetics ( ,a, b, ... ,z,aa,ab •... ,zz, .. .)
uppercase alphabetics ( ,A,B•.... Z.AA,AB, ... ,ZZ ... J
lowercase Roman ( .i,ii,iii,iv. v. vi, ... )
uppercase Roman ( .I.Il.Ilr.lV,V.Vl, .. .)
.srv name {value-expr}; set-reference-value. no break
define name as a user variable. If it does not exist, then create and initialize
it; otherwise, convert the existing variable to the type of the given value-expr.
If value-expr is' given, then evaluate it and assign it as the value of name;
otherwise, do not change any existing value. If the given expression is a
numeric expression, then the variable is a numeric variable. If the given
expression is a string expression, then the variable is a string variable .
. tab {name} {p {col -spec I [astrJ {a} {v}: ... }; table_define, no break
define a table column format according to the parameters given. If no
parameters are given, then all table column formats are cancelled. No more
than 20 table column formats may be defined at anyone time. Up to 20
columns may be defined for a table column format. The term col-spec has
2-51
AZ98-02
the form " {wI {f} {3} {I str}. See "Tables, Formatted" above for more
information on additional capabilities. The variable field parameters are:
name
the name of the table column format. If this is the only
parameter given, then the named table column format is
cancelled.
p
the table column left margin value given in terms of lO-pitch
characters as measured from the current left margin of the
page or column. These values are required and must be given
in steadily increasing order from left to right across the page.
w
the table column width value given in terms of lO-pitch
characters. If any width value is omitted, then the width for
that table column is the space from the table column left
margin to the text right margin, that is, the rest of the text
column. The separating comma is required in all cases.
f
the fill mode for the table column. The allowed values are:
f
n
a
filled (Default)
unfilled
the text alignment mode character for the table column. If
this character is given with col-spec, then the fill mode
character above must also be given. The allowed values are:
b
c
both (col-spec only and Default)
centered
left (Default with astr)
right
inside (col-spec only)
outside (col-spec only)
I
r
i
o
Istr
the leader string to be used on the last output line (of a table
entry) in the table column. It may be any character string;
however. if it contains a blank (ASCII SP), it must be given as
a quoted string. If a leader string is given, then both the fill
mode and alignment mode characters above must be given.
astr
a string to be matched for text alignment. Each text line for
the column is searched for a matching astr. If a match is
found, then the first character of the matching string is
aligned at the column position. If no match is found, the first
character of the text line is aligned at the column position
according to· the text alignment character.
v
the vertical aiignment mode character for the table column.
The text lines for the column are aligned within the available
vertical space according to this alignment character. The
allowed values are:
top (Def aul t)
centered
bottom
c
b
. tac {#}; table-column. forma! break, block break
if in table mode. finish the current output line as the last line (inserting the
leader, if any) for the current table column in the current table entry and
2-52
AZ98-02
proceed according to the value of the parameter. It is an error to give this
control when not in table mode.
Note:
This action implies that all text for a particular table column in a
table entry should be given before switching to· another table
column.
The parameter is given as an unsigned integer. If it is omitted or given as "0",
then finish the current table entry (with a block break) and return to the
formatting parameters of the contaIning column without leaving table mode. If
it is given as any other value, then switch to the formatting parameters of that
table column.
· taf; table-off, block break
finish the current table entry as for . tac above, and leave table mode. If not
in table mode, then ignore the control.
· tan name; table-on, block break
if not in table mode, then finish the current text block and enter table mode
using the named table column format for the table column formatting
parameters. If already in table mode, then finish the current table entry and
switch to the named table column format without leaving table mode.
When in table mode, input text may be given in either of the following ways:
context mode
each· input text line begins with a period and a single decimal digit (e.g.,
".ITest line."). The digit indicates the table column for which the text is
intended and it is formatted according to· the parameters fot that table
column. By convention, the digit "0" indicates the tenth column of the
format. Only the first ten table columns may be referenced with context
mode.
Whenever the use of the .tac or .tan control returns· to the formatting
parameters of the containing column, input may be given in context mode.
All control lines and any input text without the period/digit initial
characters continue to be executed in, and formatted according to the
parameters of the containing column. Their use may produce unexpected
results and/or overprinted output. In context mode, input is limited to
plain text; no special formatting features may be used.
free column mode
whenever the .tac control is used to select a table column (including the
containing column), input may be given in free column mode. Almost all
Formatter features may be used in free column mode; those few that may
not are documented and their use is an error. For example, it is an error
to give context mode input lines in free column mode; they are considered
to be unknown controls.
· tel i# in}} {title}; title-caption-line, no break
define a text caption line according to the values given in the variable field. If
no parameters are given, the caption line is a single white space line.
If a pending text caption already exists, then add the line to it; otherwise, use
the line as the caption.
The variable field parameters are:
2-53
AZ98-02
#
the amount of extra white space to be inserted ahead of the
line. It must be given as an unsigned number. The default
value is O.
n
the indention value for this line. Note that n may not be
given unless # is also given since they both may appear as
simple, unsigned numbers.
If it is omitted or given as "0", then title is aligned at the
current text left indention.
If it is given as an unsigned number, then title is aligned at
the given position relative to the page or column margin.
If it is given as a signed number, then it is used as a local
adjustment to the current text left indention.
title
the three-part title used as the caption line. Any references to
symbolic· variables in the title are evaluated when the line is
placed on the page.
· then; conditional-then, no break
begin the conditional execution group clause that is executed when the expr of
the preceding elseif or .if control is true. The end of the clause is marked by
a .elseif, .else, or .endif control and the eventual occurrence of one of them is
required. (See "Notes" under Formatting Controls above.)
· trn {abab ... }; translate, no break
each nonblank character a in the input file is replaced with its associated
character b in the output. Any number of ab pairs may be given and the ab
pairs from multiple occurrences of the control are accumulated. If a character
pair is given as aa, then transiation for a is cancelied without affecting any
other characters being translated. If no character pairs are given, then the
translation feature is disabled.
· ts expr; test, no break
evaluate expr and proceed according to the result.
If expr is given as a logical expression, then evaluate it as given. If it is given
as a string expression, then evaluate it as though it were expr = "". If it is
given as a numeric expression, then evaluate it as though it were expr = O.
If the result of evaluation is true or expr is not given, then continue
processing with the next input line; if it is false, then skip the next input line
and continue with the second input line following.
· ttl {# {n}} {title}; text-title-line, no break
define a text title line according to the values given in the variable field. If
no parameters are given, the title line is a single white space line.
If a pending text title already exists,- then add the line to it; otherwise, use the
line as the text title.
If there is no current text block, then use the line as the text title for the
next block.
If there is a current text block and it is untitled. then prepend the line onto
the current text block as a text title.
If there is a current text block and it is already titled, then insert the line
between the existing text title and the first text line in the current text block.
2-54
AZ98-02
The variable field parameters are:
#
the amount of extra white space to be inserted after the line,
It must be given as an unsigned number. The default value is
O.
n
the indention value for this line. Note that n may not be
given unless # is also given since they both may appear as
simple. unsigned numbers.
If it is omitted or given as "0", then title is aligned at the
current text left indention.
If it is given as an unsigned number. then title is aligned at
the given position relative to the page or column margin.
If it is given as a signed number, then it is used as a local
adjustment to the current text left indention.
title
the three-part title used as the title line. Any references to
symbolic variables in the title are evaluated when the line is
placed on the page.
. ty {expr}; type, no break
evaluate expr and immediately write it back to the user on the error_output
I/O switch. If expr is not given, a blank line is written .
. un {n}; undent-Ieft, format break
refer to.unl control belowf-ora description of this format control.
.unb {n}; undent-both, format break
adjust the indention for both the left and right margins for the next output
line only according to the value of the parameter. The default value for the
parameter is the current value of both· the left or right indention. Positive or
unsigned values of the parameter move the text .outward; negative values move
it inward .
. unh {n}; undent-hanging, format break
adjust the indention for the left margin for the next input text line only
according to the value of the parameter and suppress the automatic depth
advance. The line affected is formatted as unfilled and flush left at
adjusted indention point. Any following text is normally formatted according to
the current indention value and appears at the same page depth. The defauit
value for the parameter is the current value of the left indention. Positive or
unsigned values of the parameter move the text outward; negative values move
it inward .
.un! {n}; undent-Ieft, format break
adjust the indention for the left margin for the next output line only
according to the value of the parameter. The default value for the parameter is
the current value of the left indention. Positive or unsigned values of the
parameter move the text outward: negative values move it inward .
.unr {n}; undent-right, format break
adjust the indention for the right margin for the next output line only
according to the value of the parameter. The default value for the parameter is
the current value of the right indention. Positive or unsigned values of the
parameter move the text outward: negative values move it inward.
2-55
AZ98-02
·ur contro/\expr; use-reference, no break
the given parameter is evaluated once as discussed in "Advanced Features"
above and is then reprocessed as an input line.
If the parameter contains nested symbol delimiters, then the symbolic references
at the deepest level are substituted and the nesting level is reduced by 1. For
example. if the variable I contains the value "1", then the construct %%list%I%%%
becomes %listl %.
. vm {t,h,f,bJ; vertical-margin-all, no break
set vertical page margins according to the ordered set of parameters. Values are
given as numbers of lO-pitch lines. If a value is not given for a parameter,
(i.e.. its field is blank or null). then its default value is used; however. if
following values are given. the comma for the skipped value must still be
given. The variable field parameters are:
t
page top margin. The default value is 4.
Note:
The minimum value for this parameter depends on
the output device and is obtained from the device
table for the device (see Appendix C). For example.
Honeywell lineprinters (as configured for Multics)
have a minimum top margin of 3 lines.
If the value is given as an unsigned number, the top
margin is set to the value given.
If the value is given as a signed number, the top margin
is changed by the amount given.
h
page header margin. The default value is 2.
If the value is given as an unsigned number, the header
margin is set to the value given.
If the value is given as a signed number, the header
margin is changed by the amoun t given.
f
page footer margin. The default value is 2.
If the value is given as an unsigned number, the footer
margin is set to the value given.
If the value is given as a signed number. the footer
margin is changed by the amount gi ven.
b
page bottom margin. The default value is 4.
Note:
The minimum value for this parameter depends
on the output device and is obtained from the
device table for the device (see Appendix C).
For example, Honeywell1ineprinters (as configured
for Multics) have a minimum bottom margin
of 3 lines.
If the value is given as an
unsigned number. the
bottom margin is set to the value given.
If the value is given as a signed number, the bottom
margin is changed by the amount given.
2-56
AZ98-02
·vmb {n}; vertical-margin-bottom, no break
set the page bottom margin only as described for the .vm control above.
.vmf {n}; vertical-margin-footer, no break
set the page footer margin only as described for the .vrn control above.
.vmh {n}; vertical-margin-header. no break
set the page header margin only as described for the .vm control above.
.vmt {n}; vertical-margin-top, no break
set the page top margin only as described for the .vm control above.
.wit {#}; widow-text, no break
change the widow size according to the given parameter. The parameter is
given as a number of output text lines. It is an error to give a value that
results in a widow size that is negative or greater than the page length. The
default value for the parameter is 2.
If the parameter is given as an unsigned number. then set the widow size to
the value given.
If the parameter is given as a signed number, then change the widow size by
the amount given .
.wrt path {string}; write-text, no break
stri ng (as given) with an added NL (ASCII code 012) character is written to
the segment path. If string is not given. a blank line is written.
If path is not attached and open as an output file, it is attached and opened.
If it does not exist, it is created as an empty file. If it do€s exist, it is
truneated to an empty file by the open. All attachments are made through the
vfile_ I/O module (see Multics Subroutines) .
.wt; wait, no break
read one line from the user_input I/O switch and discard the input Unlike
the .rd control above, this control is executed during output rather than input.
Processing continues with the next input line.
2-57
AZ98-02
COMPREHENSIVE CONTROL SUMMARY
The following list presents a comprehensive grouping of all the compose formatting
controis, inciuding an identification of complexit)l for each. That is, one OJ rne
following categories is assigned to each control: basic, intermediate, or advanced. The
format of the list is designed for ease of duplication so that a copy can be made and
be available for posting near the users' terminal.
Code
*
+
.alb
.ale
.ali
.all
.alo
.alr
.bba
.bbf
.bbk
.bbl
.bbp
.bbt
.bcf
.bch
.bea
.bef
.bek
. bel
. bep
. bet
. bpf
.bph
.brb
.brc
.brf
. brn
.brp
.brp
.brs
. btc
.btt
.cba
.cbd
.cbf
.cbm
.cfl
.ehI
.csd
.ctd
.ecf
.ech
.else
Contr.ol Name
Group
Page
comment
continue
insert-file
align-both
align -cen ter
align-inside
align-left
align-outside .
align-right ..
block-begin-art
block-begin-f ootnote
block-begin-keep .
block-begin-literal .
block-begin-picture . .
block-begin-title .
begin -col umn -f ooter
begin-column-header
block-end-art
..
block-end-f ootnote
block-end-keep . .
block-end-literal
block-end-picture .
block-end-title ..
begin-page-footer .
begin -page-header
break-block
break-column
break-format
break-need
break-page . .
break -page . .
break-skip . .
begin-text-caption
begin-text-title
change-bar-add . .
change-bar-delete .
change-baT-off ..
change-bar-modify
co1umn -f 00 tef-line
eolumn-lieader-line
change-sym bol-delimi ter
change-ti tle-delimi ter
end-column-footer
end-column-header
conditional-else ..
.in termediate
.advanced
· in termedia te
. basic
.basic
.basic
. basic
.basic .
.basic
.intermediate
·in termediate
.intermediate
.intermediate
.in termedia te
.intermediate
.basic . . . . .
.basic . . . . .
.in termediate
·in termediate
.intermediate
.intermediate
·in termediate
·in termediate
. basic
.basic . . . . .
.basic .. .
· in termedia te
.basic
.basic . . . . .
.basic . . . . . .
·in termediate
.intermediate
.basic .. .
.basic . . . . .
.intermediate
.intermediate
.intermediate
.intermediate
.basic
1... __
2-32
2-32
2-33
2-33
2-33
2-33
2-33
2-33
2-33
2-33
2-33
2-34
2-34
2-34
2-34
2-35
2-35
2-35
2-35
2-35
2-35
2-35
2-35
2-35
2-36
2-36
2-36
2-37
2-37
2-37
2-37
2-38
2-39
2-39
2-39
2-40
2-40
2-40
2-40
·
·
·
·
·
~_
• Uc1~l'-'
·
·
·
·
·
·
·
·
·
·
·
•
.advanced
.advanced
.basic . . .
.basic ..
.advanced
2-58
·
2-42
2-42
2-42
2-42
2-42
AZ98-02
.elseif
.endif
.epf
.eph
.eqc
.err
.etc
.ett
.exc
.fi
.fif
. fin
.fnt
.frf
.fth
.ftp
.ftr
.ftu
. gl
.go
.hit
.hlf
.htd
.htf
.htn
.hy
.hyf
.hyn
.if
.ifi
.ift
.in
.inb
.indctl
.inl
.inr
.la
.Is
.pd
.pdc
.pdl
.pdw
.pn
.phl
.rd
.n
.spb
.spd
.spf
.spt
.sre
.srm
.sry
. tab
2-42
· .advanced
.advanced
· 2-42
2-42
.basic .
.basic. . .
2-42
.advanced
2-42
2-43
· .advanced
2-43
· .basic .. .
.basic .. .
· 2-43
.advanced
2-43
. basic
· 2-43
2-43
.basic .
.basic .
2-43
.intermediate
2-43
.intermediate
2-44
.intermediate
· 2-44
.in termedia te
2-44
.intermediate
· 2-44
· .in termedia te
· 2-44
.basic . .. ..
· 2-44
.advanced
2-44
·in term ed ia te
· 2-44
.intermediate
2-45
2-45
·in termedia te
.in termedia te
2-45
2-45
·in term edia te
2-45
· in termedia te
.in termedia te
2-45
.intermediate
2-45
.advanced
2-46
2-46
·in termediate
.. 2-46
.intermediate
.basic .. .
2-46
.. 2-46
.basic .. .
.advanced
2-46
.basic .. .
2-46
2-47
.basic .. .
.advanced
2-47
.basic . . . . . . . .. 2-47
.basic ..
· 2-47
· in term edia te
· 2-47
2-48
· .basic
. basic
2-48
. basic
2-48
. basic
2-49
.advanced
2-50
.intermediate
2-50
.basic .
2-50
2-50
.intermediate
.basic .
2-50
.intermediate
2-50
.advanced
2-51
.advanced
2-51
2-51
· .advanced
2-51
·in termedia te
condi tional-elseif
conditional-end
end-page-footer .
end-page-header
equation-count
error .. . . . .
end-text-caption
end-text-title ..
execute-command
fill-default
fill-off
fill-on
..
font
footnote-reference
footnote-hold .. .
f 00 tno te-paged
footnote-referenced
f ootnote-unref erenced
galley .
go-to.. . . . . . .
hit-line
..... .
header-line-f ootnote
horizontal-tab-define .
horizontal-tab-off
horizontal-tab-on
hyphenate-default
hyphenate-off
hyphenate-on
condi tional-if
insert_file ..
insert-footnotes
indent-left ..
indent-both
indent-controls
indent-left
indent-right
label
linespace
page-derine-all
page-def ine-col umn
page-def ine-Iength
page-def ine-wid th
page-footer-line
page-header-line
read
return
space-block
space-to-depth
space-f orma t
space-total .
set-reference-counter
set-reference-mode
set-ref erence-value
table-define . . . .
'0
•
2-59
AZ98-02
.tac
.taf
. tan
t,.l
• "'-'1
.then
.trn
.ts
. ttl
.ty
.un
.unb
.unh
.unl
.unr
.ur
.vm
.vmb
.vmf
.vmh
.vmt
.wit
.wrt
.wt
table-column
table-off · .....
table-on · ..
title-caption-line
conditional-then
translate · ..
..........
test
text-title-line
type
undent-left
undent-both
undent-hanging
undent-Ieft
undent-right ..
use-reference
vertical-margin -all
vertical-margin -bottom
vertical-margin -f ooter
vertical-margin -header
vertical-margin-top
widow-text
write-text
wait
.intermediate
· .intermediate
.intermediate
2-52
.. 2-53
.. 2-53
,,_C'1
~
oJ,.,
'h .... .;,.
.V~l'-'
II
....
.advanced
.advanced
.advanced
.basic . . .
.advanced
.basic
. basic
· .basic
· .basic
· .basic
. . . advanced
.basic
. . . basic
· .basic
.basic ..
. basic
.intermediate
.advanced
.advanced
2-60
·
·
·
·
·
·
2-54
2-54
2-54
2-54
2-55
2-55
2-55
2-55
2-55
2-55
2-56
2-56
2-57
2-57
2-57
2-57
2-57
2-57
2-57
AZ98-02
SECTION 3
WORD PRO COMMANDS
This section contains tools that may be used to:
•
Translate a device description file into a binary table for use by the Formatter
•
Prepare formatted documents from raw text segments for production on various
documentation devices utilizing both the compose and format_document commands
•
Produce a cross-reference index file from raw data
•
Convert runoff input segments to compose input segments
•
Display selected information from a compose device table
•
Process compose output files to an online device, or to magnetic or punched
paper tape
•
Expand an expansion input file into an expansion output file
3-1
AZ98-02
compdv
compdv
The compdv command is used to invoke the Device Table Compiler to translate a
device description file into a binary table for use by the Formatter (see "Device
Description Language" in Appendix C).
SYNTAX AS A COMMAND
compdv path {-control_args}
ARGUMENTS
path
is the pathname of the input device description file. The entryname of the file
must end with the suffix compdv, but the suffix need not be given in the
command. The output segment is created (if it does not already exist) in the
working directory with an entryname formed by replacing the suffix compdv
with comp_dsm. Multisegment files and the star convention are not supported.
CONTROL ARGUMENTS
-check
-ck
processes the input file, making all syntax checks and creating the ALM source
intermediate file, but do not invoke the ALM assembler and do not delete the
ALM source file. The default is to invoke the ALM assembler at the end of
an error-free translation and to delete the ALM source file.
-list
-Is
create an ALM assembly output listing for the translation. The default is no
listing.
3-2
AZ98-02
compose (comp)
compose (comp)
The compose command is used to prepare formatted documents from raw text
segments for production on various documentation devices including typesetters, line
printers, and user terminals. Output pages are composed from various text blocks and
controls provided in input files. Detailed control over page composition is provided by
controls in the input file.
SYNTAX AS A COMMAND
comp paths {control_argS}
ARGUMENTS
paths
are the pathnames of the input files to be formatted. The suffix com pin must
be the last component of the input file entryname; however. the suffix need
not be given in the command line. If two or more pathnames are specified,
they are treated as if compose had been invoked separately for each. Up to
200 input files may be given with one invocation of the command. Output is
produced in the order in which the pathnames are given in the command line.
Input files may be either single segments or multisegment files. Output files
for very large documents are converted to multisegment files. The star
convention is not supported.
CONTROL ARGUMENTS
all control arguments specified in the command line apply to all input file
pathnames given. Control arguments may be freely intermixed with input file
pathnames, except for -arguments which must be last in the command line.
-annotate {key key key ... }
-ann {key key key ... }
shows an font/pointsize changes identified by the optional key (where key may
only be "font" at the present time) in an extra column to the right of the
formatted text and at the output line in which they occur.
The style of a typeset document usually calls for a large number of
font/pointsize changes to improve readability. However, when one is limited to
a terminal and/or lineprinter for early checkout of the documents, it is very
difficult 'to determine if the changes are all being made correctly.
-arguments arg 1 ... argn
-ag arg 1 ... argn
all fields following are string values to be placed in the indefinite set of
program built-in variables named "CommandArgl" through "CommandArgn"
where "n" is the count of such fields. The program built-in variable
"CommandArgCount" is set to "n". If any argument is to contain blanks, it
must be given as a quoted string.
Note:
This control argument. if given. must be the last control argument
in the command line.
3-3
AZ98-02
compose (comp)
-brief
-bf
snows only the header line or tne aermed error line (i.e., the count of errors),
both at normal termination and in response to the program_interrupt command.
-change_bars {x,p,l,r,d}
-cb {x,p,l,r,d}
generates text change symbols in the output according to the parameters given.
Change symbols are shown in the text margins as determined by controls in the
text. The default for change symbol generation is OFF. All the parameters for
this control argument are optional but, if any are given, they must appear in
the order shown. If any parameter is skipped, its separating comma must still
be given. Skipped parameters retain their default values. The parameters are:
x
a change level character. If the optional change level character in any
change-bar control is less than x (in the ASCII collating sequence
sense), then no text change symbols are inserted for those controls. The
x character may be either numeric or alphabetic. The default value for
x is the SP (ASCII code 040) character.
p
a symbol placement key character. It may have the values "1" for left
margin, "rn for right margin. "i" for inside margin. or "0" for outside
margin. The default for p is "0".
I
the definition of the text change symbol to be placed to the left of
text. It must be of the form n {string} where string is any character
string and n is the separation from the text. The default value for
string is a vertical bar (I> and the default value for n is 1. The n
may be given without string, but string may not be given without n.
r
the definition of the text change symbol to be placed to the right of
text. It must be of the same form as I above.
d
the definition of the text deletion symbol. It must be of the same
form as I above except that the default for string is the asterisk (*).
-change_bars_art {x,p,/,r,d}
-cba {x,p,l,r,d}
as for -change_bars above except that the strings in the I. r, and d fields
may be given as conventional artwork symbols (see "Creating Artwork" in
Section 2). The default values for the strings are also artwork symbols.
-check
-ck
performs syntax checking on the input file(s) by processing all text and controls
but does not produce any output. The default for this feature is OFF.
-device {name}
-dv {name}
prepares output compatible with the device specified. This control argument is
used when the target device for output is not the default device for the output
mode selected. If the -output_file control argument is given, the default device
is "printer"; if it is not, the default device is "ascii". Any device for which
name .comp_dsm exists is a supported device (see "Device Table Compiler" in
Appendix C).
3-4
AZ98-02
compose (comp)
-from In}
-fm In}
starts printed output at page n. This control argument is mutually exclusive
with the -pages control argument You must give the desired structured page
number~ for example, to print the fourth page of Section 3. you must give
tLpage 3-4
The default value of n is l.
ft.
-galley {n 1} {,n2}
-gl {n 1} {,n2}
produces galley format (continuous single-column text without page headers and
footers) output for lines n 1 through n2 of the input file. The default value of
n 1 is 1 and the default value for n2 is the last line in the input file. If n2 is
not given, the comma need not be given. If n 1 is not given, a comma must
precede a given value for n2. The default for this feature is OFF.
-hyphenate {n}
-hyph In}
-hph In}
changes the default hyphenation mode from OFF to ON. The optional
parameter n is the length of the smallest separated word part Its default value
is 2.
-indent In}
-ind In}
adds n spaces at the left page margin of the output This space IS In addition
to any indention given in the text The default value of n is 0 (i.e.• at the
left-:-hand mechanical stop of the output device).
-input_file path
-if path
the name of an input file even though path may have the appearance of a
numeric parameter or a control argument.
-linespace {n}
-Is {n}
changes the default line spacing value to n. The lineSpace control uses n as a
minimum value. The default value for n is 1.
-noart
-noa
disables the c~nversion of conventional artwork constructs and inserts space into
the output at the positions that such constructs would occupy. The default for
the artwork conversion feature is ON.
-nobell
-nob
suppresses the audible BELL signal when signalling the "waiting" state to the
user when the ~stop or -wait control· argumerit is used.
-nofill
-nof
sets the default fill mode to OFF. The default fill mode is ON.
-number
-nb
prints input line numbers at the left margin of the output. The line numbers
have the form "i n" where i is the index number of an inserted file. A list of
3-5
AZ98-02
compose (comp)
inserted files showing the index numbers is written on user_output after
completion of all text processing. The default for this feature is OFF.
-num ber_brief
-nbb
prints input line numbers at the left margin of the output as for the -number
control argument but the list of inserted files is not produced.
-output_file {path}
-of {path}
directs the formatted output to a file instead of to the user's terminal. The
assumed output device is the Multics online printer but may be changed with
the -device control argument. If path is not given, then the output for all
given input files is written to individual output files whose names are formed
by replacing the suffix compin of the input file entrynames with the suffix
compout. If path is given, then output for all given input fil~ is accumulated
in that single bulk output file. The default for this feature is OFF; that is,
formatted output is written back to the user's terminal.
-pages page_, i st
-pgs page 'i st
specifies a blank-separated list of selected pages to be printed. Each element in
the page 'i st must be either a single page, n, or a range of pages, n f1, where
n is a structured page number as for -from above. The page numbers given
must steadily increase without duplication. At least one page must be specified
and up to 100 list elements may be given. This control argument is mutually
exclusive with the -from and -to control arguments. The default for this
feature is OFF.
Note:
Page number structures containing parentheses are changed by the
command processor and must be given as quoted strings.
-pages_changed {x}
-pgc {x}
specifies that, of the pages selected for prIntmg (either all pages or some subset
of pages selected through use of the -pages, -from. and -to control arguments),
only those pages containing text within the range of an active change-bar
control or within the scope of the dot_page documentation macro are actually
printed. The base pages of the dot page set (for example, page 3 of the set 3,
3.1. 3.2) are not considered part of the dot page set This control argument is
independent of the -change_bars and -change_bars_art control arguments, either
of which must be given to cause the text change marks to be created. The
optional parameter x chooses the change bar level and must match the x
parameter given with -change_bars for proper operation. The default value for
x is SP (ASCII 040) and chooses all active change levels.
-parameter {st ring}
-pm {string}
assigns string as the value of the built-in variable "Parameter". The default
value for string is an empty string.
-passes n
-pass n
processes the input file n times to permit proper evaluation of expressions
containing variables that are defined following their reference(s) in the text. No
output is produced until the last pass. The default value for n is 1.
3-6
AZ98-02
compose (comp)
-stop
-sp
waits for a newline character (ASCII code 012) from the user before beginning
each page of output and after the last page. The pause is signalled by giving
two BEL/HT sequences and returning the print head to the left margin. If
only a newline is typed, the next page is printed. If a q is typed, the
command invocation is terminated gracefully. If an r is typed, the page just
printed is reprinted. The default for this feature is OFF.
-to n
ends output after the page numbered n where n is a structured page number
as for the -from control argument above. This control argument is mutually
exclusive with the -pages control argument. The default value for n is the last
page.
-wait
-wt
waits for a newline character (ASCII code 012) before beginning the first page
of output to the terminal. but not between pages (see the -stop control
argument above). The default for this feature is OFF.
3-7
AZ98-02
compose_index (cndx)
compose_index (cndx)
This command processes raw index data gathered by compose and produces a
cross-reference index file according to a specified format
SYNTAX AS A COMMAND
cndx path {-controf_args}
ARGUMENTS
path
is the pathname of the compin file producing the raw index data. The compin
suffix need not be given.
CONTROL ARGUMENTS
-alpha_header
-ahdr
inserts centered uppercase alphabetic characters as group separators whenever the
first character of the primary key changes.
-control_file ctl path
-cf ctf path
-
uses
ctl path.cndxctl as the control file for this index. The suffix "cndxctl" is
assumed-if noi given. The default control file is path.cndxctl.
-number Ii
-nb n
one of the 10 (0 through 9) possible raw index data files. The default value is
O. See Notes below.
NOTES
The raw index data files are produced by compose when the .hit control (refer to
Section 2) is used. The default raw data file is path .O.cndx. The output file is
path_entryname.n.index in the current working directory. If the output file does
not exist, it is created; if it does, it is overwritten.
The data in the raw data file is processed into an arbitrarily chosen format, the
style of which is determined partially by constants built into the program and
partially by statements in path .cndxctl. See "Index Control Files" below.
The final set of hit strings (after all raw data processing is complete) is sorted into
an alphabetic collating sequence (i.e.. without regard to case). The handling of
certain prefix characters is provided by the use of a control directive. See
"compose_index Control Directives" below.
Index Control Files
The index control file contains compose controls and text lines that partially
determine the format of the index. and directives for compose_index that control
3-8
AZ98-02
compose_index (cndx)
the processing of the hit strings (see "compose_index Control Directives") and
complete the definition of the index format. The use of an index control file is
not required. If one does not exist, defaults in the documentation macros and the
program determine the format of the index.
The output file created by this program is to be treated just like any other section
of the document to which it applies. Therefore, the same macro package
initialization must be performed and this is done with compose controls at the
beginning of the index control file. A standard macro, IOindex, is provided for
users who do not wish to provide their own detailed index format. The format
established by this macro is the one used by all Multics and GeOS user
documen ta tion.
All hit string processing control directives are given as compose comment strings.
During processing of the index control file, any line that is not recognized as a
control directive is written to the output file for further processing by compose.
compose_i ndex Control Directives
For the control directives that follow, all input is given in lower-case without
regard to the case of the output. '
.*blind abed...
In many instances, a particular keyword will appear as such and with one or
more prefix characters. A typical example is "rawo" and "Arawo" in tty_
modes. It is desirable to have such prefixed and unprefixed keys sort together
in the index. To accomplished this, the sort algorithm may be made "blind" to
such prefix characters by the use of this directive. The set of characters abed ...
are treated specially as prefix characters such that keywords with them sort
after their unprefixed counterparts. For example:
• ~'(b 1 i nd A\$
Note:
An older form of this directive, ". *ignore" is also supported.
The following directives apply to permuted keys only .
. *phrase st r
Very frequently, it is necessary that a short phrase instead of a single word be
a key in the index. This directive provides the ability to indicate that such
phrases are to be treated as keys. Since punctutation may be wanted in the
phrase, only one str may be given in the line. For example:
.*phrase access control
.*phrase pack labels
.*phrase control cards
.*tran str 7 ,str2
Also very frequently, various grammatical forms of a root keyword or a
suffixed keyword appear in an index and should be sorted together. This
directive provides the ability to transform such keys for sorting only; the
given keys will appear in the final index. For example:
.*tran labels, label
.*tran labeled,label
3-9
AZ98-02
compose_index (cndx)
.*tran labelling, label
.*tran sys_info_$,sys_info_
.*excl excl key {~xcl key} ...
During-permutation, many unwanted hit strings may be generated, primarily due
to conjunctions. articles, prepositions, etc., in the given primary key. Further,
permutation may generate unwanted hit strings that have a primary key that is
wanted for other hits. This directive controls the exclusion of hit strings that
begi n with the partial hit string excl key. Only as much of the unwanted hit
string as is needed for unique identifICation need be given, but it must contain
the entire new primary key. For example:
.*excl to,for,from,and,but
.*excl system-info,reporting-standard
3-10
AZ98-02
convert_runoff (cv___rf)
convert_runoff (cv_rf)
The convert_runoff command converts a runoff input segment into a compose
input segment.
SYNT AX AS A COMMAND
cv_rf path
where path is the pathname of a runoff input segment to be converted. The suffix
runoff need not be given. Output is written into [wd] >entryname.compin. where
entryname is extracted from path.
NOTES
All controls .processed are at the beginning of lines or immediately after a series of
. ur controls.
Warning messages are produced for conversions that are doubtful or may be
ambiguous.
3-11
AZ98-02
display _comp_dsm (ddsm)
display_comp_dsm (ddsm)
The display_comp_dsm command displays selected information from a compose
device description table. device.comp_dsm.
SYNTAX AS A COMMAND
ddsm path {font} {-control_ args}
ARGUMENTS
path
is the pathname of a device description table. The entryname must end with
the suffix comp_dsm but the suffix need not be given in the command line.
The star convention is not supported. If this is the only argument given.
summary information on all devices defined in the table is displayed. See
Examples below.
font
is the external name of a font defined for the device. It may be given as
family or family/member. If this argument is given, then all the graphics
(Multics characters) for the named font with their widths are displayed;
otherwise, information on the device is displayed.
CONTROL ARGUMENTS
-device
-dv
displays information on the named device only (including all defined fonts).
-linelength n
-11 n
sets the line length for the display to n.
defined linelength for the user's terminal.
The default value is the system
-long
-lg
displays detailed information. If font is given, then display all the graphics
(Multics characters) for the font with the width and replacement output string
for each. If font is not given, then display all the defined parameters for the
named device.
3-12
AZ98-02
display_comp_dsm (ddsm)
EXAMPLES
-show all devices
ddsm dtc300s
Device:
dtc300s. DTC300s;
devc1ass: diablo;
Device:
vdtc. v300;
devc1ass: photocomp;
-show named device with all fonts
ddsm dtc300s -dv -11 65
Device:
dtc300s, DTC300s;
devc1ass: diablo;
family: centuryschoo1book. cs, helvetica, h;
member: I, 1m. Imedium. Ir, Iroman;
member: Ib, Ibo1d, Iboldroman, Ibr;
member: Ibi, Ibo1ditalic;
member: I;, litalic, Imediumita1ic, Imi;
f am i 1 y: pica 10;
member: I. 1m, Imedium. Ir, Iroman;
member: Ib, Ibo1d, Ibo1droman, Ibr;
member: Ibi, Ibolditalic;
member: Icaps;
member: Icaps_;
member: Ii, litalic, Imediumitalic, Imi;
bachelor: aSCii. 14font, 13exact, 14exact, text, footnote.
APL, CSR, HR;
bachelor: 10font. 13font, ASCII;
bachelor: bold, CSBR, HBR, HBB1;
bachelor: ital iC. 12font, l1exact, 12exact, CSI, HmI, asci i
bachelor: l1font, ASCII;
bachelor: CSBI, HBI;
fami ly: pica12;
member: I, 1m. Imedium. Ir, Iroman;
member: lb. Ibo1d, Iboldroman, Ibr, Icaps;
member: Ibi, Ibolditalic, Icaps_;
member: Ii, litalic. Imediumita1ic. Imi;
bachelor: pica12;
bachelor: pica12_;
bachelor: PICA12;
bachelor: PICA12_;
ddsm dtc300s -19 -11 65
Device:
dtc300s. DTC300s;
devclass: diablo;
-show long device information
1*
-table version and expected version
~ersion:
1 (1)
~/
units: pt;
attach:
syn_ user_output
comment:
Type Wh~el Identification
1 - 38101-01
PICA 10
2 - 38510
APL 10
3 - 38102-01
ELITE 12
DB: dtc300s_writer_$display
II
II ;
II
-this is the closing quote for the
3-13
AZ98-02
display _comp_dsm (ddsm)
comment: field
cl eanup: "\033\033\033\006\033 0";
defaultmargs: 48., 24., 24., 48.;
init: pica10/m 7.2;
interleave: on;
letterspace: 0;
maxpages: unlimited;
maxfiles: unlimited;
maxpagelength: unlimited;
maxpagewidth: 950.4;
minbotmarg: 0.;
min 1ead : 1. 5 ;
mi nspace : 1. 2 ;
mintopmarg: 0.;
stream: off;
taperec: unlimited;
family: centuryschoolbook, cs, helvetica, hmember: I, 1m, Imedium, Ir, Iroman;
member: Ib, Ibold, Iboldroman, Ibr;
member: Ibi, Ibolditalic;
member: Ii, litalic, Imediumitalic, Imi;
family: pica10;
member: I, 1m, Imedium, Ir, Iroman;
member: Ib, Ibold, Iboldroman, Ibr;
member: Ibi, Ibolditalic;
member: leaps;
member: Icaps_;
member: Ii, litalic, Imediumitalic, Imi;
bachelor: ascii, l4font, l3exact, l4exact, text, footnote,
APL, CSR, HR;
bachelor: 10font, 13font, ASCII;
bachelor: bold, CSBR, HBR, HBB1;
bachelor: ita 1 i c, 12font, 11exact, l2exact, CSI, HmI, asc i i
bachelor: l1font, ASCII;
bachelor: CSBI, HBI;
fam i 1 y: pica 12;
member: I, 1m, Imedium, Ir, Iroman;
member: Ib, Ibold, Iboldroman, Ibr, leaps;
member: Ibi, Ibolditalic, Icaps_;
member: Ii, litalic, Imediumitalic, Imi;
bachelor: pica12;
bachelor: pica12_;
bachelor: PICA12;
bachelor: PICA12_;
ddsm dtc300s CSR -11 65
Device:
dtc300s. DTC300s;
devclass: diablo;
bachelor: CSR;
strokes: 6
wordspace: 3,6,9
010( -6)
040( 6)
" ! " (6)
1/%1/ (6)
(6)
"&" (6)
"
J
"
3-14
-show font information
042(6)
"#" (6)
"$" (6)
" ( " (6)
" ) " (6)
"*" (6)
AZ98-02
display_comp_dsm (ddsm)
" , ,,( 6)
"+" (6)
"2" (6)
" 1" (6)
"7" (6)
"8" (6)
">" (6)
"=" (6)
"0" (6)
"c" (6)
"I" (6)
"J" (6)
liP " (6)
"0" (6)
"V" (6)
"U" (6)
" [ "( 6)
"\" (6)
"a" (6)
"b"(6)
"g" (6)
"h"(6)
"m" (6)
"n"(6)
"t"(6)
"s" (6)
liZ" (6)
"y" (6)
177( 0)
200(6)
237(6)
240(6)
301 (6)
(e)(12)
360(6)
375(6)
EN_(6)
ENd(6)
1hi-X(6) 424(6)
delete-mark(6)
1hi-{(6)
1hi-[(6)
1hi-}(6)
1hi-](6)
441(6)
450(6)
455(6)
456(6)
463(6)
464(6)
471(6)
412(6)
417(6)
500(6)
505(6)
506(6)
513(6)
514(6)
521 (6)
522(6)
527(6)
530(6)
537(6)
"-"(6)
"3"(6)
"9" (6)
"?" (6)
"E" (6)
"K" (6)
"Q" (E)
"W"(6)
" ] ,,( 6)
"."(6)
"4" (6)
" : "( 6)
"@l"(6)
"F" (6)
"L" (6)
"R" (6)
"X"(6)
",..." (6)
"e" (6)
" i "( 6)
"0" (6)
"u II (6)
"d" (6)
"j "( 6)
"p"(6)
"v"(6)
" I ,,( 6)
202(6)
EMd(12)
" {" (6)
201 (0)
254(10)
0(6)
304(9)
PS(6)
EM(6)
THN(6)
-(6)
dn-arrow(6)
di a.-r ight (6)
left-eirele(6)
right-eirele(6)
451 (6)
452(6)
457(6)
460(6)
465(6)
466(6)
413(6)
414(6)
501 (6)
502(6)
507(6)
510(6)
515(6)
516(6)
523(6)
524(6)
531( 6)
532(0)
"_" (6)
"e" (6)
"k" (6)
"q" (6)
"w" (6)
"} "( 6)
235(6)
261( 6)
320(6)
EM_(12)
I '(6)
" (6)
426(6)
dia-left(6)
dia-top(6)
«6)
431(6)
->(6)
444(6)
up-arrow(6)
453(0)
454(6)
461 (0)
462(6)
461(0)
470(6)
415(0)
476(6)
503(0)
504(6)
511(0)
512(6)
511(0)
520(6)
525(0)
526(6)
534(6)
536(6)
-show long font information
ddsm dtc300s CSR -lg -11 65
Device:
dtc300s, OTC300s;
devclass: diablo;
bachelor: CSR;
strokes: 6
wordspace: 3,6,9
010( -6, "\010") 040( 6, Ii " )
" ! ,,( 6, II ! 10 )
"#"(6,"#")
II J II (6, .. I II )
"%10(6,"%")
")"(6,")10)
"+"(6,"+")
;; /
i; ( 6, "/ ,; )
"$"(6,"$")
"("(6,"(")
II (6, .. , II )
..
Of
;, 0
ii (
6 , "0;' )
11_11(6,.11_11)
II
*
II
(6,
1I;,lc: II )
II
•
II
(6,
II . . . )
10 2 10 (6, "2 10 )
"6" (6. "6")
">" (6, "> ")
"8" (6, "8")
II FlO (6, "F ")
"J IO (6,"J")
"3" (6. "3")
"4" (6, "4")
"8 10 (6,"8")
11<11(6,11<11)
11=11(6,11=11)
II? II (6, II? II)
n@lIO(6,"@l")
IOAIO(6,"A")
"C" (6, "C")
"0"(6,"0")
"G" (6, "G")
IHI(6."H")
"K" (6, "K ")
"L"(6,"L")
IPIO(6,"P")
"E"(6,"E")
10 1 10 (6, "110 )
IOMIO(6,IOMIO)
IOQII(6,"QIO)
3-15
042 ( 6, " " " " )
"&"(6,"&")
(6, " 1 10 )
"5"(6,"5 10 )
10 9" (6, "9")
.. 1 ..
"7" (6, "7")
II ; It (6, II ; II )
"0"(6,"0")
"0" (6)
"6" (6)
"<" (6)
"8" (6)
"H"(6)
"N"(6)
"T" (6)
"Z" (6)
" \ "( 6)
"f" (6)
"1 "( 6)
"r" (6)
"X" (6)
"-." (6)
236(6)
211(0)
324(12)
EN(6)
"/" (6)
"5" (6)
" ; "( 6)
"A" (6)
"G" (6)
"M" (6)
"S"(6)
"Y" (6)
II
:
If
(6,
II
:
.. )
ION"(6,IN")
IORIO (6, "R")
AZ98-02
"5"(6,"5")
"W"(6, "W")
"["(6,"[")
"T"(6,"T")
"X"(G,"X II )
"\"(6,"\")
(6,
"U"(G,"U
"Y"(G,"Y")
N
)
"]"(6,"]")
"V"(G,"V")
"Z"(G,"Z")
p! .'"
,!
(6 ,
!~.-" !! )
Ibl(G,lb")
la"(G,"a")
"e"(G," e .,)
"f"(G,"f")
Ij"(G,lj")
II i ,,( G, " i " )
"nl(G,ln")
Im"(G,lm")
II 1 II (G, "1 " )
"ql(G,"ql)
Ir"(G,"r")
"p"(G,"p")
It"(G,"t")
"U"(G,"U")
" V "(G," V ")
"5"(6,"5")
"WI(6,IW")
"'X"(G,"X")
lyl(G,ly")
IIZIl(G, "Z")
"{"(G,"{")
"I"(G,"I")
"}"(G,"}")
"-,II(G, 11-,")
177(0,"\177")
200(G,"\033cl\033PoA1B4q\033\033C.")
201(0, "\033C_\033POY2hG1G5Q\033\033C.")
202(G,"\033CI\033POq1B4A\033\033C.")
235(G,ld\010_")
23G(G, "n\010_")
237(G, "x\010_")
240(G,I\033C[\033P2H\033\033C]\033POP4h\033\033C.")
254(10,"\033C\\033POO\033\033C_\033POWH\033\033C/\033P1W4h\033\033C.")
EMd(12,"\033P<8\033_\033P=B")
2G1(G,I\033cl\o33PC»A\033\033C-\033P085s\033\033C.")
277(0,"11)
301(G,"\033CI\033PC»A\033\033C-\033POB1D4q\033\033C.")
(C)(12,"\033P4V\033c\033P3H2AAIIHIHH3IHIIAA1IIHIHHOIHI4h")
304(9,"\033C/\033POO\033\033C_\033POJH\033\033C\\033P1JHO@IIIIII>X1@IIIIII(G,"\033P7]OPAAAAAA1IAAAAOIAA1IPOOHHHHAAAAAAAA=\")
461(0,"\033P?RO@lHHHHAAAA")
462(6,"\033PO@lAAAAAAAA=\")
463 (6. "\033PO@lAAAAAA II HH= L" )
464(6. "\033POOAAIIHH=J")
465(6,"\033P2@1AAIIOIIAA=\")
466(6,"\033PAAIIHH=j")
473(6, "\033PO@lAAII2IIAA=\")
4 74 (6, "\033P>PO@lHHIIAAAAAA=\" )
475(0,"\033P?RO@lHH!!AA")
476(6, "\033POOAAAAAAAA=\")
477(6,"\033PO@lAAAAAAIIHH=L")
500(6,"\033POOAAIIHH=J")
501(6."\033PO@lAAAAAAAA=\")
502(6,"\033PPO@lHHIIAAAAAA=\")
511( 0, "\033P?RO@>HHI IAA" )
512(6."\033PQ@lAAAAAAAA=\")
513(6,"\033PO@lAAAAAAAA=\")
514(6, "\033PO@lAAAA=Zi')
515(6,"\033PO@lAAAAAAAA=\")
516(6."\033POOAAAAAAAA=\")
517 (0, "\033P=BO@lAAAA" )
520(6,"\033PQ@lAAAAAAAA=\")
521(6, "\033P2PAAAAAAAA1XAAAAAAAA4h")
522(6. "\033P2PAAAA1XAAAA4h")
523(6. "\033P2PAAAAAAAA1XAAAAAAAA4h")
524(6,"\033P2PAAAAAAAA1XAAAAAAAA4h")
525(0,"\033P?JO@>AAAA4X1@1AAAA6L")
526(6."\033P2PAAAAAAAA1XAAAAAAAA4h")
527(6,"\033P=BO@>AAAAAAAAAAAA=\")
530(6,"\033P?JO@lAAAAAAAAAAAA4X1@1AAAAAAAAAAAA41")
531 (6. "\033P7YO@>I II I II I I 1=0" )
532 (0, "\033P=AO@lAAAAAAAA =C" )
534(6."\033PO@lHHHHH4H")
536(6,"\033P>K1@>IIIIIIIIIIIIIIII7N")
3-17
AZ98-02
expand_device_writer (xdw)
expand_device_writer (xdw)
The expano_oevice_writer command is useo LO invoke the Expander to expand an
expansion input file into an expansion output file. The language used in the expansion
input file is described above.
SYNTAX AS A COMMAND
xdw {path} {-contro'_args}
ARGUMENTS
path
is the pathname of the expansion input file. The entryname of this file must
have the suffix xdw, but the suffix need not be given in the command line. By
default. the expanded expansion output file is written to a segment in the
working directory whose name is formed by stripping the suffix from the input
file en try name. Multi-segment files and the star convention are not supported.
If no pathname is given, input may be given to the Expander by using the
-input_string control argument described below.
CONTROL ARGUMENTS
-argumen ts ...
-ag ...
all remaining parameters in the command line are arguments to be passed to
the file or input string being expanded.
-brief
-bf
does not display the expansion usage list when the expansion is complete.
(Default)
-call command /; ne
if there are no errors in processing. executes the given command line when the
expansion is complete.
-input_string st ring
-instr str; ng
expands the given str; ng as an expansion input file. By default. the expansion
is displayed and no expanded output file is created.
-long
-lg
displays the expansion usage list when the expansion is complete.
-outpuCfile path
-of path
vv'iites the expanded output into the segment with the given
forces no_print even if -print is also given.
3-18
__ .... 1- _ _ _ _
pi1 L1111i1Ult;.
AZ98-02
expand_device_writer (xdw)
-no_print
-npr
does not display the resulting expansion. This is the default when path is given
and is forced when -output_file is given.
-print
-pr
dislays (or prints) the resulting expansion. This is the default when -input_string
is given and is mutually exclusive with -output_file.
3-19
AZ98-02
format_document (fdoc)
format_document (fdoc)
The format~document command formats text segments. Output lines are built per
the embedded control lines within the input file being formatted. Although the control
lines are embedded within the input text, they do not appear in the output.
SYNTAX AS A COMMAND
fdoc path {-control _args}
ARGUMENTS
path
is the pathname of an input segment or multisegment file. The suffix fdocin
must be the last component of the entry name; however the suffix need not be
supplied in the command line.
CONTROL ARGUMENTS
-indent {n}
-ind {n}
indents the output n spaces from the left margin. This space is in addition to
any indention established by the usage of the indent control line within the text
of the input file.
-output_file {path}
-of {path}
directs the output to a file instead of to the user's terminal. If path is not
given. then the output is written to an output file whose name is formed by
replacing the suffix fdocin of the input file entry name with the suffix
fdocout. (The default for this feature is off.)
-page_num bers
-pgno
ends each page with two blank lines and a centered page number. (The default
for this feature is off.)
Formatter Description
Basically. the format_document command (a text formatter) takes an input file
which was created using a text editor (e.g., ted, qedx and edm), formats that file, and
either displays it on the terminal or writes it to a new file with a unique name. To
direct format_document to perform certain actions, the user places special lines. called
control lines, in the input file. All control lines begin with a period and must be on
a line by themselves. The format_document command makes certain assumptions about
how the document is to be formatted (i.e.. when the fo!"mat_document command
executes. it aelauns to certain COnOlllOnS in the absence of user-speClIleo control lines).
It assumes that the output is going to be on standard-sized paper which has 66 lines
per page and that the user wants the printed lines to be 65 characters wide. These
values represent an 8-1/2 by 11 inch page with one-inch margins all around. It also
assumes that the user wishes to have both the left and right margins lined up evenly
like the margins of this paragraph. When the user wants format_document to do
3-20
AZ98-02
format_document (fdoc)
something different than the standard defaults. the user must insert the necessary
control lines in the input file to accomplish the desired results.
As discussed below. line filling is the moving of words from line -to line to make
the line size as near to the prescribed length as possible. A control break is an action
that temporarily stops this process (i.e., it processes the previous line. the line just
ahead of the break) and prints th.is line as is. even if it is a short one. All of the
control lines. with the exception of .pdl and .pdw. cause control breaks. A blank line
or a line that starts with a space also causes a control break.
Following is a summary of the control lines recognized by the format_document
command .
.alb
(align both) puts extra spaces into each line so that both the left and the right
margins are even. This control line is effective only if fill (.fin) is also in
effect. (Default)
.all
(align left) does not put extra spaces into the lines. The left margin is even
and the right margin is ragged. This control line is effective only if fill (.fin)
is also in effect
.fif
(fill off) retains lines in the output file as they are in the input file no matter
how long or short
. fin
(fill on) restructures the input file lines to the current line length for the
output file by taking a word or words from the next line in order to fill the
line as close as possible to the current line length. -If a line in the input file is
longer than the current line length, move a word or words to the next line.
etc. (See the description of the .alb and .all control lines.)- (Default)
.in {n}
(indent) sets the indention level. It is possible to have format_document indent
each line a certain number of characters. If n is given with a plus or minus
sign. then n is added to or subtracted from the current indention level. If n is
given without a sign. then n becomes the indention level. An error message is
displayed if an indention level is less than zero or greater than the line length.
(The default indention level is 0.)
.pdl {n}
(page length) sets the page length. If n is given with a plus or minus sign.
then n is added to or subtracted from the current page length. If n is given
without a plus or minus sign, the page length is changed to n. The
format_document command inserts blank lines at the top and bottom of each
page. so be careful not to set the page length to a value less than 13 (or less
than 14 if you are having page numbers printed.) i~ n error message is displayed
if a page length of less than the required lines is given. (The default page
length is 66 lines.)
..
.pdw {n}
(page width) sets the page width (line length). If n is given with a plus or
minus sign. then n is added to or subtracted from the current line length. If n
is given without a plus or minus sign. the line length is changed to n. An
3-21
AZ98-02
format_document (fdoc)
error message is displayed if the set line length does not accommodate the
input file. (The default page width is 65 characters.)
.un {n}
(undent) sets the indention level for the output of the next line only. If n has
a plus sign or no sign, the line is indented n characters less than the current
indention level. If n has a minus sign, the line is indented n characters more
than the current indention level. If this seems backwards, just remember that
undent goes in the opposite direction from indent. An error message is
displayed if the indention that is caused by undenting is less than zero or more
than the line length.
EXAMPLE
The following is an exampl~ of a business letter created using the format_document
command (fdoc). Suppose you are creating a business letter that is to be printed on a
standard 8-1/2 by 11 inch piece of paper and. that you want the lines to be 60
characters long. You first create the input file with a text editor. In this example the
input file is labeled letter.fdocin. Line numbers are shown on the example for purposes
of commentary immediately following the example. All of the numbered items below
are user-entered data and are not flagged with a bullet as no system responses are
included.
3-22
AZ98-02
format_document (fdoc)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
ted
a
.pdw 60
.fif
. in 35
9341 Millennium Lane
Reston, Virginia 22061
November 24, 1981
. in
Zimmerman Widget Company
53698 Dixie Highway
Drayton Plains, Michigan 48999
Dear Sir,
.fin
.un-5
I recently purchased one of your model GX-721 widgets.
I feel that your engineering staff deserves high
praise for this new model.
It is apparent
that a great deal of thought has gone into its
design.
I am particularly pleased with the optional
conetop replacement mechanism.
.un-5
My purpose in writing this letter, however, is to
obtain information.
As you are well
aware, the filter requires a complete overhaul after
each 250 hours of use.
The service brochure indicates
that the nearest service center to my location is in
Chapel Hill, North Carolina, which is a six-hour drive
from my residence.
If you can direct me to a service
center that is more convenient to my location,
would
be grateful.
.f i f
. in 35
Sincerely yours,
Michael P. Marley
\f
w letter.fdocin
q
3-23
AZ98-02
format_document (fdoc)
line 1
Invokes the text editor.
line 2
Places the text editor in append mode.
line 3
Sets the line length (page width) to 60 characters. If this control is not present,
then the line length would be set to 65 characters by default
line 4
Turns fill mode "off". The reason for turning fill off is because the text beginning
on line 6 through line 8 is an address. If fill mode was not turned "off" then the
address would be reformatted by fdoc, words might be moved from line to line, or
extra spaces might be filled in. You do not want this to happen, so you turn fill
off. The same thing is dOtle at line 43 just prior to the closing.
line 5
Sets the indention to character position 35. Text begins at column 1 unless you
change it, and since the return address is to be on the right-hand side of the letter
you must set the indention to the location desired (character position 35 in this
case).
line 6-8
Return address.
line 9-11
Three blank lines are inserted by pressing the newline (NL) or carriage return (CR)
key three times.
line 12
Resets the indention level to 0 (the absence of a number after the control results
in a default to 0).
line 13-19
Address of the recipient, two blank lines, the salutation, and another blank line.
line 20
Turns fill mode "on" (fill was turned off by the control on line 4) as you want
the body of the letter filled.
line 21
The indention level is set to 0 by the control in line 12, but you want to indent
the next line (and only the next line) by 5 characters since it begins a paragraph.
To change the indention for only one line you use the undent control which works
in the opposite direction of the indent control. Undent subtracts the number from
the indention (i.e., if you used .un 5 it would move the indention 5 spaces to the
left). You want to move 5 spaces to the right to indent the paragraph. so you use
a negative number.
line 22-40
This is the body of the letter. Notice that there has been no attempt to control
the entered line lengths; it is entered free-form. The fdoc command formats all of
the data for you, so long as fill mode is on". Lines can be as short or as long as
you wish. even if the lines wrap around (wrap around is the situation where the
user continues entering data until the line on the terminal has reached the right
margin, at which point the system moves the cursor to character position 0 of the
'i
3-24
AZ98-02
format_document (fdoc)
next line, and data entry is continued until a newline or a carriage return is
entered).
line 41
Turns fill mode "off".
line 42
Sets the indention to character position 35 so that the letter closing, signature, and
sender's name appear on the right side of the page (lines 45-50).
line 49
Terminates append mode and returns the user to edit mode.
line 50
Writes the buffer contents to permanent storage. In this case the buffer is stored
in a segment identified as letter.fdocin.
line 51
Quits the editor and returns the user to Multics command level.
Now that your input file (1 etter. fdoc i n) is ready, you can have it formatted and
printed on the terminal for your perusal.
3-25
AZ98-02
format_document (fdoc)
fdoc letter.fdocin
9341 Millennium Lane
Reston, Virginia 22061
November 24, 1981
Zimmerman Widget Company
53698 Dixie Highway
Drayton Plains, Michigan 48999
Dear Sir,
recently purchased one of your model GX-721 widgets.
I feel that your engineering staff deserves high praise for
this new model.
It is apparent that a great deal of thought
has gone into its design.
I am particularly pleased with
the optional conetop replacement mechanism.
however, is to
My purpose in writing this letter.
As you are well
aware,
the filter
obtain information.
requires a complete overhaul after each 250 hours of use.
The service brochure indicates that the nearest service
center to my location is in Chapel Hill, North Carolina,
which is a six-hour drive from my residence.
If you can
direct me to a service center that is more convenient to my
location, I would be grateful.
Sincerely yours,
Michael P. Marley
3-26
AZ98-02
format_document (fdoc)
Assume the output looks good, and you are ready to make a final copy. Since your
lines are 60 characters long and you are going to print it on standard 8-1/2 by 11 inch
paper, and since most terminals and printers print 85 characters in 8-1/2 inches, you
will want your letter to be centered on the paper. This is where the -indent control
available within the format_document command comes into play. Your lines are 25
characters shorter than the width of the paper, so if each line begins at character
position 12 (roughly half of 25) your letter will be centered on the page. The command
line:
fdoc letter -indent 12
accomplishes this.
Let us say, for example, that you are going to save your letter in a file so that
you can print it later on another terminal or on a high-speed printer. In such a
situation, you would type
fdoc letter -indent 12 -output_file
The -output_f i 1e control argument saves the output in a file rather than printing it
on your terminal. In this example, the file is named 1et ter. fdocout. You can now
use the dprint or print commands (see the command descriptions in Multics
Commands) to print the letter.
If you have a high-quality printing terminal and wish to print this letter on a
piece of typing paper, you would type:
print letter.fdocout -stop
After entering this command, place the typing paper in the terminal, poSitIOn it so that
printing begins at the top, and then enter a carriage return (newline character). The
letter is then printed, stopping at the last line. At this point, you can remove the
paper and put in a new sheet (if the letter is more than one page). When the letter
has been printed, you can enter another carriage return, and you are back at Multics
command level.
3-27
AZ98-02
process_com pout (pco)
process_compout (pco)
The process_com pout command is used to process one or more compose output
(com"pout) files to an online device, or to a magnetic or punched paper tape. All or
portions of the files may be requested.
SYNTAX AS A COMMAND
pco paths {-control_args}
ARGUMENTS
paths
are the pathnames of input
the last component of the
argument below); however.
line. Output is produced in
command line.
files to be processed. The suffix compout must be
input file entrynames (but, see -pathname control
the suffix need not be supplied in the command
the order in which the pathnames are given in the
CONTROL ARGUMENTS
any control argument specified in the command line applies to all input file
pathnames given.
-files In} {,m}
overrides either or both of the default output file factors when wntmg output
to magnetic or paper tape. The default output file factors are found in the
header record of the input file and are set from data in the device description
table. n is the maximum number of pages per file and m is the maximum
number of files allowed on the tape.
-from n
-fm n
starts printed output at page n. This control argument is mutually exclusive
with the -pages control argument. The default value of n is "!l". See "Page
Numbers" below for a discussion of page numbers.
-mode xxx
selects any of the known alternative modes of output or specifies an entirely
new mode.
If xxx is a single word, then it may be any of the built-in modes shown
below or the name of an output mode given in the header record of the input
file. If it is a built-in mode, then the action described below is taken; if it is
a known output mode, then output is prepared according to that mode; if it is
neither. then an error message is displayed giving the names of all output
modes in the input file header.
If xxx is a quoted string containing white space, then it is used as an attach
description for the output medium irrespective of any modes specified in the
input file header.
If this control argument is not given. the mode used is the first known mode
found in the input file header; if there are none. output is written to the
user_output I/O switch.
3-28
AZ98-02
process_compout (pco)
The built-in modes are:
comment
produces a listing of the comment information in the input file header.
This comment contains output mode specifications. device setup information.
print wheel identification for "diablo" class terminals, and possible other
miscellaneous information. It is a copy of the "Comment" information in
the device description table. (See "Magnetic Tape Header Files" below for
a discussion of the information shown in this display.) For example:
pco ascii.walvip -mode comment
** From file: >udd>m>jaf>cdv>ascii.walvip.compout
mode:tape=tape_ibm_ -bk 800 -n1b -den 800 -fmt f -mode ascii
setup:***** This tape created [date_time].
*****
setup:This tape is to be sent via FedExp to
setup:
Honeywell, Multics Computer Operations
setup:
5115 N 27th Ave
setup:
Phoenix, AZ 85017
setup:
setup:Font setup:
1
2
setup:
456
3
setup:
A- 187 145 108 409 2160 534
setup: 8- 84
85
69 1716 1715
70
setup: C-8398
X
71
x
x
x
setup:
setup:Disposition of output:
f i 1 e:
file:Document: [compout][ioa_ "Document: Aa" [compout]]
file:
Return[compask" Pasteup?" no= "yes= pasted up"]
\c[compask "original/copy? " o=original original c=copy copy].
file:
Cost center: [compask" Cost center: II]
file:
Send to: [compask "
Send to: "]
file:
Destination: [compask" Destination: "]
file:
Comments: [
compask" Comments: "]
content_file: seg, tape
pco -mode comment
** From file: >exl>cd>doc>pco.compout
Type Wheel Identification
1 - 38101-01
PICA 10
2 - 38510
APL 10
3 - 38102-01
ELITE 12
?
displays all the "mode:" lines in the comment information.
setup
displays all the "setup:" and "file:" lines in the comment information.
3-29
AZ98-o2
process_compout (pco)
display
produces a directory of files as would be written to the tape. followed by
an interpreted display of all the files (see "Display ~..1ode Interpretations"
below). This output is written to the user_output I/O switch. For example:
pco ascii.walvip -mode display
File information:
Directory of files as would go on tape, by file number
#
- - - pageids present
o
Document: ascii .walvip
1
FRONT 1
2
2
**
From file: >udd>m>jaf>cdv>ascii .walvip.compout
**** FILE #1 ****
=a1=p08=f120=14500=g=.OO=k1=rx=tn3
=tl09=ta06=tm06=8=ilOOOO=j
7.181-05-261614.8ascii.walvip1
<1/4>=d<1/4>
=a1=p08=f120=14500=g=.OO=k1=rx=tn3
=tl09=ta06=tm06=8=ilOOOO=j
7.181-05-261614.8ascii.walvip2
<1/4>=d<1/4>
=if090=a1=p10=f120=14500
=ilOOOO=a1=m120.<1/8>B6 --Sups--
=il0300X=b6 &()1,-.0123456789:;<1/4>4<1/8>?=a1X=b6
=a1X=b6ABCDEFGHIJKLMNOPORSTUVWXYZ!$=a1X=b6
=a1X=b6abcdefghijklmnopqrstuvwxyz =a1X=b6
=m120,=if990=if990=if990=ilOOOO=14500=a1=p08<1/4>=d<1/4>
=if090
=s
display -long
as for the display mode above but also showing the evaluated comment
information as it would be written to the tape. For example:
3-30
AZ98-02
process_com pout (pco)
pco ascii .walvip -mode display -long
Document: ascii.walvip
Pasteup? () ! no
original
original/copy? ()
J86
Cost center: ()
JFalksen
Send to: ()
CRF
Destination: ()
Comments: () ! Do not cut, send in tube
File information:
***** This tape created 06/05/81 0950.3 mst Fri.
This tape is to be sent via FedExp to
Honeywell, Multics Computer Operations
5115 N 27th Ave
Phoenix, AZ 85017
Font setup:
1
2
A- 187 145
8- 84
85
C-8398
X
3
108
70
71
*****
4
5
6
409 2160 534
69 1716 1715
X
X
X
Disposition of output:
Document: ascii.walvip
Return original.
Cost center: J86
Send to: JFalksen
Destination: CRF
Comments: Do not cut, send in tube
Directory of files as would go on tape, by file number
#
- - - pageids present
o
Document: ascii .walvip
1
FRONT 1
2
2
**
From file:
***~
>udd>Doc>jaf>cdv>ascii.walvip.compout
FILE #1 ****
as above
dump
produces an octal/ascii dump of
example:
the records of
the input file.
For
pco pco.compout -mode dump
3-31
AZ98-02
process_compout (pco)
Record
000116
000122
000126
000132
000136
000142
000146
000152
000156
000162
000166
000172
000176
000202
000206
000212
0 3740 252
000000000002 144151141142
040040040040 040040040040
040040040040 144164143063
040040040040 040040040040
040040040040 144164143063
040040040040 040040040040
040040040040 777777777777
000000000000 000000000007
000000000166 040124171160
111144145156 164151146151
061040055040 063070061060
101040061060 012040062040
101120114040 061060012040
062055060061011105114111
072040144164 143063060060
137044144151 163160154141
154157040040 040040040040
040040040040 040040040040
060060163040 040040040040
040040040040 040040040040
060060163040 040040040040
040040040040 040040040040
777777777777 77J777777777
033033033006 033040060040
145040127150 145145154040
143141164151 157156012040
061055060061011120111103
055040063070 065061060011
063040055040 063070061060
124105040061062012104102
163137167162 151164145162
171012040040
Record 1 52700 2744
000000 061040040040 040040040040 040040040040
000004 040040040040 040040040040 040040040040
0000 10 200000000000
17 RAW HALT 2 preface 600000000000
000011 744000000000 000000000021 007011177177
000015 177177177177 177177177177 177000000000
267 1 RAW 000000000000
000020 600000000000 000000005157 033040066012
000024 137137137137 137137137137 137137137137
000030 137137033124 055137137137 137137137137
000034 137137137137 137137137137 137012012015
000040 143145163163 137143157155 160157165164
000044 157033124055 177160162157 143145163163
000050 160157165164 054040160143 157012015137
000054 137137137137 137137137137 137137137137
000060 124055137137 137137137137 137137137137
000064 137137137137 137137012012 012015033040
000070 103105123123 137103117115 120117125124
000074 117033040066 012012033124 004124150145
000100 120033160162 157143145163 163137143157
000104 164040033120 065120033143 157155155141
000110 120065120033 151163040033 120065120033
000114 040033120065 120033164157 040033120065
000120 157143145163 163040033120 065120033157
000124 120065120033 157162040033 120065120033
000130 012015143157 155160157163 145040033120
000134 165164160165 164040033120 065130033050
000140 157165164051 040033120065 130032146151
000;44 033;20065;30 033;64;57040 033;20065130
000150 033120065130 033157156055 154151156145
000154 130033144145 166151143145 054040033120
000160 162040033120 065130033164 157040033120
000164 012015155141 147156145164 151143040157
000170 156143150145 144040160141 160145162040
3-32
.... diablo
dtc300s
dtc300s
........... .
............. 0
... v Type Wheel
Identification.
1 - 38101-01.PIC
A 10. 2 - 38510.
APL 10. 3 - 3810
2-01.ELITE 12.D8
: dtc300s_writer
_$display.
040040040040
040040040040
177177177177
012012137137
137137137137
137137137137
177160162157
054040160143
137143157155
137137137137
137137137033
137137137137
066120122117
054040120103
040033120065
155160157165
156144040033
165163145144
120033160162
156145040033
155157162145
065130033157
143157155160
154145163040
03314;156040
040033120065
065130033157
065130033141
162040160165
164141160145
. . . . . . . 0.
6 ...
.T-_ _ _ _ __
.... pro
cess_compout, pc
o.T-.process_com
pout, peo .. _____
T- _____________
6PRO
CESS_COMPOUT, PC
O. 6 ... T.The .P5
P.process_compou
t .P5P.command
P5P.is .P5P.used
.P5P.to .P5P.pr
ocess .P5P.one
P5P.or .P5P.more
.. compose .P5X.o
utput .P5X.(comp
out) .P5X.files
.P5X.to .P5X.an
.P5X.on-line .P5
X.device, .P5X.o
r .P5X.to .P5X.a
.. magnetic or pu
nched paper tape
AZ98-02
process_com pout (pco)
-pages n In,n ...
-pgs n In,n ...
specifies a blank-separated list of selected pages to be printed. Each
member of the list must be a single page, {n}, or a range of pages, {n,n}.
The page numbers given must normally increase without duplication
(however, see the discussion of page numbers in "Page Numbers" below).
At least one page must be specified. This control argument is mutually
exclusive with the -from and -to control arguments. The default for this
feature is off.
-pages_changed
-pgc
specifies that only addendum pages and those pages containing text within
the range of a change-bar control (from the pages specified by the
"-pages" or "-from" and "-to" control arguments, if given) are to be
printed.
-pathname path
-pn path
is the pathname of an input file even though it may have the appearance
of a numeric parameter or a control argument, or is a compose bulk
output file that does not have the suffix compout.
-stop
-sp
waits for a newline character (ASCII code 012) from the user before
beginning each page of output If only a newline is typed, the next page is
printed. If "q" is typed, the command invocation is terminated gracefully.
If "rn is typed, the page just printed is reprinted. The default for this
feature is off.
-table
-tb
print a table listing information about all selected pages in the file. Only a
table is produced. This control argument is mutually exclusive with all
others. See "Table Option Output" below for further information.
-to n
\
ends output after page n. This control argument is mutually exclusive with
the -pages control argument. The default value for n is "$". See "Page
Numbers" below for a discussion of page numbers.
writes the
parameters
description
file. The
overridden
output to the magnetic tape whose volume name is xx. The
needed for attaching the tape are provided by the device
table and are contained in the header record of the compout
attach descriptions in the "file header may be selected or
by using the -mode control argument described above.
3-33
AZ98-02
process_compout .(pco)
-wait
-wt
waits for a newiine character ~A~Cll code 012) before beginning the first
page of output to the terminal, but not between pages (see the -stop
control argument above). The default for this feature is off.
Page Numbers
Pages may be ref erred to by several methods:
!n
means the n'th physical page in the file. However, n=O means "go back to the
beginning of the file". No page is printed for the W.
$
means the last physical page in the file.
+n
means the n'th relative (to the last page referenced) page in the file.
$-n
means the n'th relative page from the end of the file.
page-id
is the page number constructed by compose. It may be just a simple number
or a compound number such as "B-1", "3-14.2", or "i-5". The page-id must
be an exact match for that in the document. No less than/greater than
checking is possible.
~A).
page selection could be:
-pages !1,!5 +19,+2127.4,127.42 $-1,$
This means to· process four ranges: the first five pages, then the three pages beginning
with the 24th, then the 39 pages beginning with page number 127.4, then the last 2
pages of the file.
Table Option Output
The -table option prints a table of information about selected pages in a file. This
information includes the physical page number and the actual page-ida Any "changed"
pages are marked with the word "CHANGED". Front pages are flagged with the word
"FRONT". Intentionally blank pages are flagged with the word "BLANK" and missing
pages in the front/back sequence are indicated.
The format of this table is such that it can be used as a control file, either directly or
after editing to remove any unwanted data. For example:
3-34
AZ98-02
process_compout (peo)
pco flow_sheet -to !7 -table
-pathname flow sheet -pages
FRONT
!1
1*
!2
1*
!3
!4
!5
!6
!7
FRONT
1*
1* BLANK
FRONT
1*
blank
back
not
1*
FRONT
1*
1*
1-5
1-6
*1
CHANGED
CHANGED
CHANGED
1-G.1
1-6.2
1-7 *1
suppl i ed
2-1 *1
2-2 *1
*1
*1
*1
*1
Magnetic Tape Header Files
The first file on a magnetic tape (noted as file 0 in the "display" output) is the
ASCII-coded, evaluated comment information for all the document files on the tape.
This information is obtained from the device description table and is created by use of
the Comment global statement in the device description file (see "Device Table
Compiler" in Appendix C and the "comment" example listed under the control
argument "-mode" above.) The comment may contain any information, but the
following lines are important to process_com pout
Note:
In the discussion below, xxx denotes a dynamic string that may have
various-vall.ies --dependihgofi its context andlhetise-ofopehihg and closing
brackets ([]) in the string indicates that· the enclosure is evaluated as an
active function and is replaced with the string returned.
mode: name=x x x
defines an output mode. name, whose attachment is xxx. name is the mode
name given with the -mode control argument The first mode present is the
default mode. If any others are given. they represent alternatives available. A
mode defines a method of transcribing the formatted output onto some output
medium.
xxx may be tape_ibm_ or tape_ansi_, each with an attach description
acceptable to iox_ (see Multics Subroutines), or either of the following:
online
output is written to the user_output I/O switch. This is the default if no
modes are given.
punch 6
writes 6-level TTS (or reverse TIS) code to the user_output I/O switch.
assuming that the user has an appropriate paper tape punch on the
termina1.
Note:
An ASCII Standard 7-1evel paper tape punch may be used if
the upper two punch channels are disabled.
leader: xxx
x x x is punched out. in "big" letters on the paper tape leader after the file
identification. Its purpose is to convey machine setup information to the
typesetter operator when paper tape is the output medium. This information
has no effect on magnetic tape output
3-35
AZ98-02
process......compout (pco) ..
setup: xxx
all or part of the information necessary to process a magnetic tape on the
target machine. This information may include machine setup and disposition of
the tape. This keyword line may be given any number of times and the
information appears only once per tape.
file: xxx
all or part of the information necessary for the processing of each file on a
magnetic tape. This information may include the handling and disposition of
the generated output from each file. This keyword line may be given any
num ber of times and the information is repeated for each file.
content_file: {t\} seg, {I'.} tape
controls the disposition of the "contents" file as created from the comment
information discussed above. The parameters are:
seg,"seg
do/do not create the segment vo/-id.contents in the working directory.
vo/- id is the magnetic tape volume identifier given with the -volume
control argument. The segment cannot be created if no identifier is given.
tape. "tape
do/do not write the contents file as the first file on the tape.
The default is "seg,"tape" if this keyword line is not given.
pack: spec {, I; spec} ...
controls the format of the data bytes written to magnetic tape. Ordinarily. data
bytes are written as normal ASCII or EBCDIC characters (depending on the
tape attachment). However, because of the front-end processors on some target
machines, it may be necessary to use a non-standard recording mode.
spec
may be any of the following "micro-ops":
c n
copy the next n bits of the current input byte (in the compout file) to
the output data byte.
f n
move forward n bits in the current input byte (in the compout file).
b n
move backward n bits in the current input byte (in the compout file).
'b ... '
a literal bit string of any desired length to be added to the output data
byte. (The single quote is used here because the Comment: statement
requires the use of the double quote.)
micro-op separating delimiter implying use of the current input data
byte (in the compout file) for the next micro-op.
micro-op separating delimiter implying use of the next input data byte
(in the compout file) for the next micro-op.
3-36
AZ98-02
process_conipout (pco)
In the above. n is a single decimal digit and it is an error to give a value of
n or a literal bit string that results in a bit position outside the limits of either
the input or output data bytes.
Examples:
(The compout file always contains one output character per 9-bit byte
stored as the low-order bits.)
ITS code is to be written to 7-track tape with the third tape channel set
to binary O.
f3.c2. '0' ,c4
ITS code is to be written on 9-track tape in packed mode (3 characters
per 2 tape bytes).
f3,c6.f3,c3;f3,c3.c6
[compout]
when used in a file: keyword line, this active function is replaced with the
name of the current compout file with the suffix removed; it is an error to
use it anywhere else.
[compask question {responses}]
this active function is replaced with the response from the user to the given
question. (See the comment and display examples listed under the control
argument "-mode" above.)
question
any question or prompt string. It is written to user_output and processing
waits for a response from the user. Any existing response to the question
is shown in parentheses. The responses to all questions are initialized to
empty strings when the processing of the comment information begins.
responses
the valid responses to the question, including any translation of responses to
other strings. There may be any number of valid responses and, if none
are given, any response is used as typed. Any invalid response causes the
question to be repeated. Ref erring specifically to the "Pasteup?" question in
the comment example listed under the control argument -mode above, only
"yes" and "no" are valid with a "no" response being translated to an empty
string and a "yes" response being translated to "JP....sted up", For th.e
"original/copy?" question in the same file: keyword line, four responses are
valid with the two short forms being translated to the long forms.
If only a newline (NL) is typed, the existing response is used.
If a period (.) is typed, the existing response is reset to an empty string
and is used.
For example (all on one line in a cornpout comment string):
fi le:Return [compask 1111 Pasteup? 1111 no= IIl1yes= pasted Upllll]
[compask IIlIor i gina 1/copy? 1111 o=or i gina 1 or i gina 1 c=copy copy].
If two files are being processed, it causes the interaction:
Pas teup?
0
yes
3-37
AZ98-02
process_com pout (pco)
original/copy ()
0
and generates the following lines:
Return pasted up copy.
Return original.
Note that the pasteup question returns a null result if the answer is no and
the string "pasted up" if the answer is yes. The yes response must be
quoted because it contains SPs. The other question allows the user to reply
"0" instead of having to type "original", yet the result is the more
meaningful whole word.
Display Mode Interpretations
The meanings of the coded symbols that may be found in the display mode output
are defined below.
Each symbol may be followed by a parenthesized word (or words) that indicates the
devices in whose output the symbols may be found. If no word is shown, then the
symbol may appear in the display output for all devices. The words used are:
type
print
diablo
photo
hyterm
d tc300s
monospace typewriter terminals
line printers
Diablo-like incremental terminals
phototypesetters
Diablo 1620 and 1720 series terminals
DTC300 series terminals
A circumflex (1\) appearing before the word is a logical "not" and means "all except".
All Honeywell-supplied display routines validate the data being displayed. Whenever
a string is encountered that meets certain criteria that make it appear to be a string
with known syntax, it is validated against that suspected known syntax. If any errors
are discovered in the validation, one or more indented error messages are inserted into
the display; the first such error message being flagged with three asterisks "***".
Normal display output resumes on the next line starting with the character that
triggered the validation check. When the interpretation of the entire display input string
(as little as a single word or as much as an entire output page) is complete and errors
have occurred, an error count message is given. For example (from a file for a
Mergenthaler V-I-P typesetter):
=al=plO=f120=14500
=11500=m630.This is an example for use with =a3pco
=aldocumentation. It is doctored up after
composition to cause errors for the
purpose of the display.
=m570.
2nd char SHIFTed
3rd char SHIFTed
=1 LOOOO=14500=po6-=d-
=i f090
***
=s
3-38
AZ98-02
process_compout (pco)
***
Total errors: 1
, ("photo)
a backspace (ASCII code 010) or a string of n backspaces
("photo)
a carriage return (ASCII code 015) - return to left margin without a line advance
(diablo)
a direct or absolute tab to position n
an "orphan" escape character - the character following is not in any escape
sequence known to the display routine
(print)
a formfeed (ASCII code 014) - a page eject
(diablo)
set horizontal motion increment (space implied by space and backspace) depends on device resolution
n
(hyterm)
special form of horizontal motion increment for the hyterm device
("photo)
a horizontal tab (ASCII code 011)
Qoo (diablo)
a linefeed (ASCII code 012) at the end of a text line - does not imply a return to
the left margin
, (diablo)
on a line by themselves, a linefeed or a string of n linefeeds - an indication of
white lines on the page
... (type, print)
a newline (ASCII code 012) at the end of a text line - does imply a return to the
left margin
, (type, print)
on a line by themselves, a newline or a string of n newlines - an indication of
white lines on the page
<000>
octal value of any non-printing character not known to the display routine as a
control character
, (dtc300s)
set plot character - first form if character is printable, second if not
, , <"PLT> (diablo)
"plot". "superplot", and "un plot" . respectively - appear only in error messages (see
"Plot Strings" below)
(diablo)
a plot string - everything between the brackets is an interpreted form of the plot
control data (see "Plot Strings" below)
(hyterm)
a plot string terminated by a carriage return (see "Plot Strings" below)
3-39
AZ98-02
process_compout (pco)
(dtc300s)
a superplot string - everything between the braces is an interpreted form of the
superplot control data (see "Plot Strings" below)
(diablo)
set vertical motion increment (space implied by linefeed) - n depends on device
resolution
< .•. >
any control function name unique to a particular device. Among these are "QL"
and "STOP" shared by almost all typesetters, "EL" (for elevate) for Mergenthaler
V-I-P typesetters, and "CUT" for most phototypesetters.
PLOT ST RINGS
The symbols appearing in plot strings (see "PLT" symbols above) depend on the
output device and the plot mode in which it is currently operating. The Honeywell-supplied
display routines support two plot modes: "plot" and "superplot" as defined below.
Plot Modes
plot mode
the horizontal and vertical increments are automatically set (by the device
hardware) to the minimum (or defined) values. Plot strings are given as
sequences of the four motion characters (space~ backspace, linefeed, reverse-linefeed)
and any selected plot character.
superplot mode
a plot character is specified, possible "pen control" is given. and the horizontal
and vertical motion increments are chosen. Plot strings are given as sequences
of two defined characters (usually "X" and "Y") with the plot character being
printed automatically after one of them. In most devices, both the plot
character and the increments may be changed without leaving and reentering
plot mode.
Plot Mode Symbols
"c ... "
the literal text string c...
M, n(M)
one or n occurrences of a "motion" sequence M that contains (in order) a
horizontal motion, a vertical motion, and the plot character. Any two of the
three parts may be absent. The motion directions are those of the print head
relative to the paper.
The horizontal motion may be:
r
1
HT
one increment to the righ t
one increment to the left
ten columns to the right
The vertical motion may be:
3-40
AZ98-02
process_compout (pco)
d
u
d/2
u/2
one increment down
one increment up
a half-line down
a half-line up
Superplot Mode Symbols
Superplot strings contain control sequences and motion sequences. The sequences
may contain two or more characters and are separated by a colon (:).
Control Sequences
A control sequence always appears first in the plot string and contains a pen
control character and a quadrant selector.
The pen control character may be:
d
u
pen down or print
pen up or no print
The quadrant selector may be:
1
2
3
4
11
22
33
44
upper
upper
lower
lower
upper
upper
lower
lower
right
left
left
right
right, doubled
left, doubled
left, doubled
right, doubled
"doubled" means that the values given in the motion sequences following are
doubled; that is, the superplot strings {d 1 : x2} and {d 11 : xl} are equivalent
Motion Sequences
M, n(M}
one or n occurrences of a "motion" sequence M that contains (in order) a
horizontal motion and a vertical motion. Either of the parts may be
absent. The motion directions are determined by the quadrant chosen in
the control sequence.
The horizontal motion is:
xn
n increments
The vertical motion is:
yn
n increments
Extraneous Characters
Any characters appearing in superplot strings that are not known to the display
routine as superplot characters are displayed as c if they are printing characters
or 000 if they are not.
3-41
AZ98-02
SECTION 4
WORDPRO DICTIONARIES
DICTIONARY USE
WORDPRO dictionaries are used to perform hyphenation and detect spelling errors
and inconsistencies in word usage. A standard WORDPRO dictionary is provided as part
of the WORDPRO system, and is easily located by any user. However, the standard
dictionary can be augmented or replaced by the addition of user-supplied dictionaries to
the dictionary search list used by the WORDPRO system. Search lists provide a quick
way of letting Multics know where to find needed segments, (i.e., dictionaries).
Throughout this section. this list of dictionaries is ref erred to as the diet search list.
The search facility commands contain information which is unnecessary for users
performing the standard dictionary-related functions. (The search facility commands are
fully described in the Multics Commands.)
Standard WORDPRO Dictionary
The initial version of the standard WORDPRO dictionary contains about 30.000
words.
Note:
The standard WORDPRO dictionary is at the present time a preliminary
version, to be updated and revised at a future date.
Hyphenation points are selectively compiled from the most commonly used method
of dividing words. since published dictionaries do vary. Syllables are specified for all
words that are able to be hyphenated except homographs (i.e., words with identical
spellings but different meanings and often different pronunciation as in the verb resume
(to begin again) and the noun resume (a summing up». Homographs are excluded since
the differences in pronunciation affect hyphenation. In general, no technical, scientific,
medical, foreign language, or other specialized words are included in this dictionary.
User - Supplied Dictionaries
Commands are provided that enable the user to create and maintain personal or
private dictionaries (see "Summary of Commands" below). For each word entered into a
dictionary, the user can specify the desired hyphenation points and whether the word
should be trimmed (see "Unwanted Words" below for an explanation of trimming and
the no-trim attribute).
In order for the commands described in this chapter to refer to a user-supplied
dictionary it is necessary to include the dictionary in the dictionary search list. To do
this, it is necessary to invoke either the add_search_paths or the set_search_paths
command (refer to Multics Commands). Because setting the dictionary search list is
only effective for the life of a process. it is beneficial to include the command in the
user's start_up.ec.
4-1
AZ98-02
Dictionary Files
Dictionaries are stored in dictionary files. A dictionary file can be recognized by
its entryname, which always includes the suffix dict (e.g., new_words.dict). Dictionary
files are not suitable for printing. The list_dict_words command lists the contents of a
diCtionary file.
HYPHENATION
Hyphenation is simply the division of a single word into two parts separated by a
hyphen. The Multics WORDPRO compose command can perform hyphenation automatically
for the user whenever a word occurring at the end of a line does not fit in the space
remaining on that line.
When Hyphenation Is Needed
The process of composing text often makes hyphenation necessary in order to avoid
unsightly output. If the user has requested justification of text on a line, the padding
of the line with blanks sometimes generates sparsely-filled lines containing an
unacceptable amount of blank space. If the user has not requested justification, then
the hand margin often becomes very ragged.
Since hyphenation allows more characters to be put on a line, it can help solve
both of these problems. Adding more characters to a line reduces the text padding
needed when justification is performed. When justification is not performed, hyphenation
results in smoother right-hand margins.
Hyphenation Problems Solved by WORDPRO
The many problems of hyphenation are solved by WORDPRO through the
cooperation of the compose command and the WORDPRO dictionaries. The main
hyphenation task of the compose command is to identify when hyphenation is required
and then to obtain the correct hyphenation from the dictionaries. The compose
command also provides the following additional hyphenation capabilities:
•
the ability to explicitly turn hyphenation on or off within a document
•
the ability to determine the number of blank spaces on a line after filling with
complete words, and then whether extra spaces (padding) or hyphenation of a word
at the end of that line is suitable
WORDPRO Hyphenation Technique
WORDPRO uses a multiple-dictionary hyphenation technique; i.e., it searches
through more than one dictionary looking for the correct hyphenation point for a word.
A user may specify an ordered list of dictionaries to be searched to ascertain the
hyphenation points of words requiring hyphenation. This list of dictionaries is specified
in a special list called the diet search list. The default dict search list specifies one
dictionary, the standard WORDPRO dictionary (>unb>standard.dict).
An important point about the multiple-dictionary technique is that the dictionaries
are searched in order. If the word being hyphenated is not found in the first
4-2
AZ98-02
dictionary searched. then the next dictionary in the list is searched. When a word is
found in a dictionary, it is hyphenated as specified in that dictionary. No further
searching for that word is performed. If a word is not found in any of the specified
dictionaries, then that word is not hyphenated. It is by this method that users can
specify their preferred hyphenation and have it override any other division.
The multiple dictionary technique offers the WORDPRO user viable solutions to the
problems of hyphenation. A list of the most important advantages of this technique is
presented below:
•
Hyphenation is automatic (Le.. if turned on in the compose invocation). No
terminal operator interaction is required. In situations where output is directed to a
file, or the composing of a document is performed by an absentee process, any
requirement for online human interaction is unacceptable. It is also unlikely that
any terminal operator can quickly and consistently make accurate hyphenation
decisions.
•
No complicated hyphenation rules are used. It is easy to understand how
WORDPRO performs hyphenation. As stated above, a word is hyphenated as
specified in the first dictionary in which it is found.
•
Each user has complete control over hyphenation. Any Multics user may specify a
private set of WORDPRO dictionaries. There are two reasons why this attribute is
valuable for users. The first is that many published dictionaries disagree on
preferred hyphenation. Second, different users have different preferences in
hyphenation, especially how to hyphenate words that are already hyphenated (e.g.,
non-European -- should it be hyphenated only at its true point of hyphenation,
any Jlyphenationpoint, or not at all?). By adding a private dictionary to the
beginning of the diet search list, a user may specify how words contained in that
dictionary are hyphenated. If a user does not like the hyphenation specified for a
word in the standard WORDPRO dictionary, then that word may be added to a
private dictionary and hyphenated differently or not at all. If a user wants a word
hyphenated that is not contained in the standard WORDPRO dictionary, then that
word may be added to a private dictionary.
•
The problem of homographs can be solved. In the standard WORDPRO dictionary
no hyphenation is specified for homographs. The reason is that it is better not to
hyphenate a word than to hyphenate it incorrectly. A user who wants a particular
instance of a homograph hyphenated can specify the hyphenation within the
document, the only place that the correct hyphenation is known. This can be done
by using a compose control to make external calls to the search list and dictionary
commands. For example, calls can be made to add a temporary dictionary to the
dict search list, to add the homograph to this dictionary, and then to delete the
temporary dictionary from the diet search list after the homograph is processed by
compose.
•
The multiple dictionary technique fits the needs of all users. For the average user
who does not demand specialized vocabulary, the standard WORDPRO dictionary
alone is sufficient. No knowledge of dictionaries or search iists is required. Other
users may have a second or third dictionary set up for use each time they log in
and thus they also need not be concerned with how this technique works. However,
other users may require absolute and dynamic control over hyphenation. For these
users, absolute and dynamic control is available.
4-3
AZ98-02
SPELLING ERRORS
WORDPRO provides tools that help detect and correct spelling errors or inconsistencies.
The ability to eliminate misspelled, mistyped, and unwanted words from a document
provides WORDPRO users with the means to consistently produce quality documents.
Spelling Error Detection
The WORDPRO technique used to detect spelling errors consists of three steps:
1.
Make an alphabetized list of all unique words contained in a document This
list is called a word list.
2.
Remove from the wordlist all words that can be found in a dictionary or set
of dictionaries. This operation is called trimming.
3.
Print the remaining words in the word list and check them for spelling errors.
The words printed in step 3 are those words contained in a document that could
not be found in -a dictionary and thus are likely to be misspelled. Normally, for a
document containing perhaps thousands of words, only a small number of these words
need to be checked. This makes spelling error detection fast, simple, and thorough. The
search for a word in a dictionary (step 2 above) is performed in the same way as
described for hyphenation (multiple dictionaries can be used). As with hyphenation. the
list of dictionaries to be used is specified in the dict search list. Thus the default
dictionary used for spelling error detection is the same standard WORDPRO dictionary
used for hyphenation.
A separate command is provided to perform each of the three steps. However,
these three commands can be combined into a single operation, if desired, by use of
the Multics exec_com facility (described in Mu/tics Commands).
Unwanted Words
In addition to the need to identify misspelled words in a document, there is the
need to identify unwanted words. An unwanted word may be some normally-acceptable,
correctly-spelled word that should not appear in a particular document (at least not
without a careful check of the context in which the word is used).
An example of an unwanted word is the word "basic" within Honeywell Multics
documentation. This word. when used in computer documentation, may be confused
with the command used to invoke the BASIC compiler. Thus every instance of this
word in a Honeywell Multics document must be identified and approved.
The problem with unwanted words is that they are very likely contained in the
standard WORDPRO dictionary and thus are deleted from any list of misspelled words.
It is not acceptable to solve this problem by requiring a user to maintain a private
dictionary containing all of the words in the standard WORDPRO dictionary except the
few words that" are unwanted.
Instead, it is possible to add a word to a dictionary (presumably a private
dictionary that is searched before the standard dictionary) and to specify that this word
is unwanted and therefore should not be trimmed (deleted) from any list of words by
the trim_wordlist command. This no-trim attribute is denoted by preceding the word
with the circumflex character "/\" (ASCII code 136). If a word is actually spelled with
4-4
AZ98-02
a leading circumflex, then that circumflex must be followed by another circumflex or
an equal sign.
The sequence "1\1\" preceding a word indicates that the word contains a leading
circumflex and has the no-trim attribute. The sequence "1\=" preceding a word indicates
that the word contains a leading circumflex and does not have the no-trim attribute.
Wordlist Segments
Wordlists are stored in word list segments. A wordlist segment can be recognized by
its entryname, which always includes the suffix wI (e.g., new_words. wI). A wordlist
segment contains a sequence of words each separated by a newline character: thus, if a
wordlist segment is printed, each line contains exactly one word. A command is
provided to print a wordlist in a multiple-column format (see the print_wordlist
command below).
Spelling Error Correction
When a misspelled word is detected as described above, it can be easily corrected
using WORDPRO tools. The revise_words command is provided to revise all instances
of specified misspellings within a document. It is not necessary for the user to locate
the misspellings within the document or to know how many times each misspelling
occurs.
For most errors in spelling or consistency, the intended word can be recognized
from the wordlist In some cases, however. the intended word may not be recognizable
without examining the context For this purpose, the locate_words command is provided
to locate all occurrences of a given misspelling or unwanted word within a document
and to print all lines in which these words appear. For each occurrence. it is possible
to print not only the containing line, but surrounding lines as well.
4-5
AZ98-02
The add_dict_words command is used to add words to a WORDPRO dictionary.
SYNTAX AS A COMMAND
adw path {words} {-control_args}
ARGUMENTS
path
is the pathname of the dictionary to which the words are added. If path does
not have the suffix dict. one is assumed; however, dict must be the last
component of the dictionary segment name. If the dictionary does not exist, it
is created.
words
are words to add to the dictionary. At least one word is required unless
-input_file is specified. If a word is already in the dictionary with the same
hyphenation and no-trim attribute, the word is ignored without comment (see
"Notes" below). However, if the word is already in the dictionary with
different hyphenation or no-trim attribute, then a warning i~ issued and the
dictionary word is left unchanged.
CONTROL ARGUMENTS
-count
-ct
reports the number of words added and the total number of words in the
dictionary.
-force
-fc
allows a word already in the dictionary to be replaced. This feature may be
used to change the no-trim attribute or hyphenation of a word in the
dictionary.
-input_file path
-if path
adds to the dictionary words contained in the segment specified by path.
Words in this segment must be separated by newlines. This control argument
may be specified more than once.
-raw
suppresses the special interpretation otherwise given to hyphen and circumflex
characters (see "Notes" beloW).
-word string
adds the word str; ng to the dictionary even though string may look like a
control argument.
4-6
AZ98-02
NOTES
The correct hyphenation of a word can be specified when it is added to the
dictionary. Embedded hyphens indicate the hyphenation points. If no hyphenation
points are specified. it is assumed that the word cannot be hyphenated. If a word is
spelled with a hyphen. then that hyphen must be followed by another hyphen or an
equal sign. The character sequence "-" indicates that the word contains a hyphen and
that hyphenation may be performed at (after) the hyphen. The character sequence "-="
indicates that the word contains a hyphen. but the word may not be hyphenated at the
hyphen.
If the -raw control argument is specified, no special interpretation is given to
either hyphen or circumflex characters. Eaeh sueh character found within a word is
taken literally as a part of the word. Therefore, words added with the -raw control
argument cannot have the no-trim attribute or hyphenation points.
Maximum word size is limited to 256 "literal characters". Only characters contained
in the normal spelling of a word are literal characters. Thus, the special sequences "--"
and "-=" both represent the single literal character n_n. Literal hyphens may appear
anywhere within a word. Hyphenation points, however. may not appear beyond the 33rd
literal character of a word.
EXAMPLES
To add the word test to the dictionary good_words. diet in the working directory (it
is not hyphenated), type:
adw good_words test
To add the word example to the dictionary Webster. diet in the user's project
directory (it is hyphenated at the specified hyphenation points), type:
adw >udd>Project=id>Webster ex-am-ple
To add the word basic to the dictionary my_words. diet (it is hyphenated, but it is
not trimmed), type:
adw >my_words Abas-ic
To add the word i n- house to the dictionary good_ words. diet (it is hyphenated at
the point of its aetual hyphen). type:
adw good_words in--house
To add the word co-star to the dictionary good_words. diet (it is not hyphenated),
type:
adw good_words co-star -raw
To add the word
hyphenated), type:
right-hand
to
the dictionary good_words. diet (it is not
adw good_words right-=hand
To add all words in the segment new_words to the dictionary good_words.diet,
type:
adw good_words -input_fi Ie new words
4-7
AZ98-02
The count_diet_words command prints the number of words contained in a
specified dictionary.
SYNTAX AS A COMMAND
cdw path
ARGUMENTS
path
is the pathname the dictionary. If path does not have the suffix diet, one is
assumed; however, diet must be the last component of the dictionary segment
name.
4-8
AZ98-02
create_wordlist (cwO
create_wordlist (cwO
The create_word list command produces an alphabetized list of an distinct words
found in the specified text segment This list is saved in a wordlist segment that. is
created in the working directory. The wordlist segment is given the entryname of the
text segment with a suffix of wI added. The total number of words in the text segment
and the number of words put into the wordlist segment are displayed.
SYNTAX AS A COMMAND
cwl path {-control_args}
ARGUMENTS
where path is the pathname of the text segment.
CONTROL ARGUMENTS
-brief
-bf
suppresses the display of the total number of words in the text segment and
the number of words put into the wordlist segment.
-from n
-fm n
words are processed in the text segment starting from the line number specified
by n. If this control argument is not specified. then the text segment is
processed starting from the first line.
-header
-he
displays the pathname of the text segment.
-no_con trol_lines
-ncl
suppresses the display of the control lines (i.e.• lines that begin with a period).
-no_exclude
-ne
specifies that words containing only special characters or punctuation are not to
be excluded from the wordlist (see "Notes" below).
-no_sort
-ns
specifies that the words in the word list segment are not to be sorted into
alphabetical order. They are put into the wordlist segment in the order in
which they are found in the text segment and duplications are not eliminated.
(This control argument is intended for special application and should not be
used for normal word list segment creation.)
-to n
words are processed in the text segment up to and including the line number
specified by n. If this control argument is not specified, then the text segment
is processed to the last line.
4-9
AZ98-Q2
create_wordlist (cwO
NOTES
Words in the text segment are separated by the following delimiter (white space)
characters:
space
horizontal tab
vertical tab
newline
form feed
Punctuation characters are removed from the word. The characters n( [{ are
removed from the I eft side of the word. The characters U)]} .,;:?! are removed from
the right side of the word. Also, PAD characters (octal 177) are removed from the
I eft side of the word.
Additional special processing is performed on each word after all punctuation is
removed. A summary of this special processing is given below:
•
if the entire word is underscored, then the underscores are removed. If only
part of a word is underscored, then the underscores remain.
•
if the word contains no letters, i.e., consists entirely of punctuation characters
or other special characters, then the word is excluded from the wordlist The
-no_exclude control argument disables the automatic exclusion of such words.
EXAMPLES
The table below shows examples of how punctuation is removed from a word and
how special processing is performed.
WORD BEFORE
PROCESSING
example
"example"
example. )
{example}
example
exam.ple
)example(
1000
1,000
7-5.2
+1
1/2
1A
1( 2)
WORD IN
WORDLISf
example
example
example
example
example
exam.ple
)example(
(trimmed)
(trimmed)
(trimmed)
(trimmed)
(trimmed)
1A
(trimmed)
4-10
AZ98-02
delete_diet_words (ddw)
The delete_diet_words command deletes one or more words from a WORDPRO
dictionary.
SYNTAX AS A COMMAND
ddw path {words} {-contro'_args}
ARGUMENTS
path
is the pathname of the dictionary. If path does not have the suffix diet, one
is assumed; however, diet must be the last component of the dictionary segment
name.
words
are words to be deleted from the dictionary. At least one word is required
unless -input_file is specified (see below). If a word is not found in the
dictionary. a warning message is issued.
CONTROL ARGUMENTS
-brief
-bf
suppresses the warning message usually given when a word is not found in the
dictionary.
-count
-ct
reports the number of words deleted and the number of words in the
dictionary.
-input_file path
-if path
deletes from the dictionary the words contained in the segment specified by
path. Words in this segment should be separated by newlines. This control
argument may be specified more than once.
-word st ring
deletes the word str; ng from the dictionary even though str; ng may look like
a control argument.
NOTES
A word to be deleted from the dictionary must be spelled in its raw form, i.e.,
without indicating hyphenation points or the no-trim attribute (see add_dict_words
command above).
4-11
AZ98-02
delete_dict_words (ddw)
EXAMPLES
To delete the word example from the dictionary Webster.dict in the Project_id
directory. type:
ddw >udd>Project_id>Webster example
To delete the word basic from the dictionary my_words. diet in the current working
directory, type:
ddw my_words basic
To delete the words test, in-house, and Multics from the dictionary good_words.dict.
type:
ddw good_words test in-house Multics
4-12
AZ98-02
The find_diet_words command finds and displays words contained in the sequence
of dictionaries defined by the diet search list.
SYNTAX AS A COMMAND
fdw {words} {-control_args}
ARGUMENTS
words
are words to be found. At least one word must
-input_file control argument is specified.
be given unless the
CONTROL ARGUMENTS
-brief
-bf
suppresses the warning message usually given when a word is not found.
-dictionary
-diet
displays the pathname of the dictionary in which the word waS found.
-exact_match
-exm
finds only those words that match a dictionary word exactly, i.e., no special
processing is performed with respect to capitalization (see "Note" below).
-input_file path
-if path
finds words in the segment specified by path. Words in this segment must be
separated by newlines. This control argument may be specified more than once.
-output_file path
-of path
writes words found into the segment specified by path instead of displaying
words on the user's terminal. Words are separated by newlines in the output
segment.
-raw
displays the words without indicating the no-trim attribute or hyphenation
points (see add_diet_words command above). Otherwise, words are printed in
the format accepted by add_diet_words.
-word string
finds the word string even though string may look like a control argument.
NOTES
When searching for a word in a dictionary, special processing of capital letters is
performed unless the -exact_match control argument is specified. This special processing
is identical to that performed by the trim_wordlist command below.
4-13
AZ9S-02
hyphenate_word_
The hyphenate_word_ subroutine returns the character position at which a word can
be hyphenated. The word is located in a dictionary via the diet search list.
USAGE SYNTAX
declare hyphenate_word_ entry (char(*), fixed bin,
fixed bin, fixed bin(35»;
call hyphenate_word_ (string, space, break, code);
-QRdeclare hyphenate_word_ entry (char( *), fixed bin, fixed bin);
call hyphenate_word_ (string, space, break);
ARGUMENTS
stri ng (input)
the text word that is to be split.
space (input)
the number of print positions remaining in the line.
break (output)
the number of characters from the word that should be placed on the current
line; it should be at least one less than the value of space (to allow for the
hyphen), and can be 0 to specify that the word is not to be broken. Thus if
the word "calling" is to be split, and six spaces remain in the line, the
procedure should return the value 4 (adjustment is performed after hyphenation).
code (output)
a standard status code. In order to retain compatibility with an older version
of this subroutine, this argument is optional, depending upon how hyphenate_word_
is declared in the calling program. If this subroutine is called with only three
argumen ts, then no code is returned.
4-14
AZ98-02
The list_dict_words command displays a list of words in a WORDPRO dictionary ~
SYNTAX AS A COMMAND
ldw path {words} {-control_args}
ARGUMENTS
path
is the pathname of the dictionary to be listed. If path does not have the
suffix dict, one is assumed; however, dict must be the last component of the
dictionary segment name.
words
are words to be listed. If no words are specified, and if the -input_file
control argument is not specified, all words in the dictionary are listed.
CONTROL ARGUMENTS
-brief
-bf
suppresses the warning message usually given when a word is not found" in the
dictionary.
-input_file path
-if path
lists the words contained in the segment specified by path. Words in this
segment should be separated by newlines. This control argument may be
specif ied more than once.
-output_file path
-of path
writes words to be listed into the segment specified by path instead of printing
words on the user's terminal. The words are separated by newlines in the
output segment.
-Taw
displays the words without indicating the no-trim attribute or hyphenation
points. Otherwise, words are listed in the format accepted by the add_dict_words
command above.
-word string
lists the word string even though string may look like a control argument
4-15
AZ98-02
NOTES
When listing an entire dictionary, or any large number of dictionary
be convenient to use the list_dict_words command together with the
command to obtain multiple column output. This is accomplished
-output_file control argument to create a wordlist (Le., a segment whose
the suffix wI). The resulting word list can then be printed by the
command.
words. it may
print_wordlist
by using the
entryname has
print_wordlist
For example, the following command sequence displays the dictionary English.dict:
1 ist_diet_words English -output_file diet_words.wl
print_wordlist diet_words
In the above example, all words in English.dict are displayed in ASCII collating
order (Le., the order in which they are stored in the dictionary). This ordering is
different from the alphabetical ordering used for wordlists. However, the create_wordlist
command can be used to alphabetize dictionary words. For example, the following
command sequence displays the dictionary English.dict in alphabetical order. The
dictionary words are in raw format.
list_diet_words English -raw -output_file raw_words
create_word 1 ist raw_words
print_wordlist raw_words
If raw format is not desired, the command sequence below can be used:
ldw English -raw -of raw_words
cwl raw_words
ldw English -if raw_words.wl -of diet_words.wl
print_wordl ist diet words
4-16
AZ98-Q2
locate_words Ow)
locate_words Ow)
The locate_words
specified text segment
each occurrence of a
containing the word is
command locates all occurrences of a given word within a
The user can specify more than one word to be located. For
given word within the text segment, the number of the lines
displayed.
SYNTAX AS A COMMAND
lw path words {-contro'_args}
ARGUMENTS
path
is the pathname of the text segment
words
are words to be located in the text segment
CONTROL ARGUMENTS
-count
-ct
displays only the number of occurrences for each word.
-from n
-fm n
the text segment is searched starting from the line number specified by n. If
this control argument is not specified. the text segment is searched starting
f rom the first line,
-header
-he
displays the pathname of the text segment
-iines {n}
-Ii {n}
for each occurrence of a given word, the lines (and line numbers) starting n
lines before, through 17 lines after the line containing the word are displayed.
Thus, if n is 1, three lines are displayed. If n is not specified, only the line
containing the word is displayed (Default).
-long
-lg
for each occurrence of a given word, the line (and line number) of that word
is displayed.
-to n
the text segment is searched up to and including the line number specified by
n. If this control argument is not specified. the text segment is searched to the
last line.
-word string
locates the word string even though string may look like a control argument
4-17
AZ98-o2
locate_words Ow)
NOTES
The -count control argument is mutually exclusive with the -long and -lines control
arguments.
Words are found in the text segment in the same way as described for the
create_wordlist command. Words containing no letters can be found even though they
are normally excluded from a wordlist.
4-18
AZ98-02
print_word list (pwI)
print_wordlist (pwI)
The print_wordlist command displays (prints) the words contained in a word list
segment in a multiple column format (see the create_wordlist command above).
SYNTAX AS A COMMAND
pwl path {-contro'_ args}
ARGUMENTS
path
is the pathname of a word list segment. If path does not have the suffix wI,
one is assumed; however, wI must be the last component of the segment name.
CONTROL ARGUMENTS
-columns n
-cols n
specifies that the output is to contain n columns. The default number of
columns depends on the line length and the column width (see "Notes" below).
-column_width n
-cw n
specifies that the column width is n characters. The default column width is
20.
-output_file path
-of path
directs the output to the segment specified by path in a format suitable jor
printing on a line printer.
-page_length n
-pI n
specifies that the page length is n lines. The default page length is 60 if
-output_file is specified; otherwise, it is 66.
-vertical_margin n
-vm n
specifies that the vertical margin size is n lines. The default vertical margin
size is 0 if -output_file is specified; otherwise, it is 3.
NOTES
The default number of columns is the maximum number of columns that fit within
the line length. If the -output_file control argument is specified, a line length of 136
is assumed. Otherwise, the line length defined for the user_output switch is used. If
none is defined, a line length of 72 is assumed.
If the length of a word is greater than or equal to the column width, the word is
truncated. An asterisk (*) is appended to such words to indicate truncation.
Output is divided into pages. Each page has a top and bottom vertical margin
consisting of n blank lines where n is the vertical margin size. These lines are included
4-19
AZ98-02
print_wordlist (pwI)
in the page length. The column height on a page is equal to the page length minus
twice the vertical margin size. In the default case, the column height equals 60 lines
whether or not the =output_file control argument is specified. On the last page of
output, the column height is reduced to the minimum height needed to accommodate
remaining words. If the -output_file control argument is specified, each page is
terminated by an ASCII new page character (octal 014).
4-20
AZ98-o2
revise_words (rw)
revise_words (rw)
The revise_words command replaces all occurrences of a given word within a
specified text segment with a new word called the revision. The user can specify more
than one word to be revised.
SYNTAX AS A COMMAND
rw path word1 rev1 .,. {wordn revn} {-control_args}
ARGUMENTS
path
is the pathname of the text segment
word;
is a word in the text segment to be revised.
revi
is the revision (i.e.. the replacement for word i).
CONTROL ARGUMENTS
-brief
-bf
suppresses the display of the number of revisions for each wordi.
-from n
-fm n
revisions are made in the text segment starting from the line number specified
by n. Ii this control argument is not specified. the text segment is processed
starting from the first line.
-header
-he
dispiays the pathname of the text segment
-lines In}
-li In}
for each reVISIon made. the lines (and line numbers) starting n lines before.
through n lines after the line containing the revision are displayed. Thus, if n
is 1, three lines are displayed. If n is not specified, only the line containing
the revision is displayed (Default).
-long
-lg
for each word revised, the line (and line number) where the revision is made is
displayed.
-to n
revisions are made in the text segment up to and including the line number
specified by n. If this control argument is not specified, the text segment is
processed to the last line.
4-21
AZ98-o2
revise_words (rw)
-word string1 string2
replaces the word string1 with the revision string2 even though string1 may
look like a control argument.
NOTES
The -brief control argument is mutually exclusive with the -long and -lines control
arguments.
Vv·ords are found in the text segment in the same way as described for the
create_wordlist command. Words containing no letters can be revised even though they
are normally excluded from a word list.
EXAMPLES
To replace the word typpoo with the word typo wherever it occurs in documentcompin.
type:
revise_words document.compin typpoo typo
If there are two occurrences of typpoo. the command displays the message:
2 revisions for "typpoo"
4-22
AZ98-02
trim_wordlist (twI)
trim_wordlist (twI)
The trim_wordlist command trims (deletes) all words in the specified wordlist
segment that are found in one or more WORDPRO dictionaries. The dictionaries may
be specified explicitly or else the dict search list is used. The trimmed wordlist segment
replaces the original wordlist segment The number of words trimmed and the number
of words remaining in the trimmed wordlist segment are displayed.
SYNTAX AS A COMMAND
twl path {dietyaths} {-eontrol_args}
ARGUMENTS
path
is the pathname of the wordlist segment to be trimmed. If path does not have
the suffix wI, one is assumed; however, wI must be the last component of the
segment name.
diet paths
are the pathnames of dictionaries to be searched in order. If diet paths does
not have a suffix of dict, one is assumed; however, dict must be the last
component of the dictionary segment name. If no dietyaths are specified, the
dictionaries in the dict search list are used.
CONTROL ARGUMENTS
-brief
-bf
suppresses the display of the number of words trimmed and the number of
words remaining in the trimmed wordlist segment
-exact_match
-exm
trims only those words that match exactly a word found in a dictionary, i.e.,
no special processing is performed with respect to capitalization (see "Notes"
below).
NOTES
For each word processed, the dictionaries are searched in the order specified or as
defined in the dict search list. Normally, when a word is found in a dictionary, it is
trimmed. However, if the word found has the no-trim attribute, then the word is not
trimmed and no more dictionaries are searched for this word.
When searching for a word in a dictionary, special processing of capital letters is
performed unless the -exact_match control argument is specified. Most words in a
dictionary consist of all lowercase letters. These words match any representations of
themselves that are either all lowercase letters, all lowercase letters with a leading
capital letter, or all capital letters. Words in a dictionary that have a leading capital
letter only match representations of themselves that have a leading capital letter or are
all capital letters. Words in a dictionary that consist of all capital letters or mixed
4-23
AZ98-02
trim_wordlist (twI)
lowercase and capital letters only match representations of themselves that have the
identical capitalization.
The table below shows examples of different ways a word in a dictionary may be
capitalized. It also shows which representations of these words match and which do not
match.
WORD
MATCH
NO MATCH
example
example
Example
EXAMPLE
ExAmple
Multics
Multics
MULTICS
multics
MULTics
WORDPRO
WORDPRO
Wordpro
Word Pro
wordpro
non-ASCII
non-ASCII
non-ascii
Non-ASCII
NON-ASCII
4-24
AZ98-02
SECTION 5
SPEEDTYPE
SPEEDTYPING
The primary goal of Speedtype is to allow users to type input data more quickly.
Speedtyping, quite simply, is the ability to type a document using the least possible
number of key-strokes. Typing speed is therefore increased since less is typed.
Speed type can also help improve typing accuracy. Typing accuracy is improved by
defining and using symbols for words or phrases that are often mistyped. For example,
the common typo teh (intended to be th~ can be corrected automatically by having
Speedtype expand the symbol teh into the. Even better, this typo can be eliminated
entirely by typing the symbol t and have Speedtype expand it into the
Speed type is quite similar to the Multics abbrev subsystem (see the description of
the abbrev command in Multics Commands) which expands command line input.
Speed type, however, can define, maintain. and list a set of abbreviations that can be
typed as input text and then expanded.
In or-der to avoid- confusion and ambiguity in terminology between Speedtype and
abbrev, the term abbreviation is not used when discussing Speedtype. Instead, the term
symbol is used. All Speedtype commands are named to conform to this terminology.
The primary job of Speedtype is to expand text. The following paragraphs describe
the features of Speedtype that are involved in the expansion process.
Text Segments
Speedtype deals with two types of files; text segments and symbol dictionaries. A
text segment contains the input text processed by Speed type. This processing involves
searching through the text segment and expanding all defined symbols. The expanded
text is copied into an output text segment.
Speedtype processes an input text segment as just one long character string. The
resulting output text segment may also be thought of as one character string. The input
string is divided into pairs of tokens. Speed type recognizes two types of tokens:
delimiter tokens and text tokens. Certain ASCII characters are designated as delimiter
characters (in general, white space and punctuation characters other than period). All
other characters are considered text characters. Speedtype divides an input string into
pairs of tokens.
Now ... country
5-1
AZ98-02
Not shown are the special cases that may exist at the beginning and end of an input
string where one of the tokens in a pair may be missing.
Speedtype also -recognizes special delimiter and text characters if they are present.
If they are found in certain positions, special processing is performed. For example:
Icountry+.
- where
- where
'" is
I is
+ is
. is
an escape character
a prefix character(s),
a suffix character, and
termination (period)
Speedtype performs special processing on the last character of a delimiter token and
on the first and last characters of a text token. This special processing is outlined
below and discussed in detail later in this section.
Escapes
Certain delimiter characters are recognized as escape characters. If the last
character of a delimiter token is an escape character, then special processing is
performed on the following text token.
Prefixes
Certain text characters are recognized as prefix characters. If a prefix character
is found at the beginning of a text token, then special processing is performed.
Recognized prefix characters are not considered part of the symbol. Prefix
characters found within the text token cause no special processing and are
considered part of the symbol. More than one prefix character may precede
the symbol.
Capitalization
If the first character of the symbol is an uppercase letter. then the first letter
.of the expansion string representing this symbol is capitalized when copied into
the output string.
Suffixes
Certain text characters are recognized as suffix characters. If the last character
of a text token (after any trailing period is removed) is a suffix character, then
special processing is performed. A recognized suffix character is not considered
part of the symbol. Suffix characters found within the text token cause no
special processing and are considered part of the symbol. Only one suffix
character may follow the symbol.
Period
If the last character of a text token is a period
then it is stripped from
the text token. The period is copied into the output string after the text token
is processed.
fl. " ,
Symbol Dictionaries
A symbol dictionary contains all of the information needed by Speed type to
expand an input string. A symbol dictionary is similar to an abbrev prof i Ie segment
(explained in the description of the abbrev command in Mu/tics Commands). A
s ymbo 1 die t i ana r y is identified by the entryname suffix,
s ymbo 1s (e.g.,
standard_words.symbols). Speedtype allows a user to specify the symbol dictionary
5-2
AZ98-02
used. As a default, Speedtype 1.:lSeS a symbol dictionary in the user's home directory.
The default symbol dictionary has the pathname:
>udd>Project>Person_id>Person_id.symbols
A symbol dictionary contains three types of information. Speedtype commands
allow a user to set, change, and list all of this information. The three types of
information are:
Options
Several types of control information are kept in a symbol dictionary. These
Speedtype options may be set by a user. (See the option_symbols command at
the end of this section for a description of the Speedtype options.) The
Speed type options are:
Delimiters (except escapes and white space)
Escape Characters
Prefix Characters
Suffix Characters
Symbols
A symbol is a character string that represents a word or phrase. A symbol
must be unique within a symbol dictionary. Since symbols are found within
text tokens, they may not contain any delimiter characters. The first character
of a symbol may not be a prefix character, and the last character of a symbol
may not be a suffix character or a period.
Expansions
Every defined symbol has a corresponding expansion string. Expansions do not
have to be unique within a symbol dictionary. An expansion may contain any
character, including delimiter characters. All suffixing, capitalization, and
underlining is performed on expansions, not on symbols. Associated with each
expansion is information that spp...cifies how Spp...edtype is to perform· suffixing
on that expansion.
Expansion Process
Speedtype uses the general expansion algorithm described above. However, Speedtype
also performs special processing. A more detailed description of how Speedtype expands
a token pair is given below:
Delimiters
Processing of the delimiter token only involves copying it into the output
string.
Esca pe Processing
If the last character of the delimiter token is an escape character, then special
processing is performed on the following text token. Escape characters
contained within the delimiter token are not recognized as escapes. The most
important type of escape processing involves inhibiting any processing of the
following text token. Instead, the text token is just copied into the output
string.
5-3
AZ98-Q2
Finding the Symbol
If no escape inhibits the processing of the text token. then the next step is to
find the symbol contained in the text token. This involves stripping off any
prefix characters. suffix character. or trailing period. If no symbol is found
within the text token (i.e.. it consists of just prefix and/or suffix characters)
then no further processing is performed on this text token and it is copied as
is into the output string.
Decapitalization
If the text token contains a symbol. then it is placed in lowercase. This
involves testing the first character of the symbol. and if it is an uppercase
letter, translating it to lowercase. This translation is actually performed on a
temporary copy of the symbol. The original input symbol is not modified.
Expansion
Speedtype then takes the lowercase symbol and searches for it in the current
symbol dictionary. If found, the expansion for this symbol is copied into the
output string, otherwise the original input symbol (and any suffix character) is
copied.
Capitalization
If the input symbol was put in lowercase and replaced by expansion, then
Speedtype capitalizes the expansion string copied into the output string. This
involves testing the first character of the expansion string, and if it is a
lowercase letter, translating it to uppercase.
Suffix Processing
If a suffix character was stripped from the symbol. and if the symbol was
expanded, then Speed type performs suffixing on the expansion string copied into
the output string. This processing depends upon the suffix and how the suffix
is defined for this symbol.
Prefix Processing
If any prefix
performs prefix
copied into the
capitalization or
characters were stripped from the symbol, then Speedtype
processing on the symbol or the expansion string which was
output string. Prefix processing is always performed after any
suffixing.
Period Processing
If a period was stripped from the symbol, then it is added to the output string
after all other processing of the text token is performed.
Escapes
The escapes recognized by Speedtype are listed below. The actual escape characters
recognized are defined in a symbol dictionary and may be set by the user. Listed with
each escape is its name and its default character. The special processing performed for
~..,~l-.
'-'O.'-'u
or,.." .....o
\,.0)\,10.
p\,.
";r-
1~
n1-=".I"\.
a.l~V
~.o.~""",~h.o..rI
"'\"0)\,111 V\,."'.
'" (temp)
The temp (temporary) escape .is the standard Speedtype escape. It causes
Speedtype to not process the following text token. Thus this escape can be used
to prevent a symbol from being expanded and can prohibit prefix processing
5-4
AZ98-02
for the next text token. Instead, the text token is copied as 1S 1nto the output
string.. The temp escape character itself is not copied into the output string.
octal 177 (pad)
The pad escape is useful in situations where an input text segment is also used
as the output text segment and is expanded over and over. The effect of this
escape is the same as that for the temp escape. However, unlike the temp
escape, this escape character is copied into the output .ur string. The default
character used for the pad escape is the pad character (ASCII code 177). Even
though this character is copied into the output string, it is not printed. Users
are cautioned that the presence of a pad character in the text segment may
cause problems during subsequent editing.
, (perm)
The perm (permanent) escape is a convenient way for a user to enter a pad
escape. The effect of this escape is the same as the temp escape, and like the
pad escape, it is copied into the output string. However, the perm escape
character is then converted to the pad escape character.
(trans)
The function of the trans (transparent) escape is to concatenate text tokens
that are processed separately. The trans escape character is not copied into the
output string. The following text token is processed as if no escape was
recognized. Any prefix processing performed on the previous text token is
continued and performed on the next text token. Additional prefix processing
may be specified.
(space)
The function of the space escape is to generate spaces (ASCII blanks) in the
output string. The processing of this escape is conditional on the first
characters of the following text token. If the following text token begins with
one or two numeric characters (numbers from 0 to 99), then the space escape
character and these numeric characters are replaced in the output string with
the specified number of spaces. For example, "; 9' is replaced by five spaces
in the output string. The rest of the text token is then processed normally. If
the following text token does not contain a number as specified above, then the
space escape character remains unchanged in the output string and the following
text token is processed as if no escape was recognized.
Suffixes
Suffix processing is performed only on defined symbols. If a symbol is not
defined, or if the specified suffix is turned off for the symbol. then no suffix
processing is performed. Instead, the symbol and the suffix character are copied as is
';.."1",,
~ut..v
tl,...o
"u~
,"",t"""llf"
C'f" ... .: .... ,...
V"'''P'''t.. .31,,11110.
Appending a suffix to a symbol's expansion string is done in several different ways
depending upon how the suffix is defined for the symbol. The normal way is to just
addition the suffix string associated with the suffix directly to the expansion string.
However, to accommodate the many anomalies of the English language, such tricks as
dropping the last letter, doubling the last letter, adding letters, etc., may be performed
on the expansion string in order to addition a suffix string.
5-5
AZ98-02
A user has considerable control over how Speedtype performs suffixing. (See the
add_symbols command at the end of this section for a description of how Speedtype
performs suffixing.) A user may disable suffixing for a given symbol, or just disable
one or more suffixes for that symbol. A user may also specify a different way to
process a suffix for a symbol.
The suffixes currently recognized by Speedtype are listed below. The actual
characters representing the suffixes are defined in a symbol dictionary and may be set
by the user. Except for plural, the suffix string associated with each suffix is the
suffix itself. Also listed with each suffix is the default character used to represent that
suffix.
Suffix Name
plural
ed
ing
er
ly
Suffix String
Default Character
s
ed
ing
er
ly
+
*
Prefixes
Prefix processing is performed on the text token string copied into the output
string. It is performed regardless of whether symbol expansion was performed, and is
always performed after capitalization and suffixing have been performed.
The prefixes recognized by Speedtype are listed below. The actual prefix characters
recognized are defined in a symbol dictionary and may be set by the user. Listed for
each prefix is its name and its default character. The special processing performed for
each prefix is also described.
{under}
The function of the under (underline) prefix is to underline the output string.
The underlining is performed by taking each character of the output string and
adding, in a canonical way, a backspace character and an underscore character.
The resulting underlined string is in canonical form. Underlining is not
performed if the output string already contains backspace characters.
(upper)
The function of the upper (uppercase) prefix is to translate the output string
into uppercase. Each lowercase letter in the output string is translated to
uppercase. Characters that are not lowercase letters are not changed. If both
the upper and under prefixes are recognized, then regardless of the order in
which they are specified, uppercase processing is performed first.
5-6
AZ98-02
add_symbols (asb)
add_symbols (asb)
The add_symbols command adds a symbol to the current symbol dictionary. All
suffixes are enabled for the added symbol.
SYNTAX AS A COMMAND
asb symbol expansion {-control_args}
ARGUMENTS
symbol
is the symbol to be added. Its length must be 7 characters or less and it may
not contain delimiter characters. Its first character may not be a defined prefix
character or a capital letter, and its last character may not be a defined suffix
character or a period.
expansion
is the expansion string that replaces the sym bol. The length of the expansion
string must not exceed 56 characters. The expansion string may contain any
characters. If the expansion string contains spaces and/or tabs, then it must be
enclosed in quotes.
CONTROL ARGUMENTS
-force
-fc
specifies that the replacement of an eXIstIng symbol should be done without
question. If the symbol is already defined, and this argument is not specified,
then the user is asked to authorize the replacement of the symbol.
-suffix string
enables or disables suffixing for this symbol. string must be either on or off.
If string is on then suffixing is enabled and all suffixes are processed
according to the default rules described in "Notes" below. If string is off,
then all suffixes are disabled for the symbol. If this control argument is not
specified, then on is assumed.
-plural string
defines the plural suffix for this symbol. string must be on or off. or a string
that can be used as the plural of the expansion of this symbol. If string is
on, then the plural suffix is enabled for this symbol and processed according to
the default rules for the plural suffix. If string is off the plural suffix is
disabled for this symbol.
-ed string
defines the ed suffix for this symbol. This control argument follows the same
rules as the -plural control argument.
-ing string
defines the ing suffix for this symbol. This control argument follows the same
rules as the -plural control argument.
5-7
AZ98-02
add_symbols (asb)
-er string
defines the er suffix for this symbol. This control argument follows the same
rules as the -plurai control argument.
-ly string
defines the ly suffix for this symbol. This control argument follows the same
rules as the -plural control argument.
NOTES
The default rule for appending a suffix string to an expansion string is a function
of the suffix and the word type of the expansion string.
The word type of the expansion string is determined from its last characters. The
characters C and Vare used below to represent consonants and vowels. The character X
is used to represent any character. The word types recognized and the suffix strings
used are:
REF. NO.
WORD TYPES
o
other (=> none of those below)
XCe
XVe
XCy
XVy
Xch, Xsh, or Xex
1
2
3
4
5
6
cve
REF. NO.
SUFFIX STRINGS
1
s (plural)
ed
2
3
4
5
ing
er
1y
The actions performed by Speedtype when adding a suffix string to an expansion
string are:
REF. NO.
SUFFIX ACfIONS
1
add suffix string directly
drop last character, add suffix string
double last character, add suffix string
replace last character with i, add suffix string
replace last' character with ie, add suffix string
add e, add suffix string
2
3
4
5
6
The suffix action table presented below shows the action performed by Speed type
when adding a specified suffix string to an expansion string of a given word type.
5-8
AZ98-02
add_symbols (asb)
SUFFIX ACfION TABLE
SUFFIX STRING REF. NO.
Word Type
Ref. No.
1
2
3
4
5
0
1
1
1
1
1
1
1
2
2
2
2
2
1
2
1
2
1
3
5
4
1
4
1
4
1
1
1
1
1
5
6
1
1
1
6
133
I
II
5-9
I
1
I
3
1
I
I
AZ98-02
change.;.....symbols (csb)
change_symbols (csb)
The change_symbols command changes the expansion or suffixing of specified
symbol. Control arguments are processed one at a time. Specifying more than one
control argument has the same effect as issuing the command several times with one
control argument each time.
SYNTAX AS A COMMAND
csb synnbol {-control_args}
ARGUMENTS
synnbol
is the symbol changed.
dictionary.
This symbol must be defined in the current symbol
CONTROL ARGUMENTS
one or more arguments nnust be chosen from the following:
-exp string
where st ring represents the new expansion string for this symbol. This control
argument does not change the way suffixing is performed for the symbol.
-suffix string
enables or disables suffixing for this symbol. string must be. either on or off.
If string is on. then suffixing is enabled and all suffixes are processed
according to the default rules described in the "Notes" of the option_symbols
command below. If string is off, then all suffixes are disabled for the symbol.
If this control argument is not specified, then on is assumed.
-plural st ring
defines the plural suffix for this symbol. string must be on or off. or a string
that can be used as the plural of the expansion of this symbol. If string is
on, then the plural suffix is enabled for this symbol and processed according to
the default rules for the plural suffix. If string is off, the plural suffix is
disabled for this symbol.
-ed string
defines the ed suffix for this symbol. This control argument follows the same
rules as the -plural control argument.
-ing string
defines the ing suffix for this symbol. This control argument follows the same
rules as the -plural control argument.
-er string
defines the er suffix for this symbol. This control argument follows the same
rules as the -plural control argument.
-1y string
defines the ly suffix for this symbol. This control argument follows the same
rules as the -plural control argument.
5-10
AZ98-02
delete_symbols (dsb)
delete_symbols (dsb)
The delete_symbols command deletes the specified symbols from the current symbol
dictionary.
SYNTAX AS A COMMAND
dsb symbols
where symbols are the symbols to be deleted from the current symbol dictionary.
5-11
AZ98-02
expand_symbols (esb)
expand_symbols (esb)
The expand_symbols command takes an input text segment and expands it using the
options and symbols defined in the current symbol dictionary.
SYNTAX AS A COMMAND
esb input_path {put put_pat h }
ARGUMENTS
input path
is-the pathname of the input text segment.
output path
is an optional pathname of an output text segment. If no output pathname is
specified, the original contents of the input text segment are overwritten with
the expanded material.
5-12
AZ98-02
find_symbols (fsb)
find_symbols (fsb)
The find_symbols command finds and lists an of the symbols associated with
specified expansions contained in the current symbol dictionary. One. several. or all
expansions may be listed.
SYNTAX AS A COMlv7AND
fsb {expansions} {-control_args}
ARGUMENTS
expansions
are optional arguments that specify expansions to find and list. If an expansion
is represented by more than one symbol. all of its symbols are found and
listed. If any given expansion is not found, a message is printed stating that
the expansion is not defined. If no expansions are specified, all expansions in
the current symbol dictionary are listed. The expansions are listed in order
according to ASCII collating sequence.
CONTROL ARGUMENTS
-long
-lg
specifies that for each symbol listed. its expansion string with suffixing is listed
for each suffix enabled for that symbol.
-option
-op
specifies that all option information for the current symbol dictionary is to be
listed (see the option_symbols command for a complete description of option
information). If this is the only control argument specified, only the option
information is listed.
-total
-tt
specifies that the total number of symbols defined in the current symbol
dictionary is to be printed. If this is the only control argument specified, only
the total is printed.
5-13
AZ9S-02
list_symbols (Isb)
list_symbols (Isb)
The list_symbols command lists one, or several, or all of the symbols defined in
the current symbol dictionary.
SYNTAX AS A COMMAND
lsb {symbols} {-control_args}
ARGUMENTS
symbols
are optional arguments that specify the symbols to list. If any given symbol is
not found, then a message is printed stating that the symbol is not defined. If
no symbols are specified, then all symbols in the current symbol dictionary are
listed. The list is in ASCII collating sequence order.
CONTROL ARGUMENTS
-long
-lg
specifies that for each symbol listed, its expansion string with suffixing is listed
for each suffix enabled for that symbol.
-option
-op
specifies that all option information for the current symbol dictionary is to be
listed (see the option_symbols command for a description of option information).
If this is the only control argument specified, then only the option information
is listed.
-total
-tt
specifies that the total number of symbols defined in the current symbol
dictionary is to be printed. If this is the only control argument specified, then
only the total is printed.
5-14
AZ98-02
optjoD_symbols (osb)
option_symbols (osb)
The option_symbols command allows a user to change certain optional control
information in the current symbol dictionary. This information is summarized in
"Notes" below.
SYNTAX AS A COMMAND
osb {control _ args}
CONTROL ARGUMENTS
all except -delim set corresponding escape, prefix, or suffix characters recognized
by Speedtype to the character specified by X (see "Notes" below). A complete
explanation of the escape, prefix. and suffix characters is given earlier in this
section.
-delim st r ; ng
specifies a new set of delimiter characters. None of the characters in this
str; ng may be currently defined escape. prefix. or suffix characters.
-pad X
-perm X
-temp X
-trans X
-space X
-under X
-upper X
-ed X
-ing X
-er X
-ly X
NOTES
A summary of all Speedtype options is given below. The default character(s) used
to represent each option is also shown.
Delimi ters:
Escapes (see below)
White space (space, tab, newline)
"-L...'
_
"nil
r",. .
Vlners
,\J!!<> LJ U··
~~
5-15
AZ98-02
option_symbols (osb)
Escapes:
pad
perm
temp
trans
space
(octal 177)
,
Prefixes:
under
upper
Suffixes:
plural
ed
ing
er
ly
+
*
EXAMPLES
To set the temporary escape character to &. type:
osb -temp &
5-16
AZ98-02
The print_symbols_path command prints the pathname of the current symbol
dictionary.
SYNT AX AS A COMMAND
psbp
5-17
AZ98-Q2
retain_symbols (rsb)
retain_symbols (rsb)
The retain_symbols command takes an input text segment and inserts Speedtype
escape characters wherever symbols would be expanded if this text segmen t were being
processed by the expand_symbols command. All symbols in the text segment are thus
retained during future expansion.
SYNTAX AS A COMMAND
rsb input_path {output_path} {-control_args}
ARGUMENTS
input path
is-the pathname of the input text segment.
output path
is the optional pathname of an output text segment If no output pathname is
specified, the original contents of the input text segment are overwritten.
CONTROL ARGUMENTS
-perm
specifies that the perm escape character is to be used. If no control argument
is specified, -perm is assumed.
-temp
specifies that the temp escape character is to be used. Specifying this control
argument causes the symbols in the output text segment to be retained for only
one expansion.
NOTES
In addition to inserting the specified escape character wherever necessary, all
existing pad escapes are converted to the specified escape. This allows for more
convenient editing of the input text segment, since all escape characters are thus
printable. (Refer to the escape description earlier in this section.)
5-18
AZ98-02
show_symbols (ssb)
show_symbols (ssb)
The show_symbols command shows how Speedtype expands an input string. The
expansion is performed using the options and symbols in the current symbol dictionary.
The expanded string is printed on the user's terminal.
SYNTAX AS A COMMAND
ssb terms
where terms are arguments that are concatenated into the input string that is expanded.
These terms are separated in the input string by one space. If other spacing is desired,
the input string should be enclosed in quotes.
EXAMPLES
To show the expansion for the term th which is defined in the current symbol
dictionary, type:
ssb th
these
If the term th is not defined in the user's current dictionary then the system
response would be t h .
5-19
AZ98-02
use_symbols (usb)
use_symbols (usb)
The use_symbols command sets the current symbol dictionary. All Speedtype
commands then use this symbol dictionary. If this symbol dictionary does not exist, the
user is asked if it should be created.
SYNTAX AS A COMMAND
usb path
where path is the pathname of the symbol dictionary that is to be the new current
symbol dictionary. If path does not have a suffix of symbols, one is assumed; however,
symbols must be the last component of the symbols dictionary segment name.
NOTES
If other Speedtype commands are issued in a user's process before the use_symbols
command, then those commands use the default symbol dictionary in the user's home
directory. The default symbol dictionary has the pathname:
>udd>Project_id>Person_id>Person_id.symbols
5-20
AZ98-02
SECTION 6
LIST PROCESSING
List Processing involves maintenance, sorting, and selection of items in a list (e.g.,
names, words, numbers) and the production of documents that use this information.
The processing steps involve creating an input file, compiling this file into a form that
can be manipulated by the List Processing commands, and manipulating this new file
with files that define formats for the final output.
LIST PROCESSING FUNCTIONS
The main functions of List Processing are:
•
•
•
•
list maintenance (i.e., entry and update of information in a list)
sorting
selection
report generation
An example of the use of List Processing is a dental office that maintains a list of
all patients serviced by that office. The data maintained for each patient might include
the patient's name, address, phonenu-mber, date of last visit, etc. When· a patient fir-st
visits the ofJice, the patient is added to the list. This involves using a text editor (see
Section 2) to input information about the patient, which may be updated later. For
example, on each subsequent visit, the date of last visit is updated for that patient.
This dental patient list can be used to produce various documents. For example. the
dentist may want a report listing the name, address, and phone number of all patients,
sorted alphabetically by patient name. or a form letter reminding the patient to visit
the office for a checkup. Perhaps the dentist sends this letter to those selected patients
who have not visited the office for six months or more.
On Multics, List Processing is done with a set of commands that maintain and
process online lists of information. These commands can be used to produce simple
reports like the ones described above; they also provide a means by which the output
can be saved in a segment or directed to the terminal. Once in a segment. the output
can be mailed to other users using the Multics mail facility (see Multics Commands),
or can be further processed by the Compose Text Formatter (see Section 4 of this
manual) to produce reports and form letters.
LIST PROCESSING FILES
List Processing uses three types of files (1istin, lister, and listf orm), each type
identified by its entryname suffix. A description of each of these types follows.
6-1
AZ98-02
Listin File
A listin file is an ASCII file used to input and update a list. It is identified by
the entryname suffix .1 i st i n (e.g., month 1y. 11 st In). Records can be added to,
deleted from, or updated in this list simply by editing the file with a text editor.
The format of a listin file is simple. It consists of the following three parts:
1.
Header
The header specifies the record and field delimiter
comment delimiter, and the field names. It is located
file and contains the following statements:
Comment de 1 i miter: c;
Cd:
or
Record delimiter: r;
Rd:
or
Field - del imi ter: f;
Fd:
or
Field names: f n 1 ,
fnK;
Fn:
or
Records:
...
characters, the optional
at the beginning of the
c;
r;
f;
f n 1,
fnK;
The Comment_delimiter statement enables comments in the listin segment and
specifies the character or characters used to begin and end comments.
Comments may appear anywhere that white space is allowed and are ignored.
No comments may precede this statement in the listin file. If this statement is
not given, then no comments are allowed in the listin file. If the string "pU"
is specified in the Comment_delimiter statement then comments begin with" 1*"
and end with "* I"; otherwise the comment delimiter must be exactly one
character long and must be chosen from the set below, in which case that
single delimiter both begins and ends comments. It should be noted that the
usage of the PL/I-style comments in conjunction with the usage of the "*" as
either record or field delimiters can cause problems and should be avoided if
possible.
The Record_delimiter statement specifies the character used to separate records.
If this statement is not given. the default record delimiter is a dollar sign ($).
The Field_delimiter statement specifies the character used to separate fields
within a record. If this statement is not given, the default field delimiter
character is an equal sign (=). Record and field delimiters must be exactly one
character long, cannot be the same, and must be chosen from the following set:
!#$%&*=?@/\
I
The Records statement must be the last statement in the header, and is
required. It specifies the end of the header and the beginning of the record
inf ormation.
2.
Fields
The fields contain the various types of information stored in a list (e.g., first
name, last name. street address, date of employment. etc.). Because data records
are stored separately within a listin file, the field names must be given with
each data record. Within an individual data record. a field is specified by a
field delimiter character followed immediately by the field name (e.g., =lname).
Any amount of white space (space, horizontal tab, vertical tab, newline, or new
page) can follow the field name (e.g., =lname Smith). If the field value
contains any record. field. or comment delimiters, then it must be quoted (e.g ..
=amount "$1.00") and in this case any embedded quotes must be doubled. A
field value ends at the next record or field delimiter. Leading and trailing
6-2
AZ98-02
white space is removed from field values, though such 'space can be retained by
including it within quotation marks that surround the entire field value.
All of the field names used in a list must be specified in the Field_names
statement of the header. Field names may be from 1 to 32 characters in
length, must begin with an alphabetic character, and must contain only
alphabetic, numeric, and underscore characters.
3.
Data Records
The data records (hereafter simply referred to as records) contain the specific
information associated with the subject of each record. The beginning of each
record is denoted by the record delimiter character, followed by a list of
fields. A record may contain some or all of the fields defined in the header,
and fields not specified for a record are considered to be null. Duplicate fields
are not allowed within a record.
Lister File
A lister file contains the records entered and updated through a listin file and
serves as the file from which the processes of merging, trimming, sorting, selecting, and
document processing are performed. It is identified by the entryname suffix • 1 is ter
(e.g., month 1y. 1 is ter). After records have been input or updated in a listin file, the
create_list command (described below) transfers them to a lister file. When a lister
record is created, it is assigned a decimal identifier that is unique within the lister file
and remains assigned to the record .as long as the file exists. If the record is deleted,
its unique identifier is net reused. If the- create_list -command is used tore-create an
existing lister file, the unique identifiers change. The unique identifier is referred to by
the reserved field name ":uid". Since lister files are formatted binary files, they cannot
be displayed by using the print command, as listin files can and must be. They can
only be displayed by the process_list or the display _list commands, though they can be
manipulated by the append_list, copy _list, merge.,..list, modify_Ust, sort_Ust, and trim_list
commands. Other functions are provided by the describe_list and expand_list commands
(all of the lister commands are described below).
Listform File
A listform file defines the format of a document to be produced from a list of
records. It is identified by the entryname suffix .1istform(e.g., monthly.listform).
Information from a list is copied into a document in the format specified by the
listform being used. A single listform file may be used with a number of lister files,
just as one lister file can be used with several listform files.
Three sections of a document may be defined. These three sections are the Before
section, the After section, and the Record section. The Before and After sections are
optional and usefUl for organizational purposes; i.e .. the Before section may be used for
headings and introductory matter, -and the After section for closings and summary
material. The Record section, however, is necessary for processing records. The
functions of field insertion, sorting, and selection require the presence of a Record
section within the listform to correspond to the lister records. Examples of these
sections are in the sample listform files under "Sample List Processing Files" below.
6-3
AZ98-02
These three listform sections are further described as follows:
Before
This section is added to the document as a preface hefore anj,? records aie
processed. It may contain any desired text, including compose controls. The
beginning of the Before section is identified by the string III~
The end of the Before section is identified by the string lI".
Record
This section describes the document format for each lister file record processed.
It contains field value'strings to be copied from the lister file being processed
(see "Field Insertion" below), compose controls, and any desired text The
beginning of the Record section is identified by the string II11.
The end of the Record section is identified by the string "11.
After
This section is added to the document after all records are processed. It may
contain any desired text, including compose controls, The beginning of the
After section is identified by the string III~ the end of the
After section is identified by the string "I~
FIELD INSERTION
In order to insert information from the lister file into the document. a field name
enclosed in angle brackets «» is included in the record section of the listform file
(e.g., specifies that
the value string of the field "city" is to occupy 10 character positions. If the current
value string is less than the specified field width, then it is padded on the right with
blanks. If the current value string is greater than the specified field width, then it is
truncated (cut off) on the right so its length is equal to the specified field width.
An optional field alignment may also be specified if a field width
example, specifies that the value string of this field is to
within the 10-character field width. The alignment indicators "1" for
center may also be specified. If no alignment is specified, the
left-aligned.
is specified. For
be right-aligned
left and "c" for
value string is
To insert arguments into the document using the process_list command with the
-argument control argument, an argument name (enclosed in angle brackets) may be
included in the text of the before, after, or record section. The argument name is
replaced by the actual argument when the section is processed. Field widths and field
alignments may also be specified for argument insertions.
Argument names are of the form :argl, arg2.... :argN. For an example of the use
of the -argument control argument, see the example beneath the sample letter under
"Sample List Processing Files" below. Arguments specified by the -argument control
argument, but never referenced by an argument name, are diagnosed with a warning;
....
'] Ur<:l ... n1no- ';c n;cnl<:1"",.n nn th",. t",.,,",,;n<:ll
1.""_, t-l-oo
"'1.1."'" ,."' .......... " .... ,:! .fll .... ,.t.;"' .... ';co ,.'] ......10,:! n11t "hnt
v .... '"
tr.JI'"'.a.u,J,."",..
.a....... " ....... .a. ......
Arguments named in a listform segment, but never specified by the -argument control
argument are also diagnosed with a warning. A null string is used in place of the
missing argument. The -brief _errors control argument suppresses these warnings.
\,orVll11.1J.&l.11U
1 U.ll"""'J.Vl...
J.~
""1.&..11.l'-'W
Vw\.
U.
6-4
"LLJ. J..J..I..L.Lf:>
J.1Io1
........
V.I....
" ...
.I. ...
.1..1. .........
AZ98-02
Listform files are character-oriented rather than line-oriented, so placement of field
names within these files dictates the line output. Thus. when field or character strings
follow immediately after a control, no blank lines are inserted. When
the field name or character string begins on the next line, one blank line is inserted.
The following format produces no blank lines between records:
One blank line is inserted between each record by the following listform format:
One blank line can also be produced by moving the control, as in:
The unique iden tif ier of a lister record can be inserted in to a List Processing
document by specifying <:uid> in the Record section of the listform file.
Finally. by use of listform files, the current date, the present timet and the number
of records being processed with the current command invocation can be inserted in a
List Processing document. These special fields can be inserted separately in any of the
three listform sections; Before, Record, and After. using the format:
<:date>
<: time>
<:record_count>
When the process_list command is invoked with a listform file containing any of these
special fields, the specified information is automatically inserted into the document
being printed without any type of control argument in the command line.
ANGLE BRACKET ESCAPES
To place a single left angle bracket in the text. enter two left angle brackets «<).
A single right angle bracket is left as is in the document when it is used with two left
angle brackets (e.g., «Phoenix> becomes in the output). A single right angle
bracket is also left as is in the output when it is used with no left angle brackets (e.g.,
Phoenix> remains Phoenix».
SORTING
The sorting process sorts records in a file according to specific criteria. These
criteria are indicated in the -sort control argument which is used v:ith both the
sort_list and process_list commands. With the -sort control argument, a list of names
and addresses. for instance, can be printed in alphabetical order according to first
name, last name, city, street, or any field within its record.
Note:
Within this section, references to alphabetical order refer to a sorting
sequence identical to the ASCII collating sequence with the exception that
lowercase letters immediately follow the corresponding uppercase letters.
6-5
AZ98- ....
The -sort Control argument always takes a character-string argument which specifies
the record fields used to control the sort Its format is:
-sort st ring
-st string
Note that the character-string string must be surrounded by quotation marks when
the string contains blanks or reserved characters and internal quotation marks must
be doubled.
The character-string consists of one or more field specifications separated by spaces.
The first field specification defines the primary sort field; the second field specification
defines the secondary sort field; and so forth. The lister file acted upon is reordered
permanently when using the sort_list command; the lister file is not altered when using
the -sort control argument with the process_list command.
A field specification consists of two parts, a field name and optional order and
type control arguments. The order and type control arguments can be chosen from the
following:
-ascending
-asc
specifies that this field is to be sorted into ascending order. If no order
control argument is specified, then ascending order is assumed (e.g., 0123456789Aa... Zz).
-descending
-dsc
specifies that this field is to be sorted into descending order.
-alphabetic
-alp
specifies that the field is to be sorted alphabetically.
-numeric
-num
specifies that this field is to be sorted numerically by temporarily converting
each field value to a float decimal(29) value. Values that cannot be converted,
sort as if they had the value zero.
The sort performed by the -sort control argument is stable; that is, records with
equal fields stay in the same relative order, whether an ascending or descending sort is
performed. For examples of the use of the -sort control argument. see "Sample List
Processing Files" and the sort_.list command description in this section.
SELECTION
The selection feature enables a List Processing command to select from a lister file
only certain records upon which to perform its function. The command. through the
use of the control argument described below, specifies requirements for desired fields.
If a record meets the requirements it is processed: otherwise it is skipped.
For instance, from a complete list of names and addresses, separate lists could be
printed for all entries with last name beginning with any letter, all residents of one
town, all residents of one state, or all entries with the exception of those containing
some specified criteria.
The -select control argument (which can be used with the copy_list, process_list,
and trim_list commands) always takes a character-string argument. Its format is:
6-6
AZ98-02
-select st ring
-sel string
In this argument, the character-string str; ng must be surrounded by quotation
marks. Each record in the specified lister file is tested to determine whether or
not the record fulfills the selection criteria. Those that do are processed.
The -select control argument consists of one or more field comparisons. A field
comparison involves comparing a test string to the value of the specified field in the
current record. The field comparison statement always consists of three parts:
"field _name comparison_operator test_string"
where:
field name
is the name of a field contained in the lister file. The reserved field name
":any" may be used to specify any field in the record. The reserved field name
":uid" may be used to specify the unique identifier of a record.
comparison operator
specifieS what comparison is performed. The opposite comparison is performed
if the comparison operator is preceded by "not". The List Processing comparison
operators are:
con tain (s)
test string is contained in the field value. The comparison is made without
regard to case (i.e., uppercase letters compare equal to lowercase letters).
equaHs)
test string is equal to the field value. Upper-caseletters·· are distinct from
lowercase letters with this operator.
greater
field value is alphabetically greater than the test string (e.g., 0123456789Aa. .. Zz).
less
field value is alphabetically less than the test string.
nequal(s)
field value string is numerically equal to the numeric value of test string.
ngreater
field value string is numerically greater than the test string.
nless
field value string is numerically less than the test string.
test string
IS
the string that is compared to the field value string. The special test string
":null" is used to test whether or not the field is null, i.e., missing from the
current record. The special test string ":numeric" is used to test whether or not
the field value string is numeric, i.e., can be converted to a number. Null
fields are always non-numeric.
Several field comparisons may be specified by the -select control argument. Field
comparisons are combined by the logical operators "and", "or", or "not". In the absence
of parentheses. the prefix "not" operator is evaluated first, then the infix "and"
operator, then the infix "or" operator. Parentheses may be usee to specify the exact
order of evaluation. These rules are similar to the PL/I rules for Boolean expressions.
6-7
AZ98-o2
The special test strings ":null" and ":numeric" can only be used with the equal or
nequal comparison operators.
The comparison operators (not) contain, (not) greater, (not) less, (not) ngreater, or
(not), and nless ignore records that have null fields. Unless the special test string
":null" is used, (not) equal and (not) nequal also ignore records with null fields.
For examples of the use of the -select control argument, see "Sample List
Processing Files" below, and the process_list and trim_list command described later in
this section.
SAMPLE LIST PROCESSING FILES
A sample list and two specific uses (mailing list and form letter) are shown below.
Using the dental office example, the first file shown below (patients.1istin) is the
one containing a list of the patient's names, addresses. and other pertinent information.
Its appearance is exactly as entered by the user with the text editor, except for the
heading "patients.listin". which is added when the file is displayed by the print
command. The listin file can be displayed by entering the command:
print patients.listin
which results in the following display:
patients.listin
Comment_delimiter: pll;
Record_del imiter: $;
Field_delimiter: =;
Field_names: fname,lname.street.city,state.zip.phone.
date,message;
Records:
$
=fname John
=lname Doe
=street 71 Pine Street
=city Boston
=state Massachusetts
=zip 02020
=phone (617) 555-7654
=date 770520
=message you and your family well
$
=fname Jane
=lname Smith
=street 898 Smith Avenue
=city Needham
=state Massachusetts
=zip 02112
=phone (617) 555-4567
=date 750713
=message you well
$
=fname Francis
=lname Jones
I*formerly Wi lson*1
6-8
AZ98-o2
=street PO BOX 999
=city Cambridge
=state Massachusetts
=zip 02139
=phone (617) 555-7869
=date 770131
=message you we]]
The next sample file, addresses.listf orm, defines the format to be used while
processing patients.lister (created from patients.listin) to generate the address report
below (final desired output). Its appearance is exactly as entered by the user with the
text editor, except for the heading "addresses.listform", which is added when the file is
displayed by the print command. The listform file can be displayed by entering the
command:
print addresses.listform
which results in the following display:
addresses.l istform
Dental Patient Addresses
,
Tel:
Dental Associates
The create list command makes a lister file named patients.lister from the listin file
named patients.listin. The process_list command then operates on the lister file.
formatting the records according to the addresses.listform file and arranging the records
alphabetically by last name as specified by the -sort control argument. The following
command lines are used to create the lister file and process it with the listform file:
create_list patients.listin
process_list patients.l ister addresses. listform -sort lname
w-hich results in the following display:
Dental Patient Addresses
John Doe
71 Pine Street
Boston, Massachusetts 02020
Tel:
(617) 555-7654
6-9
AZ98-02
Francis Jones
PO BOX 999
Cambridge, Massachusetts 02139
Tel:
(617) 555-7869
Jane Smith
898 Smith Avenue
Needham, Massachusetts 02112
Tel: (617) 555-4567
Dental Associates
The file. letter.1istform. defines the format of the form letter. which. when
manipulated with patients.1ister, creates the Sample Letter (shown below) specifically for
Jane Smith. Using the process_list command with the -select control argument to
specify other records within patients.1ister. the same letter can be composed for any or
all patients on the list. using any available field as a criterion. Its appearance is exactly
as entered by the user with the text editor, except for the heading "letter.listform",
which is added when the file is displayed by the print command. The listform file can
be displayed by entering the command:
print letter.1 istform
which results in the following display:
letter.listform
.pdl 40
.pdw 55
• in 1 30
·f if
Dental Associates
1001 Jamaica Avenue
Boston, Mass. 02003
(617) 555-6000
.spb 2
· in 1
,
• fin
· i n1
.spb
Dear :
.spb
I hope this letter finds .
It has been over six months since your last visit to our
office. Please call and make an appointment
to have a checkup .
• spb
· i nl 30
Keep smil ing,
.spb 2
J. Ke 1 1y, D. M. D.
6-10
AZ98-02
.brp
SAMPLE LETTER
The process_list command selects Jane Smith from the patients.lister file, processes
it with the letter.listform file, and sends it to an output file named letter.compin. This
segment is then operated on by the compose command to produce the letter to Jane
Smith. The following command lines are used to create the sample output from the
lister file:
pls patients letter -sel fname equal Jane and lname equal Smith
-of letter.compin
compose letter
which results in the following display:
Dental Associates
1001 Jamaica Avenue
Boston, Mass. 02003
(617) 555-6000
Jane Smith
898 Smith Avenue
Needham, Massachusetts 02112
Dear
Jane~
I hope this letter finds you well.
It has
been
over six months since your last visit to our office.
Please call and make an appointment to have a checkup.
Keep smiling,
J. Kelly, D.M.D.
Now assume that Dr. Kelly takes in a partner (Dr. O'Brian). When composing the
reminder in the future he wants to designate from command level whose name is to be
on each letter individually. In letter.listform (see example above that displays
"}etter.1istform"), he replaces:
J. Ke 1 1y, D. M.D.
with an argument name of the form:
<:argl>
6-11
AZ98-02
The next time that a letter is needed, he types:
pls patients letter -5e1
lnam9 equal Doe
~of
letter.compin
-a9 "W. O'Brian, D.M.D."
and the letter is supplied with Dr. O'Brian's name in the signature block.
6-12
AZ98-02
append_list (als)·
The append_list command adds a record to a lister file.
SYNTAX AS A COMMAND
als path -control_args
ARGUMENTS
path
is the pathname of the lister file. The suffix lister must be the last component
of the lister segment name; however, if path does not have a suffix of lister,
one is assumed.
CONTROL ARGUMENTS
-field_name field _name string
-fn field name string
causes-the value of string to be assigned to the field indicated by field_name.
If string contains spaces. it must be enclosed in quotes. This control argument
is required and may be given more than once. Those fields for which this
control argument is not given are assigned null values.
-string st ring
-str string
uses string as a character string with no special interpretation. This is useful
for preventing string from being interpreted as a control argument It is to be
used with the -field_name control argument (e.g., "-field_name rating -string
_''In''\
-~v
,.
EXAMPLES
To append a record to an existing lister file, type:
als patients -fn fname Benjamin -fn lname Walker
6-13
AZ98-02
copy_list (cpls)
copy_list (cpls)
The copy_list command creates a new list segment from an existing list
segment. All, or selected, records of the existing list segment are copied into the new
list segment. The new list segment is created in the working directory. Any existing
copy of this segment is overwritten.
.
SYNTAX AS A COMMAND
cpls path 1 path2 {-control_args}
ARGUMENTS
path 1
is the pathname of the existing lister file. The suffix lister must be the last
component of the list segment name; however, if path 1 does not have a suffix
of lister. one is assumed.
path2
is the pathname of the new list segment. The suffix lister must be the last
component of the list segment name; however. if path2 does not have a suffix
of lister, one is assumed.
CONTROL ARGUMENTS
-select st ring
-sel string
copies records specified by string (the string argument must be enclosed in
quotes). If this control argument is not specified, then all records are copied.
(For a complete description of how to specify string, see "Selection" earlier in
this section.)
-totals
-tt
displays the number of records copied.
EXAMPLES
To copy all records that have a city field equal to Boston from patients.lister into
Boston_patien ts.lister , type:
cpls patients Boston_patients -sel IIcity equal Boston"
To copy all records that do not have a city field equal to Boston from
patients.lister into Mass_pati en ts.lister , type:
cp 1spat i ents Mass_pat i ents -se 1 lie i ty not equa 1 Boston"
For more examples of the use of this control argument. see -the trim_Est command
description.
6-14
AZ98-02
create_list (cis)
The create_list command creates a lister file from a listin file.
SYNTAX AS A COMMAND
cIs path {-contro'_arg}
ARGUMENTS
path
is the pathname of the listin file. The suffix listin must be the last component
of the listin segment name; however, if path does not have a suffix of listin,
one is assumed. A lister file is created in the working directory with the same
entryname as path, and with the entryname suffix of listin changed to lister.
Any existing copy of this lister file is overwritten.
CONTROL ARGUMENTS
control_arg
can be -totals or -tt to display the number of records in path.
NOTES
The creation of a lister file is the only List Processing operation which uses listin
files as input. All other operations use lister files as input (which are unprintable files
containing ASCII and binary information).
A listin file provides an ASCII representation of a list It is used to input and
update a list The listin files can be created and updated by using any text editor.
Example
To create patients.lister from patients.listin (which contains three data records) and
display the number of records in patients.listin, type:
cls patients -tt
create_list: 3 records.
6-15
AZ98-02
describe_list (dIs)
describe_list (dIs)
The describe_list command displays information about a lister file.
SYNTAX AS A COMMAND
dIs path {-control_args}
SYNTAX AS AN ACTIVE FUNCTION
[dIs path {-control_args}]
ARGUMENTS
path
is the pathname of the lister file. The suffix lister must be the last component
of the lister segment name; however, if path does not have a suffix of lister,
one is assumed.
CONTROL ARGUMENTS
-delimiter {record Ifield}
-dm {recordlfield}
displays the value of the record or field delimiter. If the record and field
keywords are omitted. then both delimiters are printed.
-field_name
-fn
displays the field_names in the lister file.
-select string
-sel string
specifies those records to be indicated by the -total control argument. If this
control argument is not specified, then the total number of records in the file
is used. (For a complete description of how to specify string see "Selection"
earlier in this section.)
-total
-tt
displays the total number of records.
NOTES
If no control arguments are given. or only the -select control argument is given,
then the record and field delimiters, total, and the field names are displayed.
If none or more than one of -delimiter {record I field}. -total. or -field _ name are
specified, the values are returned in the following order: record_delimiter. field=delimiter,
total, and field_names.
6-16
AZ98-02
describe_list (dIs)
Example
dls mlist
ml ist. 1 ister
Total Records:
Record_delimiter:
Field_delimiter:
Field_names:
alias,mproj;
dls mlist -sel
5
07/02/80
1606.4 mst Wed
748
$;
name,did,addr,current,years,personid,
"mproj equal SysAdmin" -total
6-17
AZ98-o2
display_list (dils)
display _list (dils)
The display_list command displays (prints) selected portions of selected lister
records.
SYNTAX AS A COMMAND
dils path {-control_ args}
SYNTAX AS AN ACTIVE FUNCTION
[dils path {-control_ args} ]
ARGUMENTS
path
is the pathname of the lister file. The suffix lister must be the last component
of the lister segment name; however, if path does not have a suffix of lister,
one is assumed.
CONTROL ARGUMENTS
-brief _errors
-bfe
suppresses the warning when no records match the selection expression.
-field name field names
-fn field names causes- the specified f i el d names to be displayed, in the order indicated. This
control argument must be -given.
-select string
-sel string
specifies those records whose fields are to be displayed. If this control
argument is not specified, then all records are used. (For a complete
description of how to specify string. type "help process_list".)
6-18
AZ98-02
expand_list (els)
expand_list (els)
The expand_list command creates a· listin segment f rom a lister segment The
number of records expanded is displayed. The operation performed by this command is
the opposite of that performed by the create_list command.
SYNTAX AS A COMMAND
els path {-control_ argsJ
ARGUMENTS
path
is the pathname of the lister segment. If the entryname suffix lister is not
specified, then it is added. A listin segment is created in the working directory
with the same entryname as path, and with the entryname suffix lister changed
to listin. Any existing copy of this listin segment is overwritten.
CONTROL ARGUMENTS
-line_length n
-11 n
specifies that the line length of the ASCII listin segment is to be n characters.
If this control argument is not specified, then only one field is placed on each
line. A field is placed on a new line only if adding the field to the current
line would exceed the specified line length. At least one field is placed on
each line.
-totals
-tt
displays the number of records expanded.
NOTES
The ASCII listin segment created by this command has the following format:
@
The first two lines specify the record and field delimiter characters.
•
Beginning on the third line are the field names. They are separated by a
comma and a space. A field name is placed at the beginning of a new line if
adding it to the current line would exceed the specified line length.
•
Each record begins with a line containing just the record delimiter character.
e
Unless -line_length is specified, each field is placed on a separate line and
indented one space.
6-19
AZ98-02
merge_list (mls)
The merge_list command combines two lister files into a single lister file. The file
resulting from the merge may be a new lister file. or it may replace an existing lister
file. The fields defined in the two lister files must be identical. and the fields to be
compared must be in ascending order. The comparisons are performed without regard
to case (uppercase letters compare equal to lowercase letters). Sorting must be done by
the sort_list command.
SYNTAX AS A COMMAND
mls mas_path upyath {out _path} {-control_ args}
ARGUMENTS
mas path
IS
the pathname of the master lister file. The suffix lister must be the last
component of the lister file name; however. if mas_path does not have a
suffix of lister. one is assumed.
upyath
is the pathname of the update lister file. The suffix lister must be the last
component of the lister file name; however. if upyath does not have a suffix
of lister, one is assumed.
out path
-is the pathname of the output lister file. The suffix lister must be the last
component of the lister file name; however, if out_paUl does not have a suffix
of lister. one is assumed. If this argument is not specified. the master lister
file is replaced.
CONTROL ARGUMENTS
-field_name f n 1 '" f ni
-fn fnT ... fni
specifies that fields fn 1 through fni are used as the controlling fields for the
merge. (Records can only be merged if they contain the same fields, though
some of those fields may be null.) The fields are compared without regard to
~ case. If this control argument is not specified, then all fields are used to
control the merge.
-totals
-tt
displays the number of' records in the master, update, and output files.
Only one of the following four control arguments (-add. -and. -or, or -subtract)
can be specified:
-add
copies into the output lister file all records from the master lister file pI us all
records from the update lister file. Thus records contained in both lister files
are listed twice in the output file. (Default)
6-20
AZ98-02
merge_list (mls)
-and
copies into the output file those records in the master lister file that are a/so
in the update lister file. That is, those records that are listed in both files are
listed once in the output file; no records from the update lister file are copied.
-or
copies into the output lister file all records in either the master lister file or
the update lister file. Duplicate records are copied only from the update lister
file and thus appear only once in the output file.
-subtract
-sub
copies into the output lister file all records in the master lister file that are
not also contained in the update lister file. Thus no duplicate records are
copied and no records from the update lister file are copied.
NOTES
The table below shows how master and update lister files are merged for each of
the four merge operations: add, and. or, and sub. The letters listed in the table body
represent individual records, with duplications of letters simply representing different
recordings of the same basic record. When records represented in both the master and
update files are listed in the output file, the letters representing them are given the
associated numeric shown in parenthesis with the identified file in order to indicate
which recording of a particular record actually went into the output file.
Operation
Master File(l)
Update File(2)
Output File
add
abc d e
d e f 9 h
abc d 1 d2
e 1 e2 f 9 h
and
a b c d e
d e f 9 h
dl el
or
abc d e
d e f 9 h
a b c d2 e2
f 9 h
sub
abc d e
d e f 9 h
abc
EXAMPLES
To copy into Boston_patients.lister all· records in patients.lister that have the city
field equal to Boston and print the total number of records, type:
copy_list patients Boston_patients -sel "city equal Boston" -tt
To delete from. patients.lister an records that have the city field equal to Boston
and print the total number of records, type:
trim_' ist patients -se' "city equal Boston" -tt
trim_list: 1 record deleted.
To merge the lister files patients.lister and Boston_patients.lister using the city name
as the controlling field for the merge and display the total number of records, first the
6-21
AZ98-02
merge_list {mls>
sort_list command is issued to sort the patients file into ascending alphabetical order by
city:
sort_list patients -sort city
and then the merge:
merge_list patients Boston_patients -tt -fn ci~y
merge_list: 3 master and 1 update records merged into 4 output records.
merge_list patients Boston_patients out_patients -tt -fn city -and
merge_list: 4 master and 1 update records merged into 1 output record.
6-22
AZ98-02
modify_list (mdls)
modify_list (mdls)
The modify_list command modifies a field or, fields in selected lister records.
SYNTAX AS A COMMAND
mdls path -control __args
ARGUMENTS
path
is the pathname of the lister file. The suffix lister must be the last component
of the lister segment name; however, if path does 'not have a suffix of lister,
one is assumed.
CONTROL ARGUMENTS
One or more arguments must be chosen from the following:
-brief_errors
-bfe
suppresses the warning when no records match the selection expression.
-field_name field name string
-fn field name string
causeS-the value of string to be assigned to the field indicated by field name.
If string contains spaces, it must be enclosed in quotes. This control argument
is required and may be given more than once.
-select string
-sel string
specifies those records to be modified. If this control argument is not
specified, then all the records are modified. (For a complete description of
how to specify string, type "help process_l is!".)
-string string
-str string
uses string as a character string with no special interpretation. This is useful
for preventing string from being interpreted as a control argument. It is to be
used with the -field_name control argument (e.g., II-f i e 1d_name rat i ng
-string -20").
-total
-tt
displays the number of records modified.
6-23
AZ98-02
process_list (pis)
process_list (pis)
The process_iist command produces a document irom all or selected records in a
lister file. The format of the document is defined in a listform file. Other text
processors, such as compose, may be used to further format the document. By default,
the document is printed on the user's terminal. Alternatively, it may be saved in a
segment. For a description of the structure of a listform file and information on field
insertion, angle bracket escapes, and the selection and sorting procedures (-select and
-sort control arguments). see those earlier portions of this section.
SYNTAX AS A COMMAND
pIs list_path {formyath} {-control_args}
ARGUMENTS
I ist path
IS
the pathname of the lister file to be processed. The suffix lister must be the
last component of the lister file name; however, if list_path does not have a
suffix of lister, one is assumed.
form path
is the pathname of the listform file that defines the format of the document.
If form path does not have a suffix of listform, one is assumed. If this
argument is not specified, a listform file in the working directory is used that
has the same entryname as list path, with the entryname suffix of lister
changed to listf orm.
CONTROL ARGUMENTS
-argumen ts st ring
-ag string
indicates that the listform segment requires arguments. If present. it must be
followed by at least one argument. All arguments following this control
argument on the command line are taken as arguments to the listform segment.
Thus, if present, this must be the last control argument on the command
line.
-brief_errors
-bfe
suppresses warnings about mIssmg or extra arguments for the -ag control
argument. Suppresses warning when no records are selected.
-extend
-ex
specifies that the document produced by this command is to be appended to
the segment specified by path (-output_file must also be given). The default is
to replace path completely.
-output_file {path}
-of {path}
specifies that the document produced by this command is saved in the segment
specified by path (see Sample Letter in "Sample List Processing Files" earlier in
6-24
AZ98-02
process_list (pIs)
this section}. If path is not specified, this output segment is placed in the
working directory with an entry name the same as form_path and the suffix
listform changed to list.
-select string
-sel string
specifies the records selected for processing. If this control argument is not
specified, then all records in the list are processed (see "Selection" earlier in
this section).
-sort st ring
-st string
sorts the records processed according to string, which is a string enclosed in
-quotes. The new ordering of the list is in effect only for the duration of the
command. The lister file is not modified. If this control argument is not
specified, then records are processed in the order in which they currently
appear in the lister file (see "Sorting" earlier in this section).
-totals
-tt
displays the number of records processed.
EXAMPLES
Since the process_list command is an intermediate step in List Processing operations,
assume that the user has already created a segment named students.listin, containing the
first name, last name, city, state, and zip of three students; this segment includes three
records, each consisting of the above mentioned five fields. Also assume that, using the
create_list command, students.lister has been created, and a format for the list,
names.listform also exists. Following is a copy of the segment students.listin:
Record_delimiter: $;
Field_del imiter: =;
Field_names: fname,lname,city,state,zip;
Records:
$
=fname
Smith
=lname
=city
Boston
--~-.-
-:::''''CI'''C
MA
=zip
02114
$
=fname Tim
=lname
Jones
=city
Cambridge
=state
MA
=zip
$
Victor
=fname
=lname
Red
=city
Cambridge
=state
MA
=zip
02139
$
6-25
AZ98-Q2
process_list (pIs)
The first record in this segment has a null fname, and the second record contains a
null zip field. As shown in the first field (fname) of the Second record, the amount of
white space between the field name and the field value is completely arbitrary (as is
the space between field value and field delimiter), and makes no difference when
processing.
The following listform segment does not utilize the optional before or after
sections, so it creates no heading or ending lines in the finelI output. Following is a
copy of the format-defining segment, names.listform:
,
To have the process_list command select (print) all records that have a last name
(lname) field less than "m" (i.e., all persons whose last name is from A to L), type:
pls students names -se1
Tim Jones
Cambridge, MA
r 725 0.401 3.782 61
"lname less m"
To have the process_list command select (print) all records that have any field
whose value is nUll, type:
pls students names -se1
Smi th
Boston, MA 02114
":any equal
:nu1l"
Tim Jones
Cambridge, MA
r 726 0.192 0.006 2
To have the process_list command select (print) all records that do not have a city
field of Boston, type:
p1s students names -se1
Tim Jones
Cambridge, MA
"City not equal Boston"
Victor Red
Cambridge, MA 02139
r 727 1.053 2.682 4
To have the process_list command select (print) al1 records and save the resuit in
the segment names.list in the working directory, type:
pls students names -of
6-26
AZ98-02
sort_list (sIs)
The sort_list command sorts the records in the specified lister file. The records are
sorted according to the fields specified in the -sort control argument (see "Sorting"
above). Fields are sorted without regard to case; that is, they are sorted into
alphabetical order and not ASCII order.
SYNTAX AS A COMMAND
sIs path -control_arg
ARGUMENTS
path
is the pathname of the lister file to be sorted. The suffix lister must be the
last component of the lister file name; however, if path does not have a suffix
of lister, one is assumed.
CONTROL ARGUMENTS
control_arg
must be -sort st ring or -st sf ring to specify how the records in the lister file
are to be sorted (see "Sorting" above). If the command is invoked without
specifying this control argument, the sort_list command responds with a
two-line message showing proper usage.
EXAMPLES
To sort the list of patient records into ascending alphabetical order by the zip
field, type:
sort_list patients -st zip
To sort the list of patient records into descending alphabetical order (most recent
first) by the date field, type:
sort_l ist patients -st "date -dsc··
Normally, an alphabetical sort cannot be used to sort dates, but when the dates are of
the form YYMMDD (year, month, day), an alphabetical sort correctly orders the dates.
To sort the list of patient records into descending alphabetical order (most recent
first) by the date field, with those records having the same date sorted into ascending
alphabetical order by the lname field, type:
sort_jist patients -st ::date -dsc ,name -asc::
To sort the list of patient records into ascending alphabetical order by the lname
(last name) field, type:
sort_list patients -st "l name fname"
If some records have equal Iname fields, they are further sorted (ascending. alphabetically)
by fname (first name).
6-27
AZ98-02
To sort the list of patient records into ascending numerical order by the zip field
(zip code address), type:
sort_list patients -st "zip -num"
6-28
AZ98-02
trim_list (tIs)
The trim_list command deletes selected records from the specified lister file.
Because selection is required for trimming any lists, the select control argument must be
used with this command (see "Selection" above).
SYNTAX AS A COMMAND
tls path -control_arg {-optional_arg}
ARGUMENTS
path
is the pathname of the lister file being trimmed. The suffix lister must be the
last component of the lister file name; however, if path does not have a suffix
of lister, one is assumed.
CONTROL ARGUMENTS
control_arg
must be -select string or -sel string
deletion. This is a required argument.
to specify the records selected for
optional_arg
can only be -totals or -tt to display the number of records deleted.
EXAMPLES
To select (i.e., delete) from the file patients.lister all records that have an fname
(first name) field equal to John and an lname (last name) field equal to Smith. type:
trim_list patients -sel IIfname equal John and lname equal Smith"
To select from patients.lister all records that have an lname equal to Doe or Jones,
type:
trim_list patients -sel IIl name equal Doe or lname equal Jones"
To select from patients.1ister all records that have a state field equal to MA or IL,
and have a zip field that is not nUll, type:
trim_list patients -sel II {state equal MA or state equal IL)
and zip not equa 1 : nu 11"
To select from patients.lister all records that have a street field that contains the
substring "PO BOX". type:
trim_l ist patients -sel "street contains IIl1pO BOX"""
Notice the extra set of quotes required for the test string (see "Selection" earlier in this
section for description of quotation marks in -select_arg).
To select from patients.1ister all records that contain any field containing the
substring "PO BOX", type:
t 1spat i ents -se 1 ": any conta ins ""PO BOX 11I1I1
6-29
AZ98-02
APPENDIX A
COMPOSE METACHARACTER TABLE
This Appendix shows the byte value assignments for the extended character set
(metacharacters) used by compose while constructing the coded page image structure.
The characters defined in this table may appear in the image, both singly and in
various combinations. The device writer procedure decodes the charaters into printable
characters and control sequences acceptable to the target device.
000
001
002
003
004
005
006
007
010
011
012
013
014
015
016
(\1
VoL
'7
I
020
021
022
023
024
025
026
&
NUL
SOH
STX
ETX
EDT
ENQ
ACK
BEL
BSP
HT
NL
046
047
050
051
052
053
054
055
056
057
VI
061 1
0622
063 3
064 4
065 5
066 6
0677
070 8
071 9
072 :
073 ;
074 <
FF
CR
RRS
'Ql)~
.LI~"U.I
DLE
DC1
DC2
DC3
DC4
NAK
SYN
'
(
)
*
+
,
.
/
0600
114
115
116
117
120
121
122
123
124
125
126
127
130
131
132
L
M
N
0
P
134
135
136
137
140
141
142
\
]
075
=
i43
030
031
032
033
034
035
036
037
040
041
042
043
044
045
076
077
100
101
102
103
104
105
106
107
110
111
112
113
>
144
145
146
147
150
151
152
153
154
155
156
157
160
161
!
"
#
$
%
R
S
T
U
V
W
X
Y
Z
1 '1'1
rL
oL......
027 ETB
CAN
oct31
SUB
ESC
FS
GS
RS
US
SP
Q
?
@
A
B
C
D
E
F
G
H
I
J
K
A-I
1\
_
'
a
b
c
d
e
f
g
h
i
j
k
1
m
n
0
P
q
162
163
164
165
166
167
170
171
172
173
174
175
176
177
200
r
s
t
u
v
w
x
Y
z
{
!
J
DEL
"')(\1
.,VoL
202
203
204
205
206
207
210
21i
212
213
214
215
216
217
220
221
222
233
224
225
226
227
AZ98-D2
230
231
232
233
234
235
236
237
240
241
242
243
244
245
246
247
250
251
252
253
254
255
256
257
260
261
262
263
264
265
266
267
270
271
272
273
274
275
276
277
300
301
302
303
304
305
306
307
mlpy
pl_mi
nabla
EMd
slash
dagger
316
317
320
321
322
323
324
325
326
327
330
331
332
333
334
335
336
337
340
341
342
343
344
345
346
347
350
351
prlI
404 sup4
405 sup5
406 sup6
PI
tmark
tfore
appro x
infin
352 theta
per pen
not_eq
PAD
dbldag
cright
delta
310
311
312
313
314
315 bullet
353
354
355
356
357
360
361
362
363
364
365
366
367
370
371
372
373
374
375
376
377
400
401
402
403
pi
square
overbar
PS
supO
supl
sup2
sup3
A-2
.407
._.
~l1n7
-r'
410
411
412
413
414
415
416
417
420
421
422
423
424
425
426
427
430
431
432
433
434
435
436
437
440
441
442
443
444
445
446
447
450
451
452
453
454
455
456
457
460
461
462
463
464
465
466
467
470
471
sup8
sup9
EM
EM_
EN
EN_
ENd
THIN
DEVIT
lquote
rquote
multiply
modmark
daro
dbot
dIvert
delmark
drvert
dtop
laro
one {
one [
lcirc
one(
472
473
474
475
476
477
500
501
502
503
504
505
506
507
510
511
512
513
514
515
516
517
520
521
522
523
524
525
raro
526 1If1
one}
one]
rcirc
one)
uparo
527
530
531
532
533
534
535
536
537
540
541
542
543
544
545
546
547
550
551
552
553
554
555
556
557
[tp
[ht
[md
[bt
[hb
[fl
]tp
] ht
] md
] bt
] hb
] fl
{tp
{ht
{md
{bt
{hb
HI
}tp
} ht
}md
} bt
Jhb
} fl
lptp
lpht
lpmd
lpbt
lphb
lpfl
rptp
rpht
rpmd
rpbt
rphb
rpfl
ltp
lht
lmd
lbt
lhb
lfl
lItp
llht
llmd
llbt
llhb
art
art
/onehi
vrule
hstrt
hUne
hterm
lslnt
rslnt
boxt1
boxt
boxtr
boxl
box+
boxr
boxbl
boxb
boxbr
loztl
10ztr
10z1
10zr
lozbl
10zbr
AZ98-02
560
561
562
563
564
565
566
567
570
571
572
573
574
575
576
577
600
601
602
603
604
605
606
607
610
611
612
613
614
615
616
624
625
626
627
630
631
632
633
634
635
636
637
640
641
642
643
644
645
646
647
650
651
652
653
654
655
656
657
660
661
662
670
671
672
673
674
675
676
677
700
701
702
703
704
705
706
7(17
710
711
712
713
714
715
716
717
720
721
722
723
724
725
726
734
735
736
737
740
741
742
743
744
745
746
747
750
751
752
753
754
755
756
757
760
761
762
763
764
765
766
767
770
771
772
~1"7
£..£..'2
vv-'
"7"'''7
"7"7"t
620
621
622
623
664
665
666
667
730
731
732
733
774
775
776 CMODE
777 GMODE
V~I
I~I
A-3
II:>
AZ98-Q2
APPENDIX B
REFERENCE TO COMMANDS/SUBROUTINES
BY FUNCTION
This appendix contains the Multics commands and subroutines that are part of the
WORDPRO system, arranged according to function.
WORDPRO COMMANDS
compdv
Translates a device description file into a binary table for use by the Formatter.
compose (comp)
Prepares formatted documents from raw text segments for various documentation
devices using an extensive list of text formatting control lines and control arguments
to the compose command.
compose_index
Produces a cross-reference index file from raw data.
convert_runoff (cv_rf)
Converts a runoff input segment into a compose input segment.
display_comp_dsm (ddsm)
Displays selected information from a compose device description table.
expand_device_writer (xdw)
Expands an expansion input file into an expansion output file.
format_document (fdoc)
Prepares formatted documents from raw text segments using a limited set of text
formatting control lines.
process_compout (pco)
Processes one or more compose output files to an online device, or to a magnetic
or punched paper tape.
DICTIONARY COMMANDS/SUBROUTINES
add_dict_words (adw)
Adds words to a WORDPRO dictionary.
count_dict_words (cdw)
Counts words in a WORDPRO dictionary.
create_word 1 i st (cwl)
Creates a wordlist segment from a text segment.
delete_dict_words (ddw)
Deletes words from a WORDPRO dictionary.
B-1
AZ98-02
find_dict_words (fdw)
Finds words in the set of WORDPRO dictionaries defined by the search facility.
hyphenate_word_
Returns the character position at which a ,word can be hyphenated.
list_diet_words (ldw)
Lists words in a WORDPRO dictionary.
locate_words (lw)
Locates all occurrences of one or more words in a text segment.
print_wordl ist (pwl)
Displays the words in a wordlist segment
revise_words (rw)
Replaces all occurrences of one or more words in a text segment with a
corresponding revision.
trim_wordl ist (twl)
Deletes all words in a wordlist segment that can be found in a specified sequence
of dictionaries.
SPEEDTYPE COMMANDS
add_symbols (asb)
Adds symbols to the current symbol dictionary.
change_symbols (csb)
Changes the expansion or suffixing of a symbol in the current symbol dictionary.
delete_symbols (dsb)
Deletes symbols from the current symbol dictionary.
expand_symbols (esb)
Expands all the symbols in a specified text segment.
find_symbols (fsb)
Finds and lists symbols in the current symbol dictionary that represent specified
expansions.
1 i st_symbol s (1 sb)
Lists symbols in the current symbol dictionary.
option_symbols (osb)
Sets options in the current symbol dictionary.
print_symbols_path (psbp)
Displays the pathname of the current symbol dictionary.
reta i n_symbo 1s (rsb)
Retains all symbols in a specified text segment by placing a Speed type escape in
front of each symbol.
sho\.·,,_symbo 1s
(ssb)
Expands an input string and displays the output string.
use_symbo 1s (usb)
Sets the current symbol dictionary.
B-2
AZ98-02
LIST PROCESSING COMMANDS
append_list (als)
Adds a record to a lister file.
copy_l ist (cpls)
Creates a new lister file from an existing lister file.
create_list (c1s)
Creates a lister file from a listin file.
describe_list
(dls)
Displays information about a lister file.
d i sp 1ay _1 is t
(d i 1s)
Displays selected portions of selected lister records.
expand_l ist (els)
Creates a listin segment from a lister segment.
merge_list (mls)
Combines two lister files into a single lister file.
modify_list
(mdls)
Modifies a field or fields in selected lister records.
process_list (p1s)
Produces a document from selected records in a lister file.
sort_1 i st (s 1s)
Sorts the records in a lister file.
trim_l ist (tls)
Deletes selected records from the specified lister file.
B-3
AZ98-02
APPENDIX C
DEVICE SUPPORT TOOLS
DEVICE WRITER SOURCE EXPANDER
The Device Writer Source Expander is a special adaptation of a general text string
manipulation facility that expands a device writer source expansion input file into an
expanded device writer source output file. A device writer source expansion input file
is a mixture of literal text and expansion constructs. The corresponding expanded device
writer source output file contains the literal text, as-is, with the expansion constructs
replaced by their corresponding strings (if any) that may be compiled with the PL/I
compiler to obtain the device writer object module. Throughout the remainder of this
section, these two files are referred to simply as the input file and the output file
and the Device Writer Source Expander is referred to simply as the Expander.
The Expander provides these f ea.tures:
•
•
•
•
•
•
•
Variables and arrays with three data storage classes
Value assignment
Expression evaluation
Iteration
Conditional execution
Internal and external expansion calling
Active function calling
The language has intentionally been made very context-senslttve in order to allow,
as much as possible, literal text to be entered as it is to be generated.
EXPANSION CONSTRUcrs
Expansion constructs are made up of expansion tokens, white space (the ASCII
"motion" characters, SP, HT, LF, etc.), and literal text Throughout the remainder of
this section, expansion constructs are referred to simply as constructs,
Expansion tokens consist of an ampersand (&) followed by zero or more
alphanumeric characters followed by one non-alphanumeric character. There are two
types of tokens: keyword tokens and terminator tokens (See "Expansion Tokens" below
for a complete list of tokens). For some tokens, the non-alphanumeric character is
taken as part of the token; for others, it is considered part of the following input text;
for still others (if it is white space). it is discarded. Throu~h.out the remainder of this
section, expansion tokens are referred to simply as tokens.
Every construct must begin with a keyword token and each has a very specific
termination condition. There are four different classes of constructs, as determined by
termination conditions.
•
•
•
Self -terminating
Matching character terminator
General terminator token
C-l
AZ98-02
•
Specific terminator token
Constructs can be nested. When they are, the beginning and ending of each nested
construct must be totally within all containing constructs. Any expansions produced by
this nesting are "protected," that is, they do not change the existing syntax of the
original containing construct.
In the following descriptions. constructs are defined with the skeleton:
kkk bodyttt
where
kkk
is the keyword token
body
is the body (literal text. possible white space, and nested bodies and
constructS) of the construct and may be null. (This term is used in
conjunction with many of the Expander features.)
The white space shown in the skeleton above preceding the body is always
discarded and is not shown the definitions following. Any white space
within the body or between the body and the terminator is either discarded
or sent to the output file as the user directs. Literal text is sent to the
output file without modification.
The use of the term "body" in the descriptions following implies that it is
evaluated as an expression (see "Expression Evaluation" below).
ttt
is the terminator token and may be null
EXPANSION DEFINITIONS
An expansion definition begins with a definition keyword token and name. ends
with a specific definition terminator token. and has a body that is a mixture of literal
text. white space. and constructs consisting of any number of lines (including zero).
There are two forms: the static form (shown first) and the dynamic form.
&expandexpansion-nameexpansion-body&expend
&defineexpansion-nameexpansion-body&dend
The expansion-name may be up to 26 characters long, must begin with an alphabetic.
and contain only alphanumerics and "_t!. and are required characters.
represents one or more characters.
A static expansion definition may exist as all or an independent part (that is. an
unnested fragment) of an expansion input file or an expansion library file. An
expansion library file consists of only static expansion definitions with possible
interspersed commentary. It produces no output other than the commentary when
expanded from Multics command level since the commentary does not contain expansion
calls. Any expansion input file may be accessed as though it were an expansion library
file in order to use static expansions defined therein (as lOng a:s Lll~ naming
requirements are met). The files may be free-standing segments or archive components
and must have the name name.xdw.
A dynamic expansion definition may exist only as a nested construct within a
static expansion definition or an expansion input file.
C-2
AZ98-Q2
The essential difference between the two forms is that the use of a static expansion
yields the expansion-body as given but the use of a dynamic expansion yields a static
expansion definition having the expanded name and expansion-body of the dynamic
expansion.
Any attempt to redefine a static expansion (by either of these two forms) in the
same invocation of the Expander is an error; however, if the new definition is
character-for-character identical with the existing definition, the attempted redefinition
is ignored.
Examples
The static form:
&expand ck
if (code "= 0)
then do;
ca 11 com_er r _ (code, lI&name ll , &1) ;
return;
end;
&expend
Using this static expansion yields the five PL/I code lines with the two nested
constructs expanded and including all the white space.
The dynamic form:
&expand exp_def
&define A
;&dend
&define B
j&dend
&expend
Using the static expansion d f yields the two new static expansions:
&expand A
j&expend
&expand B
;&expend
V ARIABLES AND ARRAYS
A variable declaration construct begins with a declaration keyword token and a
name, ends with a general terminator token, and may contain array extents and/or an
initial value. There are three data storage classes, each with its own keyword token:
Local, like PL/I automatic; keyword token "&loc"
Internal, like PL/I internal static; keyword token "&int"
External. like PL/I external static; keyword token "&ext"
All variables must be declared before they can be
available only for the current invocation of the expansion
Internal data are available at any time after declaration
expansion in which they are declared. External data are
declaration in any expansion file.
referenced. Local data are
in which they are declared.
for any invocation of the
available at any time after
Variable names may be up to 16 characters long, must begin with an alphabetic
character. and contain only alphanumerics and ':' ", There is no conflict between
C-3
AZ98-02
variable names and expansion names because of the syntax; however, scalar variable
names and array variable names conflict. When searching for a variable name, the
Expander searches the data classes in the order shown above.
Variable values are 9-bit byte character strings with a minimum size of zero and
a maximum size of 1,044,480 (as determined by maximum segment size). Numeric
values may be only decimal numbers (that is, no coded exponential forms like 2.4e4 or
10**3), have a maximum magnitude of 10848 power and a resolution of nine decimal
places. - Variables declared without initial values are initialized with a null string.
Throughout the remainder of this section, assign means that a value is given to the
variable and access means that the value of the variable replaces the reference.
An attempt to redeclare an existing variable with different attributes is an error;
however, a redeclaration with identical attributes is ignored.
All the declarations described below can define variables in any of the classes,
however. the declarations and examples show only Local variables. that is, use the
"&loc" keyword tokens. In any of them, "&int" or "&ext" may be substituted for the
n&loc".
Scalar Variables
&loc name&;
&loc name=initial-value-body&;
Scalar variables may be declared with or without initial values. If initial-value-body
is given. it is assigned as the initial value of name.
Examples
&loc stuff&;
deClares a Local scalar with no initial value.
&loc one=l&;
declares a Local scalar with an initial value.
&loc copy_it=&it&;
declares a Local scalar with the current value of another variable as its
initial value.
Array Variables
In this section, the terms "scalar reference", "subscripted reference", and "array
reference" are used in discussing references to an array variable. The definitions of
these references are given below.
Note:
Throughout this section, the braces (n) shown in reference to array
variables are required as part of the construct syntax and do not mean
that the enclosed expr is optional.
scalar
subscripted
array
&name
&name {expr J or &nanie {expr 1 :expr2}
&name{}
The term "expr", appearing here for the first time, refers to a construct that is
evaluated as an arithmetic expression (see "Arithmetic Expressions" below).
C-4
AZ98-02
The limit for the upper and lower bounds of arrays is 34,359,738,367 as determined
by the maximum positive binary integer. The limit for the extent of arrays is 130,558
as determined by maximum segment size.
Any reference to any array element outside the declared extent is an error. An
array access to an array with an empty extent is replaced with a null string.
FIXED ARRAYS
&loc name {expr 1:expr2} &;
&loc name {expr 1:expr2} =initial-value-b6dy &;
Fixed arrays may be declared with or without initial values and have non-varying
extents as determined by the upper and lower bounds given in their declarations. expr 1
specifies the lower bound and expr2 specifies the upper bound. If initial-value-body
is given, it is assigned to each element of the array being created. Fixed array elements
are assigned and accessed with subscripted or array references.
Examples
Sloe ten_nulls{l:lO}&;
declares a Local fixed array with 10 null elements.
Sloe fifty_Ss{1:S0}=S&;
declares a Local fixed array containing 50 elements with initial value 5.
Sloe holders{&first:&last}&;
declares a Local fixed array whose extent is determined by the current
values of other variables.
VARYING ARRAYS
&loc name {expr 1:expr2} var&;
A varying array is like a fixed array except that it must be declared without initial
values and the upper and lower bounds are adjusted dynamically as elements are
assigned values. expr 1 specifies the minimum lower bound and expr2 specifies the
maximum upper bound. When created, the array is empty with no extent Varying
array elements are assigned and accessed with subscripted or array references, but an
attempted access outside the current extent is an error.
Examples
Sloe some_Os{1:2S}var&;
declares a Local varying array to hold up to 25 elements.
&loe twoway_array{-&size:&size}var&;
declare a Local varying array whose maximum extent is one more than
twiCe the valUe of some other variable.
LIST ARRAYS
&loc name {expr} list&:
A list array is a set of unique elements and must be declared without initial values.
expr is the maximum number of elements the list is to hold. When created, the list is
C-5
AZ98-o2
empty with no extent. A list assignment is made with a scalar reference; the list is
searched to see if the given value is there and it is added if the search fails. A list is
accessed with subscripted or array references, but an attempted access outside the
current extent is an error. Lists are ordered according to the order in which
assignmen ts are made.
Example
&loc et_dcls{20}list&;
declares a Local list array that can hold 20 entries.
STACK ARRAYS
&loc name {expr} fifo&;
&loc name {expr} lifo&;
A stack array may be either a push-down/pop-up stack (last-in-first-out or lifo)
or a linear delay queue (first-in-first-out or fifo) and must be declared without initial
values. expr is the maximum number of elements the stack is to hold. A stack
assignment is made with a scalar reference. the given value being added as the newest
element A scalar access to a fifo stack causes the oldest element to be accessed and
deleted. A scalar access to a lifo stack causes the newest element to be accessed and
deleted. A stack array may also be accessed with subscripted references. but these
references cause no "movement" of the stack. The subscript value 0 accesses the
top-of -stack (or next-out) element. -1 accesses the next-to-top element, etc. An array
access to a stack is an error.
Examples
&loc push_stack{25}lifo&;
declares a Local push/pop stack that holds up to 25 entries.
&loc queue{lO}fifo&;
declares a Local queue with 10 elements.
V ALUE ASSIGNMENT
&let name=val ue- body &;
&let name {expr} =val ue- body &;
&let name {expr 1:expr2} =value-body &;
An assignment construct begins with an assignment keyword token and a name,
ends with a general terminator token, and may contain array subscripts or ranges
and/ or a value. A value may be assigned to a scalar, an array element, or a range of
array elements. If a range is specified, value-body is assigned to every array element
in the range.
Examples
&let feet_per_mile=5280&;
assign a value to a scalar.
&let var{2}=&var{1}&;
assigns the value of the first element of the array to the second element. •
&let array{1:5}=3&;
assign "3" as the value of the first five elements of array.
C-6
AZ98-Q2
EXPRESSION EVALUATION
An expression is a collection
expressions, literal text, and possible
a single character string representing
a string value or numeric value that
of constructs, variable accesses, other embedded
white space that is replaced by the Expander with
its value. The result of evaluating an expression is
must obey the limits mentioned earlier.
Accessing Variables
Variables are accessed by using their names as though they were keyword tokens.
The termination conditions for the constructs thus created depend on the form of
ref erence and are specified in the descriptions following.
SCALAR ACCESSES
&name
The keyword token becomes a self -terminating construct, but the construct
terminator (the non-alphanumeric character following the token) may not be n(" or "{".
The construct is replaced by the value of name. This form of access may be made to
scalars and stack arrays. Ref erencing a stack array causes the accessed value to be
removed from the stack. (See "Stack Arrays" above.)
Examples
&let Var=foo&;
then:
&Var any literal text •••
becomes foo any 1 i tera 1 text ..•
if &Var>O
becomes iff 00>0
&Var&.bar
becomes foobar
&Var&.(l)
becomes foo (1)
SUBSCRIPTED ACCESSES
&name {expr}
&name {expr 1:expr2}
&name {expr 1:expr2,string-body}
The non-alphanumeric character following natne must be "{" and becomes part of
the keyword token. The token begins a construct that is terminated by the matching
tI} tI. expr or expr 1 and expr2 are evaluated (see "Arithmetic Expressions" below) to
obtain the element or range of elements to be accessed.
The first form above may be used to access all array types and the construct is
replaced by the value of the selected array element.
Examples
C-7
AZ98-02
&fixed_array{S}
is replaced with the value of the fifth element of f i xed_ar r ax
& 1 is t {& 1as t}
is replaced by the value of that element of 1 i s t whose list position is
given by the value of 1as t .
&stack{O}
is replaced by the value of the next element to be recovered from the
stack.
The second form may be used to access fixed, varying, and list arrays and the
construct is replaced by the list of values of the selected range of array elements,
separated by a single blank character. A subscripted access to any unassigned element in
the declared extent is replaced with a null string.
Examples
&varying_array{-2:3}
is replaced with the six elements of varying_array whose subscript values
lie between -2 and 3, inclusive, separated by single blanks.
&1 ist{l:&last}
if 1as t contains the extent of 1 is t, it is replaced by the entire contents
of 1 i s t separated by single blanks.
The third form may be used to access fixed, varying, and list arrays and the
construct is replaced by the list of values of the selected range of array elements,
separated by string-body. The length of string-body is limited to 150 characters.
Literal appearances of n&" and "J" in string-body must be protected (see "Protected
Strings" below). If the selected range is empty, the construct is replaced with a null
string.
Example
&A list{l:&last" }
if 1as t contains the extent of A_l is t, it is replaced by the entire
contents of A_l is t separated by the string ", n.
ARRAY ACCESSES
&nameD
&name {,stri ng- body}
Array accesses are a special case of subscripted accesses where the subscript
expression is given as a null string rather than being evaluated to a null string, implying
an empty range. The usage of array accesses is identical to the second and third forms
of subscripted accesses above except that the range is the entire extent of the array.
An array reference to a varying array, a list , or a stack addresses the current extent,
not the declared extent.
Example
&A_i i stt"
}
is replaced by the entire contents of A_l is t, separated by the string "
(Note that result of this access is the same as that of the previous example,
but does not depend on the value of some other variable.)
C-8
AZ98-02
Accessing Arguments
An expansion may be called with a list of arguments to be used as parameters.
(See "Expansion Calling" below.) The called expansion may access these arguments with
argument access expressions. Arguments are accessed by using their argument list
position numbers as though they were keyword tokens. The termination conditions for
the constructs thus created depend on the form of reference and are specified in the
descriptions following.
SINGLE ARGUMENT ACCESSES
&n
&nn
The token forms a self -terminating construct that is replaced by the value of the
argument having the given position in the argument list. The construct terminator is the
first non-numeric character and the number may not have more than two digits. If the
reference is to an argument beyond the argument list, the construct is replaced with a
null string.
Examples
&3
is replaced with the value of the third argument.
&05
is replaced with the value of the fifth argument.
&14
is replaced with the value of the fourteenth argument
MULTIPLE ARGUMENT ACCESSES
& {expr}
& {expr 1:expr2}
& {expr 1:expr2,stri ng-body}
At times it is necessary to reference an argument via the value of a variable, or to
reference more than one argument. This is done by accessing the argument list as
though it were a nameless array (see "Array Accesses" above). The keyword token is
n& {n and begins a construct that is terminated by a matching til
ii.
Examples
&{&arQ_counter}
is replaced by the argument whose list position is given by arQ_counter.
&{2: 4}
is replaced by a list of the values of the second. third, and fourth
arguments, separated by a single blank.
&{1:3,
+ }
is replaced by an expression representing the sum of the first three
arguments.
C-9
AZ98-o2
dec 1are & {, fixed bin (17);
declare} fixed bin (17);
creates PL/I declarations for all the arguments.
ARGUMENT COUNT
&*
This keyword token forms a self-terminating construct that is replaced by the
number of elements in the argument list with which the expansion was called.
Example
& {&~,:}
is replaced with the value of the last argument regardless of how many
have been given.
Protected Strings
&&
This keyword token forms a self-terminating construct that is replaced by a single
ampersand.
Example
if flag && index> 0 then do;
creates a PL/I logic test statement.
&nstring-body&"
The keyword token is n&"" and it begins a construct that is terminated by the next
occurrence of the same keyword token and protects any literal string. The construct is
replaced by the literal, unexpanded string-body. string-body may not contain an
em bedded protected string.
Example
{&array{,&II}&1I O}
forms a blank separated list of array elements, each enclosed in braces.
Arithmetic Expressions
&!expr}
The keyword token is "&(" and begins a construct that is terminated by the
matching ")". expr may contain only decimal numeric literals, embedded arithmetic
expressions, and arithmetic and relational operators. expr is first expanded as an
expansion expression and then evaluated as an arithmetic expression. The value of the
arithmetic expression replaces the construct.
are:
+
addition
subtraction
/
division
C-IO
AZ98-02
*
multiplication
o
factor grouping
Examples
&let array{3}=&(&array{2}+1)&;
assign the third element of the array a value that is one greater than the
second element.
&array{&(2*&2+1) :&(2*&3+1)}
access a range of array elements given by the values of the second and
third arguments where the subscript expression is "2N+l".
&(&array{,+}) &;
is replaced with the sum of all the elements of the array.
Note:
This construct fails if the array is not fully populated, that is,
if it contains any null elements, since the resulting summation
construct contains a double operator that gives rise to a missing
operand error (see next example.) The result of summing an
empty array is a null value.
&(&array {,+O}) &;
the digit "0" ensures success of the construct by representing any null
elements with "+0".
The relational operators have lower precedence than the arithmetic operators, that
is, within a factor group, all arithmetic is completed before any. relations are tested.
The result of a relational test is given a numeric value "0" representing "false" or a
numeric value "1" representing "true". In the evaluation of a relational test, any term
with a non-zero value is considered true.
The relational operators supported are:
equal
A=
not-equal
>
greater
<
less
<=
less-or-equal (not-greater)
>=
greater-or-equal (not-less)
Examples
& (& 1>0)
is replaced with "1" if the first argument is positive; otherwise, it is
replaced with "0".
&(&array{1}A=0)+(&array{2}A=0»
is true if either (value = "1") or both {value = "2"} of the first t\VO
elements of array are non-zero and false (value = "0") if both are zero.
(Note here that the addition operator takes on the role of the Boolean OR
operator.)
&let flag=&((&lA=0)*(&2 =0)*(&3 =0»&;
assigns "0" to flag if any of the first three arguments is zero and "1" if
A
A
all three are simultaneously non-zero. (Note here that the multiplication
operator takes on the role of the Boolean AND operator.)
C-l1
AZ98-02
ITERATION
An iteration construct begins with an iteration keyword token, ends with a specific
construct terminator token, and contains a two-part iteration body and a test clause.
The test clause begins with a test keyword token, ends with a general terminator token,
and contains a test body.
&do body 1 &while test- body &; body2 &od
Any of body 1 , body2, or test-body may be nUll. however. if test-body is nUll. it
is considered absolutely true and the iteration never terminates. In order to establish
effective control over the iteration, either body 1 or body2 must modify the condition
tested by test-body.
test-body may be an arithmetic relational expression as described in "Expression
Evaluation" above, or may be a string expression of either of the forms:
st r i ng- body
string-body1 RELOP string-body2
where RELOP is any of the relational operators discussed in "Expression Evaluation"
above. string-body is considered false if it has any of the values "0". "F", "FALSE",
or "NO" (without regard to case); any other values are considered true. For the
purposes of comparison, the shorter of string-body1 and string-body2 is padded out
to the length of the longer with ASCII blanks and the values of the characters are
determined by the ASCII collating sequence. In the first form, white space is stripped
from both sides of string-body. In the second form. white space is stripped from the
left sides of the iestring-bodyies but is retained on the right sides.
The flow of control in the iteration proceeds as follows:
1.
body 1 is executed.
2.
test-body
is evaluated. If it is false, control proceeds to the construct
following the iteration construct terminator. If it is true. control proceeds to
step 3.
3. body2 is executed and control goes back to step 1.
Example
&1et vv=&"c&;
&do
(& {&vv}) &+
&let vv=&(&vv-l)&;
&while &(&vv>O)&;
&od
creates a parenthesized, comma-separated list of all the arguments with the
order of the arguments inverted. Note that an array access to the argument
list can be used to create a similar list with the argument in their given
order.
Note: The token &+ is a white space control token and serves only to
improve the readability of the expansion input file. See
"tv1iSCellaneou5 Features" helo'.'.'.
C-12
AZ98-o2
CONDITIONAL EXECUTION
A conditional execution construct begins with a conditional keyword token and a
test clause, ends with a specific construct terminator token. must contain a "then"
clause, may contain any number of "elseif" clauses, and may contain a single "else"
clause.
&if test-body &then then-body
&elseif test-body &then then-body
&else else-body
&fi
The test-body is the same as that for the iteration construct described above.
Note:
The format shown for this construct (multiline with indents) is for
clarity of presentation only and is not required for correct usage.
If the evaluation of any test- pody results in a true value, then the corresponding
then-body is executed and the the rest of the construct is skipped. If no test-body is
true, then the else-body is executed if present.
Examples
&if &(&(&1»0) &then &let sign=+&;
&else &let sign=-&;
&fi
captures the arithmetic sign of the first argument (Note here that the
nested arithmetic expression in the test clause ensures that the argument is
handled correctly if it is an expression rather than a value.)
&if &(&)'«2) &then
&error 4,Second argument missing.&; &return
&fi
reports a calling sequence error and returns to the caller.
Note:
See "Miscellaneous Features" below for descriptions of the
&error and &return constructs.
EXPANSION CALLING
Expansions are called (that is, execution control passed to them) by using their
names as though they were keyword tokens.
&expansion-name(arg-body1,arg-body2, ... }
The non-alphanumeric character following expansion-name must be "(" and
becomes part of the keyword token. The token begins a construct that is terminated by
the matching ")".
expansion-name may be either name or segment$name. The Expander keeps an
internal list of all expansions it has encountered during execution. When a reference to
name is made, the list is searched for that name. If the search fails, name is
promoted to name$name and an external search for that name is made using the
expansion search list. An explicit reference to segment$name causes the Expander to
forego searching the internal list and make a direct external reference to the segment,
again using the expansion search list if the segment is not known.
Up to 99 string arguments may be passed in the call, and each is limited to 500
characters after leading white space is discarded.
C-13
AZ98-02
Any",". "(", or ")" characters resulting from the expansion of an arg-body are
literal characters; that is, they do not contribute to the syntax of the call construct.
If any arg- body is enclosed in parentheses, then it is considered a list argument,
that is, a parenthesized list of values passed as a single argument.
Examples
&a_exp (abc, def)
calls expansion a_exp with the arguments "abc" and
&a exp(&
abc,def&")II
calls expansion a_expwith the argument "abc,def".
"def~'
&a_exp((abc,def»
calls expansion a_exp with the argument" (abc, def)".
&let var=abc,def&;&a_exp(&var)
calls expansion a_exp with the argument "abc, def".
&a_exp(&2,&b_expO)
calls expansion a_exp with two arguments; the second argument of the
current expansion and the expansion of "b_exp'~
&let name=&strip_suffix(&entry,.pll)&;
calls expansion str i p_suff i x with two arguments and assigns the resulting
expansion to the variable name,
ACTIVE FUNCTION CALLING
& [active-expr]
The keyword token is "& [It and begins a construct that is terminated by the
matching n]
active-expr and the active function return string is limited to 500
characters. active-expr is first expanded as an expansion expression and then processed
as an active function. The active function return string replaces the construct.
fl.
Examples
This file created by &[user person] on &[date] at &[time].
generates an audit trail time-stamp.
The path is &[string [dir &l]>[file &2]].
generates an audit trail pathname.
MISCELLANEOUS FEATURES
The miscellaneous features discussed in this section are presented alphabetically and
listed below.
•
~
•
•
•
•
•
Built-in functions
Comments
Emptying Arrays
Error Reporting
Expansion Debugging
General Terminator Token
Null Separator Tokens
C-14
AZ98-Q2
•
•
•
•
Quote Processing
Rescanning
Return
White Space Control
Built-in functions
Three built-in functions are provided.
•
•
•
Length
Substr
Usage
LENGTH FUNCTION
&length string-body &;
The keyword token begins a construct that ends with the general terminator token
and contains a string. The construct is replaced by the number of characters in
string-body.
The function is supported internally (rather than requiring an active function call)
because of its expected .high frequency of use and because the string may contain white
space.
Example
&length &1&;
is replaced by the number of characters in the first argument.
SUBSTR FUNCTION
&substr string-body, expr1 &;
&substr string-body, expr1, expr2&;
&substr string-body, expr1: expr2&;
The keyword token begins a construct that ends with the general terminator token
and contains a string and one or two subscript expressions. The length of string-body
is limited to 16384 characters. Both exprs must refer to character positions within
string-body or the input file is in error.
This function is supported internally (rather than requiring an active function call)
because of its expected high frequency of use and the extended capabilities provided.
The first form above is replaced by that part of string-body from character
position expr 1 to the end. If expr 1 is negative, then the character position is
calculated from the end of string-body rather than from the start
Examples
&substr abcdefg,3&;
is replaced by cdef g.
&substr abcdefg,-3&;
is replaced by e f g.
The second form is replaced by that part of string-body from character poslt1on
expr 1 for a total resultant string length of expr2. If expr 1 is negative, then the
C-15
AZ98-02
character position is calculated from the end of string-body rather than from the
start
I,f the number of characters in string-body following the calculated character
position is less than the magnitude of expr2, the resultant string is padded to the
required length with ASCII space (SP) characters. If expr2 is negative. the padding is
to the left of the resultant strimg; otherwise, it is to the right If no padding is
needed, then the sign of expr2 is immaterial.
Examples
&substr abedefg,2,3&;
is replaced by bed.
&substr abedefg,2,-3&;
is replaced by bed.
&substr abedefg,3,5&;
is replaced by edefg.
&substr abedefg,3,8&;
is replaced by edef g.
&substr abedefg,-3,8&;
is replaced byefg.
&substr abedefg,-3,-8&;
is replaced by efg.
The third form is replaced by that part of string-body from character position
If either expr is negative, then the corresponding
character position is calculated from the end of string-body rather than from the
start Both exprs must refer to character positions within str; ng- body or the input file
is in error. Further, epxr2 must refer to a character position to the right of that
given by expr 1.
expr 7 to character position expr2.
Examples
&substr abedefg,3:5&;
is replaced by ede.
&substr abedefg,-3:-2&;
is replaced by e f .
USAGE FUNCTION
&usage ioa-etl-stri ng&;
This function provides a means of documenting the expansions that are used in the
generation of an expansion output file. In essence, it is a means of dumping the
Expander's internal expansion reference list (see "Expansion Calling" above) in a format
determined by the user. It should be used only in "primary" expansion files (that is,
files intended for use in the command line invoking the Expander ) and not within any
expansion definition. Further. it should be the last construct in the file so as to not
lose any references.
ioa-etl-stri ng is an i oa_ control string that describes the format of the output
(see Multies Subroutines for a description of i oaJ. It is passed to i oa_$ r snn 1 with
three string arguments each of which must have a string conversion key Ca) in the
control string. The three arguments are (in the order passed):
C-16
AZ98-02
•
•
•
the pathname of the directory contalnmg the macro input file
the entryname of the expansion input file
the name of the expansion
Every expansion used appears once in the display and the order is the "natural"
order, that is, the order in which the reference first appeared.
Example
&usage /* ~a>~a -- ~a */A/&;
generates a list of all expansion file pathnames and expansion names as
PL/I comments at the end of a generated PL/I source file.
Comments
&comment comment &;
The keyword token begins a construct that ends with the general terminator token
and contains a comment. The comment is treated as a literal string; it is not expanded
and does not contribute in any way to the processing of the expansion.
Note:
In this single case, the general terminator token does not change the
existing white space supression action (see "White Space Control" below).
Emptying Arrays
&empty array-name&;
The keyword token begins a construct that ends with the general terminator token
and contains an array name. The array is emptied by setting its extent to zero and all
its elements to' null.
Error Reporting
&error sev-expr, err-body&;
The keyword token begins a construct that ends with the general terminator token
and contains a severity expression and an error message body. sev-expr must be an
arithmetic expression in the range 0-4 and is used to select one of the message forms
shown below. The formatted messages are written to the error_output I/O switch.
The error message forms are:
o
for the user's information
NOTE: EXPANSION , line .
1
a minor error that does not affect the validity of the output
WARNING EXPANSION , 1 ine .
2
a substantive error that causes the output to be invalid
ERROR SEVERITY 2 EXPANSION , 1 ine .
C-17
AZ98-02
3
a major error that prevents creation of the expansion output file but allows
processing to continue in order to report additional errors
ERROR SEVERITY 3 EXPANSION , 1 ine .
4
a fatal error that prevents further processing of the expansion input file
ERROR SEVERITY 4 EXPANSION , 1 ine .
Examples
&error O,This code does not reference any error_table_ entries.&;
&error 1, Second argument miss i ng, 1113" assumed&;
&error 2,Source syntax error. Program will not compile.&;
&error 3,Required sections not supplied.&;
&error 4,Table name not suppl ied.&;
General Terminator Token
&;
This token is used to signal the logical end of various other constructs. It does not
contribute directly to the expansion output file.
Null Separator Tokens
&.
This token acts as a terminator token and enables the copying of white space
within expansion constructs into the expansion output file. It is used to resolve
ambiguities that might otherwise exist and to allow expansion constructs to create white
space in the expansion output file. All white space between it and the next token is
copied to the expansion output file; however. white space in any comments encountered
is discarded as part of the comments.
&+
This token disables the copying of white space within expansion constructs into the
expansion output file. It is used to suppress the copying of white space intended solely
to improve the readability of the expansion input file. All white space (and comments)
between it and the next token is discarded.
Examples
&a_exp (A, 1)
&2&.0
the third construct is replaced by 10. Were the null separator token not
present, the construct \l,'ould be a reference to (nonexistent) argument 20 of
the expansion ca:ll.
C-18
AZ98-G2
& if
•••
loop: do
end loop;&fi&.
else
the white line is copied to the expansion output file to separate the ''end''
and '~1 se" statements.
&if
•.. of the people,&+
&comment end of fragment 1&;
&fi by the people, .•.
the expansion output file contains "
...."
of the people, by the people,
Quote Processing
The Expander is internally language-independent. However, because it can communicate
with the Multics operating system and may be used to generate source code for
languages supported by Multics. it must be able to manipulate quoted strings in a
manner consistent with that expected by Multics. A quoted string is any string of
characters enclosed within ASCII double-quote ~I) marks and, for this usage. limited to
16384 characters.
"e body &;
The keyword token begins a construct that ends with the general terminator token.
The .result of the construct is a string with all quote marks doubled. Note that body is
not converted from an unquoted string to a quoted string.
&unquote body&:
The keyword token begins a construct that ends with the general terminator token.
The result of the construct is a string with all doubled quote marks reduced to single
quote marks. If body is a quoted string, it is converted to an unquoted string.
Examples
Processed on : &unquote &[date_time]&;
strips the quote marks from the string returned by the date_time active
function.
ca 11 my _proc (1I"e &s tr i ng_arg&; ") ;
ensures that any quote marks within s tr i ng_arg are correctly passed to
the procedure.
Rescanning
&scan body &:
The keyword token begins a construct that ends with the general terminator token.
In this construct. body is expanded normally and then the resulting expansion is
re-expanded as though it were another body. Normally. any constructs appearing in an
C-19
AZ98-02
expansion are "protected"; that is. they are not subjected to further expansion. In some
applications, it is necessary that any such constructs be expanded.
Examples
&exp_l ("a,b,&[time] ,d")
exp_l is expanded with the single argument a, b, &[t i me] ,d and contains
any of the following expansion calls.
&exp_2(&1)
exp_2 is expanded with one argument, a, b, &[t i me] ,d.
&exp_2(&scan &1&;)
exp_2 is expanded with one argument. a, b, 08: 21 , d.
&scan &&exp 2(&1)&;
exp_2 is ~xpanded with four arguments, q b, 08: 21 , and d.
Return
&return
The keyword token becomes a self -terminating construct that causes an immediate
hal t of processing of the current expansion.
Example
&if &(&*=0) &then
&error 2,No arguments, call ignored.&; &return
&fi
terminates the processing of an expansion if no arguments are given.
WHITE SPACE CONTROL
White space is any of the ASCII motion characters; HI, SP, NL, VT. and FF.
These characters are normally discarded when they appear as shown below; however.
they may be preserved by use of the null separator tokens discussed earlier.
1.
After the expansion tokens:
&+
&error
&;
&fi
&do
&if
&else
&length
&elseif
&od!&then
"e!&unquote
&scan!&usage
&substr!&while
2.
After II ( " and I I , II in an expansion call argument list (at level 1, that is,
outside all nesting due the parenthesis usage).
3.
After the ")
4.
After "=" in &1et, &1oc
II
in the expansion of
,&
II
&(expr) I~
i nt , and&ex t .
EXPANSION TOKENS
As mentioned briefly above, expansion constructs fall into four different classes as
determined by their termination conditions. This section lists the tokens that form
C-20
AZ98-02
constructs in each of the four classes and then gives a sorted list of all tokens for
quick reference.
Self -terminating Constructs
&&
&*
&+
&.
&;
&n
&nn
&return
&name
literal &
number of arguments given
begin white space skipping
end white space skipping
general terminator token
argument reference (constant)
argument reference (constant)
expansion return
variable reference
Matching Character Terminator Constructs
&" body &"
& [ body]
&expansion-name( body )
&name{ body J
&{ body J
protected string
active function call
expansion call
array reference
parameter reference (index or list)
General Terminator Token Constructs
&comment string &;
&empty name &;
&error body &;
&length body &;
&let body &;
&loc body &;
&int body &;
&ext body &;
"e body &;
&scan body &;
&substr body &;
&unquote body &;
&usage body &;
&while body &;
comment
array emptying
error message generator
string measurement
variable value assignment
local variable declaration
internal (static) variable declaration
external (static) variable declaration
quote-mark duplication
construct rescanning
character substrings
quote-mark reduction
expansion usage reporting
do group control clause
Specific Terminator Token Constructs
&expand body &expend
&define body &dend
&do body &od
&if body &fi
&else body &fi
&elseif body XXX
&then body XXX
static expansion definition
dynamic expansion definition
limited or repetitive execution group
conditional execution group
if group control clause
if group control clause (XXX may be &elseif,
&else, or &f i)
if group control clause (XXX may be &elseif,
&else, or &f i)
C-21
AZ98-02
Sorted Token List
The following is a sorted list of all tokens without regard to class or usage.
&"
&&
&*
&+
&.
&;
&[ ]
&{ }
&comment
&define
&dend
&do
&else
&elseif
&empty
&error
&expand
&expansion-name( )
&expend
&ext
&fi
&if
&int
&length
&let
&loc
&n
&nn
&name
&name{ }
&00
"e
&return
&scan
&substr
&then
&unquote
&usage
&while
protected string
literal &
number of arguments given
begin white space skipping
end white space skipping
general terminator token
active function call
parameter reference (index or list)
comment
dynamic expansion definition
dynamic expansion definition terminator
limited or repetitive execution group
if group control clause
if group control clause
array emptying
error message generator
static expansion definition
expansion call
static expansion definition terminator
external (static) variable declaration
if group terminator
condi tional execution group
internal (static) variable declaration
string measurement
variable value assignment
local variable declaration
argument reference (constant)
argument reference (constant)
variable reference
array reference
do group terminator
quote-mark duplication
expansion return
construct rescanning
character substrings
if group control clause
quote-mark reduction
expansion usage reporting
do group control clause
Reserved Words
It is apparent that all the keywords in the lists above are reserved words and may
not be used as variable names. In addition, there are a few others that are reserved for
future extensions of the Expander. The complete list is shown below.
arg
comment
define
dend
expand
expend
ext
fi
let
loc
macro
member
C-22
substr
then
trace
unquote
AZ98-02
do
else
elseif
empty
error
hbound
if
int
1bound
length
mend
od
quote
return
scan
usage
whi Ie
Annotated Example
The following is an example of the definition, use, and result of an expansion that
could aid a PL/I programmer in managing references to the Multics system error_table_.
The expansion is called with an error_table_ entry name each time a reference to the
entry is wanted. All the different entry names are saved in a list variable and a final
call to the expansion without an entry name returns the PL/I declaration list for all
the en tries used.
&expand et_
Define the et_ expansion.
&int et_list{SO}list&;
Declare the error_table_ list Note that repeated
executions of this have no effect due to the
identical attribute feature of variable declarations.
&if &(&":=0)
If no argument is given, generate the error_table_
&then
declaration list with an array reference to et_list
dcl error table $&et {, fixed bin(3S)ext static;
dcl error=table=$} fixed bin (3S) ext static;
&e 1se
However, if there is an argument, add it to et_list
&1et et 1 is t=& 1&;
if it is not already there.
error _table_$&l
Return -the -error~table_ -reference string-for use in
the PL/I program.
&fi&expend
End if group and expansion definition.
Next, assume an expansion to generate PL/lsource code that contains the following
fragments.
if (code = &et_(badarg))
then code = &et_(notfound);
code = &et_(badarg) ;
Finally, when the above fragments are expanded, the following PL/I code results.
if (code = error_table_$badarg)
then code = error_table_$notfound;
dcl error_table_$badarg fixed bin(3S)ext static;
dcl error_table_$notfound fixed bin (3S) ext static;
end;
C-23
AZ98-02
DEVICE WRITER
The device_writer_ is the object segment that contains the procedure to convert
the coded output page image created by the Formatter into the character stream needed
by the output device. It operates as an external subroutine of the Formatter and, as
such, is required to conform to certain conventions and restrictions. To ensure this
conformance, a "skeleton" procedure is provided by a static source expansion named
comp_dev_writer.xdw (see "Device Writer Source Expander" above).
The expansion provides a uniform coding style, defines all required entrypoints and
their interfaces, makes certain error checks, includes the structure declarations for all
the Formatter internal data bases needed. defines a number of external expansion
variables (some with default initial values) that the user can change to affect the action
of the Expander in generating output. and, most importantly. provides empty code
fragments for all image-to-character-stream conversions that require explicit knowledge
of the output device.
The code fragments are all defined as external expansion variables with null initial
values. In order to activate any fragment, the user need only reassign the variable value
with some PL/I code sequence. The expansion value assignment statements are written
in a file named device writer .pI1.xdw. The final statement in the file must be an
Expander call to comp_dev_writer.xdw.
The result of expanding device_writer _.pI7.xdw is a PL/I source segment named
device _writer _,pIT. There are comments throughout that indicate where built-in pieces
of code have been selected by the value of &devclass. Comments also introduce
user-supplied code fragments. Users must acquaint themselves with the general structure
of the writer in order to write compatible code fragments.
Note:
This description is a first attempt to document a complex and sophisticated
software development tool. It is difficult to determine. a priori, the level
of detail that should be included. Therefore, the interested reader must
study the released files for the Honeywell-supported devices to gain a full
understanding of what is required in the creation of a device_writer_
module.
Variables and Code Fragments
The following describes the external expansion variables whose values may be set by
the user. The descriptions are shown as expansion constructs with a requirement/default
comment and a descriptive paragraph. The order of presentation is alphabetical;
however, the fragments may appear in any order in deVice_writer _.pI7.xdw.
&ext art_proc= PLlI-code&; optional; default = 'nUll
any coding needed to support advanced graphic features (beyond simple
plotting) in the device. This feature is not yet used by the Formatter and the
variable name "art_proc" is considered as reserved for a future extension.
&ext dcls= PLlI-code&: optional; default = null
PL/I declarations for all variables needed by the code fragments following that
a.re not already declared by comp_dev_writer.xdw. The PL/I compiler reports
any redeclarations as errors.
Note:
Understanding of the declarations and use of the expansion
variables should be a primary goal in the study of the released
Honeywell-supported device modules.
C-24
AZ98-02
&ext devclass= cl ass&; required; no default
the device class for the device. The value given here must be the same as that
given for DevClass in dev;ce.compdv.
&ext device= dev;ce&; required; no default
the name of the device for which the procedure is being generated. All the
various uses of this name (both here and in dev;ce.compdv) must be
consistent
&ext disp_rtn= PL/I-code&; optional; (see text)
the code needed to produce the interpreted display discussed under "Display
Mode Interpretations" in the description of the process_compout command (see
Section 3). The default is:
call comp_util_$display ((dev_chars»;
This is the routine used by the Formatter
in error messages.
to display input lines
&ext epilogue= PL/I-code&; optional; default = null
any coding needed to write necessary data to the output after the end of the
document
&ext file_init= PLII-code&; optional; default = null
any coding needed for initialization of the writer at the beginning of an input
file.
&ext foot_proc= PL/I-code&; optional; default = null
any coding necessary to convert footnote references into the form to be used
on this device; for example, _superior digits. comp_dev_writer.xdw supplies what
is used for device classes typewriter and diablo.
&ext image_init= PL/I-code&; optional; default = null
any coding needed for initialization at the beginning of a "window image". For
example, in vip7801_writer_, each output "page" is made up of windows that fit
on the screen.
&ext line_finish= PL/I-code&; optional; default = null
any coding needed to complete the preparation of a line for the output stream.
0
&ext line_init= PLII-code&; optional; default = null
any coding needed to begin processing of an input line image. A line image is
a coded structure in the page image and may contain only part of an output
line. for example. a title part or the text for one of several table columns.
&ext machines= types&; optional; default = terminals
used only to specialize certain descriptive comments. Other possible values are
"typesetters" and "lineprinters."
&ext multi_pitch= n&; optional; default = 0
n may have only the values "0" and n1 n. "0" means that the device is a
typc\vriter class de-voice with a fixed pitch setting. "1" means that, eVen though
the device is not a diablo class device, it does support more than one pitch
setting.
&ext notes= PL/ I-comments&; optional; default = null
any PL/I commentary. It is inserted into the PL/I source just ahead of the
opening "procedure" statement
&ext other_procs= PL/ I-code&; optional; default = null
any additional internal PL/I utility procedures needed by the writer.
o
C-25
AZ98-o2
&ext page_finish= PLII-code&; optional; default = null
any coding needed to complete (run out) a page on the device.
&ext page_init= PLlI-code&; optional: default = null
any coding needed for writer initialization at the beginning of an output page.
&ext plot= PLlI-code&; required, no default
the coding needed to do simple horizontal and/or vertical vectors and shifts on
the device. regardless of what additional graphic capability the device may have.
&ext process_text= PLlI-code&; optional; default = null
any coding that converts the text strings in line images to output device native
characters in the required code set and format.
&ext restore= PLlI-code&; optional; default = null
PL/I assignment statements that restore any saved user variables (see save
following).
&ext save= PLII-code&; optional; default = . null
blank line (for example, a line containing nothing but font changes) suppression
requires that certain variable values be saved at the beginning of an input line
so that they may be restored if the line is truly blank. This code is PL/I
assignments that add other user-defined variables to the saved data.
&ext set_font= PLlI-code&; optional; default = null
any coding needed to effect a font change in the device.
&ext set_media= PLlI-code&; optional; default = null
any coding needed to set the device to the desired font.
&ext set_ps= PLlI-code&; optional; default = null
any coding needed to effect a pointsize change in ·the device. This and set_font
may be interdependent or may be completely independent. depending on the
device.
&tabx= PLlI-code &: optional; default = null
code to support direct (or absolute) horizontal tabulation in the device.
DEVICE TABLE COMPILER
The Device Table Compiler is a language translator that translates the plain
language description of a device intended for use by the WORDPRO Text Formatter
into the required binary table form. The input to the translator consists of various
statements in the language described below and contained in an unformatted stream file
named device.compdv. where device is an arbitrarily chosen name for the device to
be supported. This input file is referred to below as the device description file. The
output is a coded binary table in a segment named device.comp dsm that is accessed
directly by the Formatter.
The process_compout command is referenced in the text below. The description of
this command may be found in Section 3. Also, in the remainder of this section, the
WORDPRO Text Formatter is referred to simply as the Formatter. Its description
may be found in Section 2.
C-26
AZ98-02
The Device Description Language
The device description file consists of nine parts that must appear in the order
shown (unless otherwise noted):
Global Values (distributed)
Symbol Declarations (optional)
Media Character Table
Media Tables
View Tables
Definitions (optional)
Font Tables
Size Tables
Device Tables
GENERAL SYNTAX
Literals
A quoted-string means a string delimited by the double quote character ("). If a
quote is needed within such a string, it must be doubled. For example:
"A quoted string"
"A ""quoted"" string"
Comments
A comment may be placed any place in the source where the syntax allows white
space to appear (except within a quoted string). A comment is any string beginning
with /* and ending with */. For example:
/* This is a comment * /
/* And this is a
multiline comment *i
Names
A narne means a strmg of not more than 32 characters beginning with an
alphabetic followed by an arbitrary series of alphanumerics and/or underscores. All
names in a device description file must be globally unique. For some usages, name is
restricted to less than 32 characters. The restrictions are given in the discussions of the
various usages. For example:
A
name
Fonts
Two different forms of fonts are supported: the "family" font and the "bachelor"
font. A "family" is a group of fonts of different styles all of which have the same
C-27
AZ98-02
typeface such as Century Schoolbook or Helvetica. A "bachelor" is a font that has no
such close relatives such as NewsCommercialPi (NCPi) or APL.
Anywhere font appears, it is a natne having the form family/member or bachelor.
Braces, Ellipses, and Vertical Lines
A term or group of terms may be enclosed in opening and closing braces (n)
and/or followed by an ellipsis (. .. ). The braces mean that the enclosure is optional, the
ellipsis means that the preceding term or term group may be repeated as desired, and
the vertical line means a choice between (or among) the terms must be made. For
example,
integer {, integer I=
.•.
J
stands for a comma-separated list of integers and equal signs of any length that must
start with an integer.
Input
input is a single character given by either of:
000
3 octal digits
"c
tt
any single-quoted character
Range
range is an inclusive ordered set of characters given as input:input. For example:
"A":"Z"
the uppercase alphabet
000:007
the first eight ASCII control characters
Output
output is a blank separated list of elements selected from the following:
000
3 octal digits
"string"
any quoted string
XXX
any declared symbol (See "Symbol Declarations" below)
nn(output}
nn repetitions of an output string
SELF
When used in media character token definitions. means the graphic being
defined. This is a reserved word~ it may not be used as a name.
C-28
AZ98-02
For example:
del: square, "Z" 010 "N";
/-;': a black square ":/
Media Characters
A mediachar is an internal token referring to some graphic symbol or control
action available in the device. It may have the form of an input, or an eight--character
name. Note that A and "A" are not the same mediachar.
Media Character List
A mediacharlist
following:
is a blank-separated list of mediachars given as any of the
000
the octal value of a mediachar that is defined as an / nput.
"xxx"
a string of med i achars, each of which is defined as an Input.
yyy
a med i achar that is defined as a name.
nn(Mediacharl ist}
nn repetitions of a mediacharlist
-Media
A media table is a named aggregate of media character tokens wherein each token
is assigned a character width value given in "strokes". A "stroke" is an arbitrary,
dimensionless number of parts into which an EM unit is divided for the purpose of
defining character widths for a device and must be at least as large as the resolution of
the device. The term comes from antiquity and refers to the number of strokes
required with a given size pen nib to get a line of some desired thickness.
Switch
switch is the setting of a binary switch bit. It may have two values; on or off.
Numbers
Numbers are given as one of the following forms.
integer
a dimensionless decimal integer, for example:
2
253
-3
units
a decimal number given in the current space measurement units (see Units
under "Global Values" below), for example:
C-29
AZ98-02
9
-1.5
97.25
SYNTAX OF THE SECTIONS
Global Values
The Global Values section is not a formally delimited section, but consists of any
number of the following statements distributed randomly throughout the file. The
statements define values that apply to all sections following their appearance. All have
local counterparts to specify different values for a particular table.
Data dependencies affect the order in which certain statements may appear. Any
such restrictions are given in the descriptions of the affected statements.
The statements all have default values that describe the default (ASCII or printer)
device. Unless otherwise noted in the text, the default values are those shown in the
individual examples.
Artproc: name {$name J ;
the entryname and optional entrypoint of the procedure that supports special
artwork features for the device. This entry is normally needed only for devices
having graphic features beyond the scope of plotting and simple typographic
rules. The default entryname is derived from the name of the device (see
Outproc below).
This interface is not yet active due to lack of a specific
application. The calling sequence is not yet defined. Its projected
use is for the processing of half-tone raster files and generalized
graphics files.
Artproc: asci i_writer_$artproc;
Note:
Attach: quoted-string;
the attach description to use for the output switch when formatted output is
not being written to a file. If not given, no online output is possible for the
device.
Attach: "syn user_output";
Cleanup: mediacharlist;
the control string that must be sent to the device to restore its normal mode
of operation when interrupted in the middle of output. This string is required
for plotting terminals to take them out of PLOT mode when interrupted.
Cleanup: 1111;
/* no cleanup needed */
Comment: quoted-string;
a string that is emitted to a compout file as a part of its header. It is used by
the process_compout command when transcribing the file onto the output
medium.
C-30
AZ98-D2
Comment:
1111;
/* null comment */
DefaultMargs: units, units. units, units;
the default values for the top, header. footer. and bottom page margins,
respectively. This feature allows for devices (such as Braille embossers) that
demand page margins other than those normally assumed for a printed
document.
DefaultMargs; 48,24,24,48; /* 4,2,2,4 lines */
DevClass: quoted-string;
the class of the device. This string is placed in the output file header for use
by the process_compout command and is used to set the DeviceClass built-in of
the Formatter.
DevClass: "typewriter ll ;
DevName: quoted-string;
the generic name of the machine within DevClass for which the Device Tables
in this file provide support. e.g., V-I-P. Dymo. APS within "photocomp" or
dtc300s. hyterm within "diablo". Note that within a generic device. such as
dtc300s. there may be different specific devices (see "Device Table" selection
below) for minor differences such as running in 12-pitch rather than 10-pitch.
This string is also used to set the DeviceName built-in of the Formatter.
DevName: "asciill;
Endpage: input
the font character to select the page eject sequence for the device. A value of
000 means that there is no eject sequence.
Endpage: 000;
Endpage: 014;
/* asci i */
/* printer */
Footproc: {name {$name J J {, font};
the optional entryname and entrypoint of the procedure to process footnote
references and the optional font for them. The default entryname is derived
from the name of the device (see "Outproc" below) and the default font is the
default font for the device. (See "The Device Writer" below for the description
of the calling sequence for this interface.)
Footproc: ascii_writer_$footproc, asci i;
FootrefSeparator: input;
the fv1ultics character to separate mUltiple footnote references at the -same plaCe
in the text.
FootrefSeparator:
1111;
/'"
parens are suff i c i ent
~':/
Interleave: switch;
the setting of the line sorting switch for the Formatter. If the switch is on,
the output in the page image structure is sorted by the Formatter so as to
appear in strictly increasing page depth order because the device does not
support reverse leading to return to the top of the page for multi-column
output. If the switch is off, the output lines appear in the page image by page
depth within the columns, each column being a sub-array in the structure. The
default value for the switch is off; it must be set on for device with DevClass
values of "typewriter", "diablo", or "printer."
Interleave: on;
/* sort output */
C-31
AZ98-02
Letterspace: integer;
the maximum amount of interletter space allowed, given in strokes.
Letterspace: 0;
/* not supported */
MaxFiles: integer Iunlimited;
the maximum number of files to be written on a reel of magnetic tape. The
process_com pout command calls for an additional reel when this value is
reached while processing "compout" files. The number of input tape reel files
for some typesetters is limited by the software in their front-end computers. If
this statement is omitted or is given with the keyword "unlimited", then the
tape may contain any number of files.
MaxFiles: unl imited;
MaxPages: integer Iunlimited;
the maximum number of pages to be contained in an output file for the
device. The process_compout command produces output files containing no
more than this number of pages. Input files for some devices are limited by
such factors as size of paper tape input reel, capacity of tape cassette or film
magazine, etc. If this statement is omitted or is given with the keyword
"unlimited", then the file may contain any number of pages.
MaxPages: unl imited;
MaxPageLength: units Iunlimited;
the maximum length of a page. If this statement is omitted or is given with
the keyword "unlimited", then the page may be as long as the user cares to
make it.
MaxPageLength: unlimited;
MaxPageWidth: units;
the maximum width of an output page.
MaxPageWidth: 979.2; /* 136 columns */
MinBotMarg: units;
the minimum page bottom margin for the device.
MinBotMarg: 0;
I'"~ asc i i "c/
/"cpr inter ,'c/
MinBotMarg+ 36;
MinLead: units;
the minimum amount of "lead" (vertical spacing) available in the device.
MinLead: 12;
/* 1 line */
MinSpace: units;
the minimum value of horizontal space available in the device.
MinSpace: 7.2;
/* 1 column */
MinTopMarg: units:
the minimum page top margin for the device.
/~" asc i i ~"/
MinTopMarg: 0;
/'''pr i nter ~':/
MinTopMarg: 36;
C-32
AZ98-02
Outproc: name {$name} ;
the entryname and optional entrypoint of the procedure that converts the coded
page image structure constructed by the Formatter into a character stream
acceptable to the device. This is the procedure that translates internal signal
bytes into device control codes. The default entryname for the device described
in device.compdv is device_writer_. (See "Device Writer" earlier in this
section for the description of the calling sequence for this interface.)
/* ascii device */
Sizes: name;
the name of the default Size Table. name must have already been defined as
the name of a Size Table section. If this statement is not given, the name of
the first Size Table defined is used.
Sizes: onesize;
Stream: switch;
the setting of the compout file type switch for the Formatter. If the switch is
set on, the compout file written when the -output_file control arg of the
Formatter is given is an ASCII stream file suitable for processing with the
print and dprint commands. If the switch is set off. the compout file is a
sequential file containing coded binary device information that must be
processed with process_com pout command. Normally. this switch is set on only
for the ASCII and printer devices, but it may be used for any other device
that has only those features commonly found in ASCII terminals or it could be
treated (by Multics) as a line printer. The default value for the switch is off;
it must be set 011 for the ASCII device.
Stream: on;
Strokes: integer;
the number of strokes to be used for width values in Media Tables.
integer Iunlimite.ci;
the length in characters of records to be used when writing to a tape. If this
statement is omitted or is given with the keyword "unlimited", then the tape
records may contain any number of characters.
TapeRec: unlimited;
TapeRe~:
Units: keyword;
the physical units in which space values are given. Space values are given as
normal decimal numbers, e.g., 2, 14.7, and 0.025. The valid keywords are:
pi
pica (lO-pitch) monospace characters and lines
el
elite (l2-pitch) monospace characters and lines
in
inches
mm
millimeters
pc
typographic picas (6 picas = 1 inch)
1't
typographic points (72 points '"" 1 inch)
pp
picas and points as a decimal number
Units: pt;
/* default is points */
Wordspace: min, avg, max, mediachar;
the defaul,t range of allowable interword space for devices described in the file.
C-33
AZ9S-02
min, avg, max
specify the mln~mum, average. and maximum values. respectively, given in
strokes. They must obey the relation:
o <=
min <= avg <= max
and are defined as:
min
the least amount of interword space
avg
the average amount of interword space. This is the amount used
f or all wordspace characters in unjustified lines.
Note:
This value must be the same as the width given for
med i achar. See "Media Tables" below.
max
the maximum amount of interword space allowed before hyphenation
or letterspacing is attempted. Note that justified lines may contain
more than max space. but only in case hyphenation and letterspacing
fail or are not allowed.
mediachar
the character string to be emitted for wordspace insertion.
For example:
Wordspace: 1,1,2,SP; /* asci i, strokes =
Wordspace: 3,6,9,SP; /* dtc300s, strokes
Symbol Declarations
Symbols that represent output character strings may be defined for convenience in
constructing media characters. All such symbols must be defined before their use.
del: name. output;
name
the name of the symbol being defined and is restricted to a maximum
length of 8 characters.
output
the character string to replace a reference to the symbol.
For example:
de 1 :
de 1 :
de 1 :
SSP,
HT,
1f,
010;
011 ;
012;
Media Character Table
The Media Character Table section contains the symbols and output values for
all media character tokens to be used in the Media Tables following. It
consists of the following media character statement.
C-34
AZ98-o2
MediaChars: rnediachar output {, mediachar output .. .};
BSP,:OlO;
dcl:
mediachar
the media character token(s) being defined given as a name,
an input. or a range.
output
the output character string to replace a reference to the token.
For example:
MediaChars:
SP
010
017
NIL
II
II
SELF,
SELF,
014
"a":"f"
SELF,
SELF,
SELF,
SELF,
177
033
016
USR
SELF,
SSP
II
II
1111;
Media Tables
The Media Table section contains the character width values for all med iachars in
all the physical media used by the fonts defined for the device. It consists of any
number of Media: statements, each having any number of width value statements. The
syntax is devised in such a way that the table may actually be formatted as a table in
the input, that is, all width values in the first column are for the first medianame, the
second column for the second med i aname, etc.
Media: med i aname {, med i aname ... };
med i achar {integer}
{, {integer I=} ... };
medianame
the name(s) of the media being defined.
mediachar
as for MediaChars: above.
integer
the width of the character given in strokes. If this value is omitted, the
character is undefined in the associated media.
If "=" is given in the second or subseqUent column, the valUe in the
preceding column is repeated. It is an error to give "=" in the first
column.
For example (from a Mergenthaler V-I-P description):
/"" (A-534 is the Mergenthaler number for Universal Greek with Math. ) */
A-2i60,
A-i8i,
A-66i4,
A-i45,
A-534,
A-i08,
/""
Strokes: 18;
Media:
mNCPi,
A01,
A02,
A03,
A04,
A05,
09,
18,
15,
10,
06,
mUGM,
14,
15,
06,
11,
mCSR,
mCSRx,
10,
10,
1O,
18,
C-35
mCSI,
mCSSR,
17,
A-409
;,.;
mCSS I ;
15,
13;
09;
09;
09;
18,
17;
AZ98-02
View Tables
A View is a switch-like "variable" (in the sense of a Multics I/O switch) through
which an attachment is made to a Media Tabie (see "viewseieci:" under Unique Local
device Values below). A View may attach to only one Media Table, but a Media Table
may have any number of Views attached to it, An example is the superior and inferior
fonts in a Mergenthaler V-I-P typesetter that are identical in all respects except that
they are on different film plaques due to their different baseline offsets. The View
Table section consists of any number of the following View: statement.
View: viewname medianame {, viewname medianame .. J;
viewname
the name of the View being defined.
medianame
the name of the Media Table to which an attachment will be made.
For example:
View: PICA mASCIO, ELITE mASCI2, APL mASCIO;
Def i nitions
A Definition is a named aggregate of MediaChars that may be used in several
different fonts. The Definitions section consists of any number of the following Def:
statement, each followed by any number of graphic definitions.
Def: def name;
graphic {viewname} definition;
defname
the name of the med i achar aggregate being defined.
graphic
may be chosen from:
input {, input ... }
keyword {, keyword ... }
any of:
C-36
AZ98-02
EM
EN
thin
EMENhyphen
EM_
EN_
PS
1\0
1\1
1\2
1\3
1\4
1\5
1\6
1\7
1\8
1\9
EM space
EN space
thin space
EM dash
EN dash
hyphen
EM -aligned dash
EN-aligned dash
punctuation space
opening double-quote
closing double-quote
superior digits
These keywords may be thought of as "built-in" symbols that must
be assigned values if they are to be included in a font. Note that
"hyphen" must be assigned a value in order that the hyphenation
mechanism in the Formatter may work.
art artname {, art art name ... }
a keyword and the conventional name of the artwork construct· or
element selected from one of the following groups.
This group contains graphics that are complete in themselves (the
so-called "one-highs").
[ opening bracket
{ opening brace
( opening
paren thesis
I
vertical bar
• bullet
m change bar
i divide
c copyright
/\ up arrowhead
~ left arrowhead
] closing bracket
} closing brace
) closing parenthesis
II
concatenate
x multiply
d delete star
\ left slant
t trademar k
v down arrowhead
~ right arrowhead
II
This group contains graphics that are parts of larger artwork constructs,
e.g., rules, boxes, diamonds, and lozenges.
DI\
D<
Clf
-rul
/rul
diamond top
diamond left
left half-circle
horizontal rule
right slant rule
Dv
D>
Crt
Irul
\rul
C-37
diamond bottom
diamond right
right half-circle
vertical rule
left slant rule
AZ98-Q2
This group contains the parts for the multiline math symbols. The
graphics for any symbol form a consistent set; if a math symbol is to
be defined, all the parts must be given.
II {II
II [II
II} II
[tp
] tp
{tp
} tp
[ht
[md
[hb
[bt
]ht
]md
]hb
]bt
{ht
{md
{hb
{bt
}ht
}md
} hb
}bt
[f 1
] f 1
{f 1
}f 1
Note:
II {II
11)"
lptp
lpht
lpmd
1phb
lpbt
1pfl
rptp
rpht
rpmd
rphb
rpbt
rpf 1
111"
tp
ht
md
hb
bt
f 1
II
1111 symbol
tp tops
ht
md
hb
bt
f 1
half tops
middles
half bottoms
bottoms
fillers
Because the left and right parentheses are used as part of
the syntax of the device description language they may not
be used in forming tokens; hence the need to use the "lp"
and "rp" constructs for their artname parts.
viewname
the name of the View: that attaches the Media Table holding the character
widths to be used in calculating the width of the definition. The default
View: is the View: attaching the Media Table for each font that refers to
this Def:.
definition
replacement for graphic chosen from:
med i achar list
the assigned width of the definition is calculated from the width
values of the elements of mediacharl ist.
med i Bchar I i Sf =; nteger
the calculated width the med i achar list (as above) is compared to
integer. If they are the same, the value is assigned as the width of
the definition; if not. an error message is generated. This form is
useful in ensuring that plot strings are the correct width for the font
in which they are to be used.
( med i achar I i sO =i nteger
integer is assigned as the width of the definition without regard to
the calculated width. This form is useful in forcing the width of plot
strings when the calculated width is known to be wrong.
For example:
Def: etc;
016:017
(SELF)=O;
/* miscellaneous chars */
/* red/black ribbon shifts */
221
3{1I.");
/~':
ell ips i s
177
NIL;
/,'c
ASC I I PAD
~'c/
~'c/
Font Table
A font table contains the width and output string for each character contained in a
font. In this context, a "character" is a 9-bit byte placed in the output page image by
the Formatter. This byte may be a normal ASCII graphic or a coded signal for some
other output sequence.
C-38
AZ98-02
A device description file may specify up to 100 fonts; each Font Table beginning
with a Font statement and ending with the beginning of any Size Table, Device Table,
or other Font Table.
A Font Table section consists of any number of Font statements, each followed by
an optional local wordspace: statement and any number of ref: statements, and graphic
definitions.
Font: fontname viewname;
{wordspace: min, avg, max, med i achar;}
{ref: defname;}
{graphic {viewname} def i nition;}
fontname
the name of the font table being defined.
viewname
the name of the default View attaching the
definitions given in this font
M~ia
Table for any graphic
wordspace: min, avg, max, output;
as for the global Wordspace described earlier but applying only to this
font
ref: defname;
a reference to some existing Def:.
graphic {viewname} definition;
as for "Definitions" above.
Size Table
A Size Table is a list of allowable pointsize values that may be used in conjunction
with any number of Fonts in any number of Device Tables. That is to say, a Size
Table may be referenced any number of times and may be used with one Font in some
Device Table and a different Font in some other Device Table. A device description
must contain at least one Size Table.
A Size Table section consists of exactly one Size statement of the form:
Size: name, units {, units} ... ;
name
is the internal reference name of the pointsize list being defined.
units
is a value to be entered into the list. At least one units value must be
given.
For example:
Size: pitchlO, 7.2;
Device Table
A Device Table describes a specific devi'ce and provides the data needed by the
Formatter to prepare output for that device. The data in the table is gathered from
default values, Global Values, Font Table references, Size Table references, and Local
C-39
AZ98-02
Device Values. There can be any number of Device Tables in a device description file,
either describing different machines that are similar enough to share many attributes, or
different configurations of the same machine.
Font Tables and Size Tables may be freely shared among Device Tables. However,
if a font "borrows" from some other font, then both the "loaner" and "borrower" Foot
Tables must be included in the Device Table. For example the Mergenthaler V-I-P
font "ascii" is based on the Clarinda font but it borrows a few characters from
NewsCommercialPi. Hence, if the "ascii" font is to be included in a Device Table for
the V-I-P, then "NCPi" must also be included. If it is not, then the Formatter reports
errors if the borrowed characters are used.
In some machines, like the Mergenthaler V-I-P, that have limited font capacity,
many Device Tables are likely to be needed to describe the many different
configurations. Other machines. such as the Autologic APS-5, that have large font
storage capacity usually need only one Device Table.
A Device Table section begins with a Device statement and ends with the beginning
of another Device Table or the end of the device description file. Global Values may
also appear within a Device Table section.
Device: name {,alias} {like device};
name
is the name to be attached to the Device Table. It is the name by which
the device is known to the Formatter and is given as a parameter with the
-device control argument (of the Formatter). name.comp_dsm. is added to
the output segment if it is not the primary entryname.
alias
is an additional name by which the device may be know, e.g., a short
name. This alias is handled identically to the primary name.
device
is the name of some previously defined Device Table that is to be used as
a model for this device. If this Device statement is followed immediately
by another Device statement or is the last statement in the device
description file, the other Device Table is referenced directly by internal
pointers. If any changes to the model are made (with Global or Local
Device Value statements), the other Device Table is copied as initial values
for a new Device Table.
GLOBAL/LOCAL DEVICE VALUES
All the device-related items discussed in Global Values above have local counterparts.
Local Device Values apply only to the Device Table in which they appear; any given
are discarded when the Device Table is completed. The syntax of the Local Device
Values is identical to the corresponding Global Values except that the keyword tokens
are spelled with all lowercase letters. These Local Device Values are all set to their
current Global or default values when a Device Table is initialized (unless device is
used),
The Global/Local Device Values statements are listed below.
artproc: name {$name}
attach: quoted-string;
cleanup: mediacharlist;
C-40
AZ98-02
comment quoted-string;
defaultmargs: units, units. units, units;
devclass: quoted-string;
devname: quoted - st ring;
end page: input;
footproc: {name{$nameJ J {, family/member Ibachelor};
footrefseparator: input;
interleave: switch;
letterspace: integer;
max files: integer Iunlimited;
maxpages: integer Iunlimited;
maxpagelength: units Iunlimited;
maxpagewidth: units;
minbotmarg: units;
minlead: units;
minspace: units;
mintopmarg: units;
outproc: name {$name} ;
sizes: name;
stream: switch;
taperec: integer Iunlimited;
units: keyword;
UNIQUE LOCAL DEVICE VALUES
The following Local Device Values have no- Global Device Value counterparts.
init initfont initsize;
initfont
the initial font for the device given either as fami Iylmember or bachelor.
initsize
the initial pointsize for the device. It must be a value in the initial Size
Table for the device (see "sizes:" above).
For example:
init: CenturySchoolbook/medium 10;
init: asci i 7e2;
family: name {, name};
the external name and optional aliases of a group of fonts of different styles
all of which have the same typeface.
For example:
family: CenturySchoolbook, CS;
member: /name {, / name, .. J fontref;
a member font in the preceding family.
/name
the external name and optional aliases of the member.
C-41
AZ98-02
fontref
the name of the Font Table containing widths and replacements for
characters in the f on t.
For example:
member: Imedium, 1m, Iroman, /r CSmed;
member: /bold, /b CSbold;
bachelor: nameL name, ... } fontref;
"bachelor" fonts that have no family/member structure.
name
as for family: above.
fontref
as for member: above.
For example:
use: GrkMath, GM UGM;
use: APL, apl APL;
viewselect view mediacharlist{, view mediacharlist, ... };
the attachment descriptions for all the fonts used in the device.
view
the name of the View through which the attachment is to be made.
mediacharl ist
the character string giving the information needed by the device writer to
construct the font change control that is sent to the device to cause it to
select and use the desired Media. The content of this information depends
on the device and the design of the device writer procedure.
For example:
viewselect: vASCI I Pwheel pitchlO "6";
viewselect: vAPL Awheel pitchlO "6";
ARTWORK PART DESCRIPTIONS
The artwork parts for incremental plotting terminals are plot strings made up of
various motion characters and the dot (.) character. When strings for such a terminal
are constructed, they should conform to the following specifications.
In these diagrams, the grid of dots represents the 48 possible dot positions in a
print position. The starting position of the pattern is the lower left corner of the grid,
that being the position at which a single "." would print in normal typing mode. If
there is a "+" in a diagram, then its position is the final position of the print head; if
not, the print head returns to the starting position, An "0" represents a grid position
where a "," must be placed. Note that the print column for vertical lines is the left
edge of the grid.
C-42
AZ98-Q2
"One-High" Math Symbols
IIJ II
II [II
00000
O •••••
o .••••
O •••••
o .....
O •••••
O •••••
o .....
O •••••
0
0
0
00000
II {II
00000
O •••••
O •••••
O •••••
O •••••
O •••••
O •••••
O •••••
O •••••
0
0
0
00000
+
II} "
000
• 0 ••••
O •••••
O •••••
O •••••
O ••••••
+
000
O ••••••
O •••••
0
0
0
000
o •••• 0.
0
000
o ••. 0
000 ••••
00000 •••
00000 •••
000 •••• +
••
0.0 •••
• 0 ••••
0.0 •••
o .•• 0 ••
o ..•• 0.+
0
O •••••
O •••••
O •••••
O •••••
o .••..
o ..•..
• • 0 ••
0.0.
0000000.
0.0.
o. .0 •• +
0
o •... ~', .... o .••.
• 0 ••••••• 0 ••••• +
00
00
000
0000000 ••••••••
• •• 0 •• 0 ••••• 0 ••
• •• 0 •• 00 ••• 00 ••
• •• 0 •• 0.0.0.0 ••
• •• 0 •• 0 •• 0 •• 0 ••
••• 0 •• 0 ••••• 0 ••
+
o •....
.0 ••••
• .0 •••
• •• 0 ••
• ••• 0.
• •••• 0
• ••••• 0
• ••••• +0
· ..
.
...
o .....•
·.
0
.0 ••••
.0 ••••
.0 ••••
.0 ••••
.0 ••••
.0 ••••
.0 ••••
.0 ••••
0
0
0
0
0
o.
0
+
0
0
0
0
0
0
0
0
0
0
0
0
0
"\"
• •••• 0
• ••• 0.
• •• 0 ••
• .0 •••
.0 •
·.
·.
•••• 000 ••••••.••
• • 00 ••• 00 ••••••
• 0 ••••••• 0 •••••
o ••.••.•.• 0 ••••
o ......••• 0 ••••
O •••••
O •••••
O •••••
0
0
0
0
"/"
· .... · .
·· .....
.. · .
o.
III III
O ••••••
O •••••
O •••••
O •••••
"d"
00 ••••
00 ••••
00 ••••
00 ••••
00 ••••
00 ••••
00 •••• +
00
00
"1"
O •••••
O •••••
O •••••
O •••••
O ••••• +
O ••••• +
0
0
0
0
0
0
000
000
•• 0 •••
.0 ••••
O ••••• +
0
0
11m"
"
000
o .••••
o .••••
.0 •• .....
+
")
000
.0 ••••
O •••••
O •••••
O ••••••
O •••••
O •••••
O •••••
. .....
0
" (1.1
· .. +
+
NOTE: The "*" in the "c" diagram represents the position of the "c" for "copyright".
It may be changed to any other letter of the users choice.
II All
. . · ..
..
· ..
O •••
000.
00000 •••
0000000 ••
.
·.
....
. . . .+
"ViI
·.
· . · .. ·...
·· ......
· .· .
·.·.
·.·.
0000000 ••
00000.
000
0
.+
11<_"
"_>"
·· .. ·· ...
.. ·
·· .. ·· ...
.
. · ..
. . ·· ..
··· ...
.....
·. .
· .. . ·...
·..
· . . ·.
·• O.. ·
·•• 00.
·.+
000
0000
000
00
0
C-43
'"
'" '"
0
00
000
0000
000
00
0
· ..... +
AZ98-02
Large Artwork Elements
"Ov ll
"0<"
o ••••• 0
• ••••• 0
• ..... 0
o •••••
• 0 ••• 0
• .0.0.
• ••• 0.
• .0 •••
• •• 0 ••
• •• 0 ••
• •• 0 ••
• • • 0 ••
· ..... +
• • 0.0.+
o
• .0 •••
• ••• 0.
• •• 0 ••
• •• 0 ••
• ••• 0. +
o
0
o
0
0000
• •• 000-0
000 •••
o ••••••
o
o
o ••.•••
••••••
••••••
•••••• +
o
o
o
o
o
o
o
o
eO .....
• .0 ••• +
o
o
o
••••••
000 ••••
• .0 •••
• •• 0 ••
• ••• 0.
• ••• 0.
• •••• 0+
o
o
o
o
o
o
000
o
o
o
000
0000
0000
"/rulll
o •••••
O •••••
O •••••
• ••••• 0
• •••• 0
o •••••
• ••• 0.
o ••.••
• •• 0 ••
O •••••
• .0 •••
.0 •••• +
...... +
000000+
o
"\rul 11
o ••••••
o .•...
.0 ••••
• .0 •••
• •• 0 ••
• ••• 0.+
o
Math Symbol Top Parts
II
[tp"
00000
O •••••
o ••.••
O •••••
O.
.. .
O ••••
O ••
O •••••
.+
O.
. ..
...
II] tp"
00000
O •••••
O •••••
O •••••
O •••••
0 •••••
0 •••••
II
II} tp"
{tp"
000
• 0 ••••
O •••••
O •••••
O.
.. ..
O. . . .
"lptp"
000
O ••••••
O •••••
O •••••
. ..
...
O •••••
O.
O.
O •••••
0 •••
v •••••
v •••••
0 ••••• +
O ••••• +
O.
'"
...
. ..
000
.0 ••••
000
O ••••••
o ••••.
o ••..•
O •••••
O •••••
O ••
O ••
O •••••
0 •••••
O •••••
O •••••
....
....,_
"
.+ 0 ••••• +
C-44
"rptp"
. ..
· ..
o .....
II
I tp"
0
O •••••
O •••••
O •••••
O ••
0 •••••
..
. ..
o •••..
O ••
0 ••••• + 0 ••••• +
III \tpll
0
0
0
0
0
0
0
0
0
0
.0 ••••
.0 ••••
.0 ••
.0 ••••
.0 ••••
.0 ••••
;0.; •
.0 •••• +
..
AZ98-02
Math Symbol Half-top Parts
" [ht"
IIJ ht"
·· .....
.....
· .....
.....
···.....
.....
00000.
o •.••.•
o •••.•
o .•.••
o..... +
II
{ht"
·· .....
·• .....
.....
• 000.
00000 •••••
o ..•••
o ..•.•
o .••.•
o .•.•• +
o ..... +
.0 ••••
o •••••
"} ht"
111
pht"
·· .....
· .....
.....
· .....
·· .....
·
..... .....
000
o ••••••
• .000.
.0 • •••
o •••••
o •••••
o •••••
"}rnd"
111
o .••••
o •••••
"rpht"
"Iht"
"I Iht"
· ..... · .....
000
·· .....
..... ·· .....
.....
o •••••
·
.....
o •••••• o •••••
o ..••. o •••••
o ••..• o •••••
o ..... + o •.•.. +
o ..... + o ..•.. +
0
0
0
0
0
·· .....
.....
· .....
.0 ••••
.0 • •••
.0 ••••
.0 ••••
.0 • ••• +
Math Symbol Middle Parts
[rnd II
IIJrnd"
o .....
o ..•..
o •••• ,.
o .....
II
II
0
0
o •••••
o •••••
o .....
o .....
o ••••.
o •••.
o ..... +
o •••••
o •••••
o .••••
o ••••.
o .•••• +
{rnd"
0
0
o .....
o .....
o •..•••
0
o .•...
• 0 ••••
o .....
0
IIl rnd "
0
o ••.••
o .•... o .•.••
o ••••• o •••••
o ••••• + o ••••• +
o .•..•
o ••.••
o .....
o .•...
o •.•••
o .•...
o ••..•
o ••..• +
o ..•..
o •.•..
o ..•..
o ••••.
o ..••.
IIlhb"
• .0 • ••
• 0 • •••
o •••..
o •••..
o ..... +
"rprnd"
0
o .....
o •••••
· .....
o .••...
prnd"
o •••••
o •••••
o •••••
o .••••
o ..•.• +
ll
"I Irnd
0
0
0
0
0
0
0
0
0
0
.0 ••••
.0 ••••
.0 •• e.0 • •••
.0 • •••
.0 • •••
.0 • •••
.0 • ••• +
&
Math Symbol Half -bottom Parts
ii
[hb'i
·· .....
.....
·· .....
.....
·· .....
.....
... . .
o •.••.
0
0
0
00000
"]hb"
II
{hbll
lI}hb"
"lphb"
"rphb"
·· .....
.....
·· .....
.....
·· .....
.....
· .....
·· .....
.....
· .....
·· .....
.....
·· .....
.....
o ...••
·· .....
.....
·· .....
· .....
.....
· .....
o •••••
··· .....
.....
.....
· .....
......,
··· ....
.....
o ••.••
·· .....
.....
· .....
.....
·· .....
..... ··· .....
.....
· ..... · .....
o ..... o •••••
0
0
0
0
0
0
0
0
0
0
0
0
"rpbt"
"l bt "
o •••••
0
0
0
00000
•
0
•••• e
0
000
·· .....
.....
·· .....
.....
0
000
0
000
"I Ihb"
000
··· .....
.....
.....
·· .....
·.....
.....
•
0
0
0
0
0
'!'
•
'!'
~
'!o
.0 ••••
0
0
0
0
Math Symbol Bottom Parts
II
[bt"
0
o ••...
o,... •.•••
.....,
.....
o •••••
o ..•..
o .....
o •••••
{bt"
"}bt"
o ..••.
o ...•.
o •••••
o •••••
v •••••
v •••••
0 •••••
0 •••••
o .•...
o •..•.
o .•. .
o •..•.
o ••..•
o ...•.
o ......
"]bt"
0
o ... . .
o .••..
o ..••.
o .....
00000.+00000 ••••• +
II
0
o •....
.0 • •••
• .000.+000
0
"lpbt"
0
.
• ••••• +
C-45
o •..•.
o ••...
o .••••
o .....
o ••...
o ...•.
• 0 • •••
• .000.+000
0
o ••..•
o .••.•
o •...•
o .•..•
o .•..•
o .•..•
o .••.••
• ••••• +
0
o ..•..
o •••••
o •.•..
o ... . .
o ..•..
o ..•..
o •..•.
o •.••. +
"I Ibt"
0
0
0
0
0
0
0
0
0
0
.0 ••••
.0 ••••
.0 ••••
.0 • •••
.0 • •••
.0 • •••
.0 • •••
.0 ••••
AZ98-D2
Math Symbol Filler Parts
lI[f111
lI]f1 11
lI{f1 11
0
0
0
o ....•
o •••••
o •••••
o .....
o .....
o .....
o ....•
o ••.••
o .. 0.••
o .....
o ••.••
o •••••
o .•...
o .....
o .....
o ....•
II} f 111
0
o •••.•
o •.•••
o .....
o .•.••
o .•...
o .....
o .....
o .....
o .....
o •••••
o •••••
o •••••
o •••••
o .....
o ...•.
0 •••••
C-46
IIlpf111
0
IIrpf111
0
o .....
o .....
o .....
o .....
o .....
o .•...
o .. 0 ••
o .....
o .....
o. 0 0 ••
o .....
o .....
o .....
o •••••
o •••••
o •••••
III If 111
II1f111
0
o •••..
o .....
o •....
o ..•••
o •••••
o .....
o. 0 •• 0
o .....
0
0
0
0
0
0
0
0
0
0
.0 ••••
.0 ••••
.0 ••••
.0 ••••
.00 •••
.0 ••••
.0 ••••
.0 ••••
AZ98-o2
APPENDIX D
GLOSSARY
The following list is WORDPRO-specific and does not duplicate common Multics
glossary terms in other documentation (see Multics Reference Manual.)
artwork
In compose, overstruck character patterns displayed as various symbols and line art
features (e.g., diagrams, flow charts, logos).
built-in symbol
A variable (number, on, off) which is built into one of the programs being used;
not affected by user unless its value is specifically changed by a related control or
control argument.
block (text block)
In compose, the basic premise for text processing; all text material is made up of
blocks of text on which compose processing takes place, using the surrounding
controls as the basis for formatting.
canonicalization
The conversion of a terminal input line into a standard (canonical) form. This is
done so that lines that appear the same on the printed page, but that may have
been typed differently (i.e., characters overstruck in a different order), appear the
same to the system (see Multics Reference Manual).
carriage return
Movement of the typing mechanism to the first column
Multics, this action is the result of the ASCII linefeed character. The terminal type
determines which keyes) the user presses to perform the equivalent action (e.g.,
RETURN, LF. or NL).
command
A program designed to be called by typing its name at a terminal. Most commands
are system-maintained, but any user program that takes only character-string input
arguments can be used as a command (see Multics Reference Manual),
command level
The process state in which lines input from a user's terminal are interpreted by the
system as a command (i.e., the line is sent to the command processor). A user is at
command level or when he or she logs in. or when a command completes.
encounters an error. or is stopped by issuing the quit signal. Command level is
normally indicated by a ready message (see Multics Reference Manual).
command processor
The program that interprets the lines input at command level and calls the
appropriate programs, after processing parentheses and active functions (see Multics
Reference Manual).
compin
A compose input file, made up of text and compose text controls (see Section 3).
D-}
AZ98-02
compose
A command that, given input (text and controls), formats it according to the
conditions set by the user (controls and control arguments) and produces the desired
output (see Section 3).
compout
A compose output file, created (compiled) from the compin file, usually consisting
of formatted text, diagrams, etc., the format having been defined by the user
through the use of the controls (see Section 3).
control (control line)
In compose or format_document, a line that defines an action to be taken while
formatting the output. It always begins with a period in the first character
position, followed by several alphabetic characters, and sometimes another string
that further describes the action to be taken.
control argument
An argument to a command that specifies what the user wants done, or what
information the user is interested in. System control arguments begin with a
hyphen, such as -all, -long, or -hold. The meaning of each control argument
accepted by a specific command is given as part of the description of the
command. Many control arguments have standard abbreviations such as -lg for
-long. A list of the abbreviations of the most frequently used control arguments is
found in Appendix A of SDN - Standards. (System commands are described in
Multics Commands.)
crash
An unplanned termination of system availability caused by problems in hardware
and/or software.
deiimiter
(1) In Speed type , a character used to delimit between text tokens (symbols including
prefix and suffix characters).
(2) In List Processing, a user-specified character used to delimit between each
record in the list, between each field of each record, and between commerits and
data.
dictionary
In WORDPRO, an online list of words (supplied by the system or created by the
user) with which any file can be compared, to find misspelled words, unwanted
words, etc.
edm
A text editor that allows users to create input segments (text), then edit these
segments making substitutions and changes.
emacs
A text editor that allows users to create input segments (text), then edit these
segments making substitutions and changes.
equation
In compose, also called , a three-part title of the form:
! partli part2! part3!.
escape character
In Speed type , a signal that the text token immediately following is to be processed
specially, i.e., not expanded (see token).
D-2
AZ98-02
expand
In Speed ty pe, the process by which symbols are lengthened from their shorthand-type
abbreviated form to the correctly-spelled words and phrases..
fdocin
A format_document input file made up of any types of text and format_document
text controls (see format_document).
fdocout
A format_document output file usually consisting of formatted text that has been
defined by the user through use of controls (see control above).
file
A term that stands for segment and/or multisegment file.
format_document
A command that. given input (text and controls), formats the segment according to
the conditions set by the user (controls and command control arguments) producing
the desired output
help files
See info segments.
indent
In compose and format_document, a control that indents all following text until
another control indicates undent or indent zero (reset the indent to zero).
inf 0 segments
The segments whose contents are printed by invoking the help command. These
segments. sometimes called help files, give information about . the system. The
system info segments are kept in the directory >documentation>info_segments
(>doc>info). The info segments that are peculiar to an installation are kept in
>doc>iml_info_segments. (see the help command. in Multics Commands).
line art
Graphic constructs (pictures, tables, etc). that can be created online.
List Processing
A group of related commands that enable the user to create lists, define formats to
be used in the manipulation of these lists, and produce personalized form letters,
billing statements, reminders, etc.
lister
In List Processing. an unprintable file containing ASCII and binary information; a
compiled version of the list that can be processed by the various list processing
commands.
listform
In List Processing, a file that shows (defines) the format of a document to be
produced; a combination of a lister file and a listform file, processed together,
creates the desired end result
listin
In List Processing, an ASCII file containing components (records), each of which is
a complete entry in the list, that can be entered and updated using any text editor
(see record below).
project
An arbitrary set of users grouped together for accounting and access control
purposes.
D-3
AZ98-o2
Project_id
The name assigned to a project.
n,ptiv
'1."-' ..... .n..
A text editor that allows users to create input segments (text). then edit these
segments making substitutions and changes.
quit request
Several commands that read input from the keyboard and use the typed request
"quit" or "q" to indicate that the user is done. This is not the same as issuing the
quit signal.
quit signal
A method used to interrupt a running program. The quit condition is raised by
pressing the key on a terminal. such as ATTN. BRK. INTERRUPT. This condition
normally causes the printing of QUIT followed by establishment of a new command
level (see Multics Reference Manual).
quote
A character used to delimit strings in commands and source programs. On Multics
this is the double-quote (octal 042). not to be confused with the single quote or
apostrophe (octal 047).
ready message
A message that is displayed each time a user is at command level. The display of
this message may be inhibited. or the user may define a personal ready message.
The standard system ready message tells the time of day and the number of CPU
seconds. memory units. and page faults since the last ready message plus the current
listener level (if greater than 1).
record
In List Processing, a division of a list containing all fields of related information
grouped together. as one item in the list (e.g.. name. address. city. and state of one
person grouped together
recursion
The ability of a procedure to invoke itself.
Speedtype
A tool that enables the user to "type shorthand"; users can specify abbreviations
(symbols) for lengthy or frequently used words and/or phrases, enabling faster
input. to be later automatically expanded.
star convention
A method used by many commands to specify a group of segments and/or
directories using one name (see Multics Reference Manual).
subsystem
A collection of programs that provide a special environment for some particular
purpose, such as editing, calculation, or data management It may perform its own
command processing. file handling, and accounting. A subsystem is said to be
closed if: (1) all necessary operations can be handled within the subsystem and (2)
no way exists to use the normal Multics environment from within the subsystem.
suffix
The last component of an entryname (components are separated by a period
that usually specifies the type of segment (e.g... pll and .list). A segment without a
suffix is usually an object segment or data segment (see Multics Reference
(.»
Manual).
D-4
AZ98-D2
teco
A text editor that allows users to create input segments (text), then edit these
segments making substitutions and changes.
ted
A text editor that allows users to create input segments (text), then edit these
segments making substitutions and changes.
tOken (text token)
In Speed type, the symbol used in place of a word or phrase, with a prefix, suffix.
underline, or capitalize character attached.
undent
In compose or format_document, a control used in conjunction with the indent
control.
D-5
AZ98-02
INDEX
commands (cont)
display comp dsm (ddsm) 3-12
display-list (illls) 6-18
expantl-device writer (xdw) 3-18
expand-list (eIS) 6-19
expand-symbols (esb) 5-12
find dfct words (fdw) 4-13
find=symools (fsb) 5-13
format document (fdoc) 3-20
list diet words Odw) 4-15
list-symools Usb) 5-14
locate words Ow) 4-18
merge-list (mls) 6-20
modlfy list (mdls) 6-23
option symbols (osb) 5-15
print symbols path (psbp) 5-17
prillt-wordliSC{1fwl) LJ-20
process com~u ~o) 3-28
process-list (pIs) 6-24
retain SYmbols (rsb) 5-18
revise-words (rw) 4-22
show symbols (ssb) 5-19
sort TIst (sIs) 6-27
trim list (tIs) 6-29
trim- word list (tw1) 4-24
use_symbols (usb) 5-20
compdv command 3-2
compose (comp) command 3-3
compose metacharacter table A-1
compose Text Formatter 2-1
also see Formatter
compose_index (cndx) command 3-8
comprehensive control summary 2-58
control summary
comprehensive 2-58
convert_runoff (cv_rn command 3-11
copy_list (cpls) command 6-14
count_dict_words (cdw) command 4-8
create_list (cIs) command 6-15
create_wordlist (cwI) command 4-9
delete_dict_words (ddw) command 4-11
delete_symbols (dsb) command 5-11
describe_list (dIs) command 6-16
abbreviations
adw (add dict words command)
als (ap~nd list command)
asb (add symbols command)
.
cdw (count dict words command)
cls (create 11st command)
cndx (compose index command)
comp (compose command)
cRls (copy_list command)
csb (change symbols command)
cv rf (convert runoff command)
cwl (create wordlist command)
ddsm (dispfav comp dsm command)
ddw (delete ¥dict words command)
dils (display-list -command)
dIs (descrioe list)
dsb (delete symbols command)
els (expana lIst command)
esb (e~pancr symbols command)
fdoc (format document command)
f dw ~find diet words command)
fifo first=in-first-out)
fsb ( ind symbols command)
ldw (list chct words command)
lif 0 (last-in-first-out)
lsb (list symbols command)
lw (locate words command)
mdls (modify list command)
mIs {merge liSt command) .
osb option-symbols command)
~o proceSS compout command)
pIs (process list command)
psbp (Print symbols path command)
pwl (print word list command)
rsb Jre~in-=-sympols comm~pd)
rw irevlse woros commano)
sIs (sort lrst command)
ssb (show symbols command)
tIs (trim Hst command)
twl (trim wordlist command)
usb (use symbols command)
xdw (expand_device_writer command)
add_dict_words (adw) command 4-6
add_symbols (asb) command 5-7
append_list (als) command 6-13
change_symbols (csb) command 5-10
commands
add dict words (adw) 4-6
add-symools (asb) 5-7
appena list (als) 6-13
chane:e -svmbols (csb) 5-10
comovj:'2
- -compose (comp) 3-3
compose index (cndx) 3-8
convert runoff (cv rn 3-11
copy liSt (cpls) 6-14
count dict words (cdw) 4-8
create-list leIs) 6-15
create-wordlist (cwl) 4-9
delete- dict woros (ddw) 4-11
delete-symools (dsb) 5-11
descrioe_list (dIs) 6-16
Device SURport Tools
Device Table Compiler C-26
Device Writer C-14
Device Writer Source Expander C-1
Device Table Compiler C-26
device description language C-27
artwork part descnptions C-42
general syntax C-21
global/local device values C-40
see syntax
unique local device values C-41
i-I
AZ98-o2
Expander (cont,)
relational operators C-11
protected strings C-IO
features C-l
general terminator token C-18
Iteration C-12
miscellaneous features C-14
null separator tokens C-18
quote p'rocessing C-19
rescannlng C-19
return C-20
value assignment C-6
variables and arrays C-3
access C-4
arra'y variables C-4
fixed arrays C-5
list arrays C-5
stack arrays C-6
,varying arrays C-5
assIgn C-~
scalar variables C-4
white space control C-20
find_diet_words (fdw) command 4-13
find_symbols (fsb) command 5-13
format_document «doc) command 3-20
formatter 2-1
command
compose (comp) 3-3
compose 2-1
artwork 2-26
built-in variables 2-22
comprehensive control summary 2-58
formatter controls 2-31
formatting features 2-2, 2-10
general syntax 2-1
elementary 3-20
commands
format document (fdoc) 3-20
control lines 3-20
default 3-20
glossary D-1
hyphenate_word_ subroutine 4-14
List Processing 6-1
angle braCKet escapes 6-5
commands
append list (als) 6-13
copy liSt (cpls) 6-14
create list (cIs) 6-15
descrioe list (dIs) 6-16
display TIst (dils) 6-18
expana-list (els) 6-19
merge list (mls) 6-20
modify list (mdls) 6-23
process-list (pIs) 6-24
sort lisT (sIs) 6-27
trim list (tIs) 6-29
field inSertion 6-4
files 6-2
lister 6-3
listform 6-3
listin 6-2
functions 6-1
sample files 6-8
selection 6-6
sorting 6-5
list_diet_words (Idw) command 4-15
Device Writer Source Expander C-1
see Expander
dict search list 4-1
dictionaries 4-1
commands
add dict words (adw) 4-6
count diet words (cdw) 4-8
create- womlist (cwI) 4-9
delete- dict words (ddw) 4-11
find dict words (fdw) 4-13
list aict words Odw) 4-15
locate words Ow) 4-18
print word list (pwl) 4-20
revise words (rw) 4-22
trim wordlist (twI) 4-24
hyphenation 4-2
problems 4-2
technique 4-2
when needed 4-2
spelling errors 4-4
correction 4-5
detection 4-4
unwanted words 4-4
word list segments 4-5
subroutine
hyphenate_word_ 4-14
use of 4-1
files 4-2
standard 4-1
user-supplied 4-1
display_comp_dsm (ddsm) command 3-12
display_list (dils) command 6-18
expand_device_writer (xdw) command 3-18
expand_iist (eIs) command 6-19
expand_symbols (esb) command 5-12
Expander C-1
active function calling C-14
built-in functions C-15
commands
expand_device_writer (xdw) 3-18
comments C-17
conditional execution C-13
constructs C-1
nesting C-2
termination condition C-1
tokens C-1
emptying arrays C-17
error reporting C-17
expansion calhng C-13
expansion definItion C-2
dynamic C-2
static C-2
expansion tokens C-20
examples C-23
expression evaluation C-7
accessing arguments C-9
arg count C-10
multiple arg accesses C-9
simde arg accesses C-9
accessing variables C-7
array accesses C-8
scalar accesses C-7
subscripted accesses C-7
arithmetic expression C-10
arithmetic operators C-10
arithmetic expressions
i-2
AZ98-02
speedtype (cont.)
esca~ 5-4
expansion ~rocess 5-3
pref ixes 5-6
suffixes 5-5
symbol dictionaries 5-2
speeOtyping 5-1
text segment 5-1
text segvlents 5-1
tokens 5-1
subroutine
hyphenate_word_ 4-14
syntax C-30
definitions C-36
device table C-39
font table C-38
global values C-30
media character table C-34
media tables C-35
size table C-39
symbol declarations C-34
VIew tables C- 36
text formatter 3-20
see formatter
trim_list (tIs) command 6-29
trim_wordlist (twI) command 4-24
use_symbols (usb) command 5-20
WORDPRO
definition of 1-1
glossary -D-1-Wordpro commands 3-1
compdv 3-2
compose (comp) 3-3
compose index(c,ndx) 3-8
convert
runoffA ..\cv
rO 3-11
A; .. _1 ..
_ IAA",_ \
list_symbols Usb) command 5-14
locate_words Ow) command 4-18
merge_list (mls) command 6-20
metacharacter table (compose) A-I
modify _list (mdls) command 6-23
option_symbols (osb) command 5-15
print_symbols_path (psbp) command 5-17
print_wordlist (pwl) command 4-20
process_compout (peo) command 3-28
process_list (pIs) command 6-24
reference to commands/subroutines by function
B-1
Dictionary B-1
List Processing B-3
SjJeedtype B-2
Wordpro B-1
retain_symbols (rsb) command 5-18
revise_words (rw) command 4-22
show_symbols (ssb) command 5-19
sort_list (sIs) command 6-27
speed type 5-1
commands
add symbols (asb) 5-7
change_symbols (csb) 5-10
delete ·symbols--(dsbr· 5;..11
expana svmbols (esb) 5-12
find symbols (fsb) 5-13
list symbols Usb} 5-14
option symbols (osb) 5-15
print Symbols path (J?Sbp) 5-17
Tp-t~ln ~vmhol~ fr~h) ",-l'R
show-symbols- (ss1,)' 5--19- ,
use_symbols (usb) 5-20
features 5-1
n
UJi)}J14l
-,.,, _ _
~JJ.1p
~_1"
~.1U
\UUi).1UI
.;:J- i6.
expana-device writer (xdw) 3-18
format-document (fdoc) 3-20
process-:='compout (pco) 3-28
i-3
AZ98-o2
HONEYWELL INFORMATION SYSTEMS
Technical Publications Remarks Form
TITLE
ORDER
MULTICS WORDPRO
REFERENCE MANUAL
No·1
DATED
I
ERRORS IN PUBLICATION
SUGGESTION.S
fORIMPRQV~MENTTO P~Q~IC".nQI'4.
Your comments will be investigated by appropriate technical personnel
and action will be taken as required. Receipt of all forms will be
acknowledged; however, if you require a detailed reply, check here.
0
FROM: NAME -------------------------------------------TITLE _ _ _ _ _ _ _ _ _ _ _ _ _ _.___________
COMPANY --------------ADDRESS _________________________________________
DATE
AZ98-02
JULY 1983
I
I
I
I
I
I ~
I oJ
I c.:J
I Z
PLEASE FOLD AND TAPENOTE: U. S. Postal Service will not deliver stapled forms
.:3
~
o
oJ
III " I
ou.
NO POSTAGE
NECESSARY
IF MAILED
IN THE
UNITED STATES
BUSINESS REPLY MAIL
FIRST CLASS PERMIT NO. 39531 WALTHAM, MA02154
POSTAGE WILL BE PAID BY ADDRESSEE
~~
HONEYWELL INFORMATION SYSTEMS
200 SMITH STREET
WALTHAM, MA 02154
ATTN: PUBLICATIONS, MS486
I
I
I
I
I
I
I
I
I
I
oJ
!
~
UJ
Z
~g
~
o
oJ
o
u.
Honeyweii
Together. we can find the answers.
Honeywell
Honeywellinfonnation Systems
US.A.: 200 Smith St., MS486, Waltham, MA02154
~~~2.~~~~~:~.
'3.d;:..~~~~~:.O~
nu., DIt:lllIUIU,
I VYO
lid',. ':Ar.~~~,?,
VIC1 rnt:lII, t:..v It:..£t IVIlld.IIV
,", ___ A . . .
u.n. •.
\.::nt:~l
VVt;:::R
' V H r U • .,, _ _ _
IVlIUUI~A
~un
~
Mexico: Avenida Nuevo Leon 250, Mexico 11, D.F. Japan: 2-2 Kanda Jimbo-cho Chiyoda-ku, Tokyo
Australia: 124 Walker St.. North Sydney, N.S.w. 2060 S.E. Asia: Mandarin Plaza, Tsimshatsui East, H.K.
39364, 5C1283, Printed in U.SA
AZ98-02
</pre><hr>Source Exif Data: <br /><pre>File Type : PDF
File Type Extension : pdf
MIME Type : application/pdf
PDF Version : 1.3
Linearized : No
XMP Toolkit : Adobe XMP Core 4.2.1-c043 52.372728, 2009/01/18-15:56:37
Producer : Adobe Acrobat 9.55 Paper Capture Plug-in
Modify Date : 2014:11:13 00:11:12-08:00
Create Date : 2014:11:13 00:11:12-08:00
Metadata Date : 2014:11:13 00:11:12-08:00
Format : application/pdf
Document ID : uuid:d2afb87d-4ab1-f947-acca-b3c9a473453c
Instance ID : uuid:385b76c3-ad85-5e47-9a23-d8ad8a1f07cd
Page Layout : SinglePage
Page Mode : UseOutlines
Page Count : 248
</pre>
<small>EXIF Metadata provided by <a href="https://exif.tools/">EXIF.tools</a></small>
<div id="ezoic-pub-ad-placeholder-110">
<script async src="//pagead2.googlesyndication.com/pagead/js/adsbygoogle.js"></script>
<!-- usermanual link ad -->
<ins class="adsbygoogle"
style="display:block"
data-ad-client="ca-pub-0545639743190253"
data-ad-slot="6172135303"
data-ad-format="link"></ins>
<script>
(adsbygoogle = window.adsbygoogle || []).push({});
</script>
</div>
</div>
<div id="catlinks" class="catlinks catlinks-allhidden" data-mw="interface"></div> <div class="visualClear"></div>
</div>
</div>
<div id="mw-navigation">
<h2>Navigation menu</h2>
<div id="mw-head">
<div id="p-personal" role="navigation" class="" aria-labelledby="p-personal-label">
<!-- <div id="p-search" role="search">
<form action="https://usermanual.wiki/search.php" id="searchform">
<div id="simpleSearch">
<input type="search" name="search" placeholder="Search UserManual.wiki" title="Search UserManual.wiki [ctrl-option-f]" accesskey="f" id="searchInput" tabindex="1" autocomplete="off"><input type="hidden" value="Special:Search" name="title"><input type="submit" name="go" value="Go" title="Find a User Manual" id="searchButton" class="searchButton"> </div>
</form>
</div>-->
<ul>
<li id="pt-mycontris"><a href="https://usermanual.wiki/upload" title="Upload User Manual" accesskey="y">Upload a User Manual</a></li>
</ul>
</div>
<div id="left-navigation">
<div id="p-namespaces" role="navigation" class="vectorTabs" aria-labelledby="p-namespaces-label">
<h3 id="p-namespaces-label">Versions of this User Manual:</h3>
<ul>
<li id="ca-nstab-main"><span><a href="https://usermanual.wiki/Document/AZ9802wordproRefManJul83.1029668715" title="User Manual Wiki" accesskey="c">Wiki Guide</a></span></li> <li id="ca-nstab-main"><span><a href="https://usermanual.wiki/Document/AZ9802wordproRefManJul83.1029668715/html" title="HTML" accesskey="c">HTML</a></span></li> <li id="ca-nstab-main" class="selected" ><span><a href="https://usermanual.wiki/Document/AZ9802wordproRefManJul83.1029668715/help" title="Discussion / FAQ / Help" accesskey="c">Download & Help</a></span></li>
</ul>
</div>
</div>
<div id="right-navigation">
<div id="p-views" role="navigation" class="vectorTabs" aria-labelledby="p-views-label">
<h3 id="p-views-label">Views</h3>
<ul>
<li id="ca-view"><span><a href="#">User Manual</a></span></li>
<li class="selected" id="ca-edit"><span><a href="https://usermanual.wiki/Document/AZ9802wordproRefManJul83.1029668715/help" title="Ask a question" accesskey="e">Discussion / Help</a></span></li>
</ul>
</div>
</div>
</div>
<div id="mw-panel">
<div id="p-logo" role="banner"><a class="mw-wiki-logo" href="https://usermanual.wiki/Main_Page" title="Visit the main page"></a></div>
<div class="portal" role="navigation" id="p-navigation" aria-labelledby="p-navigation-label">
<h3 id="p-navigation-label">Navigation</h3>
</div>
<div class="portal" role="navigation" id="p-tb" aria-labelledby="p-tb-label">
</div>
</div>
</div>
<div id="footer" role="contentinfo">
<ul id="footer-info">
<li id="footer-info-lastmod">© 2026 UserManual.wiki</li>
</ul>
<ul id="footer-places">
<li id="footer-places-privacy"><a href="https://usermanual.wiki/ContactUs" title="UserManual.wiki:Contact Us">Contact Us</a></li>
<li id="footer-places-about"><a href="https://usermanual.wiki/DMCA" title="UserManual.wiki:DMCA">DMCA</a></li>
</ul>
<ul id="footer-icons" class="noprint">
<li id="footer-poweredbyico">
</li>
</ul>
</div>
</div></body></html>
<script src="/cdn-cgi/scripts/7d0fa10a/cloudflare-static/rocket-loader.min.js" data-cf-settings="5b272a59d9670721340742aa-|49" defer></script>