The GNU Troff Manual

Table of Contents

Next: , Previous: , Up: (dir)   [Contents][Index]

GNU troff

This manual documents GNU troff version 1.23.0.

Copyright © 1994–2023 Free Software Foundation, Inc.

Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.3 or any later version published by the Free Software Foundation; with no Invariant Sections, no Front-Cover Texts, and no Back-Cover Texts. A copy of the license is included in the section entitled “GNU Free Documentation License”.


Next: , Previous: , Up: Top   [Contents][Index]

1 Introduction

GNU roff (or groff) is a programming system for typesetting documents. It is highly flexible and has been used extensively for over thirty years.


Next: , Previous: , Up: Introduction   [Contents][Index]

1.1 Background

M. Douglas McIlroy, formerly of AT&T Bell Laboratories and present at the creation of the Unix operating system, offers an authoritative historical summary.

The prime reason for Unix was the desire of Ken [Thompson], Dennis [Ritchie], and Joe Ossanna to have a pleasant environment for software development. The fig leaf that got the nod from … management was that an early use would be to develop a “stand-alone” word-processing system for use in typing pools and secretarial offices. Perhaps they had in mind “dedicated”, as distinct from “stand-alone”; that’s what eventuated in various cases, most notably in the legal/patent department and in the AT&T CEO’s office.

Both those systems were targets of opportunity, not foreseen from the start. When Unix was up and running on the PDP-11, Joe got wind of the legal department having installed a commercial word processor. He went to pitch Unix as an alternative and clinched a trial by promising to make roff able to number lines by tomorrow in order to fulfill a patent-office requirement that the commercial system did not support.

Modems were installed so legal-department secretaries could try the Research machine. They liked it and Joe’s superb customer service. Soon the legal department got a system of their own. Joe went on to create nroff and troff. Document preparation became a widespread use of Unix, but no stand-alone word-processing system was ever undertaken.

A history relating groff to its predecessors roff, nroff, and troff is available in the roff(7) man page.


Next: , Previous: , Up: Introduction   [Contents][Index]

1.2 What Is groff?

groff (GNU roff) is a typesetting system that reads plain text input files that include formatting commands to produce output in PostScript, PDF, HTML, DVI, or other formats, or for display to a terminal. Formatting commands can be low-level typesetting primitives, macros from a supplied package, or user-defined macros. All three approaches can be combined.

A reimplementation and extension of the typesetter from AT&T Unix, groff is present on most POSIX systems owing to its long association with Unix manuals (including man pages). It and its predecessor are notable for their production of several best-selling software engineering texts. groff is capable of producing typographically sophisticated documents while consuming minimal system resources.


Next: , Previous: , Up: Introduction   [Contents][Index]

1.3 groff Capabilities

GNU troff is a typesetting document formatter; it provides a wide range of low-level text and page operations within the framework of a programming language. These operations compose to generate footnotes, tables of contents, mathematical equations, diagrams, multi-column text, and other elements of typeset works. Here is a survey of formatter features; all are under precise user control.


Next: , Previous: , Up: Introduction   [Contents][Index]

1.4 Macro Packages

Elemental typesetting functions can be be challenging to use directly with complex documents. A macro facility specifies how certain routine operations, such as starting paragraphs, or printing headers and footers, should be performed in terms of those low-level instructions. Macros can be specific to one document or collected together into a macro package for use by many. Several macro packages available; the most widely used are provided with groff. They are man, mdoc, me, mm, mom, and ms.


Next: , Previous: , Up: Introduction   [Contents][Index]

1.5 Preprocessors

An alternative approach to complexity management, particularly when constructing tables, setting mathematics, or drawing diagrams, lies in preprocessing. A preprocessor employs a domian-specific language to ease the generation of tables, equations, and so forth in terms that are convenient for human entry. Each preprocessor reads a document and translates the parts of it that apply to it into GNU troff input. Command-line options to groff tell it which preprocessors to use.

groff provides preprocessors for laying out tables (gtbl), typesetting equations (geqn), drawing diagrams (gpic and ggrn), inserting bibliographic references (grefer), and drawing chemical structures (gchem). An associated program that is useful when dealing with preprocessors is gsoelim.1

groff also supports grap, a preprocessor for drawing graphs. A free implementation of it can be obtained separately.

Unique to groff is the preconv preprocessor that enables groff to handle documents in a variety of input encodings.

Other preprocessors exist, but no free implementations are known. An example is ideal, which draws diagrams using a mathematical constraint language.


Next: , Previous: , Up: Introduction   [Contents][Index]

1.6 Output Devices

GNU troff’s output is in a device-independent page description language, which is then read by an output driver that translates this language into a file format or byte stream that a piece of (possibly emulated) hardware understands. groff features output drivers for PostScript devices, terminal emulators (and other simple typewriter-like machines), X11 (for previewing), TeX DVI, HP LaserJet 4/PCL5 and Canon LBP printers (which use CaPSL), HTML, XHTML, and PDF.


Next: , Previous: , Up: Introduction   [Contents][Index]

1.7 Installation

Locate installation instructions in the files INSTALL, INSTALL.extra, and INSTALL.REPO in the groff source distribution. Being a GNU project, groff supports the familiar ‘./configure && make’ command sequence.


Next: , Previous: , Up: Introduction   [Contents][Index]

1.8 Conventions Used in This Manual

We apply the term “groff” to the language documented here, the GNU implementation of the overall system, the project that develops that system, and the command of that name. In the first sense, groff is an extended dialect of the roff language, for which many similar implementations exist.

The roff language features several major categories for which many items are predefined. Presentations of these items feature the form in which the item is most commonly used on the left, and, aligned to the right margin, the name of the category in brackets.

Register: \n[example]

The register ‘example’ is one that that groff doesn’t predefine. You can create it yourself, though; see Setting Registers.

To make this document useful as a reference and not merely amiable bedtime reading, we tend to present these syntax items in exhaustive detail when they arise. References to topics discussed later in the text are frequent; skip material you don’t understand yet.

We use Texinfo’s “result” (⇒) and error→ notations to present output written to the standard output and standard error streams, respectively. Diagnostic messages from the GNU troff formatter and other programs are examples of the latter, but the formatter can also be directed to write user-specified messages to the standard error stream. The notation then serves to identify the output stream and does not necessarily mean that an error has occurred.2

$ echo "Twelve o'clock and" | groff -Tascii | sed '/^$/d'
    ⇒ Twelve o'clock and
$ echo '.tm all is well.' | groff > /dev/null
    error→ all is well.

Sometimes we use ⇒ somewhat abstractly to represent formatted text that you will need to use a PostScript or PDF viewer program (or a printer) to observe. While arguably an abuse of notation, we think this preferable to requiring the reader to understand the syntax of these page description languages.

We also present diagnostic messages in an abbreviated form, often omitting the name of the program issuing them, the input file name, and line number or other positional information when such data do not serve to illuminate the topic under discussion.

Most examples are of roff language input that would be placed in a text file. Occasionally, we start an example with a ‘$’ character to indicate a shell prompt, as seen above.

You are encouraged to try the examples yourself, and to alter them to better learn groff’s behavior. Our examples frequently need to direct the formatter to set a line length (with ‘.ll’) that will fit within the page margins of this manual. We mention this so that you know why it is there before we discuss the ll request formally.3


Previous: , Up: Introduction   [Contents][Index]

1.9 Credits

We adapted portions of this manual from existing documents. James Clark’s man pages were an essential resource; we have updated them in parallel with the development of this manual. We based the tutorial for macro users on Eric Allman’s introduction to his me macro package (which we also provide, little altered from 4.4BSD). Larry Kollar contributed much of the material on the ms macro package.


Next: , Previous: , Up: Top   [Contents][Index]

2 Invoking groff

This chapter focuses on how to invoke the groff front end. This front end takes care of the details of constructing the pipeline among the preprocessors, gtroff and the postprocessor.

It has become a tradition that GNU programs get the prefix ‘g’ to distinguish them from their original counterparts provided by the host (see Environment). Thus, for example, geqn is GNU eqn. On operating systems like GNU/Linux or the Hurd, which don’t contain proprietary versions of troff, and on MS-DOS/MS-Windows, where troff and associated programs are not available at all, this prefix is omitted since GNU troff is the only incarnation of troff used. Exception: ‘groff’ is never replaced by ‘roff’.

In this document, we consequently say ‘gtroff’ when talking about the GNU troff program. All other implementations of troff are called AT&T troff, which is the common origin of almost all troff implementations4 (with more or less compatible changes). Similarly, we say ‘gpic’, ‘geqn’, and so on.


Next: , Previous: , Up: Invoking groff   [Contents][Index]

2.1 Options

groff normally runs the gtroff program and a postprocessor appropriate for the selected device. The default device is ‘ps’ (but it can be changed when groff is configured and built). It can optionally preprocess with any of gpic, geqn, gtbl, ggrn, grap, gchem, grefer, gsoelim, or preconv.

This section documents only options to the groff front end. Many of the arguments to groff are passed on to gtroff; therefore, those are also included. Arguments to preprocessors and output drivers can be found in the man pages gpic(1), geqn(1), gtbl(1), ggrn(1), grefer(1), gchem(1), gsoelim(1), preconv(1), grotty(1), grops(1), gropdf(1), grohtml(1), grodvi(1), grolj4(1), grolbp(1), and gxditview(1).

The command-line format for groff is:

groff [ -abceghijklpstvzCEGNRSUVXZ ] [ -dcs ] [ -Darg ]
      [ -ffam ] [ -Fdir ] [ -Idir ] [ -Karg ]
      [ -Larg ] [ -mname ] [ -Mdir ] [ -nnum ]
      [ -olist ] [ -Parg ] [ -rcn ] [ -Tdev ]
      [ -wname ] [ -Wname ] [ files… ]

The command-line format for gtroff is as follows.

gtroff [ -abcivzCERU ] [ -dcs ] [ -ffam ] [ -Fdir ]
       [ -mname ] [ -Mdir ] [ -nnum ] [ -olist ]
       [ -rcn ] [ -Tname ] [ -wname ] [ -Wname ]
       [ files… ]

Obviously, many of the options to groff are actually passed on to gtroff.

Options without an argument can be grouped behind a single -. A filename of - denotes the standard input. Whitespace is permitted between an option and its argument.

The grog command can be used to guess the correct groff command to format a file. See its man page grog(1); type ‘man grog’ at the command line to view it.

groff’s command-line options are as follows.

-a

Generate a plain text approximation of the typeset output. The read-only register .A is set to 1. See Built-in Registers. This option produces a sort of abstract preview of the formatted output.

The above description should not be considered a specification; the details of -a output are subject to change.

-b

Write a backtrace reporting the state of gtroff’s input parser to the standard error stream with each diagnostic message. The line numbers given in the backtrace might not always be correct, because gtroff’s idea of line numbers can be confused by requests that append to macros.

-c

Start with color output disabled.

-C

Enable AT&T troff compatibility mode; implies -c. See Implementation Differences, for the list of incompatibilities between groff and AT&T troff.

-dctext
-dstring=text

Define roff string c or string as t or text. c must be one character; string can be of arbitrary length. Such string assignments happen before any macro file is loaded, including the startup file. Due to getopt_long limitations, c cannot be, and string cannot contain, an equals sign, even though that is a valid character in a roff identifier.

-Denc

Set fallback input encoding used by preconv to enc; implies -k.

-e

Run geqn preprocessor.

-E

Inhibit gtroff error messages. This option does not suppress messages sent to the standard error stream by documents or macro packages using tm or related requests.

-ffam

Use fam as the default font family. See Font Families.

-Fdir

Search in directory dir for the selected output device’s directory of device and font description files. See the description of GROFF_FONT_PATH in Environment below for the default search locations and ordering.

-g

Run ggrn preprocessor.

-G

Run grap preprocessor; implies -p.

-h

Display a usage message and exit.

-i

Read the standard input after all the named input files have been processed.

-Idir

Search the directory dir for files named in several contexts; implies -g and -s.

This option may be specified more than once; the directories are searched in the order specified. If you want to search the current directory before others, add ‘-I .’ at the desired place. The current working directory is otherwise searched last. -I works similarly to, and is named for, the “include” option of Unix C compilers.

-I options are passed to gsoelim, gtroff, and output drivers; with the flag letter changed to -M, they are also passed to ggrn.

-j

Run gchem preprocessor. Implies -p.

-k

Run preconv preprocessor. Refer to its man page for its behavior if neither of groff’s -K or -D options is also specified.

-Kenc

Set input encoding used by preconv to enc; implies -k.

-l

Send the output to a spooler for printing. The print directive in the device description file specifies the default command to be used; see Device and Font Description Files. See options -L and -X.

-Larg

Pass arg to the print spooler program. If multiple args are required, pass each with a separate -L option. groff does not prefix an option dash to arg before passing it to the spooler program.

-mname

Process the file name.tmac prior to any input files. If not found, tmac.name is attempted. name (in both arrangements) is presumed to be a macro file; see the description of GROFF_TMAC_PATH in Environment below for the default search locations and ordering. This option and its argument are also passed to geqn, grap, and ggrn.

-Mdir

Search directory dir for macro files; see the description of GROFF_TMAC_PATH in Environment below for the default search locations and ordering. This option and its argument are also passed to geqn, grap, and ggrn.

-nnum

Number the first page num.

-N

Prohibit newlines between eqn delimiters: pass -N to geqn.

-olist

Output only pages in list, which is a comma-separated list of page ranges; ‘n’ means page n, ‘m-n’ means every page between m and n, ‘-n’ means every page up to n, ‘n-’ means every page from n on. gtroff stops processing and exits after formatting the last page enumerated in list.

-p

Run gpic preprocessor.

-Parg

Pass arg to the postprocessor. If multiple args are required, pass each with a separate -P option. groff does not prefix an option dash to arg before passing it to the postprocessor.

-rcnumeric-expression
-rregister=expr

Set roff register c or register to the value numeric-expression (see Numeric Expressions). c must be one character; register can be of arbitrary length. Such register assignments happen before any macro file is loaded, including the startup file. Due to getopt_long limitations, c cannot be, and register cannot contain, an equals sign, even though that is a valid character in a roff identifier.

-R

Run grefer preprocessor. No mechanism is provided for passing arguments to grefer because most grefer options have equivalent language elements that can be specified within the document.

gtroff also accepts a -R option, which is not accessible via groff. This option prevents the loading of the troffrc and troffrc-end files.

-s

Run gsoelim preprocessor.

-S

Operate in “safer” mode; see -U below for its opposite. For security reasons, safer mode is enabled by default.

-t

Run gtbl preprocessor.

-Tdev

Direct gtroff to format the input for the output device dev. groff then calls an output driver to convert gtroff’s output to a form appropriate for dev. The following output devices are available.

ps

For PostScript printers and previewers.

pdf

For PDF viewers or printers.

dvi

For TeX DVI format.

X75

For a 75dpi X11 previewer.

X75-12

For a 75dpi X11 previewer with a 12-point base font in the document.

X100

For a 100dpi X11 previewer.

X100-12

For a 100dpi X11 previewer with a 12-point base font in the document.

ascii

For typewriter-like devices using the (7-bit) ASCII (ISO 646) character set.

latin1

For typewriter-like devices that support the Latin-1 (ISO 8859-1) character set.

utf8

For typewriter-like devices that use the Unicode (ISO 10646) character set with UTF-8 encoding.

cp1047

For typewriter-like devices that use the EBCDIC encoding IBM code page 1047.

lj4

For HP LaserJet4-compatible (or other PCL5-compatible) printers.

lbp

For Canon CaPSL printers (LBP-4 and LBP-8 series laser printers).

html
xhtml

To produce HTML and XHTML output, respectively. This driver consists of two parts, a preprocessor (pre-grohtml) and a postprocessor (post-grohtml).

The predefined GNU troff string .T contains the name of the output device; the read-only register .T is set to 1 if this option is used (which is always true if groff is used to call GNU troff). See Built-in Registers.

The postprocessor to be used for a device is specified by the postpro command in the device description file. (See Device and Font Description Files.) This can be overridden with the -X option.

-U

Operate in unsafe mode, which enables the open, opena, pi, pso, and sy requests. These requests are disabled by default because they allow an untrusted input document to write to arbitrary file names and run arbitrary commands. This option also adds the current directory to the macro package search path; see the -m option above. -U is passed to gpic and gtroff.

-v

Write version information for groff and all programs run by it to the standard output stream; that is, the given command line is processed in the usual way, passing -v to the formatter and any pre- or postprocessors invoked.

-V

Output the pipeline that would be run by groff (as a wrapper program) to the standard output stream, but do not execute it. If given more than once, the pipeline is both written to the standard error stream and run.

-wcategory

Enable warnings in category. Categories are listed in Warnings.

-Wcategory

Inhibit warnings in category. Categories are listed in Warnings.

-X

Use gxditview instead of the usual postprocessor to (pre)view a document on an X11 display. Combining this option with -Tps uses the font metrics of the PostScript device, whereas the -TX75 and -TX100 options use the metrics of X11 fonts.

-z

Suppress formatted output from gtroff.

-Z

Disable postprocessing. gtroff output will appear on the standard output stream (unless suppressed with -z; see gtroff Output for a description of this format.


Next: , Previous: , Up: Invoking groff   [Contents][Index]

2.2 Environment

There are also several environment variables (of the operating system, not within gtroff) that can modify the behavior of groff.

GROFF_BIN_PATH

This search path, followed by PATH, is used for commands executed by groff.

GROFF_COMMAND_PREFIX

If this is set to X, then groff runs Xtroff instead of gtroff. This also applies to tbl, pic, eqn, grn, chem, refer, and soelim. It does not apply to grops, grodvi, grotty, pre-grohtml, post-grohtml, preconv, grolj4, gropdf, and gxditview.

The default command prefix is determined during the installation process. If a non-GNU troff system is found, prefix ‘g’ is used, none otherwise.

GROFF_ENCODING

The value of this variable is passed to the preconv preprocessor’s -e option to select the character encoding of input files. This variable’s existence implies the groff option -k. If set but empty, groff calls preconv without an -e option. groff’s -K option overrides GROFF_ENCODING. See the preconv(7) man page; type ‘man preconv’ at the command line to view it.

GROFF_FONT_PATH

A list of directories in which to seek the selected output device’s directory of device and font description files. GNU troff will search directories given as arguments to any specified -F options before these, and a built-in list of directories after them. See Font Directories and the troff(1) or gtroff(1) man pages.

GROFF_TMAC_PATH

A list of directories in which to seek macro files. GNU troff will search directories given as arguments to any specified -M options before these, and a built-in list of directories after them. See Macro Directories and the troff(1) or gtroff(1) man pages.

GROFF_TMPDIR

The directory in which groff creates temporary files. If this is not set and TMPDIR is set, temporary files are created in that directory. Otherwise temporary files are created in a system-dependent default directory (on Unix and GNU/Linux systems, this is usually /tmp). grops, grefer, pre-grohtml, and post-grohtml can create temporary files in this directory.

GROFF_TYPESETTER

Sets the default output device. If empty or not set, a build-time default (often ps) is used. The -Tdev option overrides GROFF_TYPESETTER.

SOURCE_DATE_EPOCH

A timestamp (expressed as seconds since the Unix epoch) to use as the output creation timestamp in place of the current time. The time is converted to human-readable form using localtime(3) when the formatter starts up and stored in registers usable by documents and macro packages (see Built-in Registers).

TZ

The time zone to use when converting the current time (or value of SOURCE_DATE_EPOCH) to human-readable form; see tzset(3).

MS-DOS and MS-Windows ports of groff use semicolons, rather than colons, to separate the directories in the lists described above.


Next: , Previous: , Up: Invoking groff   [Contents][Index]

2.3 Macro Directories

A macro file must have a name in the form name.tmac or tmac.name and be placed in a tmac directory to be found by the -mname command-line option.5 Together, these directories constitute the tmac path. Each directory is searched in the following order until the desired macro file is found or the list is exhausted.


Next: , Previous: , Up: Invoking groff   [Contents][Index]

2.4 Font Directories

groff enforces few restrictions on how font description files are named. For its family/style mechanism to work (see Font Families), the names of fonts within a family should start with the family name, followed by the style. For example, the Times family uses ‘T’ for the family name and ‘R’, ‘B’, ‘I’, and ‘BI’ to indicate the styles ‘roman’, ‘bold’, ‘italic’, and ‘bold italic’, respectively. Thus the final font names are ‘TR’, ‘TB’, ‘TI’, and ‘TBI’.

Font description files are kept in font directories, which together constitute the font path. The search procedure always appends the directory devname, where name is the name of the output device. Assuming TeX DVI output, and /foo/bar as a font directory, the font description files for grodvi must be in /foo/bar/devdvi. Each directory in the font path is searched in the following order until the desired font description file is found or the list is exhausted.


Next: , Previous: , Up: Invoking groff   [Contents][Index]

2.5 Paper Format

In groff, the page dimensions for the formatter GNU troff and for output devices are handled separately. See Page Layout, for vertical manipulation of the page size, and See Line Layout, for horizontal changes. The papersize macro package, normally loaded by troffrc at startup, provides an interface for configuring page dimensions by convenient names, like ‘letter’ or ‘a4’; see groff_tmac(5). The default used by the formatter depends on its build configuration, but is usually one of the foregoing, as geographically appropriate.

It is up to each macro package to respect the page dimensions configured in this way.

For each output device, the size of the output medium can be set in its DESC file. Most output drivers also recognize a command-line option -p to override the default dimensions and an option -l to use landscape orientation. See DESC File Format, for a description of the papersize keyword, which takes an argument of the same form as -p. The output driver’s man page, such as grops(1), may also be helpful.

groff uses the command-line option -P to pass options to postprocessors; for example, use the following for PostScript output on A4 paper in landscape orientation.

groff -Tps -dpaper=a4l -P-pa4 -P-l -ms foo.ms > foo.ps

Previous: , Up: Invoking groff   [Contents][Index]

2.6 Invocation Examples

roff systems are best known for formatting man pages. Once a man librarian program has located a man page, it may execute a groff command much like the following.

groff -t -man -Tutf8 /usr/share/man/man1/groff.1

The librarian will also pipe the output through a pager, which might not interpret the SGR terminal escape sequences groff emits for boldface, underlining, or italics; see the grotty(1) man page for a discussion.

To process a roff input file using the preprocessors gtbl and gpic and the me macro package in the way to which AT&T troff users were accustomed, one would type (or script) a pipeline.

gpic foo.me | gtbl | gtroff -me -Tutf8 | grotty

Using groff, this pipe can be shortened to an equivalent command.

groff -p -t -me -T utf8 foo.me

An even easier way to do this is to use grog to guess the preprocessor and macro options and execute the result by using the command substitution feature of the shell.

$(grog -Tutf8 foo.me)

Each command-line option to a postprocessor must be specified with any required leading dashes ‘-’ because groff passes the arguments as-is to the postprocessor; this permits arbitrary arguments to be transmitted. For example, to pass a title to the gxditview postprocessor, the shell commands

groff -X -P -title -P 'trial run' mydoc.t

and

groff -X -Z mydoc.t | gxditview -title 'trial run' -

are equivalent.


Next: , Previous: , Up: Top   [Contents][Index]

3 Tutorial for Macro Users

Most users of the roff language employ a macro package to format their documents. Successful macro packages ease the composition process; their users need not have mastered the full formatting language, nor understand features like diversions, traps, and environments. This chapter aims to familiarize you with basic concepts and mechanisms common to many macro packages (like “displays”). If you prefer a meticulous and comprehensive presentation, try GNU troff Reference instead.


Next: , Previous: , Up: Tutorial for Macro Users   [Contents][Index]

3.1 Basics

Let us first survey some basic concepts necessary to use a macro package fruitfully.6 References are made throughout to more detailed information.

GNU troff reads an input file prepared by the user and outputs a formatted document suitable for publication or framing. The input consists of text, or words to be printed, and embedded commands (requests and escape sequences), which tell GNU troff how to format the output. See Formatter Instructions.

The word argument is used in this chapter to mean a word or number that appears on the same line as a request, and which modifies the meaning of that request. For example, the request

.sp

spaces one line, but

.sp 4

spaces four lines. The number 4 is an argument to the sp request, which says to space four lines instead of one. Arguments are separated from the request and from each other by spaces (not tabs). See Invoking Requests.

The primary function of GNU troff is to collect words from input lines, fill output lines with those words, adjust the line to the right-hand margin by widening spaces, and output the result. For example, the input:

Now is the time
for all good men
to come to the aid
of their party.
Four score and seven
years ago, etc.

is read, packed onto output lines, and justified to produce:

  ⇒ Now is the time for all good men to come to the aid of
  ⇒ their party.  Four score and seven years ago, etc.

Sometimes a new output line should be started even though the current line is not yet full—for example, at the end of a paragraph. To do this it is possible to force a break, starting a new output line. Some requests cause a break automatically, as do (normally) blank input lines and input lines beginning with a space or tab.

Not all input lines are text lines—words to be formatted. Some are control lines that tell a macro package (or GNU troff directly) how to format the text. Control lines start with a dot (‘.’) or an apostrophe (‘'’) as the first character, and can be followed by a macro call.

The formatter also does more complex things, such as automatically numbering pages, skipping over page boundaries, putting footnotes in the correct place, and so forth.

Here are a few hints for preparing text for input to GNU troff.

We offer further advice in Input Conventions.

GNU troff permits alteration of the distance between lines of text. This is termed vertical spacing and is expressed in the same units as the type size—the point. The default is 10-point type on 12-point spacing. To get double-spaced text you would set the vertical spacing to 24 points. Some, but not all, macro packages expose a macro or register to configure the vertical spacing.

A number of requests allow you to change the way the output is arranged on the page, sometimes called the layout of the output page. Most macro packages don’t supply macros for performing these (at least not without performing other actions besides), as they are such basic operations. The macro packages for writing man pages, man and mdoc, don’t encourage explicit use of these requests at all.

The request ‘.sp N leaves N lines of blank space. N can be omitted (skipping a single line) or can be of the form Ni (for N inches) or Nc (for N centimeters). For example, the input:

.sp 1.5i
My thoughts on the subject
.sp

leaves one and a half inches of space, followed by the line “My thoughts on the subject”, followed by a single blank line (more measurement units are available; see Measurements).

If you seek precision in spacing, be advised when using a macro package that it might not honor sp requests as you expect; it can use a formatter feature called no-space mode to prevent excess space from accumulating. Macro packages typically offer registers to control spacing between paragraphs, before section headings, and around displays (discussed below); use these facilities preferentially. See Manipulating Spacing.

Text lines can be centered by using the ce request. The line after ce is centered (horizontally) on the page. To center more than one line, use ‘.ce N (where N is the number of lines to center), followed by the N lines. To center many lines without counting them, type:

.ce 1000
lines to center
.ce 0

The ‘.ce 0 request tells GNU troff to center zero more lines, in other words, stop centering.

GNU troff also offers the rj request for right-aligning text. It works analogously to ce and is convenient for setting epigraphs.

The bp request starts a new page; this necessarily implies an ordinary (line) break.

All of these requests cause a break; that is, they always start a new line. To start a new line without performing any other action, use br. If you invoke them with the apostrophe ‘'’, the no-break control character, the (initial) break they normally perform is suppressed. ‘'br’ does nothing.


Previous: , Up: Tutorial for Macro Users   [Contents][Index]

3.2 Common Features

GNU troff provides low-level operations for formatting a document. Many routine operations are undertaken in nearly all documents that require a series of such primitive operations to be performed. These common tasks are grouped into macros, which are then collected into a macro package.

Macro packages come in two varieties: “major” or “full-service” ones that manage page layout, and “minor” or “auxiliary” ones that do not, instead fulfilling narrow, specific tasks. Find a list in the groff_tmac(5) man page. Type ‘man groff_tmac’ at the command line to view it.

We survey several capabilities of full-service macro package below. Each package employs its own macros to exercise them. For details, consult its man page or, for ms, see ms.


Next: , Previous: , Up: Common Features   [Contents][Index]

3.2.1 Paragraphs

Paragraphs can be separated and indented in various ways. Some start with a blank line and have a first-line indentation, like most of the ones in this manual. Block paragraphs omit the indentation.

  ⇒ Some  men  look  at constitutions with sanctimonious
  ⇒ reverence,  and  deem  them  like  the  ark  of  the
  ⇒ covenant, too sacred to be touched.

We also frequently encounter tagged paragraphs, which begin with a tag or label at the left margin and indent the remaining text.

  ⇒ one  This  is the first paragraph.  Notice how the
  ⇒      first line of the resulting  paragraph  lines
  ⇒      up with the other lines in the paragraph.

If the tag is too wide for the indentation, the line is broken.

  ⇒ longlabel
  ⇒      The  label does not align with the subsequent
  ⇒      lines, but they align with each other.

A variation of the tagged paragraph is the itemized or enumerated paragraph, which might use punctuation or a digit for a tag, respectively. These are frequently used to construct lists.

  ⇒ o    This  list  item  starts with a bullet.  When
  ⇒      producing output for a device using the ASCII
  ⇒      character set, an 'o' is formatted instead.

Often, use of the same macro without a tag continues such a discussion.

  ⇒ -xyz  This option is recognized but ignored.
  ⇒
  ⇒       It had a security hole that we don't discuss.

Next: , Previous: , Up: Common Features   [Contents][Index]

3.2.2 Sections and Chapters

The simplest kind of section heading is unnumbered, set in a bold or italic style, and occupies a line by itself. Others possess automatically numbered multi-level headings and/or different typeface styles or sizes at different levels. More sophisticated macro packages supply macros for designating chapters and appendices.


Next: , Previous: , Up: Common Features   [Contents][Index]

3.2.3 Headers and Footers

Headers and footers occupy the top and bottom of each page, respectively, and contain data like the page number and the article or chapter title. Their appearance is not affected by the running text. Some packages allow for different titles on even- and odd-numbered pages (for printed, bound material).

Headers and footers are together called titles, and comprise three parts: left-aligned, centered, and right-aligned. A ‘%’ character appearing anywhere in a title is automatically replaced by the page number. See Page Layout.


Next: , Previous: , Up: Common Features   [Contents][Index]

3.2.4 Page Layout

Most macro packages let the user specify the size of the page margins. The top and bottom margins are typically handled differently than the left and right margins; the latter two are derived from the page offset, indentation, and line length. See Line Layout. Commonly, packages support registers to tune these values.


Next: , Previous: , Up: Common Features   [Contents][Index]

3.2.5 Displays and Keeps

Displays are sections of text set off from the surrounding material (typically paragraphs), often differing in indentation, and/or spacing. Tables, block quotations, and figures are displayed. Equations and code examples, when not much shorter than an output line, often are. Lists may or may not be. Packages for setting man pages support example displays but not keeps.

A keep is a group of output lines, often a display, that is formatted on a single page if possible; it causes a page break to happen early so as to not interrupt the kept material.

Floating keeps can move, or “float”, relative to the text around them in the input. They are useful for displays that are captioned and referred to by name, as with “See figure 3”. Depending on the package, a floating keep appears at the bottom of the current page if it fits, and at the top of the next otherwise. Alternatively, floating keeps might be deferred to the end of a section. Using a floating keep can avoid the large vertical spaces that may precede a tall keep of the ordinary sort when it won’t fit on the page.


Next: , Previous: , Up: Common Features   [Contents][Index]

3.2.6 Footnotes and Endnotes

Footnotes and endnotes are forms of delayed formatting. They are recorded at their points of relevance in the input, but not formatted there. Instead, a mark cues the reader to check the “foot”, or bottom, of the current page, or in the case of endnotes, an annotation list later in the document. Macro packages that support these features also supply a means of automatically numbering either type of annotation.


Next: , Previous: , Up: Common Features   [Contents][Index]

3.2.7 Table of Contents

A package may handle a table of contents by directing section heading macros to save section heading text and the page number where it occurs for use in a later entry for a table of contents. It writes the collected entries at the end of the document, once all are known, upon request. A row of dots (a leader) bridges the text on the left with its location on the right. Other collections might work in this manner, providing lists of figures or tables.

A table of contents is often found at the end of a GNU troff document because the formatter processes the document in a single pass. The gropdf output driver supports a PDF feature that relocates pages at the time the document is rendered; see the gropdf(1) man page. Type ‘man gropdf’ at the command line to view it.


Next: , Previous: , Up: Common Features   [Contents][Index]

3.2.8 Indexing

An index is similar to a table of contents, in that entry labels and locations must be collected, but poses a greater challenge because it needs to be sorted before it is output. Here, processing the document in multiple passes is inescapable, and tools like the makeindex program are necessary.


Next: , Previous: , Up: Common Features   [Contents][Index]

3.2.9 Document Formats

Some macro packages supply stock configurations of certain documents, like business letters and memoranda. These often also have provision for a cover sheet, which may be rigid in its format. With these features, it is even more important to use the package’s macros in preference to the formatter requests presented earlier, where possible.


Next: , Previous: , Up: Common Features   [Contents][Index]

3.2.10 Columnation

Macro packages apart from man and mdoc for man page formatting offer a facility for setting multiple columns on the page.


Next: , Previous: , Up: Common Features   [Contents][Index]

3.2.11 Font and Size Changes

The formatter’s requests and escape sequences for setting the typeface and size are not always intuitive, so all macro packages provide macros to make these operations simpler. They also make it more convenient to change typefaces in the middle of a word and can handle italic corrections automatically. See Italic Corrections.


Next: , Previous: , Up: Common Features   [Contents][Index]

3.2.12 Predefined Text

Most macro packages supply predefined strings to set prepared text like the date, or to perform operations like super- and subscripting.


Next: , Previous: , Up: Common Features   [Contents][Index]

3.2.13 Preprocessor Support

All macro packages provide support for various preprocessors and may extend their functionality by defining macros to set their contents in displays. Examples include TS and TE for gtbl, EQ and EN for geqn, and PS and PE for gpic.


Previous: , Up: Common Features   [Contents][Index]

3.2.14 Configuration and Customization

Packages provide means of customizing many of the details of how the package behaves. These range from setting the default type size to changing the appearance of section headers.


Next: , Previous: , Up: Top   [Contents][Index]

4 Macro Packages

This chapter surveys the “major” macro packages that come with groff. One, ms, is presented in detail.

Major macro packages are also sometimes described as full-service due to the breadth of features they provide and because more than one cannot be used by the same document; for example

groff -m man foo.man -m ms bar.doc

doesn’t work. Option arguments are processed before non-option arguments; the above (failing) sample is thus reordered to

groff -m man -m ms foo.man bar.doc

Many auxiliary, or “minor”, macro packages are also available. They may in general be used with any full-service macro package and handle a variety of tasks from character encoding selection, to language localization, to inlining of raster images. See the groff_tmac(5) man page for a list. Type ‘man groff_tmac’ at the command line to view it.


Next: , Previous: , Up: Major Macro Packages   [Contents][Index]

4.1 man

The man macro package is the most widely used and probably the most important ever developed for troff. It is easy to use, and a vast majority of manual pages (“man pages”) are written in it.

groff’s implementation is documented in the groff_man(7) man page. Type ‘man groff_man’ at the command line to view it.


Up: man   [Contents][Index]

4.1.1 Optional man extensions

Use the file man.local for local extensions to the man macros or for style changes.

Custom headers and footers

In groff versions 1.18.2 and later, you can specify custom headers and footers by redefining the following macros in man.local.

Macro: .PT

Control the content of the headers. Normally, the header prints the command name and section number on either side, and the optional fifth argument to TH in the center.

Macro: .BT

Control the content of the footers. Normally, the footer prints the page number and the third and fourth arguments to TH.

Use the FT register to specify the footer position. The default is -0.5i.

Ultrix-specific man macros

The groff source distribution includes a file named man.ultrix, containing macros compatible with the Ultrix variant of man. Copy this file into man.local (or use the mso request to load it) to enable the following macros.

Macro: .CT key

Print ‘<CTRL/key>’.

Macro: .CW

Print subsequent text using a “constant-width” (monospaced) typeface (Courier roman).

Macro: .Ds

Begin a non-filled display.

Macro: .De

End a non-filled display started with Ds.

Macro: .EX [indent]

Begin a non-filled display using a monospaced typeface (Courier roman). Use the optional indent argument to indent the display.

Macro: .EE

End a non-filled display started with EX.

Macro: .G [text]

Set text in Helvetica. If no text is present on the line where the macro is called, then the text of the next line appears in Helvetica.

Macro: .GL [text]

Set text in Helvetica oblique. If no text is present on the line where the macro is called, then the text of the next line appears in Helvetica Oblique.

Macro: .HB [text]

Set text in Helvetica bold. If no text is present on the line where the macro is called, then all text up to the next HB appears in Helvetica bold.

Macro: .TB [text]

Identical to HB.

Macro: .MS title sect [punct]

Set a man page reference in Ultrix format. The title is in Courier instead of italic. Optional punctuation follows the section number without an intervening space.

Macro: .NT [C] [title]

Begin a note. Print the optional title, or the word “Note”, centered on the page. Text following the macro makes up the body of the note, and is indented on both sides. If the first argument is C, the body of the note is printed centered (the second argument replaces the word “Note” if specified).

Macro: .NE

End a note begun with NT.

Macro: .PN path [punct]

Set the path name in a monospaced typeface (Courier roman), followed by optional punctuation.

Macro: .Pn [punct] path [punct]

If called with two arguments, identical to PN. If called with three arguments, set the second argument in a monospaced typeface (Courier roman), bracketed by the first and third arguments in the current font.

Macro: .R

Switch to roman font and turn off any underlining in effect.

Macro: .RN

Print the string ‘<RETURN>’.

Macro: .VS [4]

Start printing a change bar in the margin if the number 4 is specified. Otherwise, this macro does nothing.

Macro: .VE

End printing the change bar begun by VS.

Simple example

The following example man.local file alters the SH macro to add some extra vertical space before printing the heading. Headings are printed in Helvetica bold.

.\" Make the heading fonts Helvetica
.ds HF HB
.
.\" Put more space in front of headings.
.rn SH SH-orig
.de SH
.  if t .sp (u;\\n[PD]*2)
.  SH-orig \\$*
..

Next: , Previous: , Up: Major Macro Packages   [Contents][Index]

4.2 mdoc

groff’s implementation of the BSD doc package for man pages is documented in the groff_mdoc(7) man page. Type ‘man groff_mdoc’ at the command line to view it.


Next: , Previous: , Up: Major Macro Packages   [Contents][Index]

4.3 me

groff’s implementation of the BSD me macro package is documented using itself. A tutorial, meintro.me, and reference, meref.me, are available in groff’s documentation directory. A groff_me(7) man page is also available and identifies the installation path for these documents. Type ‘man groff_me’ at the command line to view it.

A French translation of the tutorial is available as meintro_fr.me and installed parallel to the English version.


Next: , Previous: , Up: Major Macro Packages   [Contents][Index]

4.4 mm

groff’s implementation of the AT&T memorandum macro package is documented in the groff_mm(7) man page. Type ‘man groff_mm’ at the command line) to view it.

A Swedish localization of mm is also available; see groff_mmse(7).


Next: , Previous: , Up: Major Macro Packages   [Contents][Index]

4.5 mom

The main documentation files for the mom macros are in HTML format. Additional, useful documentation is in PDF format. See the groff(1) man page, section “Installation Directories”, for their location.

The mom macros are in active development between groff releases. The most recent version, along with up-to-date documentation, is available at http://www.schaffter.ca/mom/mom-05.html.

The groff_mom(7) man page (type ‘man groff_mom’ at the command line) contains a partial list of available macros, however their usage is best understood by consulting the HTML documentation.


Previous: , Up: Major Macro Packages   [Contents][Index]

4.6 ms

The ms (“manuscript”) package is suitable for the preparation of letters, memoranda, reports, and books. These groff macros feature cover page and table of contents generation, automatically numbered headings, several paragraph styles, a variety of text styling options, footnotes, and multi-column page layouts. ms supports the tbl, eqn, pic, and refer preprocessors for inclusion of tables, mathematical equations, diagrams, and standardized bibliographic citations. This implementation is mostly compatible with the documented interface and behavior of AT&T Unix Version 7 ms. Many extensions from 4.2BSD (Berkeley) and Tenth Edition Research Unix have been recreated.


Next: , Previous: , Up: ms   [Contents][Index]

4.6.1 Introduction

The ms macros are the oldest surviving package for roff systems.7 While the man package was designed for brief reference documents, the ms macros are also suitable for longer works intended for printing and possible publication.


Next: , Previous: , Up: ms Introduction   [Contents][Index]

4.6.1.1 Basic information

ms documents are plain text files; prepare them with your preferred text editor. If you’re in a hurry to start, know that ms needs one of its macros called at the beginning of a document so that it can initialize. A macro is a formatting instruction to ms. Put a macro call on a line by itself. Use ‘.PP’ if you want your paragraph’s first line to be indented, or ‘.LP’ if you don’t.

After that, start typing normally. It is a good practice to start each sentence on a new line, or to put two spaces after sentence-ending punctuation, so that the formatter knows where the sentence boundaries are. You can separate paragraphs with further paragraphing macros, or with blank lines, and you can indent with tabs. When you need one of the features mentioned earlier (see ms), return to this part of the manual.

Format the document with the groff command. nroff can be useful for previewing.

$ editor radical.ms
$ nroff -ww -z -ms radical.ms # check for errors
$ nroff -ms radical.ms | less -R
$ groff -T ps -ms radical.ms > radical.ps
$ see radical.ps

Our radical.ms document might look like this.

.LP
Radical novelties are so disturbing that they tend to be
suppressed or ignored, to the extent that even the
possibility of their existence in general is more often
denied than admitted.

→That's what Dijkstra said, anyway.

ms exposes many aspects of document layout to user control via groff’s registers and strings, which store numbers and text, respectively. Measurements in groff are expressed with a suffix called a scaling unit.

i

inches

c

centimeters

p

points (1/72 inch)

P

picas (1/6 inch)

v

vees; current vertical spacing

m

ems; width of an “M” in the current font

n

ens; one-half em

Set registers with the nr request and strings with the ds request. Requests are like macro calls; they go on lines by themselves and start with the control character, a dot (.). The difference is that they directly instruct the formatter program, rather than the macro package. We’ll discuss a few as applicable. It is wise to specify a scaling unit when setting any register that represents a length, size, or distance.

.nr PS 10.5p \" Use 10.5-point type.
.ds FAM P    \" Use Palatino font family.

In the foregoing, we see that \" begins a comment. This is an example of an escape sequence, the other kind of formatting instruction. Escape sequences can appear anywhere. They begin with the escape character (\) and are followed by at least one more character. ms documents tend to use only a few of groff’s many requests and escape sequences; see Request Index and Escape Sequence Index or the groff(7) man page for complete lists.

\"

Begin comment; ignore remainder of line.

\n[reg]

Interpolate value of register reg.

\*[str]

Interpolate contents of string str.

\*s

abbreviation of \*[s]; the name s must be only one character

\[char]

Interpolate glyph of special character named char.

\&

dummy character

\~

Insert an unbreakable space that is adjustable like a normal space.

\|

Move horizontally by one-sixth em (“thin space”).

Prefix any words that start with a dot ‘.’ or neutral apostrophe ‘'’ with \& if they are at the beginning of an input line (or might become that way in editing) to prevent them from being interpreted as macro calls or requests. Suffix ‘.’, ‘?’, and ‘!’ with \& when needed to cancel end-of-sentence detection.

My exposure was \&.5 to \&.6 Sv of neutrons, said Dr.\&
Wallace after the criticality incident.

Next: , Previous: , Up: ms   [Contents][Index]

4.6.2 Document Structure

The ms macro package expects a certain amount of structure: a well-formed document contains at least one paragraphing or heading macro call. Longer documents have a structure as follows.

Document type

Calling the RP macro at the beginning of your document puts the document description (see below) on a cover page. Otherwise, ms places the information (if any) on the first page, followed immediately by the body text. Some document types found in other ms implementations are specific to AT&T or Berkeley, and are not supported by groff ms.

Format and layout

By setting registers and strings, you can configure your document’s typeface, margins, spacing, headers and footers, and footnote arrangement. See ms Document Control Settings.

Document description

A document description consists of any of: a title, one or more authors’ names and affiliated institutions, an abstract, and a date or other identifier. See ms Document Description Macros.

Body text

The main matter of your document follows its description (if any). ms supports highly structured text consisting of paragraphs interspersed with multi-level headings (chapters, sections, subsections, and so forth) and augmented by lists, footnotes, tables, diagrams, and similar material. See ms Body Text.

Tables of contents

Macros enable the collection of entries for a table of contents (or index) as the material they discuss appears in the document. You then call a macro to emit the table of contents at the end of your document. The table of contents must necessarily follow the rest of the text since GNU troff is a single-pass formatter; it thus cannot determine the page number of a division of the text until it has been set and output. Since ms was designed for the production of hard copy, the traditional procedure was to manually relocate the pages containing the table of contents between the cover page and the body text. Today, page resequencing is more often done in the digital domain. An index works similarly, but because it typically needs to be sorted after collection, its preparation requires separate processing.


Next: , Previous: , Up: ms   [Contents][Index]

4.6.3 Document Control Settings

ms exposes many aspects of document layout to user control via groff requests. To use them, you must understand how to define registers and strings.

Request: .nr reg value

Set register reg to value. If reg doesn’t exist, GNU troff creates it.

Request: .ds name contents

Set string name to contents.

A list of document control registers and strings follows. For any parameter whose default is unsatisfactory, define its register or string before calling any ms macro other than RP.

Margin settings

Register: \n[PO]

Defines the page offset (i.e., the left margin).

Effective: next page.

Default: Varies by output device and paper format; 1i is used for typesetters using U.S. letter paper, and zero for terminals. See Paper Format.

Register: \n[LL]

Defines the line length (i.e., the width of the body text).

Effective: next paragraph.

Default: Varies by output device and paper format; 6.5i is used for typesetters using U.S. letter paper (see Paper Format) and 65n on terminals.

Register: \n[LT]

Defines the title line length (i.e., the header and footer width). This is usually the same as LL, but need not be.

Effective: next paragraph.

Default: Varies by output device and paper format; 6.5i is used for typesetters using U.S. letter paper (see Paper Format) and 65n on terminals.

Register: \n[HM]

Defines the header margin height at the top of the page.

Effective: next page.

Default: 1i.

Register: \n[FM]

Defines the footer margin height at the bottom of the page.

Effective: next page.

Default: 1i.

Titles (headers, footers)

String: \*[LH]

Defines the text displayed in the left header position.

Effective: next header.

Default: empty.

String: \*[CH]

Defines the text displayed in the center header position.

Effective: next header.

Default: ‘-\n[%]-’.

String: \*[RH]

Defines the text displayed in the right header position.

Effective: next header.

Default: empty.

String: \*[LF]

Defines the text displayed in the left footer position.

Effective: next footer.

Default: empty.

String: \*[CF]

Defines the text displayed in the center footer position.

Effective: next footer.

Default: empty.

String: \*[RF]

Defines the text displayed in the right footer position.

Effective: next footer.

Default: empty.

Text settings

Register: \n[PS]

Defines the type size of the body text.

Effective: next paragraph.

Default: 10p.

Register: \n[VS]

Defines the vertical spacing (type size plus leading).

Effective: next paragraph.

Default: 12p.

Register: \n[HY]

Defines the automatic hyphenation mode used with the hy request. Setting HY to 0 is equivalent to using the nh request. This is a Tenth Edition Research Unix extension.

Effective: next paragraph.

Default: 6.

String: \*[FAM]

Defines the font family used to typeset the document. This is a GNU extension.

Effective: next paragraph.

Default: defined by the output device; often ‘T’ (see ms Body Text)

Paragraph settings

Register: \n[PI]

Defines the indentation amount used by the PP, IP (unless overridden by an optional argument), XP, and RS macros.

Effective: next paragraph.

Default: 5n.

Register: \n[PD]

Defines the space between paragraphs.

Effective: next paragraph.

Default: 0.3v (1v on low-resolution devices).

Register: \n[QI]

Defines the indentation amount used on both sides of a paragraph set with the QP or between the QS and QE macros.

Effective: next paragraph.

Default: 5n.

Register: \n[PORPHANS]

Defines the minimum number of initial lines of any paragraph that must be kept together to avoid isolated lines at the bottom of a page. If a new paragraph is started close to the bottom of a page, and there is insufficient space to accommodate PORPHANS lines before an automatic page break, then a page break is forced before the start of the paragraph. This is a GNU extension.

Effective: next paragraph.

Default: 1.

Heading settings

Register: \n[PSINCR]

Defines an increment in type size to be applied to a heading at a lesser depth than that specified in GROWPS. The value of PSINCR should be specified in points with the p scaling unit and may include a fractional component; for example, ‘.nr PSINCR 1.5p sets a type size increment of 1.5p. This is a GNU extension.

Effective: next heading.

Default: 1p.

Register: \n[GROWPS]

Defines the heading depth above which the type size increment set by PSINCR becomes effective. For each heading depth less than the value of GROWPS, the type size is increased by PSINCR. Setting GROWPS to any value less than 2 disables the incremental heading size feature. This is a GNU extension.

Effective: next heading.

Default: 0.

Register: \n[HORPHANS]

Defines the minimum number of lines of an immediately succeeding paragraph that should be kept together with any heading introduced by the NH or SH macros. If a heading is placed close to the bottom of a page, and there is insufficient space to accommodate both the heading and at least HORPHANS lines of the following paragraph, before an automatic page break, then the page break is forced before the heading. This is a GNU extension.

Effective: next paragraph.

Default: 1.

String: \*[SN-STYLE]

Defines the style used to print numbered headings. See Headings in ms. This is a GNU extension.

Effective: next heading.

Default: alias of SN-DOT

Footnote settings

Register: \n[FI]

Defines the footnote indentation. This is a Berkeley extension.

Effective: next footnote.

Default: 2n.

Register: \n[FF]

Defines the format of automatically numbered footnotes, and those for which the FS request is given a marker argument, at the bottom of a column or page. This is a Berkeley extension.

0

Set an automatic number8 as a superscript (on typesetter devices) or surrounded by square brackets (on terminals). The footnote paragraph is indented as with PP if there is an FS argument or an automatic number, and as with LP otherwise. This is the default.

1

As 0, but set the marker as regular text and follow an automatic number with a period.

2

As 1, but without indentation (like LP).

3

As 1, but set the footnote paragraph with the marker hanging (like IP).

Effective: next footnote.

Default: 0.

Register: \n[FPS]

Defines the footnote type size.

Effective: next footnote.

Default: \n[PS] - 2p.

Register: \n[FVS]

Defines the footnote vertical spacing.

Effective: next footnote.

Default: \n[FPS] + 2p.

Register: \n[FPD]

Defines the footnote paragraph spacing. This is a GNU extension.

Effective: next footnote.

Default: \n[PD] / 2.

String: \*[FR]

Defines the ratio of the footnote line length to the current line length. This is a GNU extension.

Effective: next footnote in single-column arrangements, next page otherwise.

Default: 11/12.

Display settings

Register: \n[DD]

Sets the display distance—the vertical spacing before and after a display, a tbl table, an eqn equation, or a pic image. This is a Berkeley extension.

Effective: next display boundary.

Default: 0.5v (1v on low-resolution devices).

Register: \n[DI]

Sets the default amount by which to indent a display started with DS and ID without arguments, to ‘.DS I’ without an indentation argument, and to equations set with ‘.EQ I’. This is a GNU extension.

Effective: next indented display.

Default: 0.5i.

Other settings

Register: \n[MINGW]

Defines the default minimum width between columns in a multi-column document. This is a GNU extension.

Effective: next page.

Default: 2n.

Register: \n[TC-MARGIN]

Defines the width of the field in which page numbers are set in a table of contents entry; the right margin thus moves inboard by this amount. This is a GNU extension.

Effective: next PX call.

Default: \w'000'


Next: , Previous: , Up: ms   [Contents][Index]

4.6.4 Document Description Macros

Only the simplest document lacks a title.9 As its level of sophistication (or complexity) increases, it tends to acquire a date of revision, explicitly identified authors, sponsoring institutions for authors, and, at the rarefied heights, an abstract of its content. Define these data by calling the macros below in the order shown; DA or ND can be called to set the document date (or other identifier) at any time before (a) the abstract, if present, or (b) its information is required in a header or footer. Use of these macros is optional, except that TL is mandatory if any of RP, AU, AI, or AB is called, and AE is mandatory if AB is called.

Macro: .RP [no-repeat-info] [no-renumber]

Use the “report” (AT&T: “released paper”) format for your document, creating a separate cover page. The default arrangement is to place most of the document description (title, author names and institutions, and abstract, but not the date) at the top of the first page. If the optional no-repeat-info argument is given, ms produces a cover page but does not repeat any of its information subsequently (but see the DA macro below regarding the date). Normally, RP sets the page number following the cover page to 1. Specifying the optional no-renumber argument suppresses this alteration. Optional arguments can occur in any order. no is recognized as a synonym of no-repeat-info for AT&T compatibility.

Macro: .TL

Specify the document title. ms collects text on input lines following this call into the title until reaching AU, AB, or a heading or paragraphing macro call.

Macro: .AU

Specify an author’s name. ms collects text on input lines following this call into the author’s name until reaching AI, AB, another AU, or a heading or paragraphing macro call. Call it repeatedly to specify multiple authors.

Macro: .AI

Specify the preceding author’s institution. An AU call is usefully followed by at most one AI call; if there are more, the last AI call controls. ms collects text on input lines following this call into the author’s institution until reaching AU, AB, or a heading or paragraphing macro call.

Macro: .DA [x …]

Typeset the current date, or any arguments x, in the center footer, and, if RP is also called, left-aligned at the end of the description information on the cover page.

Macro: .ND [x …]

Typeset the current date, or any arguments x, if RP is also called, left-aligned at the end of the document description on the cover page. This is groff ms’s default.

Macro: .AB [no]

Begin the abstract. ms collects text on input lines following this call into the abstract until reaching an AE call. By default, ms places the word “ABSTRACT” centered and in italics above the text of the abstract. The optional argument no suppresses this heading.

Macro: .AE

End the abstract.

An example document description, using a cover page, follows.

.RP
.TL
The Inevitability of Code Bloat
in Commercial and Free Software
.AU
J.\& Random Luser
.AI
University of West Bumblefuzz
.AB
This report examines the long-term growth of the code
bases in two large,
popular software packages;
the free Emacs and the commercial Microsoft Word.
While differences appear in the type or order of
features added,
due to the different methodologies used,
the results are the same in the end.
.PP
The free software approach is shown to be superior in
that while free software can become as bloated as
commercial offerings,
free software tends to have fewer serious bugs and the
added features are more in line with user demand.
.AE

…the rest of the paper…

Next: , Previous: , Up: ms   [Contents][Index]

4.6.5 Body Text

A variety of macros, registers, and strings can be used to structure and style the body of your document. They organize your text into paragraphs, headings, footnotes, and inclusions of material such as tables and figures.


Next: , Previous: , Up: ms Body Text   [Contents][Index]

4.6.5.1 Text settings

The FAM string, a GNU extension, sets the font family for body text; the default is ‘T’. The PS and VS registers set the type size and vertical spacing (distance between text baselines), respectively. The font family and type size are ignored on terminal devices. Setting these parameters before the first call of a heading, paragraphing, or (non-date) document description macro also applies them to headers, footers, and (for FAM) footnotes.

Which font families are available depends on the output device; as a convention, T selects a serif family (“Times”), H a sans-serif family (“Helvetica”), and C a monospaced family (“Courier”). The man page for the output driver documents its font repertoire. Consult the groff(1) man page for lists of available output devices and their drivers.

The hyphenation mode (as used by the hy request) is set from the HY register. Setting HY to ‘0’ is equivalent to using the nh request. This is a Tenth Edition Research Unix extension.


Next: , Previous: , Up: ms Body Text   [Contents][Index]

4.6.5.2 Typographical symbols

ms provides a few strings to obtain typographical symbols not easily entered with the keyboard. These and many others are available as special character escape sequences—see the groff_char(7) man page.

String: \*[-]

Interpolate an em dash.

String: \*[Q]
String: \*[U]

Interpolate typographer’s quotation marks where available, and neutral double quotes otherwise. \*Q is the left quote and \*U the right.


Next: , Previous: , Up: ms Body Text   [Contents][Index]

4.6.5.3 Paragraphs

Paragraphing macros break, or terminate, any pending output line so that a new paragraph can begin. Several paragraph types are available, differing in how indentation applies to them: to left, right, or both margins; to the first output line of the paragraph, all output lines, or all but the first. All paragraphing macro calls cause the insertion of vertical space in the amount stored in the PD register, except at page or column breaks. Alternatively, a blank input line breaks the output line and vertically spaces by one vee.

Macro: .LP

Set a paragraph without any (additional) indentation.

Macro: .PP

Set a paragraph with a first-line left indentation in the amount stored in the PI register.

Macro: .IP [marker [width]]

Set a paragraph with a left indentation. The optional marker is not indented and is empty by default. It has several applications; see Lists in ms. width overrides the indentation amount stored in the PI register; its default unit is ‘n’. Once specified, width applies to further IP calls until specified again or a heading or different paragraphing macro is called.

Macro: .QP

Set a paragraph indented from both left and right margins by the amount stored in the QI register.

Macro: .QS
Macro: .QE

Begin (QS) and end (QE) a region where each paragraph is indented from both margins by the amount stored in the QI register. The text between QS and QE can be structured further by use of other paragraphing macros.

Macro: .XP

Set an “exdented” paragraph—one with a left indentation in the amount stored in the PI register on every line except the first (also known as a hanging indent). This is a Berkeley extension.

The following example illustrates the use of paragraphing macros.

.NH 2
Cases used in the 2001 study
.LP
Two software releases were considered for this report.
.PP
The first is commercial software;
the second is free.
.IP \[bu]
Microsoft Word for Windows,
starting with version 1.0 through the current version
(Word 2000).
.IP \[bu]
GNU Emacs,
from its first appearance as a standalone editor through
the current version (v20).
See [Bloggs 2002] for details.
.QP
Franklin's Law applied to software:
software expands to outgrow both RAM and disk space over
time.
.SH
Bibliography
.XP
Bloggs, Joseph R.,
.I "Everyone's a Critic" ,
Underground Press, March 2002.
A definitive work that answers all questions and
criticisms about the quality and usability of free
software.

Next: , Previous: , Up: ms Body Text   [Contents][Index]

4.6.5.4 Headings

Use headings to create a sequential or hierarchical structure for your document. The ms macros print headings in bold using the same font family and, by default, type size as the body text. Headings are available with and without automatic numbering. Text on input lines following the macro call becomes the heading’s title. Call a paragraphing macro to end the heading text and start the section’s content.

Macro: .NH [depth]
Macro: .NH S heading-depth-index

Set an automatically numbered heading.

ms produces a numbered heading the form a.b.c…, to any depth desired, with the numbering of each depth increasing automatically and being reset to zero when a more significant level is increased. “1” is the most significant or coarsest division of the document. Only non-zero values are output. If depth is omitted, it is taken to be ‘1’.

If you specify depth such that an ascending gap occurs relative to the previous NH call—that is, you “skip a depth”, as by ‘.NH 1’ and then ‘.NH 3’—groff ms emits a warning on the standard error stream.

Alternatively, you can give NH a first argument of S, followed by integers to number the heading depths explicitly. Further automatic numbering, if used, resumes using the specified indices as their predecessors. This feature is a Berkeley extension.

An example may be illustrative.

.NH 1
Animalia
.NH 2
Arthropoda
.NH 3
Crustacea
.NH 2
Chordata
.NH S 6 6 6
Daimonia
.NH 1
Plantae

The above results in numbering as follows; the vertical space that normally precedes each heading is omitted.

1.  Animalia
1.1.  Arthropoda
1.1.1.  Crustacea
1.2.  Chordata
6.6.6.  Daimonia
7.  Plantae
String: \*[SN-STYLE]
String: \*[SN-DOT]
String: \*[SN-NO-DOT]
String: \*[SN]

After NH is called, the assigned number is made available in the strings SN-DOT (as it appears in a printed heading with default formatting, followed by a terminating period) and SN-NO-DOT (with the terminating period omitted). These are GNU extensions.

You can control the style used to print numbered headings by defining an appropriate alias for the string SN-STYLE. By default, SN-STYLE is aliased to SN-DOT. If you prefer to omit the terminating period from numbers appearing in numbered headings, you may define the alias as follows.

.als SN-STYLE SN-NO-DOT

Any such change in numbering style becomes effective from the next use of NH following redefinition of the alias for SN-STYLE. The formatted number of the current heading is available in the SN string (a feature first documented by Berkeley), which facilitates its inclusion in, for example, table captions, equation labels, and XS/XA/XE table of contents entries.

Macro: .SH [depth]

Set an unnumbered heading.

The optional depth argument is a GNU extension indicating the heading depth corresponding to the depth argument of NH. It matches the type size at which the heading is set to that of a numbered heading at the same depth when the GROWPS and PSINCR heading size adjustment mechanism is in effect.

If the GROWPS register is set to a value greater than the level argument to NH or SH, the type size of a heading produced by these macros increases by PSINCR units over the size specified by PS multiplied by the difference of GROWPS and level. The value stored in PSINCR is interpreted in groff basic units; the p scaling unit should be employed when assigning a value specified in points. For example, the sequence

.nr PS 10
.nr GROWPS 3
.nr PSINCR 1.5p
.NH 1
Carnivora
.NH 2
Felinae
.NH 3
Felis catus
.SH 2
Machairodontinae

will cause “1. Carnivora” to be printed in 13-point text, followed by “1.1. Felinae” in 11.5-point text, while “1.1.1. Felis catus” and all more deeply nested heading levels will remain in the 10-point text specified by the PS register. “Machairodontinae” is printed at 11.5 points, since it corresponds to heading level 2.

The HORPHANS register operates in conjunction with the NH and SH macros to inhibit the printing of isolated headings at the bottom of a page; it specifies the minimum number of lines of an immediately subsequent paragraph that must be kept on the same page as the heading. If insufficient space remains on the current page to accommodate the heading and this number of lines of paragraph text, a page break is forced before the heading is printed. Any display macro call or tbl, pic, or eqn region between the heading and the subsequent paragraph suppresses this grouping. See ms keeps and displays and ms Insertions.


Next: , Previous: , Up: ms Body Text   [Contents][Index]

4.6.5.5 Typeface and decoration

The ms macros provide a variety of ways to style text. Attend closely to the ordering of arguments labeled pre and post, which is not intuitive. Support for pre arguments is a GNU extension.10

Macro: .B [text [post [pre]]]

Style text in bold, followed by post in the previous font style without intervening space, and preceded by pre similarly. Without arguments, ms styles subsequent text in bold until the next paragraphing, heading, or no-argument typeface macro call.

Macro: .R [text [post [pre]]]

As B, but use the roman style (upright text of normal weight) instead of bold. Argument recognition is a GNU extension.

Macro: .I [text [post [pre]]]

As B, but use an italic or oblique style instead of bold.

Macro: .BI [text [post [pre]]]

As B, but use a bold italic or bold oblique style instead of upright bold. This is a Tenth Edition Research Unix extension.

Macro: .CW [text [post [pre]]]

As B, but use a constant-width (monospaced) roman typeface instead of bold. This is a Tenth Edition Research Unix extension.

Macro: .BX [text]

Typeset text and draw a box around it. On terminal devices, reverse video is used instead. If you want text to contain space, use unbreakable space or horizontal motion escape sequences (\~, \SP, \^, \|, \0 or \h).

Macro: .UL [text [post]]

Typeset text with an underline. post, if present, is set after text with no intervening space.

Macro: .LG

Set subsequent text in larger type (two points larger than the current size) until the next type size, paragraphing, or heading macro call. You can specify this macro multiple times to enlarge the type size as needed.

Macro: .SM

Set subsequent text in smaller type (two points smaller than the current size) until the next type size, paragraphing, or heading macro call. You can specify this macro multiple times to reduce the type size as needed.

Macro: .NL

Set subsequent text at the normal type size (the amount in the PS register).

pre and post arguments are typically used to simplify the attachment of punctuation to styled words. When pre is used, a hyphenation control escape sequence \% that would ordinarily start text must start pre instead to have the desired effect.

The CS course's students found one C language keyword
.CW static ) \%(
most troublesome.

The foregoing example produces output as follows.

The CS course’s students found one C language keyword (static)
most troublesome.

You can use the output line continuation escape sequence \c to achieve the same result (see Line Continuation). It is also portable to older ms implementations.

The CS course's students found one C language keyword
\%(\c
.CW \%static )
most troublesome.

groff ms also offers strings to begin and end super- and subscripting. These are GNU extensions.

String: \*[{]
String: \*[}]

Begin and end superscripting, respectively.

String: \*[<]
String: \*[>]

Begin and end subscripting, respectively.

Rather than calling the CW macro, in groff ms you might prefer to change the font family to Courier by setting the FAM string to ‘C’. You can then use all four style macros above, returning to the default family (Times) with ‘.ds FAM T’. Because changes to FAM take effect only at the next paragraph, CW remains useful to “inline” a change to the font family, similarly to the practice of this document in noting syntactical elements of ms and groff.


Next: , Previous: , Up: ms Body Text   [Contents][Index]

4.6.5.6 Lists

The marker argument to the IP macro can be employed to present a variety of lists; for instance, you can use a bullet glyph (\[bu]) for unordered lists, a number (or auto-incrementing register) for numbered lists, or a word or phrase for glossary-style or definition lists. If you set the paragraph indentation register PI before calling IP, you can later reorder the items in the list without having to ensure that a width argument remains affixed to the first call.

The following is an example of a bulleted list.

.nr PI 2n
A bulleted list:
.IP \[bu]
lawyers
.IP \[bu]
guns
.IP \[bu]
money
A bulleted list:

• lawyers

• guns

• money

The following is an example of a numbered list.

.nr step 0 1
.nr PI 3n
A numbered list:
.IP \n+[step]
lawyers
.IP \n+[step]
guns
.IP \n+[step]
money
A numbered list:

1. lawyers

2. guns

3. money

Here we have employed the nr request to create a register of our own, ‘step’. We initialized it to zero and assigned it an auto-increment of 1. Each time we use the escape sequence ‘\n+[PI]’ (note the plus sign), the formatter applies the increment just before interpolating the register’s value. Preparing the PI register as well enables us to rearrange the list without the tedium of updating macro calls.

The next example illustrates a glossary-style list.

A glossary-style list:
.IP lawyers 0.4i
Two or more attorneys.
.IP guns
Firearms,
preferably large-caliber.
.IP money
Gotta pay for those
lawyers and guns!
A glossary-style list:

lawyers
      Two or more attorneys.

guns  Firearms, preferably large-caliber.

money
      Gotta pay for those lawyers and guns!

In the previous example, observe how the IP macro places the definition on the same line as the term if it has enough space. If this is not what you want, there are a few workarounds we will illustrate by modifying the example. First, you can use a br request to force a break after printing the term or label.

.IP guns
.br
Firearms,

Second, you could apply the \p escape sequence to force a break. The space following the escape sequence is important; if you omit it, groff prints the first word of the paragraph text on the same line as the term or label (if it fits) then breaks the line.

.IP guns
\p Firearms,

Finally, you may append a horizontal motion to the marker with the \h escape sequence; using the same amount as the indentation will ensure that the marker is too wide for groff to treat it as “fitting” on the same line as the paragraph text.

.IP guns\h'0.4i'
Firearms,

In each case, the result is the same.

A glossary-style list:

lawyers
      Two or more attorneys.

guns
      Firearms, preferably large-caliber.

money
      Gotta pay for those lawyers and guns!

Next: , Previous: , Up: ms Body Text   [Contents][Index]

4.6.5.7 Indented regions

You may need to indent a region of text while otherwise formatting it normally. Indented regions can be nested; you can change \n[PI] before each call to vary the amount of inset.

Macro: .RS

Begin a region where headings, paragraphs, and displays are indented (further) by the amount stored in the PI register.

Macro: .RE

End the (next) most recent indented region.

This feature enables you to easily line up text under hanging and indented paragraphs. For example, you may wish to structure lists hierarchically.

.IP \[bu] 2
Lawyers:
.RS
.IP \[bu]
Dewey,
.IP \[bu]
Cheatham,
and
.IP \[bu]
and Howe.
.RE
.IP \[bu]
Guns
• Lawyers:

  •  Dewey,

  •  Cheatham, and

  •  Howe.

• Guns

Next: , Previous: , Up: ms Body Text   [Contents][Index]

4.6.5.8 Keeps, boxed keeps, and displays

On occasion, you may want to keep several lines of text, or a region of a document, together on a single page, preventing an automatic page break within certain boundaries. This can cause a page break to occur earlier than it normally would. For example, you may want to keep two paragraphs together, or a paragraph that refers to a table, list, or figure adjacent to the item it discusses. ms provides the KS and KE macros for this purpose.

You can alternatively specify a floating keep: if a keep cannot fit on the current page, ms holds its contents and allows material following the keep (in the source document) to fill the remainder of the current page. When the page breaks, whether by reaching the end or bp request, ms puts the floating keep at the beginning of the next page. This is useful for placing large graphics or tables that do not need to appear exactly where they occur in the source document.

Macro: .KS
Macro: .KF
Macro: .KE

KS begins a keep, KF a floating keep, and KE ends a keep of either kind.

As an alternative to the keep mechanism, the ne request forces a page break if there is not at least the amount of vertical space specified in its argument remaining on the page (see Page Control). One application of ne is to reserve space on the page for a figure or illustration to be included later.

A boxed keep has a frame drawn around it.

Macro: .B1
Macro: .B2

B1 begins a keep with a box drawn around it. B2 ends a boxed keep.

Boxed keep macros cause breaks; if you need to box a word or phrase within a line, see the BX macro in Typeface and decoration. Box lines are drawn as close as possible to the text they enclose so that they are usable within paragraphs. If you wish to box one or more paragraphs, you may improve the appearance by calling B1 after the first paragraphing macro, and by adding a small amount of vertical space before calling B2.

.LP
.B1
.I Warning:
Happy Fun Ball may suddenly accelerate to dangerous
speeds.
.sp \n[PD]/2 \" space by half the inter-paragraph distance
.B2

If you want a boxed keep to float, you will need to enclose the B1 and B2 calls within a pair of KF and KE calls.

Displays turn off filling; lines of verse or program code are shown with their lines broken as in the source document without requiring br requests between lines. Displays can be kept on a single page or allowed to break across pages. The DS macro begins a kept display of the layout specified in its first argument; non-kept displays are begun with dedicated macros corresponding to their layout.

Macro: .DS L
Macro: .LD

Begin (DS: kept) left-aligned display.

Macro: .DS [I [indent]]
Macro: .ID [indent]

Begin (DS: kept) display indented by indent if specified, and by the amount of the DI register otherwise.

Macro: .DS B
Macro: .BD

Begin a (DS: kept) a block display: the entire display is left-aligned, but indented such that the longest line in the display is centered on the page.

Macro: .DS C
Macro: .CD

Begin a (DS: kept) centered display: each line in the display is centered.

Macro: .DS R
Macro: .RD

Begin a (DS: kept) right-aligned display. This is a GNU extension.

Macro: .DE

End any display.

The distance stored in the DD register is inserted before and after each pair of display macros; this is a Berkeley extension. In groff ms, this distance replaces any adjacent inter-paragraph distance or subsequent spacing prior to a section heading. The DI register is a GNU extension; its value is an indentation applied to displays created with ‘.DS’ and ‘.ID’ without arguments, to ‘.DS I’ without an indentation argument, and to indented equations set with ‘.EQ’. Changes to either register take effect at the next display boundary.


Next: , Previous: , Up: ms Body Text   [Contents][Index]

4.6.5.9 Tables, figures, equations, and references

The ms package is often used with the tbl, pic, eqn, and refer preprocessors. Mark text meant for preprocessors by enclosing it in pairs of tokens as follows, with nothing between the dot and the macro name. The preprocessors match these tokens only at the start of an input line.

Macro: .TS [H]
Macro: .TE

Demarcate a table to be processed by the tbl preprocessor. The optional argument H to TS instructs ms to repeat table rows (often column headings) at the top of each new page the table spans, if applicable; calling the TH macro marks the end of such rows. The GNU tbl(1) man page provides a comprehensive reference to the preprocessor and offers examples of its use.

Macro: .PS
Macro: .PE
Macro: .PF

PS begins a picture to be processed by the gpic preprocessor; either of PE or PF ends it, the latter with “flyback” to the vertical position at its top. You can create pic input manually or with a program such as xfig.

Macro: .EQ [align [label]]
Macro: .EN

Demarcate an equation to be processed by the eqn preprocessor. The equation is centered by default; align can be ‘C’, ‘L’, or ‘I’ to (explicitly) center, left-align, or indent it by the amount stored in the DI register, respectively. If specified, label is set right-aligned.

Macro: .[
Macro: .]

Demarcate a bibliographic citation to be processed by the refer preprocessor. The GNU refer(1) man page provides a comprehensive reference to the preprocessor and the format of its bibliographic database. Type ‘man refer’ at the command line to view it.

When refer emits collected references (as might be done on a “Works Cited” page), it interpolates the REFERENCES string as an unnumbered heading (SH).

The following is an example of how to set up a table that may print across two or more pages.

.TS H
allbox;
Cb | Cb .
Part→Description
_
.TH
.T&
GH-1978→Fribulating gonkulator
…the rest of the table follows…
.TE

Attempting to place a multi-page table inside a keep can lead to unpleasant results, particularly if the tbl allbox option is used.

Mathematics can be typeset using the language of the eqn preprocessor.

.EQ C (\*[SN-NO-DOT]a)
p ~ = ~ q sqrt { ( 1 + ~ ( x / q sup 2 ) }
.EN

This input formats a labelled equation. We used the SN-NO-DOT string to base the equation label on the current heading number, giving us more flexibility to reorganize the document.

Use groff options to run preprocessors on the input: -e for geqn, -p for gpic, -R for grefer, and -t for gtbl.


Previous: , Up: ms Body Text   [Contents][Index]

4.6.5.10 Footnotes

A footnote is typically anchored to a place in the text with a marker, which is a small integer, a symbol such as a dagger, or arbitrary user-specified text.

String: \*[*]

Place an automatic number, an automatically generated numeric footnote marker, in the text. Each time this string is interpolated, the number it produces increments by one. Automatic numbers start at 1. This is a Berkeley extension.

Enclose the footnote text in FS and FE macro calls to set it at the nearest available “foot”, or bottom, of a text column or page.

Macro: .FS [marker]
Macro: .FE

Begin (FS) and end (FE) a footnote. FS calls FS-MARK with any supplied marker argument, which is then also placed at the beginning of the footnote text. If marker is omitted, the next pending automatic footnote number enqueued by interpolation of the * string is used, and if none exists, nothing is prefixed.

You may not desire automatically numbered footnotes in spite of their convenience. You can indicate a footnote with a symbol or other text by specifying its marker at the appropriate place (for example, by using \[dg] for the dagger glyph) and as an argument to the FS macro. Such manual marks should be repeated as arguments to FS or as part of the footnote text to disambiguate their correspondence. You may wish to use \*{ and \*} to superscript the marker at the anchor point, in the footnote text, or both.

groff ms provides a hook macro, FS-MARK, for user-determined operations to be performed when the FS macro is called. It is passed the same arguments as FS itself. An application of FS-MARK is anchor placement for a hyperlink reference, so that a footnote can link back to its referential context.11 By default, this macro has an empty definition. FS-MARK is a GNU extension.

Footnotes can be safely used within keeps and displays, but you should avoid using automatically numbered footnotes within floating keeps. You can place a second \** interpolation between a \** and its corresponding FS call as long as each FS call occurs after the corresponding \** and occurrences of FS are in the same order as corresponding occurrences of \**.

Footnote text is formatted as paragraphs are, using analogous parameters. The registers FI, FPD, FPS, and FVS correspond to PI, PD, PS, and CS, respectively; FPD, FPS, and FVS are GNU extensions.

The FF register controls the formatting of automatically numbered footnote paragraphs and those for which FS is given a marker argument. See ms Document Control Settings.

The default footnote line length is 11/12ths of the normal line length for compatibility with the expectations of historical ms documents; you may wish to set the FR string to ‘1’ to align with contemporary typesetting practices. In the past,12 an FL register was used for the line length in footnotes; however, setting this register at document initialization time had no effect on the footnote line length in multi-column arrangements.13

FR should be used in preference to the old FL register in contemporary documents. The footnote line length is effectively computed as ‘column-width * \*[FR]’. If an absolute footnote line length is required, recall that arithmetic expressions in roff input are evaluated strictly from left to right, with no operator precedence (parentheses are honored).

.ds FR 0+3i \" Set footnote line length to 3 inches.

Next: , Previous: , Up: ms Body Text   [Contents][Index]

4.6.5.11 Language and localization

groff ms provides several strings that you can customize for your own purposes, or redefine to adapt the macro package to languages other than English. It is already localized for Czech, German, French, Italian, and Swedish. Load the desired localization macro package after ms; see the groff_tmac(5) man page.

$ groff -ms -mfr bienvenue.ms

The following strings are available.

String: \*[REFERENCES]

Contains the string printed at the beginning of a references (bibliography) page produced with GNU refer(1). The default is ‘References’.

String: \*[ABSTRACT]

Contains the string printed at the beginning of the abstract. The default is ‘\f[I]ABSTRACT\f[]’; it includes font selection escape sequences to set the word in italics.

String: \*[TOC]

Contains the string printed at the beginning of the table of contents. The default is ‘Table of Contents’.

String: \*[MONTH1]
String: \*[MONTH2]
String: \*[MONTH3]
String: \*[MONTH4]
String: \*[MONTH5]
String: \*[MONTH6]
String: \*[MONTH7]
String: \*[MONTH8]
String: \*[MONTH9]
String: \*[MONTH10]
String: \*[MONTH11]
String: \*[MONTH12]

Contain the full names of the calendar months. The defaults are in English: ‘January’, ‘February’, and so on.


Next: , Previous: , Up: ms   [Contents][Index]

4.6.6 Page layout

ms’s default page layout arranges text in a single column with the page number between hyphens centered in a header on each page except the first, and produces no footers. You can customize this arrangement.


Next: , Previous: , Up: ms Page Layout   [Contents][Index]

4.6.6.1 Headers and footers

There are multiple ways to produce headers and footers. One is to define the strings LH, CH, and RH to set the left, center, and right headers, respectively; and LF, CF, and RF to set the left, center, and right footers. This approach suffices for documents that do not distinguish odd- and even-numbered pages.

Another method is to call macros that set headers or footers for odd- or even-numbered pages. Each such macro takes a delimited argument separating the left, center, and right header or footer texts from each other. You can replace the neutral apostrophes (') shown below with any character not appearing in the header or footer text. These macros are Berkeley extensions.

Macro: .OH 'left'center'right'
Macro: .EH 'left'center'right'
Macro: .OF 'left'center'right'
Macro: .EF 'left'center'right'

The OH and EH macros define headers for odd- (recto) and even-numbered (verso) pages, respectively; the OF and EF macros define footers for them.

With either method, a percent sign % in header or footer text is replaced by the current page number. By default, ms places no header on a page numbered “1” (regardless of its number format).

Macro: .P1

Typeset the header even on page 1. To be effective, this macro must be called before the header trap is sprung on any page numbered “1”; in practice, unless your page numbering is unusual, this means that you should call it early, before TL or any heading or paragraphing macro. This is a Berkeley extension.

For even greater flexibility, ms is designed to permit the redefinition of the macros that are called when the groff traps that ordinarily cause the headers and footers to be output are sprung. PT (“page trap”) is called by ms when the header is to be written, and BT (“bottom trap”) when the footer is to be. The groff page location trap that ms sets up to format the header also calls the (normally undefined) HD macro after PT; you can define HD if you need additional processing after setting the header (for example, to draw a line below it). The HD hook is a Berkeley extension. Any such macros you (re)define must implement any desired specialization for odd-, even-, or first numbered pages.


Next: , Previous: , Up: ms Page Layout   [Contents][Index]

4.6.6.2 Tab stops

Use the ta request to define tab stops as needed. See Tabs and Fields.

Macro: .TA

Reset the tab stops to the ms default (every 5 ens). Redefine this macro to create a different set of default tab stops.


Next: , Previous: , Up: ms Page Layout   [Contents][Index]

4.6.6.3 Margins

Control margins using the registers summarized in “Margin settings” in ms Document Control Settings above. There is no setting for the right margin; the combination of page offset \n[PO] and line length \n[LL] determines it.


Next: , Previous: , Up: ms Page Layout   [Contents][Index]

4.6.6.4 Multiple columns

ms can set text in as many columns as reasonably fit on the page. The following macros force a page break if a multi-column layout is active when they are called. The MINGW register stores the default minimum gutter width; it is a GNU extension. When multiple columns are in use, keeps and the HORPHANS and PORPHANS registers work with respect to column breaks instead of page breaks.

Macro: .1C

Arrange page text in a single column (the default).

Macro: .2C

Arrange page text in two columns.

Macro: .MC [column-width [gutter-width]]

Arrange page text in multiple columns. If you specify no arguments, it is equivalent to the 2C macro. Otherwise, column-width is the width of each column and gutter-width is the minimum distance between columns.


Next: , Previous: , Up: ms Page Layout   [Contents][Index]

4.6.6.5 Creating a table of contents

Because roff formatters process their input in a single pass, material on page 50, for example, cannot influence what appears on page 1—this poses a challenge for a table of contents at its traditional location in front matter, if you wish to avoid manually maintaining it. ms enables the collection of material to be presented in the table of contents as it appears, saving its page number along with it, and then emitting the collected contents on demand toward the end of the document. The table of contents can then be resequenced to its desired location by physically rearranging the pages of a printed document, or as part of post-processing—with a sed(1) script to reorder the pages in troff’s output, with pdfjam(1), or with gropdf(1)’s ‘.pdfswitchtopage’ feature, for example.

Define an entry to appear in the table of contents by bracketing its text between calls to the XS and XE macros. A typical application is to call them immediately after NH or SH and repeat the heading text within them. The XA macro, used within ‘.XS’/‘.XE’ pairs, supplements an entry—for instance, when it requires multiple output lines, whether because a heading is too long to fit or because style dictates that page numbers not be repeated. You may wish to indent the text thus wrapped to correspond to its heading depth; this can be done in the entry text by prefixing it with tabs or horizontal motion escape sequences, or by providing a second argument to the XA macro. XS and XA automatically associate the page number where they are called with the text following them, but they accept arguments to override this behavior. At the end of the document, call TC or PX to emit the table of contents; TC resets the page number to ‘i’ (Roman numeral one), and then calls PX. All of these macros are Berkeley extensions.

Macro: .XS [page-number]
Macro: .XA [page-number [indentation]]
Macro: .XE

Begin, supplement, and end a table of contents entry. Each entry is associated with page-number (otherwise the current page number); a page-number of ‘no’ prevents a leader and page number from being emitted for that entry. Use of XA within XS/XE is optional; it can be repeated. If indentation is present, a supplemental entry is indented by that amount; ens are assumed if no unit is indicated. Text on input lines between XS and XE is stored for later recall by PX.

Macro: .PX [no]

Switch to single-column layout. Unless no is specified, center and interpolate the TOC string in bold and two points larger than the body text. Emit the table of contents entries.

Macro: .TC [no]

Set the page number to 1, the page number format to lowercase Roman numerals, and call PX (with a no argument, if present).

Here’s an example of typical ms table of contents preparation. We employ horizontal escape sequences \h to indent the entries by sectioning depth.

.NH 1
Introduction
.XS
Introduction
.XE

.NH 2
Methodology
.XS
\h'2n'Methodology
.XA
\h'4n'Fassbinder's Approach
\h'4n'Kahiu's Approach
.XE

.NH 1
Findings
.XS
Findings
.XE

.TC

The remaining features in this subsubsection are GNU extensions. groff ms obviates the need to repeat heading text after XS calls. Call XN and XH after NH and SH, respectively.

Macro: .XN heading-text
Macro: .XH depth heading-text

Format heading-text and create a corresponding table of contents entry. XN computes the indentation from the depth of the preceding NH call; XH requires a depth argument to do so.

groff ms encourages customization of table of contents entry production.

Macro: .XN-REPLACEMENT heading-text
Macro: .XH-REPLACEMENT depth heading-text

These hook macros implement XN and XH, respectively. They call XN-INIT and pass their heading-text arguments to XH-UPDATE-TOC.

Macro: .XN-INIT
Macro: .XH-UPDATE-TOC depth heading-text

The XN-INIT hook macro does nothing by default. XH-UPDATE-TOC brackets heading-text with XS and XE calls, indenting it by 2 ens per level of depth beyond the first.

We could therefore produce a table of contents similar to that in the previous example with fewer macro calls. (The difference is that this input follows the “Approach” entries with leaders and page numbers.)

.NH 1
.XN Introduction

.NH 2
.XN Methodology
.XH 3 "Fassbinder's Approach"
.XH 3 "Kahiu's Approach"

.NH 1
.XN Findings

To get the section number of the numbered headings into the table of contents entries, we might define XN-REPLACEMENT as follows. (We obtain the heading depth from groff ms’s internal register nh*hl.)

.de XN-REPLACEMENT
.XN-INIT
.XH-UPDATE-TOC \\n[nh*hl] \\$@
\&\\*[SN] \\$*
..

You can change the style of the leader that bridges each table of contents entry with its page number; define the TC-LEADER special character by using the char request. A typical leader combines the dot glyph ‘.’ with a horizontal motion escape sequence to spread the dots. The width of the page number field is stored in the TC-MARGIN register.


Next: , Previous: , Up: ms   [Contents][Index]

4.6.7 Differences from AT&T ms

The groff ms macros are an independent reimplementation, using no AT&T code. Since they take advantage of the extended features of groff, they cannot be used with AT&T troff. groff ms supports features described above as Berkeley and Tenth Edition Research Unix extensions, and adds several of its own.

Register: \n[GS]

The register GS is set to 1 by the groff ms macros, but is not used by the AT&T ms package. Documents that need to determine whether they are being formatted with groff ms or another implementation should test this register.


Previous: , Up: Differences from AT&T ms   [Contents][Index]

4.6.7.1 Unix Version 7 ms macros not implemented by groff ms

Several macros described in the Unix Version 7 ms documentation are unimplemented by groff ms because they are specific to the requirements of documents produced internally by Bell Laboratories, some of which also require a glyph for the Bell System logo that groff does not support. These macros implemented several document type formats (EG, IM, MF, MR, TM, TR), were meaningful only in conjunction with the use of certain document types (AT, CS, CT, OK, SG), stored the postal addresses of Bell Labs sites (HO, IH, MH, PY, WH), or lacked a stable definition over time (UX). To compatibly render historical ms documents using these macros, we advise your documents to invoke the rm request to remove any such macros it uses and then define replacements with an authentically typeset original at hand.17 For informal purposes, a simple definition of UX should maintain the readability of the document’s substance.

.rm UX
.ds UX Unix\"

Next: , Previous: , Up: ms   [Contents][Index]

4.6.8 Legacy Features

groff ms retains some legacy features solely to support formatting of historical documents; contemporary ones should not use them because they can render poorly. See the groff_char(7) man page.

AT&T accent mark strings

AT&T ms defined accent mark strings as follows.

String: \*[']

Apply acute accent to subsequent glyph.

String: \*[`]

Apply grave accent to subsequent glyph.

String: \*[:]

Apply dieresis (umlaut) to subsequent glyph.

String: \*[^]

Apply circumflex accent to subsequent glyph.

String: \*[~]

Apply tilde accent to subsequent glyph.

String: \*[C]

Apply caron to subsequent glyph.

String: \*[,]

Apply cedilla to subsequent glyph.

Berkeley accent mark and glyph strings

Berkeley ms offered an AM macro; calling it redefined the AT&T accent mark strings (except for ‘\*C’), applied them to the preceding glyph, and defined additional strings, some for spacing glyphs.

Macro: .AM

Enable alternative accent mark and glyph-producing strings.

String: \*[']

Apply acute accent to preceding glyph.

String: \*[`]

Apply grave accent to preceding glyph.

String: \*[:]

Apply dieresis (umlaut) to preceding glyph.

String: \*[^]

Apply circumflex accent to preceding glyph.

String: \*[~]

Apply tilde accent to preceding glyph.

String: \*[,]

Apply cedilla to preceding glyph.

String: \*[/]

Apply stroke (slash) to preceding glyph.

String: \*[v]

Apply caron to preceding glyph.

String: \*[_]

Apply macron to preceding glyph.

String: \*[.]

Apply underdot to preceding glyph.

String: \*[o]

Apply ring accent to preceding glyph.

String: \*[?]

Interpolate inverted question mark.

String: \*[!]

Interpolate inverted exclamation mark.

String: \*[8]

Interpolate small letter sharp s.

String: \*[q]

Interpolate small letter o with hook accent (ogonek).

String: \*[3]

Interpolate small letter yogh.

String: \*[d-]

Interpolate small letter eth.

String: \*[D-]

Interpolate capital letter eth.

String: \*[th]

Interpolate small letter thorn.

String: \*[Th]

Interpolate capital letter thorn.

String: \*[ae]

Interpolate small ligature.

String: \*[Ae]

Interpolate capital ligature.

String: \*[oe]

Interpolate small oe ligature.

String: \*[OE]

Interpolate capital OE ligature.


Previous: , Up: ms   [Contents][Index]

4.6.9 Naming Conventions

The following conventions are used for names of macros, strings, and registers. External names available to documents that use the groff ms macros contain only uppercase letters and digits.

Internally, the macros are divided into modules. Conventions for identifier names are as follows.

Thus the groff ms macros reserve the following names.


Next: , Previous: , Up: Top   [Contents][Index]

5 GNU troff Reference

This chapter covers all of the facilities of the GNU troff formatting engine. Users of macro packages may skip it if not interested in details.


Next: , Previous: , Up: GNU troff Reference   [Contents][Index]

5.1 Text

AT&T troff was designed to take input as it would be composed on a typewriter, including the teletypewriters used as early computer terminals, and relieve the user drafting a document of concern with details like line length, hyphenation breaking, and the achievement of straight margins. Early in its development, the program gained the ability to prepare output for a phototypesetter; a document could then be prepared for output to either a teletypewriter, a phototypesetter, or both. GNU troff continues this tradition of permitting an author to compose a single master version of a document which can then be rendered for a variety of output formats or devices.

roff input files contain text interspersed with instructions to control the formatter. Even in the absence of such instructions, GNU troff still processes its input in several ways, by filling, hyphenating, breaking, and adjusting it, and supplementing it with inter-sentence space.


Next: , Previous: , Up: Text   [Contents][Index]

5.1.1 Filling

When GNU troff starts up, it obtains information about the device for which it is preparing output.18 An essential property is the length of the output line, such as “6.5 inches”.

GNU troff interprets plain text files employing the Unix line-ending convention. It reads input a character at a time, collecting words as it goes, and fits as many words together on an output line as it can—this is known as filling. To GNU troff, a word is any sequence of one or more characters that aren’t spaces or newlines. The exceptions separate words.19 To disable filling, see Manipulating Filling and Adjustment.

It is a truth universally acknowledged
that a single man in possession of a
good fortune must be in want of a wife.
    ⇒ It is a truth universally acknowledged that a
    ⇒ single man in possession of a good fortune must
    ⇒ be in want of a wife.

Next: , Previous: , Up: Text   [Contents][Index]

5.1.2 Sentences

A passionate debate has raged for decades among writers of the English language over whether more space should appear between adjacent sentences than between words within a sentence, and if so, how much, and what other circumstances should influence this spacing.20 GNU troff follows the example of AT&T troff; it attempts to detect the boundaries between sentences, and supplies additional inter-sentence space between them.

Hello, world!
Welcome to groff.
    ⇒ Hello, world!  Welcome to groff.

GNU troff flags certain characters (normally ‘!’, ‘?’, and ‘.’) as potentially ending a sentence. When GNU troff encounters one of these end-of-sentence characters at the end of an input line, or one of them is followed by two (unescaped) spaces on the same input line, it appends an inter-word space followed by an inter-sentence space in the output.

R. Harper subscribes to a maxim of P. T. Barnum.
    ⇒ R. Harper subscribes to a maxim of P. T. Barnum.

In the above example, inter-sentence space is not added after ‘P.’ or ‘T.’ because the periods do not occur at the end of an input line, nor are they followed by two or more spaces. Let’s imagine that we’ve heard something about defamation from Mr. Harper’s attorney, recast the sentence, and reflowed it in our text editor.

I submit that R. Harper subscribes to a maxim of P. T.
Barnum.
    ⇒ I submit that R. Harper subscribes to a maxim of
    ⇒ P. T.  Barnum.

“Barnum” doesn’t begin a sentence! What to do? Let us meet our first escape sequence, a series of input characters that give instructions to GNU troff instead of being used to construct output device glyphs.21 An escape sequence begins with the backslash character \ by default, an uncommon character in natural language text, and is always followed by at least one other character, hence the term “sequence”.

The dummy character escape sequence \& can be used after an end-of-sentence character to defeat end-of-sentence detection on a per-instance basis. We can therefore rewrite our input more defensively.

I submit that R.\& Harper subscribes to a maxim of P.\&
T.\& Barnum.
    ⇒ I submit that R. Harper subscribes to a maxim of
    ⇒ P. T. Barnum.

Adding text caused our input to wrap; now, we don’t need \& after ‘T.’ but we do after ‘P.’. Consistent use of the escape sequence ensures that potential sentence boundaries are robust to editing activities. Further advice along these lines will follow in Input Conventions.

Normally, the occurrence of a visible non-end-of-sentence character (as opposed to a space or tab) immediately after an end-of-sentence character cancels detection of the end of a sentence. For example, it would be incorrect for GNU troff to infer the end of a sentence after the dot in ‘3.14159’. However, several characters are treated transparently after the occurrence of an end-of-sentence character. That is, GNU troff does not cancel end-of-sentence detection when it processes them. This is because such characters are often used as footnote markers or to close quotations and parentheticals. The default set is ‘"’, ‘'’, ‘)’, ‘]’, ‘*’, \[dg], \[dd], \[rq], and \[cq]. The last four are examples of special characters, escape sequences whose purpose is to obtain glyphs that are not easily typed at the keyboard, or which have special meaning to GNU troff (like \ itself).22

\[lq]The idea that the poor should have leisure has always
been shocking to the rich.\[rq]
(Bertrand Russell, 1935)
    ⇒ "The idea that the poor should have
    ⇒ leisure has always been shocking to
    ⇒ the rich."  (Bertrand Russell, 1935)

The sets of characters that potentially end sentences or are transparent to sentence endings are configurable. See the cflags request in Using Symbols. To change the additional inter-sentence space amount—even to remove it entirely—see Manipulating Filling and Adjustment.


Next: , Previous: , Up: Text   [Contents][Index]

5.1.3 Hyphenation

When an output line is nearly full, it is uncommon for the next word collected from the input to exactly fill it—typically, there is room left over only for part of the next word. The process of splitting a word so that it appears partially on one line (with a hyphen to indicate to the reader that the word has been broken) with its remainder on the next is hyphenation. Hyphenation points can be manually specified; GNU troff also uses a hyphenation algorithm and language-specific pattern files (based on those used in TeX) to decide which words can be hyphenated and where.

Hyphenation does not always occur even when the hyphenation rules for a word allow it; it can be disabled, and when not disabled there are several parameters that can prevent it in certain circumstances. See Manipulating Hyphenation.


Next: , Previous: , Up: Text   [Contents][Index]

5.1.4 Breaking

Once an output line is full, the next word (or remainder of a hyphenated one) is placed on a different output line; this is called a break. In this manual and in roff discussions generally, a “break” if not further qualified always refers to the termination of an output line. When the formatter is filling text, it introduces breaks automatically to keep output lines from exceeding the configured line length. After an automatic break, GNU troff adjusts the line if applicable (see below), and then resumes collecting and filling text on the next output line.

Sometimes, a line cannot be broken automatically. This usually does not happen with natural language text unless the output line length has been manipulated to be extremely short, but it can with specialized text like program source code. We can use perl at the shell prompt to contrive an example of failure to break the line. We also employ the -z option to suppress normal output.

$ perl -e 'print "#" x 80, "\n";' | nroff -z
    error→ warning: cannot break line

The remedy for these cases is to tell GNU troff where the line may be broken without hyphens. This is done with the non-printing break point escape sequence ‘\:’; see Manipulating Hyphenation.

What if the document author wants to stop filling lines temporarily, for instance to start a new paragraph? There are several solutions. A blank input line not only causes a break, but by default it also outputs a one-line vertical space (effectively a blank output line). This behavior can be modified; see Blank Line Traps. Macro packages may discourage or disable the blank line method of paragraphing in favor of their own macros.

A line that begins with one or more spaces causes a break. The spaces are output at the beginning of the next line without being adjusted (see below); however, this behavior can be modified (see Leading Space Traps). Again, macro packages may provide other methods of producing indented paragraphs. Trailing spaces on text lines are discarded.23

What if the file ends before enough words have been collected to fill an output line? Or the output line is exactly full but not yet broken, and there is no more input? GNU troff interprets the end of input as a break. Certain requests also cause breaks, implicitly or explicitly. This is discussed in Manipulating Filling and Adjustment.


Next: , Previous: , Up: Text   [Contents][Index]

5.1.5 Adjustment

After GNU troff performs an automatic break, it may then adjust the line, widening inter-word spaces until the text reaches the right margin. Extra spaces between words are preserved. Leading and trailing spaces are handled as noted above. Text can be aligned to the left or right margin only, or centered; see Manipulating Filling and Adjustment.


Next: , Previous: , Up: Text   [Contents][Index]

5.1.6 Tabs and Leaders

GNU troff translates input horizontal tab characters (“tabs”) and Control+A characters (“leaders”) into movements to the next tab stop. Tabs simply move to the next tab stop; leaders place enough periods to fill the space. Tab stops are by default located every half inch measured from the drawing position corresponding to the beginning of the input line; see Page Geometry. Tabs and leaders do not cause breaks and therefore do not interrupt filling. Below, we use arrows → and bullets • to indicate input tabs and leaders, respectively.

1
→ 2 → 3 • 4
→ • 5
⇒ 1         2       3.......4         ........5

Tabs and leaders lend themselves to table construction.24 The tab and leader glyphs can be configured, and further facilities for sophisticated table composition are available; see Tabs and Fields. There are many details to track when using such low-level features, so most users turn to the tbl(1) preprocessor to lay out tables.


Next: , Previous: , Up: Text   [Contents][Index]

5.1.7 Requests and Macros

We have now encountered almost all of the syntax there is in the roff language, with an exception already noted in passing. A request is an instruction to the formatter that occurs after a control character, which is recognized at the beginning of an input line. The regular control character is a dot (.). Its counterpart, the no-break control character, a neutral apostrophe ('), suppresses the break that is implied by some requests. These characters were chosen because it is uncommon for lines of text in natural languages to begin with them. If you require a formatted period or apostrophe (closing single quotation mark) where GNU troff is expecting a control character, prefix the dot or neutral apostrophe with the dummy character escape sequence, ‘\&’.

An input line beginning with a control character is called a control line. Every line of input that is not a control line is a text line.25

Requests often take arguments, words (separated from the request name and each other by spaces) that specify details of the action GNU troff is expected to perform. If a request is meaningless without arguments, it is typically ignored.

GNU troff’s requests and escape sequences comprise the control language of the formatter. Of key importance are the requests that define macros. Macros are invoked like requests, enabling the request repertoire to be extended or overridden.26

A macro can be thought of as an abbreviation you can define for a collection of control and text lines. When the macro is called by giving its name after a control character, it is replaced with what it stands for. The process of textual replacement is known as interpolation.27 Interpolations are handled as soon as they are recognized, and once performed, a roff formatter scans the replacement for further requests, macro calls, and escape sequences.

In roff systems, the de request defines a macro.28

.de DATE
2020-11-14
..

The foregoing input produces no output by itself; all we have done is store some information. Observe the pair of dots that ends the macro definition. This is a default; you can specify your own terminator for the macro definition as the second argument to the de request.

.de NAME ENDNAME
Heywood Jabuzzoff
.ENDNAME

In fact, the ending marker is itself the name of a macro to be called, or a request to be invoked, if it is defined at the time its control line is read.

.de END
Big Rip
..
.de START END
Big Bang
.END
.START
    ⇒ Big Rip Big Bang

In the foregoing example, “Big Rip” printed before “Big Bang” because its macro was called first. Consider what would happen if we dropped END from the ‘.de START’ line and added .. after .END. Would the order change?

Let us consider a more elaborate example.

.de DATE
2020-10-05
..
.
.de BOSS
D.\& Kruger,
J.\& Peterman
..
.
.de NOTICE
Approved:
.DATE
by
.BOSS
..
.
Insert tedious regulatory compliance paragraph here.

.NOTICE

Insert tedious liability disclaimer paragraph here.

.NOTICE
    ⇒ Insert tedious regulatory compliance paragraph here.
    ⇒
    ⇒ Approved: 2020-10-05 by D. Kruger, J. Peterman
    ⇒
    ⇒ Insert tedious liability disclaimer paragraph here.
    ⇒
    ⇒ Approved: 2020-10-05 by D. Kruger, J. Peterman

The above document started with a series of control lines. Three macros were defined, with a de request declaring each macro’s name, and the “body” of the macro starting on the next line and continuing until a line with two dots ‘..’ marked its end. The text proper began only after the macros were defined; this is a common pattern. Only the NOTICE macro was called “directly” by the document; DATE and BOSS were called only by NOTICE itself. Escape sequences were used in BOSS, two levels of macro interpolation deep.

The advantage in typing and maintenance economy may not be obvious from such a short example, but imagine a much longer document with dozens of such paragraphs, each requiring a notice of managerial approval. Consider what must happen if you are in charge of generating a new version of such a document with a different date, for a different boss. With well-chosen macros, you only have to change each datum in one place.

In practice, we would probably use strings (see Strings) instead of macros for such simple interpolations; what is important here is to glimpse the potential of macros and the power of recursive interpolation.

We could have defined DATE and BOSS in the opposite order; perhaps less obviously, we could also have defined them after NOTICE. “Forward references” like this are acceptable because the body of a macro definition is not (completely) interpreted, but stored instead (see Copy Mode). While a macro is being defined (or appended to), requests are not interpreted and macros not interpolated, whereas some commonly used escape sequences are interpreted. roff systems also support recursive macro calls, as long as you have a way to break the recursion (see Conditionals and Loops). Maintainable roff documents tend to arrange macro definitions to minimize forward references.


Next: , Previous: , Up: Text   [Contents][Index]

5.1.8 Macro Packages

Macro definitions can be collected into macro files, roff input files designed to produce no output themselves but instead ease the preparation of other roff documents. There is no syntactical difference between a macro file and any other roff document; only its purpose distinguishes it. When a macro file is installed at a standard location and suitable for use by a general audience, it is often termed a macro package.29 Macro packages can be loaded by supplying the -m option to GNU troff or a groff front end. Alternatively, a document requiring a macro package can load it with the mso (“macro source”) request.


Next: , Previous: , Up: Text   [Contents][Index]

5.1.9 Input Encodings

The groff command’s -k option calls the preconv preprocessor to perform input character encoding conversions. Input to the GNU troff formatter itself, on the other hand, must be in one of two encodings it can recognize.

cp1047

The code page 1047 input encoding works only on EBCDIC platforms (and conversely, the other input encodings don’t work with EBCDIC); the file cp1047.tmac is loaded at startup.

latin1

ISO Latin-1, an encoding for Western European languages, is the default input encoding on non-EBCDIC platforms; the file latin1.tmac is loaded at startup.

Any document that is encoded in ISO 646:1991 (a descendant of USAS X3.4-1968 or “US-ASCII”), or, equivalently, uses only code points from the “C0 Controls” and “Basic Latin” parts of the Unicode character set is also a valid ISO Latin-1 document; the standards are interchangeable in their first 128 code points.30

Other encodings are supported by means of macro packages.

latin2

To use ISO Latin-2, an encoding for Central and Eastern European languages, invoke ‘.mso latin2.tmac at the beginning of your document or supply ‘-mlatin2’ as a command-line argument to groff.

latin5

To use ISO Latin-5, an encoding for the Turkish language, invoke ‘.mso latin5.tmac at the beginning of your document or supply ‘-mlatin5’ as a command-line argument to groff.

latin9

ISO Latin-9 succeeds Latin-1; it includes a Euro sign and better glyph coverage for French. To use this encoding, invoke ‘.mso latin9.tmac at the beginning of your document or supply ‘-mlatin9’ as a command-line argument to groff.

Some characters from an input encoding may not be available with a particular output driver, or their glyphs may not have representation in the font used. For terminal devices, fallbacks are defined, like ‘EUR’ for the Euro sign and ‘(C)’ for the copyright sign. For typesetter devices, you may need to “mount” fonts that support glyphs required by the document. See Font Positions.

Because a Euro glyph was not historically defined in PostScript fonts, groff comes with a font called freeeuro.pfa that provides the Euro in several styles. Standard PostScript fonts contain the glyphs from Latin-5 and Latin-9 that Latin-1 lacks, so these encodings are supported for the ps and pdf output devices as groff ships, while Latin-2 is not.

Unicode supports characters from all other input encodings; the utf8 output driver for terminals therefore does as well. The DVI output driver supports the Latin-2 and Latin-9 encodings if the command-line option -mec is used as well. 31


Previous: , Up: Text   [Contents][Index]

5.1.10 Input Conventions

Since GNU troff fills text automatically, it is common practice in the roff language to avoid visual composition of text in input files: the esthetic appeal of the formatted output is what matters. Therefore, roff input should be arranged such that it is easy for authors and maintainers to compose and develop the document, understand the syntax of roff requests, macro calls, and preprocessor languages used, and predict the behavior of the formatter. Several traditions have accrued in service of these goals.

We conclude this section with an example sufficiently long to illustrate most of the above suggestions in practice. For the purpose of fitting the example between the margins of this manual with the font used for its typeset version, we have shortened the input line length to 56 columns. As before, an arrow → indicates a tab character.

.\"   nroff this_file.roff | less
.\"   groff -T ps this_file.roff > this_file.ps
→The theory of relativity is intimately connected with
the theory of space and time.
.
I shall therefore begin with a brief investigation of
the origin of our ideas of space and time,
although in doing so I know that I introduce a
controversial subject.  \" remainder of paragraph elided
.
.

→The experiences of an individual appear to us arranged
in a series of events;
in this series the single events which we remember
appear to be ordered according to the criterion of
\[lq]earlier\[rq] and \[lq]later\[rq], \" punct swapped
which cannot be analysed further.
.
There exists,
therefore,
for the individual,
an I-time,
or subjective time.
.
This itself is not measurable.
.
I can,
indeed,
associate numbers with the events,
in such a way that the greater number is associated with
the later event than with an earlier one;
but the nature of this association may be quite
arbitrary.
.
This association I can define by means of a clock by
comparing the order of events furnished by the clock
with the order of a given series of events.
.
We understand by a clock something which provides a
series of events which can be counted,
and which has other properties of which we shall speak
later.
.\" Albert Einstein, _The Meaning of Relativity_, 1922

Next: , Previous: , Up: GNU troff Reference   [Contents][Index]

5.2 Page Geometry

roff systems format text under certain assumptions about the size of the output medium, or page. For the formatter to correctly break a line it is filling, it must know the line length, which it derives from the page width (see Line Layout). For it to decide whether to write an output line to the current page or wait until the next one, it must know the page length (see Page Layout).

A device’s resolution converts practical units like inches or centimeters to basic units, a convenient length measure for the output device or file format. The formatter and output driver use basic units to reckon page measurements. The device description file defines its resolution and page dimensions (see DESC File Format).

A page is a two-dimensional structure upon which a roff system imposes a rectangular coordinate system with its upper left corner as the origin. Coordinate values are in basic units and increase down and to the right. Useful ones are therefore always positive and within numeric ranges corresponding to the page boundaries.

While the formatter (and, later, output driver) is processing a page, it keeps track of its drawing position, which is the location at which the next glyph will be written, from which the next motion will be measured, or where a geometric object will commence rendering. Notionally, glyphs are drawn from the text baseline upward and to the right.33 The text baseline is a (usually invisible) line upon which the glyphs of a typeface are aligned. A glyph therefore “starts” at its bottom-left corner. If drawn at the origin, a typical letter glyph would lie partially or wholly off the page, depending on whether, like “g”, it features a descender below the baseline.

Such a situation is nearly always undesirable. It is furthermore conventional not to write or draw at the extreme edges of the page. Therefore the initial drawing position of a roff formatter is not at the origin, but below and to the right of it. This rightward shift from the left edge is known as the page offset.34 The downward shift leaves room for a text output line.

Text is arranged on a one-dimensional lattice of text baselines from the top to the bottom of the page. Vertical spacing is the distance between adjacent text baselines. Typographic tradition sets this quantity to 120% of the type size. The initial drawing position is one unit of vertical spacing below the page top. Typographers term this unit a vee.

Vertical spacing has an impact on page-breaking decisions. Generally, when a break occurs, the formatter moves the drawing position to the next text baseline automatically. If the formatter were already writing to the last line that would fit on the page, advancing by one vee would place the next text baseline off the page. Rather than let that happen, roff formatters instruct the output driver to eject the page, start a new one, and again set the drawing position to one vee below the page top; this is a page break.

When the last line of input text corresponds to the last output line that fits on the page, the break caused by the end of input will also break the page, producing a useless blank one. Macro packages keep users from having to confront this difficulty by setting “traps” (see Traps); moreover, all but the simplest page layouts tend to have headers and footers, or at least bear vertical margins larger than one vee.


Next: , Previous: , Up: GNU troff Reference   [Contents][Index]

5.3 Measurements

The formatter sometimes requires the input of numeric parameters to specify measurements. These are specified as integers or decimal fractions with an optional scaling unit suffixed. A scaling unit is a letter that immediately follows the last digit of a number. Digits after the decimal point are optional. Measurement expressions include ‘10.5p’, ‘11i’, and ‘3.c’.

Measurements are scaled by the scaling unit and stored internally (with any fractional part discarded) in basic units. The device resolution can therefore be obtained by storing a value of ‘1i’ to a register. The only constraint on the basic unit is that it is at least as small as any other unit.

u

Basic unit.

i

Inch; defined as 2.54 centimeters.

c

Centimeter; a centimeter is about 0.3937 inches.

p

Point; a typesetter’s unit used for measuring type size. There are 72 points to an inch.

P

Pica; another typesetter’s unit. There are 6 picas to an inch and 12 points to a pica.

s
z

See Using Fractional Type Sizes, for a discussion of these units.

f

GNU troff defines this unit to scale decimal fractions in the interval [0, 1] to 16-bit unsigned integers. It multiplies a quantity by 65,536. See Colors, for usage.

The magnitudes of other scaling units depend on the text formatting parameters in effect. These are useful when specifying measurements that need to scale with the typeface or vertical spacing.

m

Em; an em is equal to the current type size in points. It is named thus because it is approximately the width of the letter ‘M’.

n

En; an en is one-half em.

v

Vee; recall Page Geometry.

M

Hundredth of an em.


Next: , Previous: , Up: Measurements   [Contents][Index]

5.3.1 Motion Quanta

An output device’s basic unit u is not necessarily its smallest addressable length; u can be smaller to avoid problems with integer roundoff. The minimum distances that a device can work with in the horizontal and vertical directions are termed its motion quanta. Measurements are rounded to applicable motion quanta. Half-quantum fractions round toward zero.

Register: \n[.H]
Register: \n[.V]

These read-only registers interpolate the horizontal and vertical motion quanta, respectively, of the output device in basic units.

For example, we might draw short baseline rules on a terminal device as follows. See Drawing Geometric Objects.

.tm \n[.H]
    error→ 24
.nf
\l'36u' 36u
\l'37u' 37u
    ⇒ _ 36u
    ⇒ __ 37u

Previous: , Up: Measurements   [Contents][Index]

5.3.2 Default Units

A general-purpose register (one created or updated with the nr request; see see Registers) is implicitly dimensionless, or reckoned in basic units if interpreted in a measurement context. But it is convenient for many requests and escape sequences to infer a scaling unit for an argument if none is specified. An explicit scaling unit (not after a closing parenthesis) can override an undesirable default. Effectively, the default unit is suffixed to the expression if a scaling unit is not already present. GNU troff’s use of integer arithmetic should also be kept in mind (see Numeric Expressions).

The ll request interprets its argument in ems by default. Consider several attempts to set a line length of 3.5 inches when the type size is 10 points on a terminal device with a resolution of 240 basic units and horizontal motion quantum of 24. Some expressions become zero; the request clamps them to that quantum.

.ll 3.5i      \" 3.5i (= 840u)
.ll 7/2       \" 7u/2u -> 3u -> 3m -> 0, clamped to 24u
.ll (7 / 2)u  \" 7u/2u -> as above
.ll 7/2i      \" 7u/2i -> 7u/480u -> 0 -> as above
.ll 7i/2      \" 7i/2u -> 1680u/2m -> 1680u/24u -> 35u
.ll 7i/2u     \" 3.5i (= 840u)

The safest way to specify measurements is to attach a scaling unit. To multiply or divide by a dimensionless quantity, use ‘u’ as its scaling unit.


Next: , Previous: , Up: GNU troff Reference   [Contents][Index]

5.4 Numeric Expressions

A numeric expression evaluates to an integer: it can be as simple as a literal ‘0’ or it can be a complex sequence of register and string interpolations interleaved with measurements and operators.

GNU troff provides a set of mathematical and logical operators familiar to programmers—as well as some unusual ones—but supports only integer arithmetic.35 The internal data type used for computing results is usually a 32-bit signed integer, which suffices to represent magnitudes within a range of 2 billion.36

Arithmetic infix operators perform a function on the numeric expressions to their left and right; they are + (addition), - (subtraction), * (multiplication), / (truncating division), and % (modulus). Truncating division rounds to the integer nearer to zero, no matter how large the fractional portion. Overflow and division (or modulus) by zero are errors and abort evaluation of a numeric expression.

Arithmetic unary operators operate on the numeric expression to their right; they are - (negation) and + (assertion—for completeness; it does nothing). The unary minus must often be used with parentheses to avoid confusion with the decrementation operator, discussed below.

Observe the rounding behavior and effect of negative operands on the modulus and truncating division operators.

.nr T 199/100
.nr U 5/2
.nr V (-5)/2
.nr W 5/-2
.nr X 5%2
.nr Y (-5)%2
.nr Z 5%-2
T=\n[T] U=\n[U] V=\n[V] W=\n[W] X=\n[X] Y=\n[Y] Z=\n[Z]
    ⇒ T=1 U=2 V=-2 W=-2 X=1 Y=-1 Z=1

The sign of the modulus of operands of mixed signs is determined by the sign of the first. Division and modulus operators satisfy the following property: given a dividend a and a divisor b, a quotient q formed by ‘(a / b)’ and a remainder r by ‘(a % b)’, then qb + r = a.

GNU troff’s scaling operator, used with parentheses as (c;e), evaluates a numeric expression e using c as the default scaling unit. If c is omitted, scaling units are ignored in the evaluation of e. This operator can save typing by avoiding the attachment of scaling units to every operand out of caution. Your macros can select a sensible default unit in case the user neglects to supply one.

.\" Indent by amount given in first argument; assume ens.
.de Indent
.  in (n;\\$1)
..

Without the scaling operator, the foregoing macro would, if called with a unitless argument, cause indentation by the in request’s default scaling unit (ems). The result would be twice as much indentation as expected.

GNU troff also provides a pair of operators to compute the extrema of two operands: >? (maximum) and <? (minimum).

.nr slots 5
.nr candidates 3
.nr salaries (\n[slots] <? \n[candidates])
Looks like we'll end up paying \n[salaries] salaries.
    ⇒ Looks like we'll end up paying 3 salaries.

Comparison operators comprise < (less than), > (greater than), <= (less than or equal), >= (greater than or equal), and = (equal). == is a synonym for =. When evaluated, a comparison is replaced with ‘0’ if it is false and ‘1’ if true. In the roff language, positive values are true, others false.

We can operate on truth values with the logical operators & (logical conjunction or “and”) and : (logical disjunction or “or”). They evaluate as comparison operators do.

A logical complementation (“not”) operator, !, works only within if, ie, and while requests. Furthermore, ! is recognized only at the beginning of a numeric expression not contained by another numeric expression. In other words, it must be the “outermost” operator. Including it elsewhere in the expression produces a warning in the ‘number’ category (see Warnings), and its expression evaluates false. This unfortunate limitation maintains compatibility with AT&T troff. Test a numeric expression for falsity by comparing it to a false value.37

.nr X 1
.nr Y 0
.\" This does not work as expected.
.if (\n[X])&(!\n[Y]) .nop A: X is true, Y is false
.
.\" Use this construct instead.
.if (\n[X])&(\n[Y]<=0) .nop B: X is true, Y is false
    error→ warning: expected numeric expression, got '!'
    ⇒ B: X is true, Y is false

The roff language has no operator precedence: expressions are evaluated strictly from left to right, in contrast to schoolhouse arithmetic. Use parentheses ( ) to impose a desired precedence upon subexpressions.

.nr X 3+5*4
.nr Y (3+5)*4
.nr Z 3+(5*4)
X=\n[X] Y=\n[Y] Z=\n[Z]
    ⇒ X=32 Y=32 Z=23

For many requests and escape sequences that cause motion on the page, the unary operators + and - work differently when leading a numeric expression. They then indicate a motion relative to the drawing position: positive is down in vertical contexts, right in horizontal ones.

+ and - are also treated differently by the following requests and escape sequences: bp, in, ll, lt, nm, nr, pl, pn, po, ps, pvs, rt, ti, \H, \R, and \s. Here, leading plus and minus signs serve as incrementation and decrementation operators, respectively. To negate an expression, subtract it from zero or include the unary minus in parentheses with its argument. See Setting Registers, for examples.

A leading | operator indicates a motion relative not to the drawing position but to a boundary. For horizontal motions, the measurement specifies a distance relative to a drawing position corresponding to the beginning of the input line. By default, tab stops reckon movements in this way. Most escape sequences do not; | tells them to do so.

Mind the \h'1.2i'gap.
.br
Mind the \h'|1.2i'gap.
.br
Mind the
\h'|1.2i'gap.
    ⇒ Mind the             gap.
    ⇒ Mind the    gap.
    ⇒ Mind the             gap.

One use of this feature is to define macros whose scope is limited to the output they format.

.\" underline word $1 with trailing punctuation $2
.de Underline
.  nop \\$1\l'|0\[ul]'\\$2
..
Typographical emphasis is best used
.Underline sparingly .

In the above example, ‘|0’ specifies a negative motion from the current position (at the end of the argument just emitted, \$1) to the beginning of the input line. Thus, the \l escape sequence in this case draws a line from right to left. A macro call occurs at the beginning of an input line;38 if the | operator were omitted, then the underline would be drawn at zero distance from the current position, producing device-dependent, and likely undesirable, results. On the ‘ps’ output device, it underlines the period.

For vertical motions, the | operator specifies a distance from the first text baseline on the page or in the current diversion,39 using the current vertical spacing.

A
.br
B \Z'C'\v'|0'D
    ⇒ A D
    ⇒ B C

In the foregoing example, we’ve used the \Z escape sequence (see Page Motions) to restore the drawing position after formatting ‘C’, then moved vertically to the first text baseline on the page.

Escape sequence: \B'anything'

Interpolate 1 if anything is a valid numeric expression, and 0 otherwise. The delimiter need not be a neutral apostrophe; see Delimiters.

You might use \B along with the if request to filter out invalid macro or string arguments. See Conditionals and Loops.

.\" Indent by amount given in first argument; assume ens.
.de Indent
.  if \B'\\$1' .in (n;\\$1)
..

A register interpolated as an operand in a numeric expression must have an Arabic format; luckily, this is the default. See Assigning Register Formats.

Because spaces separate arguments to requests, spaces are not allowed in numeric expressions unless the (sub)expression containing them is surrounded by parentheses. See Invoking Requests, and Conditionals and Loops.

.nf
.nr a 1+2 + 2+1
\na
    error→ expected numeric expression, got a space
    ⇒ 3
.nr a 1+(2 + 2)+1
\na
    ⇒ 6

The nr request (see Setting Registers) expects its second and optional third arguments to be numeric expressions; a bare + does not qualify, so our first attempt got a warning.


Next: , Previous: , Up: GNU troff Reference   [Contents][Index]

5.5 Identifiers

An identifier labels a GNU troff datum such as a register, name (macro, string, or diversion), typeface, color, special character, character class, environment, or stream. Valid identifiers consist of one or more ordinary characters. An ordinary character is an input character that is not the escape character, a leader, tab, newline, or invalid as GNU troff input.

Invalid input characters are a subset of control characters (from the sets “C0 Controls” and “C1 Controls” as Unicode describes them). When GNU troff encounters one in an identifier, it produces a warning in category ‘input’ (see Warnings). They are removed during interpretation: an identifier ‘foo’, followed by an invalid character and then ‘bar’, is processed as ‘foobar’.

On a machine using the ISO 646, 8859, or 10646 character encodings, invalid input characters are 0x00, 0x08, 0x0B, 0x0D0x1F, and 0x800x9F. On an EBCDIC host, they are 0x000x01, 0x08, 0x09, 0x0B, 0x0D0x14, 0x170x1F, and 0x300x3F.40 Some of these code points are used by GNU troff internally, making it non-trivial to extend the program to accept UTF-8 or other encodings that use characters from these ranges.41

Thus, the identifiers ‘br’, ‘PP’, ‘end-list’, ‘ref*normal-print’, ‘|’, ‘@_’, and ‘!"#$%'()*+,-./’ are all valid. Discretion should be exercised to prevent confusion. Identifiers starting with ‘(’ or ‘[’ require care.

.nr x 9
.nr y 1
.nr (x 2
.nr [y 3
.nr sum1 (\n(x + \n[y])
    error→ a space character is not allowed in an escape
    error→   sequence parameter
A:2+3=\n[sum1]
.nr sum2 (\n((x + \n[[y])
B:2+3=\n[sum2]
.nr sum3 (\n[(x] + \n([y)
C:2+3=\n[sum3]
    ⇒ A:2+3=1 B:2+3=5 C:2+3=5

An identifier with a closing bracket (‘]’) in its name can’t be accessed with bracket-form escape sequences that expect an identifier as a parameter. For example, ‘\[foo]]’ accesses the glyph ‘foo’, followed by ‘]’ in whatever the surrounding context is, whereas ‘\C'foo]'’ formats a glyph named ‘foo]’. Similarly, the identifier ‘(’ can’t be interpolated except with bracket forms.

If you begin a macro, string, or diversion name with either of the characters ‘[’ or ‘]’, you foreclose use of the grefer preprocessor, which recognizes ‘.[’ and ‘.]’ as bibliographic reference delimiters.

Escape sequence: \A'anything'

Interpolate 1 if anything is a valid identifier, and 0 otherwise. The delimiter need not be a neutral apostrophe; see Delimiters. Because invalid input characters are removed (see above), invalid identifiers are empty or contain spaces, tabs, or newlines.

You can employ \A to validate a macro argument before using it to construct another escape sequence or identifier.

.\" usage: .init-coordinate-pair name val1 val2
.\" Create a coordinate pair where name!x=val1 and
.\" name!y=val2.
.de init-coordinate-pair
.  if \A'\\$1' \{\
.    if \B'\\$2' .nr \\$1!x \\$2
.    if \B'\\$3' .nr \\$1!y \\$3
.  \}
..
.init-coordinate-pair center 5 10
The center is at (\n[center!x], \n[center!y]).
.init-coordinate-pair "poi→nt" trash garbage \" ignored
.init-coordinate-pair point trash garbage \" ignored
    ⇒ The center is at (5, 10).

In this example, we also validated the numeric arguments; the registers ‘point!x’ and ‘point!y’ remain undefined. See Numeric Expressions for the \B escape sequence.

How GNU troff handles the interpretation of an undefined identifier depends on the context. There is no way to invoke an undefined request; such syntax is interpreted as a macro call instead. If the identifier is interpreted as a string, macro, or diversion, GNU troff emits a warning in category ‘mac’, defines it as empty, and interpolates nothing. If the identifier is interpreted as a register, GNU troff emits a warning in category ‘reg’, initializes it to zero, and interpolates that value. See Warnings, Interpolating Registers, and Strings. Attempting to use an undefined typeface, special character, color, character class, environment, or stream generally provokes an error diagnostic.

Identifiers for requests, macros, strings, and diversions share one name space; special characters and character classes another. No other object types do.

.de xxx
.  nop foo
..
.di xxx
bar
.br
.di
.
.xxx
    ⇒ bar

The foregoing example shows that GNU troff reuses the identifier ‘xxx’, changing it from a macro to a diversion. No warning is emitted, and the previous contents of ‘xxx’ are lost.


Next: , Previous: , Up: GNU troff Reference   [Contents][Index]

5.6 Formatter Instructions

To support documents that require more than filling, automatic line breaking and hyphenation, adjustment, and supplemental inter-sentence space, the roff language offers two means of embedding instructions to the formatter.

One is a request, which begins with a control character and takes up the remainder of the input line. Requests often perform relatively large-scale operations such as setting the page length, breaking the line, or starting a new page. They also conduct internal operations like defining macros.

The other is an escape sequence, which begins with the escape character and can be embedded anywhere in the input, even in arguments to requests and other escape sequences. Escape sequences interpolate special characters, strings, or registers, and handle comparatively minor formatting tasks like sub- and superscripting.

Some operations, such as font selection and type size alteration, are available via both requests and escape sequences.


Next: , Previous: , Up: Formatter Instructions   [Contents][Index]

5.6.1 Control Characters

The mechanism of using roff’s control characters to invoke requests and call macros was introduced in Requests and Macros. Control characters are recognized only at the beginning of an input line, or at the beginning of the branch of a control structure request; see Conditionals and Loops.

A few requests cause a break implicitly; use the no-break control character to prevent the break. Break suppression is its sole behavioral distinction. Employing the no-break control character to invoke requests that don’t cause breaks is harmless but poor style. See Manipulating Filling and Adjustment.

The control ‘.’ and no-break control ‘'’ characters can each be changed to any ordinary character42 with the cc and c2 requests, respectively.

Request: .cc [o]

Recognize the ordinary character o as the control character. If o is absent or invalid, the default control character ‘.’ is selected. The identity of the control character is associated with the environment (see Environments).

Request: .c2 [o]

Recognize the ordinary character o as the no-break control character. If o is absent or invalid, the default no-break control character ‘'’ is selected. The identity of the no-break control character is associated with the environment (see Environments).

When writing a macro, you might wish to know which control character was used to call it.

Register: \n[.br]

This read-only register interpolates 1 if the currently executing macro was called using the normal control character and 0 otherwise. If a macro is interpolated as a string, the .br register’s value is inherited from the context of the string interpolation. See Strings.

Use this register to reliably intercept requests that imply breaks.

.als bp*orig bp
.de bp
.  ie \\n[.br] .bp*orig
.  el          'bp*orig
..

Testing the .br register outside of a macro definition makes no sense.


Next: , Previous: , Up: Formatter Instructions   [Contents][Index]

5.6.2 Invoking Requests

A control character is optionally followed by tabs and/or spaces and then an identifier naming a request or macro. The invocation of an unrecognized request is interpreted as a macro call. Defining a macro with the same name as a request replaces the request. Deleting a request name with the rm request makes it unavailable. The als request can alias requests, permitting them to be wrapped or non-destructively replaced. See Strings.

There is no inherent limit on argument length or quantity. Most requests take one or more arguments, and ignore any they do not expect. A request may be separated from its arguments by tabs or spaces, but only spaces can separate an argument from its successor. Only one between arguments is necessary; any excess is ignored. GNU troff does not allow tabs for argument separation.43

Generally, a space within a request argument is not relevant, not meaningful, or is supported by bespoke provisions, as with the tl request’s delimiters (see Page Layout). Some requests, like ds, interpret the remainder of the control line as a single argument. See Strings.

Spaces and tabs immediately after a control character are ignored. Commonly, authors structure the source of documents or macro files with them.

.de center
.  if \\n[.br] \
.    br
.  ce \\$1
..
.
.
.de right-align
.→if \\n[.br] \
.→→br
.→rj \\$1
..

If you assign an empty blank line trap, you can separate macro definitions (or any input lines) with blank lines.

.de do-nothing
..
.blm do-nothing  \" activate blank line trap

.de center
.  if \\n[.br] \
.    br
.  ce \\$1
..


.de right-align
.→if \\n[.br] \
.→→br
.→rj \\$1
..

.blm             \" deactivate blank line trap

See Blank Line Traps.


Next: , Previous: , Up: Formatter Instructions   [Contents][Index]

5.6.3 Calling Macros

If a macro of the desired name does not exist when called, it is created, assigned an empty definition, and a warning in category ‘mac’ is emitted. Calling an undefined macro does end a macro definition naming it as its end macro (see Writing Macros).

To embed spaces within a macro argument, enclose the argument in neutral double quotes ". Horizontal motion escape sequences are sometimes a better choice for arguments to be formatted as text.

Consider calls to a hypothetical section heading macro ‘uh’.

.uh The Mouse Problem
.uh "The Mouse Problem"
.uh The\~Mouse\~Problem
.uh The\ Mouse\ Problem

The first line calls uh with three arguments: ‘The’, ‘Mouse’, and ‘Problem’. The remainder call the uh macro with one argument, ‘The Mouse Problem’. The last solution, using escaped spaces, can be found in documents prepared for AT&T troff. It can cause surprise when text is adjusted, because \SP inserts a fixed-width, non-breaking space. GNU troff’s \~ escape sequence inserts an adjustable, non-breaking space.44

The foregoing raises the question of how to embed neutral double quotes or backslashes in macro arguments when those characters are desired as literals. In GNU troff, the special character escape sequence \[rs] produces a backslash and \[dq] a neutral double quote.

In GNU troff’s AT&T compatibility mode, these characters remain available as \(rs and \(dq, respectively. AT&T troff did not consistently define these special characters, but its descendants can be made to support them. See Device and Font Description Files.

If even that is not feasible, options remain. To obtain a literal escape character in a macro argument, you can simply type it if you change or disable the escape character first. See Using Escape Sequences. Otherwise, you must escape the escape character repeatedly to a context-dependent extent. See Copy Mode.

For the (neutral) double quote, you have recourse to an obscure syntactical feature of AT&T troff. Because a double quote can begin a macro argument, the formatter keeps track of whether the current argument was started thus, and doesn’t require a space after the double quote that ends it.45 In the argument list to a macro, a double quote that isn’t preceded by a space doesn’t start a macro argument. If not preceded by a double quote that began an argument, this double quote becomes part of the argument. Furthermore, within a quoted argument, a pair of adjacent double quotes becomes a literal double quote.

.de eq
.  tm arg1:\\$1 arg2:\\$2 arg3:\\$3
.  tm arg4:\\$4 arg5:\\$5 arg6:\\$6
.. \" 4 backslashes on the next line
.eq a" "b c" "de"f\\\\g" h""i "j""k"
    error→ arg1:a" arg2:b c arg3:de
    error→ arg4:f\g" arg5:h""i arg6:j"k

Apart from the complexity of the rules, this traditional solution has the disadvantage that double quotes don’t survive repeated argument expansion in AT&T troff or GNU troff’s compatibility mode. This can frustrate efforts to pass such arguments intact through multiple macro calls.

.cp 1
.de eq
.  tm arg1:\\$1 arg2:\\$2 arg3:\\$3
.  tm arg4:\\$4 arg5:\\$5 arg6:\\$6
..
.de xe
.  eq \\$1 \\$2 \\$3 \\$4 \\$5 \\$6
.. \" 8 backslashes on the next line
.xe a" "b c" "de"f\\\\\\\\g" h""i "j""k"
    error→ arg1:a" arg2:b arg3:c
    error→ arg4:de arg5:f\g" arg6:h""i

Outside of compatibility mode, GNU troff doesn’t exhibit this problem because it tracks the nesting depth of interpolations. See Implementation Differences.


Next: , Previous: , Up: Formatter Instructions   [Contents][Index]

5.6.4 Using Escape Sequences

Whereas requests must occur on control lines, escape sequences can occur intermixed with text and may appear in arguments to requests, macros, and other escape sequences. An escape sequence is introduced by the escape character, a backslash \ (but see the ec request below). The next character selects the escape’s function.

Escape sequences vary in length. Some take an argument, and of those, some have different syntactical forms for a one-character, two-character, or arbitrary-length argument. Others accept only an arbitrary-length argument. In the former scheme, a one-character argument follows the function character immediately, an opening parenthesis ‘(’ introduces a two-character argument (no closing parenthesis is used), and an argument of arbitrary length is enclosed in brackets ‘[]’. In the latter scheme, the user selects a delimiter character. A few escape sequences are idiosyncratic, and support both of the foregoing conventions (\s), designate their own termination sequence (\?), consume input until the next newline (\!, \", \#), or support an additional modifier character (\s again, and \n). As with requests, use of some escape sequences in source documents may interact poorly with a macro package you use; consult its documentation to learn of “safe” sequences or alternative facilities it provides to achieve the desired result.

If an escape character is followed by a character that does not identify a defined operation, the escape character is ignored (producing a diagnostic of the ‘escape’ warning category, which is not enabled by default) and the following character is processed normally.

$ groff -Tps -ww
.nr N 12
.ds co white
.ds animal elephant
I have \fI\nN \*(co \*[animal]s,\f[]
said \P.\&\~Pseudo Pachyderm.
    error→ warning: escape character ignored before 'P'
    ⇒ I have 12 white elephants, said P. Pseudo Pachyderm.

Escape sequence interpolation is of higher precedence than escape sequence argument interpretation. This rule affords flexibility in using escape sequences to construct parameters to other escape sequences.

.ds family C\" Courier
.ds style I\" oblique
Choice a typeface \f(\*[family]\*[style]wisely.
    ⇒ Choose a typeface wisely.

In the above, the syntax form ‘\f(’ accepts only two characters for an argument; the example works because the subsequent escape sequences are interpolated before the selection escape sequence argument is processed, and strings family and style interpolate one character each.46

The escape character is nearly always interpreted when encountered; it is therefore desirable to have a way to interpolate it, disable it, or change it.

Escape sequence: \e

Interpolate the escape character.

The \[rs] special character escape sequence formats a backslash glyph. In macro and string definitions, the input sequences \\ and \E defer interpretation of escape sequences. See Copy Mode.

Request: .eo

Disable the escape mechanism except in copy mode. Once this request is invoked, no input character is recognized as starting an escape sequence in interpretation mode.

Request: .ec [o]

Recognize the ordinary character o as the escape character. If o is absent or invalid, the default escape character ‘\’ is selected.

Switching escape sequence interpretation off to define a macro and back on afterward can obviate the need to double the escape character within the definition. See Writing Macros. This technique is not available if your macro needs to interpolate values at the time it is defined—but many do not.

.\" simplified `BR` macro from the man(7) macro package
.eo
.de BR
.  ds result \&
.  while (\n[.$] >= 2) \{\
.    as result \fB\$1\fR\$2\"
.    shift 2
.  \}
.  if \n[.$] .as result \fB\$1\"
\*[result]
.  rm result
.  ft R
..
.ec
Request: .ecs
Request: .ecr

The ecs request stores the escape character for recall with ecr. ecr sets the escape character to ‘\’ if none has been saved.

Use these requests together to temporarily change the escape character.

Using a different escape character, or disabling it, when calling macros not under your control will likely cause errors, since GNU troff has no mechanism to “intern” macros—that is, to convert a macro definition into a form independent of its representation.47 When a macro is called, its contents are interpreted literally.


Previous: , Up: Formatter Instructions   [Contents][Index]

5.6.5 Delimiters

Some escape sequences that require parameters use delimiters. The neutral apostrophe ' is a popular choice and shown in this document. The neutral double quote " is also commonly seen. Letters, numerals, and leaders can be used. Punctuation characters are likely better choices, except for those defined as infix operators in numeric expressions; see below.

\l'1.5i\[bu]' \" draw 1.5 inches of bullet glyphs

The following escape sequences don’t take arguments and thus are allowed as delimiters: \SP, \%, \|, \^, \{, \}, \', \`, \-, \_, \!, \?, \), \/, \,, \&, \:, \~, \0, \a, \c, \d, \e, \E, \p, \r, \t, and \u. However, using them this way is discouraged; they can make the input confusing to read.

A few escape sequences, \A, \b, \o, \w, \X, and \Z, accept a newline as a delimiter. Newlines that serve as delimiters continue to be recognized as input line terminators.

A caf\o
e\(aa
in Paris
    ⇒ A caf in Paris

Use of newlines as delimiters in escape sequences is also discouraged.

Finally, the escape sequences \D, \h, \H, \l, \L, \N, \R, \s, \S, \v, and \x prohibit many delimiters.

Delimiter syntax is complex and flexible primarily for historical reasons; the foregoing restrictions need be kept in mind mainly when using groff in AT&T compatibility mode. GNU troff keeps track of the nesting depth of escape sequence interpolations, so the only characters you need to avoid using as delimiters are those that appear in the arguments you input, not any that result from interpolation. Typically, ' works fine. See Implementation Differences.

$ groff -Tps
.de Mw
.  nr wd \w'\\$1'
.  tm "\\$1" is \\n(wd units wide.
..
.Mw Wet'suwet'en
.Mw Wet+200i
.cp 1 \" turn on compatibility mode
.Mw Wet'suwet'en
.Mw Wet'
.Mw Wet+200i
    error→ "Wet'suwet'en" is 54740 units wide.
    error→ "Wet'+200i" is 42610 units wide.
    error→ "Wet'suwet'en" is 15860 units wide.
    error→ "Wet'" is 15860 units wide.
    error→ "Wet'+200i" is 14415860 units wide.

We see here that in compatibility mode, the part of the argument after the ' delimiter escapes from its context and, if nefariously crafted, influences the computation of the wd register’s value in a surprising way.


Next: , Previous: , Up: GNU troff Reference   [Contents][Index]

5.7 Comments

One of the most common forms of escape sequence is the comment.48

Escape sequence: \"

Start a comment. Everything up to the next newline is ignored.

This may sound simple, but it can be tricky to keep the comments from interfering with the appearance of the output. If the escape sequence is to the right of some text or a request, that portion of the line is ignored, but spaces preceding it are processed normally by GNU troff. This affects only the ds and as requests and their variants.

One possibly irritating idiosyncrasy is that tabs should not be used to vertically align comments in the source document. Tab characters are not treated as separators between a request name and its first argument, nor between arguments.

A comment on a line by itself is treated as a blank line, because after eliminating the comment, that is all that remains.

Test
\" comment
Test
    ⇒ Test
    ⇒
    ⇒ Test

To avoid this, it is common to combine the empty request with the comment escape sequence as ‘.\"’, causing the input line to be ignored.

Another commenting scheme sometimes seen is three consecutive single quotes (''') at the beginning of a line. This works, but GNU troff emits a warning diagnostic (if enabled) about an undefined macro (namely ‘''’).

Escape sequence: \#

Start a comment; everything up to and including the next newline is ignored. This groff extension was introduced to avoid the problems described above.

Test
\# comment
Test
    ⇒ Test Test
Request: .ig [end]

Ignore input until, in the current conditional block (if any),49 the macro end is called at the start of a control line, or the control line ‘..’ is encountered if end is not specified. ig is parsed as if it were a macro definition, but its contents are discarded, not stored.50

hand\c
.de TX
fasting
..
.ig TX
This is part of a large block of input that has been
temporarily(?) commented out.
We can restore it simply by removing the .ig request and
the call of its end macro.
.TX
    ⇒ handfasting

Next: , Previous: , Up: GNU troff Reference   [Contents][Index]

5.8 Registers

In the roff language, numbers can be stored in registers. Many built-in registers exist, supplying anything from the date to details of formatting parameters. You can also define your own. See Identifiers, for information on constructing a valid name for a register.


Next: , Previous: , Up: Registers   [Contents][Index]

5.8.1 Setting Registers

Define registers and update their values with the nr request or the \R escape sequence.

Request: .nr ident value
Escape sequence: \R'ident value'

Set register ident to value. If ident doesn’t exist, GNU troff creates it. In the \R escape sequence, the delimiter need not be a neutral apostrophe; see Delimiters. It also does not produce an input token in GNU troff. See Gtroff Internals.

.nr a (((17 + (3 * 4))) % 4)
\n[a]
.\R'a (((17 + (3 * 4))) % 4)'
\n[a]
    ⇒ 1 1

(Later, we will discuss additional forms of nr and \R that can change a register’s value after it is dereferenced but before it is interpolated. See Auto-increment.)

The complete transparency of \R can cause surprising effects if you use registers like .k, which get evaluated at the time they are accessed.

.ll 1.6i
.
aaa bbb ccc ddd eee fff ggg hhh\R':k \n[.k]'
.tm :k == \n[:k]
    ⇒ :k == 126950
.
.br
.
aaa bbb ccc ddd eee fff ggg hhh\h'0'\R':k \n[.k]'
.tm :k == \n[:k]
    ⇒ :k == 15000

If you process this with the PostScript device (-Tps), there will be a line break eventually after ggg in both input lines. However, after processing the space after ggg, the partially collected line is not overfull yet, so GNU troff continues to collect input until it sees the space (or in this case, the newline) after hhh. At this point, the line is longer than the line length, and the line gets broken.

In the first input line, since the \R escape sequence leaves no traces, the check for the overfull line hasn’t been done yet at the point where \R gets handled, and you get a value for the .k register that is even greater than the current line length.

In the second input line, the insertion of \h'0' to cause a zero-width motion forces GNU troff to check the line length, which in turn causes the start of a new output line. Now .k returns the expected value.

nr and \R each have two additional special forms to increment or decrement a register.

Request: .nr ident +value
Request: .nr ident -value
Escape sequence: \R'ident +value'
Escape sequence: \R'ident -value'

Increment (decrement) register ident by value. In the \R escape sequence, the delimiter need not be a neutral apostrophe; see Delimiters.

.nr a 1
.nr a +1
\na
    ⇒ 2

A leading minus sign in value is always interpreted as a decrementation operator, not an algebraic sign. To assign a register a negative value or the negated value of another register, you can force GNU troff to interpret ‘-’ as a negation or minus, rather than decrementation, operator: enclose it with its operand in parentheses or subtract it from zero.

.nr a 7
.nr b 3
.nr a -\nb
\na
    ⇒ 4
.nr a (-\nb)
\na
    ⇒ -3
.nr a 0-\nb
\na
    ⇒ -3

If a register’s prior value does not exist (the register was undefined), an increment or decrement is applied as if to 0.

Request: .rr ident

Remove register ident. If ident doesn’t exist, the request is ignored. Technically, only the name is removed; the register’s contents are still accessible under aliases created with aln, if any.

Request: .rnn ident1 ident2

Rename register ident1 to ident2. If ident1 doesn’t exist, the request is ignored. Renaming a built-in register does not otherwise alter its properties.

Request: .aln new old

Create an alias new for an existing register old, causing the names to refer to the same stored object. If old is undefined, a warning in category ‘reg’ is produced and the request is ignored. See Warnings, for information about the enablement and suppression of warnings.

To remove a register alias, invoke rr on its name. A register’s contents do not become inaccessible until it has no more names.


Next: , Previous: , Up: Registers   [Contents][Index]

5.8.2 Interpolating Registers

Register contents are interpolated with the \n escape sequence.

Escape sequence: \ni
Escape sequence: \n(id
Escape sequence: \n[ident]

Interpolate register with name ident (one-character name i, two-character name id). \n is interpreted even in copy mode (see Copy Mode). If the register is undefined, it is created and assigned a value of ‘0’, that value is interpolated, and a warning in category ‘reg’ is emitted. See Warnings, for information about the enablement and suppression of warnings.

.nr a 5
.nr as \na+\na
\n(as
    ⇒ 10
.nr a1 5
.nr ab 6
.ds str b
.ds num 1
\n[a\n[num]]
    ⇒ 5
\n[a\*[str]]
    ⇒ 6

Next: , Previous: , Up: Registers   [Contents][Index]

5.8.3 Auto-increment

Registers can also be incremented or decremented by a configured amount at the time they are interpolated. The value of the increment is specified with a third argument to the nr request, and a special interpolation syntax is used to alter and then retrieve the register’s value. Together, these features are called auto-increment.51

Request: .nr ident value incr

Set register ident to value and its auto-incrementation amount to to incr. The \R escape sequence doesn’t support an incr argument.

Auto-incrementation is not completely automatic; the \n escape sequence in its basic form never alters the value of a register. To apply auto-incrementation to a register, interpolate it with ‘\n’.

Escape sequence: \n+i
Escape sequence: \n-i
Escape sequence: \n+(id
Escape sequence: \n-(id
Escape sequence: \n+[ident]
Escape sequence: \n-[ident]

Increment or decrement ident (one-character name i, two-character name id) by the register’s auto-incrementation value and then interpolate the new register value. If ident has no auto-incrementation value, interpolate as with \n.

.nr a 0 1
.nr xx 0 5
.nr foo 0 -2
\n+a, \n+a, \n+a, \n+a, \n+a
.br
\n-(xx, \n-(xx, \n-(xx, \n-(xx, \n-(xx
.br
\n+[foo], \n+[foo], \n+[foo], \n+[foo], \n+[foo]
    ⇒ 1, 2, 3, 4, 5
    ⇒ -5, -10, -15, -20, -25
    ⇒ -2, -4, -6, -8, -10

To change the increment value without changing the value of a register, assign the register’s value to itself by interpolating it, and specify the desired increment normally. Apply an increment of ‘0’ to disable auto-incrementation of the register.


Next: , Previous: , Up: Registers   [Contents][Index]

5.8.4 Assigning Register Formats

A writable register’s value can be interpolated in several number formats. By default, conventional Arabic numerals are used. Other formats see use in sectioning and outlining schemes and alternative page numbering arrangements.

Request: .af reg fmt

Use number format fmt when interpolating register reg. Valid number formats are as follows.

0

Arabic numerals 0, 1, 2, and so on. Any decimal digit is equivalent to ‘0’; the formatter merely counts the digits specified. Multiple Arabic numerals in fmt cause interpolations to be zero-padded on the left if necessary to at least as many digits as specified (interpolations never truncate a register value). A register with format ‘00’ interpolates values 1, 2, 3 as ‘01’, ‘02’, ‘03’. The default format for all writable registers is ‘0’.

I

Uppercase Roman numerals: 0, I, II, III, IV, ...

i

Lowercase Roman numerals: 0, i, ii, iii, iv, ...

A

Uppercase letters: 0, A, B, C, …, Z, AA, AB, ...

a

Lowercase letters: 0, a, b, c, …, z, aa, ab, ...

Omitting fmt causes a warning in category ‘missing’. See Warnings, for information about the enablement and suppression of warnings. Specifying an unrecognized format is an error.

Zero values are interpolated as ‘0’ in non-Arabic formats. Negative quantities are prefixed with ‘-’ irrespective of format. In Arabic formats, the sign supplements the field width. If reg doesn’t exist, it is created with a zero value.

.nr a 10
.af a 0           \" the default format
\na,
.af a I
\na,
.af a 321
.nr a (-\na)
\na,
.af a a
\na
    ⇒ 10, X, -010, -j

The representable extrema in the ‘i’ and ‘I’ formats correspond to Arabic 39,999. GNU troff uses ‘w’ and ‘z’ to represent 5,000 and 10,000 in Roman numerals, respectively, following the convention of AT&T troff—currently, the correct glyphs for Roman numerals five thousand (U+2181) and ten thousand (U+2182) are not used.

Assigning the format of a read-only register is an error. Instead, copy the read-only register’s value to, and assign the format of, a writable register.

Escape sequence: \gr
Escape sequence: \g(rg
Escape sequence: \g[reg]

Interpolate the format of the register reg (one-character name r, two-character name rg). Zeroes represent Arabic formats. If reg is not defined, reg is not created and nothing is interpolated. \g is interpreted even in copy mode (see Copy Mode).

GNU troff interprets only Arabic numerals. The Roman numeral or alphabetic formats cannot be used as operands to arithmetic operators in expressions (see Numeric Expressions). For instance, it may be desirable to test the page number independently of its format.

.af % i \" front matter
.de header-trap
.  \" To test the page number, we need it in Arabic.
.  ds saved-page-number-format \\g%\"
.  af % 0
.  nr page-number-in-decimal \\n%
.  af % \\*[saved-page-number-format]
.  ie \\n[page-number-in-decimal]=1 .do-first-page-stuff
.  el \{\
.    ie o .do-odd-numbered-page-stuff
.    el   .do-even-numbered-page-stuff
.  \}
.  rm saved-page-number-format
..
.wh 0 header-trap

Previous: , Up: Registers   [Contents][Index]

5.8.5 Built-in Registers

Predefined registers whose identifiers start with a dot are read-only. Many are Boolean-valued, interpolating a true or false value testable with the if, ie, or while requests. Some read-only registers are string-valued, meaning that they interpolate text.

Caution: Built-in registers are subject to removal like others; once removed, they can be recreated only as normal writable registers and will not reflect formatter state.

A register name (without the dot) is often associated with a request of the same name. A complete listing of all built-in registers can be found in Register Index.

We present here a few built-in registers that are not described elsewhere in this manual; they have to do with invariant properties of GNU troff, or obtain information about the formatter’s command-line options, processing progress, or the operating environment.

\n[.A]

Approximate output is being formatted (Boolean-valued); see groff -a option (Groff Options).

\n[.c]
\n[c.]

Input line number. ‘c.’ is a writable synonym, affecting subsequent interpolations of both ‘.c’ and ‘c.’.

\n[.F]

Name of input file (string-valued).

\n[.g]

Always true in GNU troff (Boolean-valued). Documents can use this to ask the formatter if it claims groff compatibility.

\n[.P]

Output page selection status (Boolean-valued); see groff -o option (Groff Options).

\n[.R]

Count of available unused registers; always 10,000 in GNU troff.52

\n[.T]

Indicator of output device selection (Boolean-valued); see groff -T option (Groff Options).

\n[.U]

Unsafe mode enablement status (Boolean-valued); see groff -U option (Groff Options).

\n[.x]

Major version number of the running GNU troff formatter. For example, if the version number is 1.23.0, then .x contains ‘1’.

\n[.y]

Minor version number of the running GNU troff formatter. For example, if the version number is 1.23.0, then .y contains ‘23’.

\n[.Y]

Revision number of the running GNU troff formatter. For example, if the version number is 1.23.0, then .Y contains ‘0’.

\n[$$]

Process identifier (PID) of the GNU troff program in its operating environment.

Date- and time-related registers are set per the local time as determined by localtime(3) when the formatter launches. This initialization can be overridden by SOURCE_DATE_EPOCH and TZ; see Environment.

\n[seconds]

Count of seconds elapsed in the minute (0–60).

\n[minutes]

Count of minutes elapsed in the hour (0–59).

\n[hours]

Count of hours elapsed since midnight (0–23).

\n[dw]

Day of the week (1–7; 1 is Sunday).

\n[dy]

Day of the month (1–31).

\n[mo]

Month of the year (1–12).

\n[year]

Gregorian year.

\n[yr]

Gregorian year minus 1900. This register is incorrectly documented in the AT&T troff manual as storing the last two digits of the current year. That claim stopped being true in 2000. Old troff input that looks like:

'\" The year number is a surprise after 1999.
This document was formatted in 19\n(yr.

can be corrected to:

This document was formatted in \n[year].

or, for portability across many roff programs, to the following.

.nr y4 1900+\n(yr
This document was formatted in \n(y4.

Next: , Previous: , Up: GNU troff Reference   [Contents][Index]

5.9 Manipulating Filling and Adjustment

When an output line is pending (see below), a break moves the drawing position to the beginning of the next text baseline, interrupting filling. Various ways of causing breaks were shown in Breaking. The br request likewise causes a break. Several other requests imply breaks: bp, ce, cf, fi, fl, in, nf, rj, sp, ti, and trf. If the no-break control character is used with any of these requests, GNU troff suppresses the break; instead the requested operation takes effect at the next break. ‘'br’ does nothing.

.ll 55n
This line is normally filled and adjusted.
.br
A line's alignment is decided
'ce \" Center the next input line (no break).
when it is output.
This line returns to normal filling and adjustment.
    ⇒ This line is normally filled and adjusted.
    ⇒    A line's alignment is decided when it is output.
    ⇒ This line returns to normal filling and adjustment.

Output line properties like page offset, indentation, adjustment, and even the location of its text baseline, are not determined until the line has been broken. An output line is said to be pending if some input has been collected but an output line corresponding to it has not yet been written; such an output line is also termed partially collected. If no output line is pending, it is as if a break has already happened; additional breaks, whether explicit or implicit, have no effect. If the vertical drawing position is negative—as it is when the formatter starts up—a break starts a new page (even if no output line is pending) unless an end-of-input macro is being interpreted. See End-of-input Traps.

Request: .br

Break the line: emit any pending output line without adjustment.

foo bar
.br
baz
'br
qux
    ⇒ foo bar
    ⇒ baz qux

Sometimes you want to prevent a break within a phrase or between a quantity and its units.

Escape sequence: \~

Insert an unbreakable space that is adjustable like an ordinary space. It is discarded from the end of an output line if a break is forced.

Set the output speed to\~1.
There are 1,024\~bytes in 1\~KiB.
J.\~F.\~Ossanna wrote the original CSTR\~#54.

By default, GNU troff fills text and adjusts it to reach the output line length. The nf request disables filling; the fi request renables it.

Request: .fi
Register: \n[.u]

Enable filling of output lines; a pending output line is broken. The read-only register .u is set to 1. The filling enablement status, sometimes called fill mode, is associated with the environment (see Environments). See Line Continuation, for interaction with the \c escape sequence.

Request: .nf

Disable filling of output lines: the output line length (see Line Layout) is ignored and output lines are broken where the input lines are. A pending output line is broken and adjustment is suppressed. The read-only register .u is set to 0. The filling enablement status is associated with the environment (see Environments). See Line Continuation, for interaction with the \c escape sequence.

Request: .ad [mode]
Register: \n[.j]

Enable output line adjustment in mode, taking effect when the pending (or next) output line is broken. Adjustment is suppressed when filling is. mode can have one of the following values.

b
n

Adjust “normally”: if the output line does not consume the distance between the indentation and the configured output line length, GNU troff stretches adjustable spaces within the line until that length is reached. When the indentation is zero, this mode spreads the line to both the left and right margins. This is the GNU troff default.

c

Center filled text. Contrast with the ce request, which centers text without filling it.

l

Align text to the left without adjusting it.

r

Align text to the right without adjusting it.

mode can also be a value previously stored in the .j register. Using ad without an argument is the same as ‘.ad \n[.j]’; unless filling is disabled, GNU troff resumes adjusting lines in the same way it did before adjustment was disabled by invocation of the na request.

The adjustment mode and enablement status are encoded in the read-only register .j. These parameters are associated with the environment (see Environments).

The value of .j for any adjustment mode is an implementation detail and should not be relied upon as a programmer’s interface. Do not write logic to interpret or perform arithmetic on it.

.ll 48n
.de AD
.  br
.  ad \\$1
..
.de NA
.  br
.  na
..
left
.AD r
.nr ad \n(.j
right
.AD c
center
.NA
left
.AD
center
.AD \n(ad
right
    ⇒ left
    ⇒                                            right
    ⇒                      center
    ⇒ left
    ⇒                      center
    ⇒                                            right
Request: .na

Disable output line adjustment. This produces the same output as left-alignment, but the value of the adjustment mode register .j is altered differently. The adjustment mode and enablement status are associated with the environment (see Environments).

Request: .brp
Escape sequence: \p

Break, adjusting the line per the current adjustment mode. \p schedules a break with adjustment at the next word boundary. The escape sequence is itself neither a break nor a space of any kind; it can thus be placed in the middle of a word to cause a break at the end of that word.

Breaking with immediate adjustment can produce ugly results since GNU troff doesn’t have a sophisticated paragraph-building algorithm, as TeX has, for example. Instead, GNU troff fills and adjusts a paragraph line by line.

.ll 4.5i
This is an uninteresting sentence.
This is an uninteresting sentence.\p
This is an uninteresting sentence.

is formatted as follows.

This  is  an uninteresting sentence.  This is
an          uninteresting           sentence.
This is an uninteresting sentence.

To clearly present the next couple of requests, we must introduce the concept of “productive” input lines. A productive input line is one that directly produces formatted output. Text lines produce output,53 as do control lines containing requests like tl or escape sequences like \D. Macro calls are not directly productive, and thus not counted, but their interpolated contents can be. Empty requests, and requests and escape sequences that define registers or strings or alter the formatting environment (as with changes to the size, face, height, slant, or color of the type) are not productive. We will also preview the output line continuation escape sequence, \c, which “connects” two input lines that would otherwise be counted separately. 54

.de hello
Hello, world!
..
.ce \" center output of next productive input line
.
.nr junk-reg 1
.ft I
Chorus: \c
.ft
.hello
Went the day well?
  ⇒                  Chorus: Hello, world!
  ⇒ Went the day well?
Request: .ce [n]
Register: \n[.ce]

Break (unless the no-break control character is used), center the output of the next n productive input lines with respect to the line length and indentation without filling, then break again regardless of the invoking control character. If the argument is not positive, centering is disabled. Omitting the argument implies an n of ‘1’. The count of lines remaining to be centered is stored in the read-only register .ce and is associated with the environment (see Environments).

While the ‘.ad c request also centers text, it fills the text as well.

.de FR
This is a small text fragment that shows the differences
between the `.ce' and the `.ad c' requests.
..
.ll 4i
.ce 1000
.FR
.ce 0

.ad c
.FR
    ⇒ This is a small text fragment that shows
    ⇒              the differences
    ⇒ between the ‘.ce’ and the ‘.ad c’ requests.
    ⇒
    ⇒ This is a small text fragment that shows
    ⇒  the differences between the ‘.ce’ and
    ⇒         the ‘.ad c’ requests.

The previous example illustrates a common idiom of turning centering on for a quantity of lines far in excess of what is required, and off again after the text to be centered. This technique relieves humans of counting lines for requests that take a count of input lines as an argument.

Request: .rj [n]
Register: \n[.rj]

Break (unless the no-break control character is used), align the output of the next n productive input lines to the right margin without filling, then break again regardless of the control character. If the argument is not positive, right-alignment is disabled. Omitting the argument implies an n of ‘1’. The count of lines remaining to be right-aligned is stored in the read-only register .rj and is associated with the environment (see Environments).

.ll 49n
.rj 3
At first I hoped that such a technically unsound
project would collapse but I soon realized it was
doomed to success. \[em] C. A. R. Hoare
    ⇒  At first I hoped that such a technically unsound
    ⇒ project would collapse but I soon realized it was
    ⇒              doomed to success. -- C. A. R. Hoare
Request: .ss word-space-size [additional-sentence-space-size]
Register: \n[.ss]
Register: \n[.sss]

Set the sizes of spaces between words and sentences55 in twelfths of font’s space width (typically one-fourth to one-third em for Western scripts). The default for both parameters is 12. Negative values are erroneous. The first argument is a minimum; if an output line undergoes adjustment, such spaces may increase in width. The optional second argument sets the amount of additional space separating sentences on the same output line. If omitted, this amount is set to word-space-size. The request is ignored if there are no parameters.

Additional inter-sentence space is used only if the output line is not full when the end of a sentence occurs in the input. If a sentence ends at the end of an input line, then both an inter-word space and an inter-sentence space are added to the output; if two spaces follow the end of a sentence in the middle of an input line, then the second space becomes an inter-sentence space in the output. Additional inter-sentence space is not adjusted, but the inter-word space that always precedes it may be. Further input spaces after the second, if present, are adjusted as normal.

The read-only registers .ss and .sss hold the minimal inter-word space and additional inter-sentence space amounts, respectively. These parameters are part of the environment (see Environments), and rounded down to the nearest multiple of 12 on terminals.

The ss request can insert discardable horizontal space; that is, space that is discarded at a break. For example, some footnote styles collect the notes into a single paragraph with large gaps between each note.

.ll 48n
1.\~J. Fict. Ch. Soc. 6 (2020), 3\[en]14.
.ss 12 48 \" applies to next sentence ending
Reprints no longer available through FCS.
.ss 12 \" go back to normal
2.\~Better known for other work.
    ⇒ 1.  J.  Fict. Ch. Soc. 6 (2020), 3-14.  Reprints
    ⇒ no longer available through FCS.      2.  Better
    ⇒ known for other work.

If undiscardable space is required, use the \h escape sequence.


Next: , Previous: , Up: GNU troff Reference   [Contents][Index]

5.10 Manipulating Hyphenation

When filling, GNU troff hyphenates words as needed at user-specified and automatically determined hyphenation points. The machine-driven determination of hyphenation points in words requires algorithms and data, and is susceptible to conventions and preferences. Before tackling such automatic hyphenation, let us consider how hyphenation points can be set explicitly.

Explicitly hyphenated words such as “mother-in-law” are eligible for breaking after each of their hyphens. Relatively few words in a language offer such obvious break points, however, and automatic detection of syllabic (or phonetic) boundaries for hyphenation is not perfect,56 particularly for unusual words found in technical literature. We can instruct GNU troff how to hyphenate specific words if the need arises.

Request: .hw word …

Define each hyphenation exception word with each hyphen ‘-’ in the word indicating a hyphenation point. For example, the request

.hw in-sa-lub-rious alpha

marks potential hyphenation points in “insalubrious”, and prevents “alpha” from being hyphenated at all.

Besides the space character, any character whose hyphenation code is zero can be used to separate the arguments of hw (see the hcode request below). In addition, this request can be used more than once.

Hyphenation points specified with hw are not subject to the within-word placement restrictions imposed by the hy request (see below).

Hyphenation exceptions specified with the hw request are associated with the hyphenation language (see the hla request below) and environment (see Environments); invoking the hw request in the absence of a hyphenation language is an error.

The request is ignored if there are no parameters.

These are known as hyphenation exceptions in the expectation that most users will avail themselves of automatic hyphenation; these exceptions override any rules that would normally apply to a word matching a hyphenation exception defined with hw.

Situations also arise when only a specific occurrence of a word needs its hyphenation altered or suppressed, or when a URL or similar string needs to be breakable in sensible places without hyphenation.

Escape sequence: \%
Escape sequence: \:

To tell GNU troff how to hyphenate words as they occur in input, use the \% escape sequence; it is the default hyphenation character. Each instance within a word indicates to GNU troff that the word may be hyphenated at that point, while prefixing a word with this escape sequence prevents it from being otherwise hyphenated. This mechanism affects only that occurrence of the word; to change the hyphenation of a word for the remainder of input processing, use the hw request.

GNU troff regards the escape sequences \X and \Y as starting a word; that is, the \% escape sequence in, say, ‘\X'...'\%foobar or ‘\Y'...'\%foobar no longer prevents hyphenation of ‘foobar’ but inserts a hyphenation point just prior to it; most likely this isn’t what you want. See Postprocessor Access.

\: inserts a non-printing break point; that is, a word can break there, but the soft hyphen glyph (see below) is not written to the output if it does. This escape sequence is an input word boundary, so the remainder of the word is subject to hyphenation as normal.

You can combine \: and \% to control breaking of a file name or URL, or to permit hyphenation only after certain explicit hyphens within a word.

The \%Lethbridge-Stewart-\:\%Sackville-Baggins divorce
was, in retrospect, inevitable once the contents of
\%/var/log/\:\%httpd/\:\%access_log on the family web
server came to light, revealing visitors from Hogwarts.
Request: .hc [char]

Change the hyphenation character to char. This character then works as the \% escape sequence normally does, and thus no longer appears in the output.57 Without an argument, hc resets the hyphenation character to \% (the default). The hyphenation character is associated with the environment (see Environments).

Request: .shc [c]

Set the soft hyphen character, inserted when a word is hyphenated automatically or at a hyphenation character, to the ordinary or special character c.58 If the argument is omitted, the soft hyphen character is set to the default, \[hy]. If no glyph for c exists in the font in use at a potential hyphenation point, then the line is not broken there. Neither character definitions (specified with the char and similar requests) nor translations (specified with the tr request) are applied to c.

Several requests influence automatic hyphenation. Because conventions vary, a variety of hyphenation modes is available to the hy request; these determine whether hyphenation will apply to a word prior to breaking a line at the end of a page (more or less; see below for details), and at which positions within that word automatically determined hyphenation points are permissible. The places within a word that are eligible for hyphenation are determined by language-specific data and lettercase relationships. Furthermore, hyphenation of a word might be suppressed due to a limit on consecutive hyphenated lines (hlm), a minimum line length threshold (hym), or because the line can instead be adjusted with additional inter-word space (hys).

Request: .hy [mode]
Register: \n[.hy]

Set automatic hyphenation mode to mode, an integer encoding conditions for hyphenation; if omitted, ‘1’ is implied. The hyphenation mode is available in the read-only register ‘.hy’; it is associated with the environment (see Environments). The default hyphenation mode depends on the localization file loaded when GNU troff starts up; see the hpf request below.

Typesetting practice generally does not avail itself of every opportunity for hyphenation, but the details differ by language and site mandates. The hyphenation modes of AT&T troff were implemented with English-language publishing practices of the 1970s in mind, not a scrupulous enumeration of conceivable parameters. GNU troff extends those modes such that finer-grained control is possible, favoring compatibility with older implementations over a more intuitive arrangement. The means of hyphenation mode control is a set of numbers that can be added up to encode the behavior sought.59 The entries in the following table are termed values; the sum of the desired values is the mode.

0

disables hyphenation.

1

enables hyphenation except after the first and before the last character of a word.

The remaining values “imply” 1; that is, they enable hyphenation under the same conditions as ‘.hy 1’, and then apply or lift restrictions relative to that basis.

2

disables hyphenation of the last word on a page,60 even for explicitly hyphenated words.

4

disables hyphenation before the last two characters of a word.

8

disables hyphenation after the first two characters of a word.

16

enables hyphenation before the last character of a word.

32

enables hyphenation after the first character of a word.

Apart from value 2, restrictions imposed by the hyphenation mode are not respected for words whose hyphenations have been specified with the hyphenation character (‘\%’ by default) or the hw request.

Nonzero values in the previous table are additive. For example, mode 12 causes GNU troff to hyphenate neither the last two nor the first two characters of a word. Some values cannot be used together because they contradict; for instance, values 4 and 16, and values 8 and 32. As noted, it is superfluous to add 1 to any non-zero even mode.

The automatic placement of hyphens in words is determined by pattern files, which are derived from TeX and available for several languages. The number of characters at the beginning of a word after which the first hyphenation point should be inserted is determined by the patterns themselves; it can’t be reduced further without introducing additional, invalid hyphenation points (unfortunately, this information is not part of a pattern file—you have to know it in advance). The same is true for the number of characters at the end of a word before the last hyphenation point should be inserted. For example, you can supply the following input to ‘echo $(nroff)’.

.ll 1
.hy 48
splitting

You will get

s- plit- t- in- g

instead of the correct ‘split- ting’. English patterns as distributed with GNU troff need two characters at the beginning and three characters at the end; this means that value 4 of hy is mandatory. Value 8 is possible as an additional restriction, but values 16 and 32 should be avoided, as should mode 1. Modes 4 and 6 are typical.

A table of left and right minimum character counts for hyphenation as needed by the patterns distributed with GNU troff follows; see the groff_tmac(5) man page for more information on GNU troff’s language macro files.

languagepattern nameleft minright min
Czechcs22
Englishen23
Frenchfr23
German traditionaldet22
German reformedden22
Italianit22
Swedishsv12

Hyphenation exceptions within pattern files (i.e., the words within a TeX \hyphenation group) obey the hyphenation restrictions given by hy.

Request: .nh

Disable automatic hyphenation; i.e., set the hyphenation mode to 0 (see above). The hyphenation mode of the last call to hy is not remembered.

Request: .hpf pattern-file
Request: .hpfa pattern-file
Request: .hpfcode a b [c d] …

Read hyphenation patterns from pattern-file, which is sought in the same way that macro files are with the mso request or the -mname command-line option to groff. The pattern-file should have the same format as (simple) TeX pattern files. More specifically, the following scanning rules are implemented.

The hpfa request appends a file of patterns to the current list.

The hpfcode request defines mapping values for character codes in pattern files. It is an older mechanism no longer used by GNU troff’s own macro files; for its successor, see hcode below. hpf or hpfa apply the mapping after reading the patterns but before replacing or appending to the active list of patterns. Its arguments are pairs of character codes—integers from 0 to 255. The request maps character code a to code b, code c to code d, and so on. Character codes that would otherwise be invalid in GNU troff can be used. By default, every code maps to itself except those for letters ‘A’ to ‘Z’, which map to those for ‘a’ to ‘z’.

The set of hyphenation patterns is associated with the language set by the hla request (see below). The hpf request is usually invoked by a localization file loaded by the troffrc file.61

A second call to hpf (for the same language) replaces the hyphenation patterns with the new ones. Invoking hpf or hpfa causes an error if there is no hyphenation language. If no hpf request is specified (either in the document, in a file loaded at startup, or in a macro package), GNU troff won’t automatically hyphenate at all.

Request: .hcode c1 code1 [c2 code2] …

Set the hyphenation code of character c1 to code1, that of c2 to code2, and so on. A hyphenation code must be an ordinary character (not a special character escape sequence) other than a digit or a space. The request is ignored if given no arguments.

For hyphenation to work, hyphenation codes must be set up. At startup, GNU troff assigns hyphenation codes to the letters ‘a’–‘z’ (mapped to themselves), to the letters ‘A’–‘Z’ (mapped to ‘a’–‘z’), and zero to all other characters. Normally, hyphenation patterns contain only lowercase letters which should be applied regardless of case. In other words, they assume that the words ‘FOO’ and ‘Foo’ should be hyphenated exactly as ‘foo’ is. The hcode request extends this principle to letters outside the Unicode basic Latin alphabet; without it, words containing such letters won’t be hyphenated properly even if the corresponding hyphenation patterns contain them.

For example, the following hcode requests are necessary to assign hyphenation codes to the letters ‘’, needed for German.

.hcode     
.hcode     
.hcode     
.hcode  

Without these assignments, GNU troff treats the German word ‘Kindergrten’ (the plural form of ‘kindergarten’) as two words ‘kinderg’ and ‘rten’ because the hyphenation code of the umlaut a is zero by default, just like a space. There is a German hyphenation pattern that covers ‘kinder’, so GNU troff finds the hyphenation ‘kin-der’. The other two hyphenation points (‘kin-der-gr-ten’) are missed.

Request: .hla lang
Register: \n[.hla]

Set the hyphenation language to lang. Hyphenation exceptions specified with the hw request and hyphenation patterns and exceptions specified with the hpf and hpfa requests are associated with the hyphenation language. The hla request is usually invoked by a localization file, which is turn loaded by the troffrc or troffrc-end file; see the hpf request above.

The hyphenation language is available in the read-only string-valued register ‘.hla’; it is associated with the environment (see Environments).

Request: .hlm [n]
Register: \n[.hlm]
Register: \n[.hlc]

Set the maximum quantity of consecutive hyphenated lines to n. If n is negative, there is no maximum. If omitted, n is -1. This value is associated with the environment (see Environments). Only lines output from a given environment count toward the maximum associated with that environment. Hyphens resulting from \% are counted; explicit hyphens are not.

The .hlm read-only register stores this maximum. The count of immediately preceding consecutive hyphenated lines is available in the read-only register .hlc.

Request: .hym [length]
Register: \n[.hym]

Set the (right) hyphenation margin to length. If the adjustment mode is not ‘b’ or ‘n’, the line is not hyphenated if it is shorter than length. Without an argument, the hyphenation margin is reset to its default value, 0. The default scaling unit is ‘m’. The hyphenation margin is associated with the environment (see Environments).

A negative argument resets the hyphenation margin to zero, emitting a warning in category ‘range’.

The hyphenation margin is available in the .hym read-only register.

Request: .hys [hyphenation-space]
Register: \n[.hys]

Suppress hyphenation of the line in adjustment modes ‘b’ or ‘n’ if it can be justified by adding no more than hyphenation-space extra space to each inter-word space. Without an argument, the hyphenation space adjustment threshold is set to its default value, 0. The default scaling unit is ‘m’. The hyphenation space adjustment threshold is associated with the environment (see Environments).

A negative argument resets the hyphenation space adjustment threshold to zero, emitting a warning in category ‘range’.

The hyphenation space adjustment threshold is available in the .hys read-only register.


Next: , Previous: , Up: GNU troff Reference   [Contents][Index]

5.11 Manipulating Spacing

A break causes the formatter to update the vertical drawing position at which the new text baseline is aligned. You can alter this location.

Request: .sp [distance]

Break and move the next text baseline down by distance, or until springing a page location trap.62 If invoked with the no-break control character, sp moves the pending output line’s text baseline by distance. A negative distance will not reduce the position of the text baseline below zero. Inside a diversion, any distance argument is ignored. The default scaling unit is ‘v’. If distance is not specified, ‘1v’ is assumed.

.pl 5v \" Set page length to 5 vees.
.de xx
\-\-\-
.  br
..
.wh 0 xx \" Set a trap at the top of the page.
foo on page \n%
.sp 2v
bar on page \n%
.sp 50v \" This will cause a page break.
baz on page \n%
.pl \n(nlu \" Truncate page to current position.
    ⇒ ---
    ⇒ foo on page 1
    ⇒
    ⇒
    ⇒ bar on page 1
    ⇒ ---
    ⇒ baz on page 2

You might use the following macros to set the baseline of the next output text at a given distance from the top or the bottom of the page. We subtract one line height (\n[.v]) because the | operator moves to one vee below the page top (recall Numeric Expressions).

.de y-from-top-down
.  sp |\\$1-\\n[.v]u
..
.
.de y-from-bot-up
.  sp |\\n[.p]u-\\$1-\\n[.v]u
..

A call to ‘.y-from-bot-up 10c’ means that the next text baseline will be 10 cm from the bottom edge of the paper.

Request: .ls [count]
Register: \n[.L]

Set the line spacing; add count-1 blank lines after each line of text. With no argument, GNU troff uses the previous value before the last ls call. The default is 1.

The read-only register .L contains the current line spacing; it is associated with the environment (see Environments).

The ls request is a coarse mechanism. See Changing the Type Size, for the requests vs and pvs as alternatives to ls.

Escape sequence: \x'spacing'
Register: \n[.a]

Sometimes, an output line requires additional vertical spacing, for instance to allow room for a tall construct like an inline equation with exponents or subscripts (particularly if they are iterated). The \x escape sequence takes a delimited measurement (like ‘\x'3p'’) to increase the vertical spacing of the pending output line. The default scaling unit is ‘v’. If the measurement is positive, extra vertical space is inserted below the current line; a negative measurement adds space above. If \x is applied to the pending output line multiple times, the maxima of the positive and negative adjustments are separately applied. The delimiter need not be a neutral apostrophe; see Delimiters.

The .a read-only register contains the extra vertical spacing after the text baseline of the most recently emitted output line. (In other words, it is the largest positive argument to \x encountered on that line.) This quantity is exposed via a register because if an output line requires this “extra post-vertical line spacing”, and the subsequent output line requires “extra pre-vertical line spacing” (a negative argument to \x), then applying both can lead to excessive spacing between the output lines. Text that is piling high on line n might not require (as much) extra pre-vertical line spacing if line n-1 carries extra post-vertical line spacing.

Use of \x can be necessary in combination with the bracket-building escape sequence \b,63 as the following example shows.

.nf
This is a test of \[rs]b (1).
This is a test of \[rs]b (2).
This is a test of \b'xyz'\x'-1m'\x'1m' (3).
This is a test of \[rs]b (4).
This is a test of \[rs]b (5).
    ⇒ This is a test of \b (1).
    ⇒ This is a test of \b (2).
    ⇒                   x
    ⇒ This is a test of y (3).
    ⇒                   z
    ⇒ This is a test of \b (4).
    ⇒ This is a test of \b (5).

Without \x, the backslashes on the lines marked ‘(2)’ and ‘(4)’ would be overprinted.

Request: .ns
Request: .rs
Register: \n[.ns]

Enable no-space mode. Vertical spacing, whether by sp requests or blank input lines, is disabled. The bp request to advance to the next page is also disabled, unless it is accompanied by a page number (see Page Control). No-space mode ends automatically when text64 is formatted for output 65 or the rs request is invoked, which ends no-space mode. The read-only register .ns interpolates a Boolean value indicating the enablement of no-space mode.

A paragraphing macro might ordinarily insert vertical space to separate paragraphs. A section heading macro could invoke ns to suppress this spacing for the first paragraph in a section.


Next: , Previous: , Up: GNU troff Reference   [Contents][Index]

5.12 Tabs and Fields

A tab character (ISO code point 9, EBCDIC code point 5) causes a horizontal movement to the next tab stop, if any.

Escape sequence: \t

Interpolate a tab in copy mode; see Copy Mode.

Request: .ta [[n1 n2nn ]T r1 r2rn]
Register: \n[.tabs]

Change tab stop positions. This request takes a series of tab specifiers as arguments (optionally divided into two groups with the letter ‘T’) that indicate where each tab stop is to be, overriding any previous settings. The default scaling unit is ‘m’. Invoking ta without an argument removes all tab stops. GNU troff’s startup value is ‘T 0.5i.

Tab stops can be specified absolutely—as distances from the left margin. The following example sets six tab stops, one every inch.

.ta 1i 2i 3i 4i 5i 6i

Tab stops can also be specified using a leading ‘+’, which means that the specified tab stop is set relative to the previous tab stop. For example, the following is equivalent to the previous example.

.ta 1i +1i +1i +1i +1i +1i

GNU troff supports an extended syntax to specify repeating tab stops. These stops appear after a ‘T’ argument. Their values are always taken as distances relative to the previous tab stop. This is the idiomatic way to specify tab stops at equal intervals in groff. The following is, yet again, the same as the previous examples. It does more, in fact, since it defines an infinite number of tab stops at one-inch intervals.

.ta T 1i

Now we are ready to interpret the full syntax given above. The ta request sets tabs at positions n1, n2, …, nn, then at nn+r1, nn+r2, …, nn+rn, then at nn+rn+r1, nn+rn+r2, …, nn+rn+rn, and so on.

For example, ‘4c +6c T 3c 5c 2c’ is equivalent to ‘4c 10c 13c 18c 20c 23c 28c 30c …’.

Text written to a tab column (i.e., between two tab stops, or between a tab stop and an output line boundary) may be aligned to the right or left, or centered in the column. This alignment is determined by appending ‘R’, ‘L’, or ‘C’ to the tab specifier. The default is ‘L’.

.ta 1i 2iC 3iR

The beginning of an output line is not a tab stop; the text that begins an output line is placed according to the configured alignment and indentation; see Manipulating Filling and Adjustment and Line Layout.

A tab stop is converted into a non-breakable horizontal movement that cannot be adjusted.

.ll 2i
.ds foo a\tb\tc
.ta T 1i
\*[foo]
    error→ warning: cannot break line
    ⇒ a         b         c

The above creates a single output line that is a bit longer than two inches (we use a string to show exactly where the tab stops are). Now consider the following.

.ll 2i
.ds bar a\tb c\td
.ta T 1i
\*[bar]
    error→ warning: cannot adjust line
    ⇒ a         b
    ⇒ c       d

GNU troff first converts the line’s tab stops into unbreakable horizontal movements, then breaks after ‘b’. This usually isn’t what you want.

Superfluous tab characters—those that do not correspond to a tab stop—are ignored except for the first, which delimits the characters belonging to the last tab stop for right-alignment or centering.

.ds Z   foo\tbar\tbaz
.ds ZZ  foo\tbar\tbazqux
.ds ZZZ foo\tbar\tbaz\tqux
.ta 2i 4iR
\*[Z]
.br
\*[ZZ]
.br
\*[ZZZ]
.br
    ⇒ foo                 bar              baz
    ⇒ foo                 bar           bazqux
    ⇒ foo                 bar              bazqux

The first line right-aligns “baz” within the second tab stop. The second line right-aligns “bazqux” within it. The third line right-aligns only “baz” because of the additional tab character, which marks the end of the text occupying the last tab stop defined.

Tab stops are associated with the environment (see Environments).

The read-only register .tabs contains a string representation of the current tab settings suitable for use as an argument to the ta request.66

.ds tab-string \n[.tabs]
\*[tab-string]
    ⇒ T120u
Request: .tc [c]

Set the tab repetition character to the ordinary or special character c; normally, no glyph is written when moving to a tab stop (and some output devices may output space characters to achieve this motion). A tab repetition character causes the formatter to write as many instances of c as are necessary to occupy the interval from the horizontal drawing position to the next tab stop. With no argument, GNU troff reverts to the default behavior. The tab repetition character is associated with the environment (see Environments). Only a single character of c is recognized; any excess is ignored.

Request: .linetabs n
Register: \n[.linetabs]

If n is missing or non-zero, activate line-tabs; deactivate it otherwise (the default). Active line-tabs cause GNU troff to compute tab distances relative to the start of the output line instead of the input line.

.de Tabs
.  ds x a\t\c
.  ds y b\t\c
.  ds z c
.  ta 1i 3i
\\*x
\\*y
\\*z
..
.Tabs
.br
.linetabs
.Tabs
    ⇒ a         b         c
    ⇒ a         b                   c

Line-tabs activation is associated with the environment (see Environments). The read-only register .linetabs interpolates 1 if line-tabs are active, and 0 otherwise.


Next: , Previous: , Up: Tabs and Fields   [Contents][Index]

5.12.1 Leaders

Sometimes it is desirable to fill a tab stop with a given glyph, but also use tab stops normally on the same output line. An example is a table of contents entry that uses dots to bridge the entry name with its page number, which is itself aligned between tab stops. The roff language provides leaders for this purpose.67

A leader character (ISO and EBCDIC code point 1, also known as SOH or “start of heading”), behaves similarly to a tab character: it moves to the next tab stop. The difference is that for this movement, the default fill character is a period ‘.’.

Escape sequence: \a

Interpolate a leader in copy mode; see Copy Mode.

Request: .lc [c]

Set the leader repetition character to the ordinary or special character c. Recall Tabs and Leaders: when encountering a leader character in the input, the formatter writes as many dots ‘.’ as are necessary until reaching the next tab stop; this is the leader definition character. Omitting c unsets the leader character. With no argument, GNU troff treats leaders the same as tabs. The leader repetition character is associated with the environment (see Environments). Only a single c is recognized; any excess is ignored.

A table of contents, for example, may define tab stops after a section number, a title, and a gap to be filled with leader dots. The page number follows the leader, after a right-aligned final tab stop wide enough to house the largest page number occurring in the document.

.ds entry1 19.\tThe Prophet\a\t98
.ds entry2 20.\tAll Astir\a\t101
.ta .5i 4.5i +.5iR
.nf
\*[entry1]
\*[entry2]
    ⇒ 19.  The Prophet.............................   98
    ⇒ 20.  All Astir...............................  101

Previous: , Up: Tabs and Fields   [Contents][Index]

5.12.2 Fields

Fields are a more general way of laying out tabular data. A field is defined as the data between a pair of delimiting characters. It contains substrings that are separated by padding characters. The width of a field is the distance on the input line from the position where the field starts to the next tab stop. A padding character inserts an adjustable space similar to TeX’s \hss command (thus it can even be negative) to make the sum of all substring lengths plus the adjustable space equal to the field width. If more than one padding character is inserted, the available space is evenly distributed among them.

Request: .fc [delim-char [padding-char]]

Define a delimiting and a padding character for fields. If the latter is missing, the padding character defaults to a space character. If there is no argument at all, the field mechanism is disabled (which is the default). In contrast to, e.g., the tab repetition character, delimiting and padding characters are not associated with the environment (see Environments).

.fc # ^
.ta T 3i
#foo^bar^smurf#
.br
#foo^^bar^smurf#
    ⇒ foo         bar          smurf
    ⇒ foo            bar       smurf

Next: , Previous: , Up: GNU troff Reference   [Contents][Index]

5.13 Character Translations

A translation is a mapping of an input character to an output glyph. The mapping occurs at output time, i.e., the input character gets assigned the metric information of the mapped output character right before input tokens are converted to nodes (see Gtroff Internals, for more on this process).

Request: .tr abcd
Request: .trin abcd

Translate character a to glyph b, character c to glyph d, and so on. If there is an odd number of characters in the argument, the last one is translated to a fixed-width space (the same one obtained by the \SP escape sequence).

The trin request is identical to tr, but when you unformat a diversion with asciify it ignores the translation. See Diversions, for details about the asciify request.

Some notes:

Request: .trnt abcd

trnt is the same as the tr request except that the translations do not apply to text that is transparently throughput into a diversion with \!. See Diversions.

For example,

.tr ab
.di x
\!.tm a
.di
.x

prints ‘b’ to the standard error stream; if trnt is used instead of tr it prints ‘a’.


Next: , Previous: , Up: GNU troff Reference   [Contents][Index]

5.14 troff and nroff Modes

Historically, nroff and troff were two separate programs; the former for terminal output, the latter for typesetters. GNU troff merges both functions into one executable68 that sends its output to a device driver (grotty for terminal devices, grops for PostScript, and so on) which interprets this intermediate output format. When discussing AT&T troff, it makes sense to talk about nroff mode and troff mode since the differences are hard-coded. GNU troff takes information from device and font description files without handling requests specially if a terminal output device is used, so such a strong distinction is unnecessary.

Usually, a macro package can be used with all output devices. Nevertheless, it is sometimes necessary to make a distinction between terminal and non-terminal devices: GNU troff provides two built-in conditions ‘n’ and ‘t’ for the if, ie, and while requests to decide whether GNU troff shall behave like nroff or like troff.

Request: .troff

Make the ‘t’ built-in condition true (and the ‘n’ built-in condition false) for if, ie, and while conditional requests. This is the default if GNU troff (not groff) is started with the -R switch to avoid loading of the startup files troffrc and troffrc-end. Without -R, GNU troff stays in troff mode if the output device is not a terminal (e.g., ‘ps’).

Request: .nroff

Make the ‘n’ built-in condition true (and the ‘t’ built-in condition false) for if, ie, and while conditional requests. This is the default if GNU troff uses a terminal output device; the code for switching to nroff mode is in the file tty.tmac, which is loaded by the startup file troffrc.

See Conditionals and Loops, for more details on built-in conditions.


Next: , Previous: , Up: GNU troff Reference   [Contents][Index]

5.15 Line Layout

The following drawing shows the dimensions that gtroff uses for placing a line of output onto the page. They are labeled with the request that manipulates each dimension.

     -->| in |<--
        |<-----------ll------------>|
   +----+----+----------------------+----+
   |    :    :                      :    |
   +----+----+----------------------+----+
-->| po |<--
   |<--------paper width---------------->|

These dimensions are:

po

Page offset—this is the leftmost position of text on the final output, defining the left margin.

in

Indentation—this is the distance from the left margin where text is printed.

ll

Line length—this is the distance from the left margin to right margin.

The right margin is not explicitly configured; the combination of page offset and line length provides the information necessary to derive it.

A simple demonstration:

.ll 3i
This is text without indentation.
The line length has been set to 3\~inches.
.in +.5i
.ll -.5i
Now the left and right margins are both increased.
.in
.ll
Calling .in and .ll without parameters restores
the previous values.
    ⇒ This  is text without indenta-
    ⇒ tion.   The  line  length  has
    ⇒ been set to 3 inches.
    ⇒      Now   the  left  and
    ⇒      right  margins   are
    ⇒      both increased.
    ⇒ Calling  .in  and  .ll without
    ⇒ parameters restores the previ-
    ⇒ ous values.
Request: .po [offset]
Request: .po +offset
Request: .po -offset
Register: \n[.o]

Set page offset to offset (or increment or decrement its current value by offset). If invoked without an argument, the page offset is restored to the value before the previous po request. This request does not cause a break; the page offset in effect when an output line is broken prevails (see Manipulating Filling and Adjustment). The initial value is 1i and the default scaling unit is ‘m’. On terminal devices, the page offset is set to zero by a driver-specific macro file, tty.tmac. The current page offset can be found in the read-only register ‘.o’. This request is incorrectly documented in the AT&T troff manual as using a default scaling unit of ‘v’.

.po 3i
\n[.o]
    ⇒ 720
.po -1i
\n[.o]
    ⇒ 480
.po
\n[.o]
    ⇒ 720
Request: .in [indent]
Request: .in +indent
Request: .in -indent
Register: \n[.i]

Set indentation to indent (or increment or decrement the current value by indent). This request causes a break. Initially, there is no indentation.

If in is called without an argument, the indentation is reset to the previous value before the last call to in. The default scaling unit is ‘m’.

If a negative indentation value is specified (which is not allowed), gtroff emits a warning in category ‘range’ and sets the indentation to zero.

The effect of in is delayed until a partially collected line (if it exists) is output. A temporary indentation value is reset to zero also.

The current indentation (as set by in) can be found in the read-only register ‘.i’. The indentation is associated with the environment (see Environments).

Request: .ti offset
Request: .ti +offset
Request: .ti -offset
Register: \n[.in]

Temporarily indent the next output line by offset. If an increment or decrement value is specified, adjust the temporary indentation relative to the value set by the in request.

This request causes a break; its value is associated with the environment (see Environments). The default scaling unit is ‘m’. A call of ti without an argument is ignored.

If the total indentation value is negative (which is not allowed), gtroff emits a warning in category ‘range’ and sets the temporary indentation to zero. ‘Total indentation’ is either offset if specified as an absolute value, or the temporary plus normal indentation, if offset is given as a relative value.

The effect of ti is delayed until a partially collected line (if it exists) is output.

The read-only register .in is the indentation that applies to the current output line.

The difference between .i and .in is that the latter takes into account whether a partially collected line still uses the old indentation value or a temporary indentation value is active.

Request: .ll [length]
Request: .ll +length
Request: .ll -length
Register: \n[.l]
Register: \n[.ll]

Set the line length to length (or increment or decrement the current value by length). Initially, the line length is set to 6.5i. The effect of ll is delayed until a partially collected line (if it exists) is output. The default scaling unit is ‘m’.

If ll is called without an argument, the line length is reset to the previous value before the last call to ll. If a negative line length is specified (which is not allowed), gtroff emits a warning in category ‘range’ and sets the line length to zero. The line length is associated with the environment (see Environments).

The current line length (as set by ll) can be found in the read-only register ‘.l’. The read-only register .ll is the line length that applies to the current output line.

Similar to .i and .in, the difference between .l and .ll is that the latter takes into account whether a partially collected line still uses the old line length value.


Next: , Previous: , Up: GNU troff Reference   [Contents][Index]

5.16 Line Continuation

When filling is enabled, input and output line breaks generally do not correspond. The roff language therefore distinguishes input and output line continuation.

Escape sequence: \RET

\RET (a backslash immediately followed by a newline) suppresses the effects of that newline in the input. The next input line thus retains the classification of its predecessor as a control or text line. \RET is useful for managing line lengths in the input during document maintenance; you can break an input line in the middle of a request invocation, macro call, or escape sequence. Input line continuation is invisible to the formatter, with two exceptions: the | operator recognizes the new input line (see Numeric Expressions), and the input line counter register .c is incremented.

.ll 50n
.de I
.  ft I
.  nop \\$*
.  ft
..
Our film class watched
.I The Effect of Gamma Rays on Man-in-the-Moon
Marigolds. \" whoops, the input line wrapped
.br
.I My own opus begins on line \n[.c] \
and ends on line \n[.c].
    ⇒ Our film class watched The Effect of Gamma Rays onMan-in-the-Moon Marigolds.
    ⇒ My own opus begins on line 11 and ends on line 12.
Escape sequence: \c
Register: \n[.int]

\c continues an output line. Nothing after it on the input line is formatted. In contrast to \RET, a line after \c remains a new input line, so a control character is recognized at its beginning. The visual results depend on whether filling is enabled; see Manipulating Filling and Adjustment.

An intervening control line that causes a break overrides \c, flushing out the pending output line in the usual way.

The .int register contains a positive value if the last output line was continued with \c; this datum is associated with the environment (see Environments).69


Next: , Previous: , Up: GNU troff Reference   [Contents][Index]

5.17 Page Layout

The formatter permits configuration of the page length and page number.

Request: .pl [length]
Request: .pl +length
Request: .pl -length
Register: \n[.p]

Change (increase or decrease) the page length per the numeric expression length. The default scaling unit is ‘v’. A negative length is valid, but an uncommon application: it prevents page location traps from being sprung,70 and each output line is placed on a new page. If length is invalid, GNU troff emits a warning in category ‘number’. If length is absent or invalid, ‘11i’ is assumed.

The read-only register ‘.p’ interpolates the current page length.

Request: .pn num
Request: .pn +num
Request: .pn -num
Register: \n[.pn]

Change (increase or decrease) the page number of the next page per the numeric expression num. If num is invalid, GNU troff emits a warning in category ‘number’ and ignores the request. Without an argument, pn is ignored.

The read-only register .pn interpolates num if set by pn on the current page, or the current page number plus 1.

The formatter offers special support for typesetting headers and footers, collectively termed titles. Titles have an independent line length, and their placement on the page is not restricted.

Request: .tl 'left'center'right'

Format an output line as a title consisting of left, center, and right, each aligned accordingly. The delimiter need not be a neutral apostrophe: tl accepts the same delimiters as most escape sequences; see Delimiters. If not used as the delimiter, any page number character character is replaced with the current page number; the default is ‘%’; see the the pc request below. Without an argument, tl is ignored. tl writes the title line immediately, ignoring any partially collected line.

It is not an error to omit delimiters after the first. For example, ‘.tl /Thesis is interpreted as ‘.tl /Thesis///: it sets a title line comprising only the left-aligned word ‘Thesis’.

Request: .lt [length]
Request: .lt +length
Request: .lt -length
Register: \n[.lt]

Change (increase or decrease) the line length used by titles per the numeric expression length. The default scaling unit is ‘m’. If length is negative, GNU emits a warning in category ‘range’ and treats length as ‘0’. If length is invalid, GNU troff emits a warning in category ‘number’ and ignores the request. The formatter’s default title length is ‘6.5i’. With no argument, the title length is restored to the previous value. The title length is is associated with the environment (see Environments).

The read-only register ‘.lt’ interpolates the title line length.

Request: .pc [char]

Set the page number character to char. With no argument, the page number character is disabled. pc does not affect the register %.

The following example exercises title features.

.lt 50n
This is my partially collected
.tl 'Isomers 2023'%'Dextrose Edition'
line.
    ⇒ Isomers 2023             1        Dextrose Edition
    ⇒ This is my partially collected line.

We most often see titles used in page header and footer traps. See Traps.


Next: , Previous: , Up: GNU troff Reference   [Contents][Index]

5.18 Page Control

Discretionary page breaks can prevent the unwanted separation of content. A new page number takes effect during page ejection; see The Implicit Page Trap.

Request: .bp [page-number]
Request: .bp +page-number
Request: .bp -page-number
Register: \n[%]

Break the page and change (increase or decrease) the next page number per the numeric expression page-number. If page-number is invalid, GNU troff emits a warning in category ‘number’ and ignores the argument. This request causes a break. A page break advances the vertical drawing position to the bottom of the page, springing traps. See Page Location Traps. bp has effect only if invoked within the top-level diversion.71 This request is incorrectly documented in the AT&T troff manual as having a default scaling unit of ‘v’.

The register % interpolates the current page number.

.de BP
'  bp \" schedule page break once current line is output
..
Request: .ne [space]

Force a page break if insufficient vertical space is available (assert “needed” space). ne tests the distance to the next page location trap; see Page Location Traps, and breaks the page if that amount is less than space. The default scaling unit is ‘v’. If space is invalid, GNU troff emits a warning in category ‘number’ and ignores the argument. If space is not specified, ‘1v’ is assumed.

We can require space for at least the first two output lines of a paragraph, preventing its first line from being widowed at the page bottom.

.ne 2v
Considering how common illness is,
how tremendous the spiritual change that it brings,
how astonishing,
when the lights of health go down,
the undiscovered countries that are then disclosed,
what wastes and deserts of the soul a slight attack
of influenza brings to view,

This method is reliable only if no output line is pending when ne is invoked. When macro packages are used, this is often not the case: their paragraphing macros perform the break. You may need to experiment with placing the ne after the paragraphing macro, or br and ne before it.

ne is also useful to force grouping of section headings with their subsequent paragraphs, or tables with their captions and/or explanations. Macro packages often use ne with diversions to implement keeps and displays; see Diversions. They may also offer parameters for widow and orphan management.

Request: .sv [space]
Request: .os

Require vertical space as ne does, but also save it for later output by the os request. If space is available before the next page location trap, it is output immediately. Both requests ignore a partially collected line, taking effect at the next break. sv and os ignore no-space mode (recall Manipulating Spacing). While the sv request allows negative values for space, os ignores them. The default scaling unit is ‘v’. If space is not specified, ‘1v’ is assumed.

Register: \n[nl]

nl interpolates or sets the vertical drawing position. When the formatter starts, the first page transition hasn’t happened yet, and nl is negative. If a header trap has been planted on the page (typically at vertical position 0), you can assign a negative value to nl to spring it if that page has already started (see Page Location Traps).

.de HD
.  sp
.  tl ''Goldbach Solution''
.  sp
..
.
First page.
.bp
.wh 0 HD \" plant header trap at top of page
.nr nl (-1)
Second page.
    ⇒ First page.
    ⇒
    ⇒ (blank lines elided)
    ⇒
    ⇒                         Goldbach Solution
    ⇒
    ⇒ (blank lines elided)
    ⇒
    ⇒ Second page.

Without resetting nl to a negative value, the trap just planted would be active beginning with the next page, not the current one.

See Diversions, for a comparison of nl with the .h and .d registers.


Next: , Previous: , Up: GNU troff Reference   [Contents][Index]

5.19 Using Fonts

In digital typography, a font is a collection of characters in a specific typeface that a device can render as glyphs at a desired size.72 A roff formatter can change typefaces at any point in the text. The basic faces are a set of styles combining upright and slanted shapes with normal and heavy stroke weights: ‘R’, ‘I’, ‘B’, and ‘BI’—these stand for roman, italic, bold, and bold-italic. For linguistic text, GNU troff groups typefaces into families containing each of these styles.73 A text font is thus often a family combined with a style, but it need not be: consider the ps and pdf devices’ ZCMI (Zapf Chancery Medium italic)—often, no other style of Zapf Chancery Medium is provided. On typesetting devices, at least one special font is available, comprising unstyled glyphs for mathematical operators and other purposes.

Like AT&T troff, GNU troff does not itself load or manipulate a digital font file;74 instead it works with a font description file that characterizes it, including its glyph repertoire and the metrics (dimensions) of each glyph.75 This information permits the formatter to accurately place glyphs with respect to each other. Before using a font description, the formatter associates it with a mounting position, a place in an ordered list of available typefaces. So that a document need not be strongly coupled to a specific font family, in GNU troff an output device can associate a style in the abstract sense with a mounting position. Thus the default family can be combined with a style dynamically, producing a resolved font name.

Fonts often have trademarked names, and even Free Software fonts can require renaming upon modification. groff maintains a convention that a device’s serif font family is given the name ‘T’ (“Times”), its sans-serif family ‘H’ (“Helvetica”), and its monospaced family ‘C’ (“Courier”). Historical inertia has driven groff’s font identifiers to short uppercase abbreviations of font names, as with ‘TR’, ‘TI’, ‘TB’, ‘TBI’, and a special font ‘S’.

The default family used with abstract styles can be changed at any time; initially, it is ‘T’. Typically, abstract styles are arranged in the first four mounting positions in the order shown above. The default mounting position, and therefore style, is always ‘1’ (‘R’). By issuing appropriate formatter instructions, you can override these defaults before your document writes its first glyph.

Terminal output devices cannot change font families and lack special fonts. They support style changes by overstriking, or by altering ISO 6429/ECMA-48 graphic renditions (character cell attributes).


Next: , Previous: , Up: Using Fonts   [Contents][Index]

5.19.1 Selecting Fonts

We use font to refer to any of several means of identifying a font: by mounting position (‘3’), by abstract style (‘B’), or by its identifier (‘TB’).

Request: .ft [font]
Escape sequence: \ff
Escape sequence: \f(fn
Escape sequence: \f[font]
Register: \n[.fn]

The ft request selects the typeface font. If the argument is absent or ‘P’, it selects the previously chosen font. If font is a non-negative integer, it is interpreted as mounting position; the font mounted there is selected. If that position refers to an abstract style, it is combined with the default family (see fam and \F below) to make a resolved font name. If the mounting position is not a style and no font is mounted there, GNU troff emits a warning in category ‘font’ and ignores the request.

If font matches a style name, it is combined with the current family to make a resolved font name. Otherwise, font is assumed to already be a resolved font name.

The resolved font name is subject to translation (see request ftr below). Next, the (possibly translated) font name’s mounting position is looked up; if not mounted, font is sought on the file system as a font description file and, if located, automatically mounted at the next available position (see register .fp below). If the font was mounted using an identifier different from its font description file name (see request fp below), that file name is then looked up. If a font description file for the resolved font name is not found, GNU troff emits a warning in category ‘font’ and ignores the request.

The \f escape sequence is similar, using one-character name (or mounting position) f, two-character name fn, or a name font of arbitrary length. \f[]’ selects the previous font. The syntax form ‘\fP’ is supported for backward compatibility, and ‘\f[P]’ for consistency.

eggs, bacon,
.ft I
spam,
.ft
and sausage.
.br
eggs, bacon, \fIspam,\fP and sausage.
    ⇒ eggs, bacon, spam, and sausage
    ⇒ eggs, bacon, spam, and sausage

The current and previously selected fonts are properties of the environment (see Environments).

The read-only string-valued register .fn contains the resolved font name of the selected font.

\f doesn’t produce an input token in GNU troff; it thus can be used in requests that expect a single-character argument. We can assign a font to a margin character as follows (see Miscellaneous).

.mc \f[I]x\f[]
Request: .ftr f [g]

Translate font f to font g. Whenever a font named f is referred to in a \f escape sequence, in the F and S conditional operators, or in the ft, ul, bd, cs, tkf, special, fspecial, fp, or sty requests, font g is used. If g is missing or equal to f the translation is undone.

Font translations cannot be chained.

.ftr XXX TR
.ftr XXX YYY
.ft XXX
    error→ warning: can't find font 'XXX'
Request: .fzoom f [zoom]
Register: \n[.zoom]

Set magnification of font f to factor zoom, which must be a non-negative integer multiple of 1/1000th. This request is useful to adjust the optical size of a font in relation to the others. In the example below, font CR is magnified by 10% (the zoom factor is thus 1.1).

.fam P
.fzoom CR 1100
.ps 12
Palatino and \f[CR]Courier\f[]

A missing or zero value of zoom is the same as a value of 1000, which means no magnification. f must be a resolved font name, not an abstract style.

The magnification of a font is completely transparent to GNU troff; a change of the zoom factor doesn’t cause any effect except that the dimensions of glyphs, (word) spaces, kerns, etc., of the affected font are adjusted accordingly.

The zoom factor of the current font is available in the read-only register ‘.zoom’, in multiples of 1/1000th. It returns zero if there is no magnification.


Next: , Previous: , Up: Using Fonts   [Contents][Index]

5.19.2 Font Families

To accommodate the wide variety of fonts available, GNU troff distinguishes font families and font styles. A resolved font name is the catenation of a font family and a style. Selecting an abstract style causes GNU troff to combine it with the default font family.

You can thus compose a document using abstract styles exclusively for its body or running text, selecting a specific family only for titles or examples, for instance, and change the default family on the command line (recall Groff Options).

Fonts for the devices ps, pdf, dvi, lj4, lbp, and the X11 devices support this mechanism. By default, GNU troff uses the Times family with the four styles ‘R’, ‘I’, ‘B’, and ‘BI’.

Request: .fam [family]
Register: \n[.fam]
Escape sequence: \Ff
Escape sequence: \F(fm
Escape sequence: \F[family]

Set the default font family, used in combination with abstract styles to construct a resolved font name, to family (one-character name f, two-character name fm). If no argument is given, GNU troff selects the previous font family; if there none, is it falls back to the device’s default76 or its own (‘T’).

The \F escape sequence works similarly. In disanalogy to \f, ‘\FP’ makes ‘P’ the default family. Use ‘\F[]’ to select the previous default family. The default font family is available in the read-only string-valued register .fam; it is associated with the environment (see Environments).

spam,     \" startup defaults are T (Times) R (roman)
.fam H    \" make Helvetica the default family
spam,     \" family H + style R = HR
.ft B     \" family H + style B = HB
spam,
.ft CR    \" Courier roman (default family not changed)
spam,
.ft       \" back to Helvetica bold
spam,
.fam T    \" make Times the default family
spam,     \" family T + style B = TB
.ft AR    \" font AR (not a style)
baked beans,
.ft R     \" family T + style R = TR
and spam.

\F doesn’t produce an input token in GNU troff. As a consequence, it can be used in requests like mc (which expects a single character as an argument) to change the font family on the fly.

.mc \F[P]x\F[]
Request: .sty n style
Register: \n[.sty]

Associate an abstract style style with mounting position n, which must be a non-negative integer. If the requests cs, bd, tkf, uf, or fspecial are applied to an abstract style, they are instead applied to the member of the current family corresponding to that style.

The default family can be set with the -f option (see Groff Options). The styles command in the DESC file controls which font positions (if any) are initially associated with abstract styles rather than fonts.

Caution: The style argument is not validated. Errors may occur later, when the formatter attempts to construct a resolved font name, or format a character for output.

.nr BarPos \n[.fp]
.sty \n[.fp] Bar
.fam Foo
.ft \n[BarPos]
.tm .f=\n[.f]
A
    error→ error: no font family named 'Foo' exists
    error→ .f=41
    error→ error: cannot format glyph: no current font

When an abstract style has been selected, the read-only string-valued register ‘.sty’ interpolates its name; this datum is associated with the environment (see Environments). Otherwise, ‘.sty’ interpolates nothing.


Next: , Previous: , Up: Using Fonts   [Contents][Index]

5.19.3 Font Positions

To support typeface indirection through abstract styles, and for compatibility with AT&T troff, the formatter maintains a list of font positions at which fonts required by a document are mounted. An output device’s description file DESC typically configures a set of pre-mounted fonts; see Device and Font Description Files. A font need not be explicitly mounted before it is selected; GNU troff will search GROFF_FONT_PATH for it by name and mount it at the first free mounting position on demand.

Request: .fp pos id [font-description-file-name]
Register: \n[.f]
Register: \n[.fp]

Mount a font under the name id at mounting position pos, a non-negative integer. When the formatter starts up, it reads the output device’s description to mount an initial set of faces, and selects font position 1. Position 0 is unused by default. Unless the font-description-file-name argument is given, id should be the name of a font description file stored in a directory corresponding to the selected output device. GNU troff does not traverse directories to locate the font description file.

The optional third argument enables font names to be aliased, which can be necessary in compatibility mode since AT&T troff syntax affords no means of identifying fonts with names longer than two characters, like ‘TBI’ or ‘ZCMI’, in a font selection escape sequence. See Compatibility Mode. You can also alias fonts on mounting for convenience or abstraction. (See below regarding the .fp register.)

.fp \n[.fp] SC ZCMI
Send a \f(SChand-written\fP thank-you note.
.fp \n[.fp] Emph TI
.fp \n[.fp] Strong TB
Are \f[Emph]these names\f[] \f[Strong]comfortable\f[]?

DESC’, ‘P’, and non-negative integers are not usable as font identifiers.

The position of the currently selected font (or abstract style) is available in the read-only register ‘.f’. It is associated with the environment (see Environments).

You can copy the value of .f to another register to save it for later use.

.nr saved-font \n[.f]
text involving many font changes
.ft \n[saved-font]

The index of the next (non-zero) free font position is available in the read-only register ‘.fp’. Fonts not listed in the DESC file are automatically mounted at position ‘\n[.fp]’ when selected with the ft request or \f escape sequence. When mounting a font at a position explicitly with the fp request, this same practice should be followed, although GNU troff does not enforce this strictly.


Next: , Previous: , Up: Using Fonts   [Contents][Index]

5.19.4 Using Symbols

A glyph is a graphical representation of a character. While a character is an abstraction of semantic information, a glyph is something that can be seen on screen or paper. A character has many possible representation forms (for example, the character ‘A’ can be written in an upright or slanted typeface, producing distinct glyphs). Sometimes, a sequence of characters map to a single glyph: this is a ligature—the most common is ‘fi’.

Space characters never become glyphs in GNU troff. If not discarded (as when trailing on text lines), they are represented by horizontal motions in the output.

A symbol is simply a named glyph. Within gtroff, all glyph names of a particular font are defined in its font file. If the user requests a glyph not available in this font, gtroff looks up an ordered list of special fonts. By default, the PostScript output device supports the two special fonts ‘SS’ (slanted symbols) and ‘S’ (symbols) (the former is looked up before the latter). Other output devices use different names for special fonts. Fonts mounted with the fonts keyword in the DESC file are globally available. To install additional special fonts locally (i.e., for a particular font), use the fspecial request.

Here are the exact rules how gtroff searches a given symbol:

See Device and Font Description Files, and Special Fonts, for more details.

The groff_char(7) man page houses a complete list of predefined special character names, but the availability of any as a glyph is device- and font-dependent. For example, say

man -Tdvi groff_char > groff_char.dvi

to obtain those available with the DVI device and default font configuration.77 If you want to use an additional macro package to change the fonts used, groff (or gtroff) must be run directly.

groff -Tdvi -mec -man groff_char.7 > groff_char.dvi

Special character names not listed in groff_char(7) are derived algorithmically, using a simplified version of the Adobe Glyph List (AGL) algorithm, which is described in https://github.com/adobe-type-tools/agl-aglfn. The (frozen) set of names that can’t be derived algorithmically is called the groff glyph list (GGL).

Escape sequence: \(nm
Escape sequence: \[name]
Escape sequence: \[base-glyph combining-component …]

Typeset a special character name (two-character name nm) or a composite glyph consisting of base-glyph overlaid with one or more combining-components. For example, ‘\[A ho]’ is a capital letter “A” with a “hook accent” (ogonek).

There is no special syntax for one-character names—the analogous form ‘\n’ would collide with other escape sequences. However, the four escape sequences \', \-, \_, and \`, are translated on input to the special character escape sequences \[aa], \[-], \[ul], and \[ga], respectively.

A special character name of length one is not the same thing as an ordinary character: that is, the character a is not the same as \[a].

If name is undefined, a warning in category ‘char’ is produced and the escape is ignored. See Warnings, for information about the enablement and suppression of warnings.

GNU troff resolves \[] with more than a single component as follows:

No check for the existence of any component (similar to tr request) is done.

Examples:

\[A ho]

A’ maps to u0041, ‘ho’ maps to u02DB, thus the final glyph name would be u0041_02DB. This is not the expected result: the ogonek glyph ‘ho’ is a spacing ogonek, but for a proper composite a non-spacing ogonek (U+0328) is necessary. Looking into the file composite.tmac, one can find ‘.composite ho u0328, which changes the mapping of ‘ho’ while a composite glyph name is constructed, causing the final glyph name to be u0041_0328.

\[^E u0301]
\[^E aa]
\[E a^ aa]
\[E ^ ']

^E’ maps to u0045_0302, thus the final glyph name is u0045_0302_0301 in all forms (assuming proper calls of the composite request).

It is not possible to define glyphs with names like ‘A ho within a groff font file. This is not really a limitation; instead, you have to define u0041_0328.

Escape sequence: \C'xxx'

Typeset the glyph of the special character xxx. Normally, it is more convenient to use \[xxx], but \C has some advantages: it is compatible with AT&T device-independent troff (and therefore available in compatibility mode79) and can interpolate special characters with ‘]’ in their names. The delimiter need not be a neutral apostrophe; see Delimiters.

Request: .composite id1 id2

Map special character name id1 to id2 if id1 is used in \[...] with more than one component. See above for examples. This is a strict rewriting of the special character name; no check is performed for the existence of a glyph for either. A set of default mappings for many accents can be found in the file composite.tmac, loaded by the default troffrc at startup.

Escape sequence: \N'n'

Typeset the glyph with code n in the current font (n is not the input character code). The number n can be any non-negative decimal integer. Most devices only have glyphs with codes between 0 and 255; the Unicode output device uses codes in the range 0–65535. If the current font does not contain a glyph with that code, special fonts are not searched. The \N escape sequence can be conveniently used in conjunction with the char request:

.char \[phone] \f[ZD]\N'37'

The code of each glyph is given in the fourth column in the font description file after the charset command. It is possible to include unnamed glyphs in the font description file by using a name of ‘---’; the \N escape sequence is the only way to use these.

No kerning is applied to glyphs accessed with \N. The delimiter need not be a neutral apostrophe; see Delimiters.

A few escape sequences are also special characters.

Escape sequence: \'

An escaped neutral apostrophe is a synonym for \[aa] (acute accent).

Escape sequence: \`

An escaped grave accent is a synonym for \[ga] (grave accent).

Escape sequence: \-

An escaped hyphen-minus is a synonym for \[-] (minus sign).

Escape sequence: \_

An escaped underscore (“low line”) is a synonym for \[ul] (underrule). On typesetting devices, the underrule is font-invariant and drawn lower than the underscore ‘_’.

Request: .cflags n c1 c2 …

Assign properties encoded by the number n to characters c1, c2, and so on.

Input characters, including special characters introduced by an escape, have certain properties associated with them.80 These properties can be modified with this request. The first argument is the sum of the desired flags and the remaining arguments are the characters to be assigned those properties. Spaces between the cn arguments are optional. Any argument cn can be a character class defined with the class request rather than an individual character. See Character Classes.

The non-negative integer n is the sum of any of the following. Some combinations are nonsensical, such as ‘33’ (1 + 32).

1

Recognize the character as ending a sentence if followed by a newline or two spaces. Initially, characters ‘.?!’ have this property.

2

Enable breaks before the character. A line is not broken at a character with this property unless the characters on each side both have non-zero hyphenation codes. This exception can be overridden by adding 64. Initially, no characters have this property.

4

Enable breaks after the character. A line is not broken at a character with this property unless the characters on each side both have non-zero hyphenation codes. This exception can be overridden by adding 64. Initially, characters ‘\-\[hy]\[em]’ have this property.

8

Mark the glyph associated with this character as overlapping other instances of itself horizontally. Initially, characters ‘\[ul]\[rn]\[ru]\[radicalex]\[sqrtex]’ have this property.

16

Mark the glyph associated with this character as overlapping other instances of itself vertically. Initially, the character ‘\[br]’ has this property.

32

Mark the character as transparent for the purpose of end-of-sentence recognition. In other words, an end-of-sentence character followed by any number of characters with this property is treated as the end of a sentence if followed by a newline or two spaces. This is the same as having a zero space factor in TeX. Initially, characters ‘"')]*\[dg]\[dd]\[rq]\[cq]’ have this property.

64

Ignore hyphenation codes of the surrounding characters. Use this in combination with values 2 and 4 (initially, no characters have this property).

For example, if you need an automatic break point after the en-dash in numeric ranges like “3000–5000”, insert

.cflags 68 \[en]

into your document. However, this practice can lead to bad layout if done thoughtlessly; in most situations, a better solution instead of changing the cflags value is to insert \: right after the hyphen at the places that really need a break point.

The remaining values were implemented for East Asian language support; those who use alphabetic scripts exclusively can disregard them.

128

Prohibit a line break before the character, but allow a line break after the character. This works only in combination with flags 256 and 512 and has no effect otherwise. Initially, no characters have this property.

256

Prohibit a line break after the character, but allow a line break before the character. This works only in combination with flags 128 and 512 and has no effect otherwise. Initially, no characters have this property.

512

Allow line break before or after the character. This works only in combination with flags 128 and 256 and has no effect otherwise. Initially, no characters have this property.

In contrast to values 2 and 4, the values 128, 256, and 512 work pairwise. If, for example, the left character has value 512, and the right character 128, no break will be automatically inserted between them. If we use value 6 instead for the left character, a break after the character can’t be suppressed since the neighboring character on the right doesn’t get examined.

Request: .char c [contents]
Request: .fchar c [contents]
Request: .fschar f c [contents]
Request: .schar c [contents]

Define a new character or glyph c to be contents, which can be empty. More precisely, char defines a groff object (or redefines an existing one) that is accessed with the name c on input, and produces contents on output. Every time glyph c needs to be printed, contents is processed in a temporary environment and the result is wrapped up into a single object. Compatibility mode is turned off and the escape character is set to \ while contents is processed. Any emboldening, constant spacing, or track kerning is applied to this object rather than to individual glyphs in contents.

An object defined by these requests can be used just like a normal glyph provided by the output device. In particular, other characters can be translated to it with the tr or trin requests; it can be made the leader character with the lc request; repeated patterns can be drawn with it using the \l and \L escape sequences; and words containing c can be hyphenated correctly if the hcode request is used to give the object a hyphenation code.

There is a special anti-recursion feature: use of the object within its own definition is handled like a normal character (not defined with char).

The tr and trin requests take precedence if char accesses the same symbol.

.tr XY
X
    ⇒ Y
.char X Z
X
    ⇒ Y
.tr XX
X
    ⇒ Z

The fchar request defines a fallback glyph: gtroff only checks for glyphs defined with fchar if it cannot find the glyph in the current font. gtroff carries out this test before checking special fonts.

fschar defines a fallback glyph for font f: gtroff checks for glyphs defined with fschar after the list of fonts declared as font-specific special fonts with the fspecial request, but before the list of fonts declared as global special fonts with the special request.

Finally, the schar request defines a global fallback glyph: gtroff checks for glyphs defined with schar after the list of fonts declared as global special fonts with the special request, but before the already mounted special fonts.

See Character Classes.

Request: .rchar c …
Request: .rfschar f c …

Remove definition of each ordinary or special character c, undoing the effect of a char, fchar, or schar request. Those supplied by font description files cannot be removed. Spaces and tabs may separate c arguments.

The request rfschar removes glyph definitions defined with fschar for font f.


Next: , Previous: , Up: Using Fonts   [Contents][Index]

5.19.5 Character Classes

Classes are particularly useful for East Asian languages such as Chinese, Japanese, and Korean, where the number of needed characters is much larger than in European languages, and where large sets of characters share the same properties.

Request: .class name c1 c2 …

Define a character class (or simply “class”) name comprising the characters c1, c2, and so on.

A class thus defined can then be referred to in lieu of listing all the characters within it. Currently, only the cflags request can handle references to character classes.

In the request’s simplest form, each cn is a character (or special character).

.class [quotes] ' \[aq] \[dq] \[oq] \[cq] \[lq] \[rq]

Since class and glyph names share the same name space, it is recommended to start and end the class name with [ and ], respectively, to avoid collisions with existing character names defined by GNU troff or the user (with char and related requests). This practice applies the presence of ] in the class name to prevent the use of the special character escape form \[], thus you must use the \C escape to access a class with such a name.

You can also use a character range notation consisting of a start character followed by ‘-’ and then an end character. Internally, GNU troff converts these two symbol names to Unicode code points (according to the groff glyph list [GGL]), which then give the start and end value of the range. If that fails, the class definition is skipped.

Furthermore, classes can be nested.

.class [prepunct] , : ; > }
.class [prepunctx] \C'[prepunct]' \[u2013]-\[u2016]

The class ‘[prepunctx]’ thus contains the contents of the class [prepunct] as defined above (the set ‘, : ; > }’), and characters in the range between U+2013 and U+2016.

If you want to include ‘-’ in a class, it must be the first character value in the argument list, otherwise it gets misinterpreted as part of the range syntax.

It is not possible to use class names as end points of range definitions.

A typical use of the class request is to control line-breaking and hyphenation rules as defined by the cflags request. For example, to inhibit line breaks before the characters belonging to the prepunctx class defined in the previous example, you can write the following.

.cflags 2 \C'[prepunctx]'

See the cflags request in Using Symbols, for more details.


Next: , Previous: , Up: Using Fonts   [Contents][Index]

5.19.6 Special Fonts

Special fonts are those that gtroff searches when it cannot find the requested glyph in the current font. The Symbol font is usually a special font.

gtroff provides the following two requests to add more special fonts. See Using Symbols, for a detailed description of the glyph searching mechanism in gtroff.

Usually, only non-TTY devices have special fonts.

Request: .special [s1 s2 …]
Request: .fspecial f [s1 s2 …]

Use the special request to define special fonts. Initially, this list is empty.

Use the fspecial request to designate special fonts only when font f is active. Initially, this list is empty.

Previous calls to special or fspecial are overwritten; without arguments, the particular list of special fonts is set to empty. Special fonts are searched in the order they appear as arguments.

All fonts that appear in a call to special or fspecial are loaded.

See Using Symbols, for the exact search order of glyphs.


Next: , Previous: , Up: Using Fonts   [Contents][Index]

5.19.7 Artificial Fonts

There are a number of requests and escape sequences for artificially creating fonts. These are largely vestiges of the days when output devices did not have a wide variety of fonts, and when nroff and troff were separate programs. Most of them are no longer necessary in GNU troff. Nevertheless, they are supported.

Escape sequence: \H'height'
Escape sequence: \H'+height'
Escape sequence: \H'-height'
Register: \n[.height]

Change (increment, decrement) the height of the current font, but not the width. If height is zero, restore the original height. Default scaling unit is ‘z’.

The read-only register .height contains the font height as set by \H.

Currently, only the -Tps and -Tpdf devices support this feature.

\H doesn’t produce an input token in GNU troff. As a consequence, it can be used in requests like mc (which expects a single character as an argument) to change the font on the fly:

.mc \H'+5z'x\H'0'

In compatibility mode, gtroff behaves differently: If an increment or decrement is used, it is always taken relative to the current type size and not relative to the previously selected font height. Thus,

.cp 1
\H'+5'test \H'+5'test

prints the word ‘test’ twice with the same font height (five points larger than the current font size).

Escape sequence: \S'slant'
Register: \n[.slant]

Slant the current font by slant degrees. Positive values slant to the right. Only integer values are possible.

The read-only register .slant contains the font slant as set by \S.

Currently, only the -Tps and -Tpdf devices support this feature.

\S doesn’t produce an input token in GNU troff. As a consequence, it can be used in requests like mc (which expects a single character as an argument) to change the font on the fly:

.mc \S'20'x\S'0'

This escape is incorrectly documented in the AT&T troff manual; the slant is always set to an absolute value.

Request: .ul [lines]

The ul request normally underlines subsequent lines if a TTY output device is used. Otherwise, the lines are printed in italics (only the term ‘underlined’ is used in the following). The single argument is the quantity of input lines to be underlined; with no argument, the next line is underlined. If lines is zero or negative, stop the effects of ul (if it was active). Requests and empty lines do not count for computing the number of underlined input lines, even if they produce some output like tl. Lines inserted by macros (e.g., invoked by a trap) do count.

At the beginning of ul, the current font is stored and the underline font is activated. Within the span of a ul request, it is possible to change fonts, but after the last line affected by ul the saved font is restored.

This number of lines still to be underlined is associated with the environment (see Environments). The underline font can be changed with the uf request.

The ul request does not underline spaces.

Request: .cu [lines]

The cu request is similar to ul but underlines spaces as well (if a TTY output device is used).

Request: .uf font

Set the underline font (globally) used by ul and cu. By default, this is the font at position 2. font can be either a non-negative font position or the name of a font.

Request: .bd font [offset]
Request: .bd font1 font2 [offset]
Register: \n[.b]

Embolden font by overstriking its glyphs offset by offset units minus one.

Two syntax forms are available.

Request: .cs font [width [em-size]]

Switch to and from constant glyph space mode. If activated, the width of every glyph is width/36 ems. The em size is given absolutely by em-size; if this argument is missing, the em value is taken from the current font size (as set with the ps request) when the font is effectively in use. Without second and third argument, constant glyph space mode is deactivated.

Default scaling unit for em-size is ‘z’; width is an integer.


Next: , Previous: , Up: Using Fonts   [Contents][Index]

5.19.8 Ligatures and Kerning

Ligatures are groups of characters that are run together, i.e, producing a single glyph. For example, the letters ‘f’ and ‘i’ can form a ligature ‘fi’ as in the word ‘file’. This produces a cleaner look (albeit subtle) to the printed output. Usually, ligatures are not available in fonts for TTY output devices.

Most PostScript fonts support the fi and fl ligatures. The C/A/T typesetter that was the target of AT&T troff also supported ‘ff’, ‘ffi’, and ‘ffl’ ligatures. Advanced typesetters or ‘expert’ fonts may include ligatures for ‘ft’ and ‘ct’, although GNU troff does not support these (yet).

Only the current font is checked for ligatures and kerns; neither special fonts nor special charcters defined with the char request (and its siblings) are taken into account.

Request: .lg [flag]
Register: \n[.lg]

Switch the ligature mechanism on or off; if the parameter is non-zero or missing, ligatures are enabled, otherwise disabled. Default is on. The current ligature mode can be found in the read-only register .lg (set to 1 or 2 if ligatures are enabled, 0 otherwise).

Setting the ligature mode to 2 enables the two-character ligatures (fi, fl, and ff) and disables the three-character ligatures (ffi and ffl).

Pairwise kerning is another subtle typesetting mechanism that modifies the distance between a glyph pair to improve readability. In most cases (but not always) the distance is decreased. Typewriter-like fonts and fonts for terminals where all glyphs have the same width don’t use kerning.

Request: .kern [flag]
Register: \n[.kern]

Switch kerning on or off. If the parameter is non-zero or missing, enable pairwise kerning, otherwise disable it. The read-only register .kern is set to 1 if pairwise kerning is enabled, 0 otherwise.

If the font description file contains pairwise kerning information, glyphs from that font are kerned. Kerning between two glyphs can be inhibited by placing \& between them: ‘V\&A’.

See Font Description File Format.

Track kerning expands or reduces the space between glyphs. This can be handy, for example, if you need to squeeze a long word onto a single line or spread some text to fill a narrow column. It must be used with great care since it is usually considered bad typography if the reader notices the effect.

Request: .tkf f s1 n1 s2 n2

Enable track kerning for font f. If the current font is f the width of every glyph is increased by an amount between n1 and n2 (n1, n2 can be negative); if the current type size is less than or equal to s1 the width is increased by n1; if it is greater than or equal to s2 the width is increased by n2; if the type size is greater than or equal to s1 and less than or equal to s2 the increase in width is a linear function of the type size.

The default scaling unit is ‘z’ for s1 and s2, ‘p’ for n1 and n2.

The track kerning amount is added even to the rightmost glyph in a line; for large values it is thus recommended to increase the line length by the same amount to compensate.


Next: , Previous: , Up: Using Fonts   [Contents][Index]

5.19.9 Italic Corrections

When typesetting adjacent glyphs from typefaces of different slants, the space between them may require adjustment.

Escape sequence: \/

Apply an italic correction: modify the spacing of the preceding glyph so that the distance between it and the following glyph is correct if the latter is of upright shape. For example, if an italic ‘f’ is followed immediately by a roman right parenthesis, then in many fonts the top right portion of the ‘f’ overlaps the top left of the right parenthesis, which is ugly. Use this escape sequence whenever an oblique glyph is immediately followed by an upright glyph without any intervening space.

Escape sequence: \,

Apply a left italic correction: modify the spacing of the following glyph so that the distance between it and the preceding glyph is correct if the latter is of upright shape. For example, if a roman left parenthesis is immediately followed by an italic ‘f’, then in many fonts the bottom left portion of the ‘f’ overlaps the bottom of the left parenthesis, which is ugly. Use this escape sequence whenever an upright glyph is followed immediately by an oblique glyph without any intervening space.


Previous: , Up: Using Fonts   [Contents][Index]

5.19.10 Dummy Characters

As discussed in Requests and Macros, the first character on an input line is treated specially. Further, formatting a glyph has many consequences on formatter state (see Environments). Occasionally, we want to escape this context or embrace some of those consequences without actually rendering a glyph to the output.

Escape sequence: \&

Interpolate a dummy character, which is constitutive of output but invisible.81 Its presence alters the interpretation context of a subsequent input character, and enjoys several applications.

The dummy character escape sequence sees use in macro definitions as a means of ensuring that arguments are treated as text even if they begin with spaces or control characters.

.de HD \" typeset a simple bold heading
.  sp
.  ft B
\&\\$1 \" exercise: remove the \&
.  ft
.  sp
..
.HD .\|.\|.\|surprised?

One way to think about the dummy character is to imagine placing the symbol ‘&’ in the input at a certain location; if doing so has all the side effects on formatting that you desire except for sticking an ugly ampersand in the midst of your text, the dummy character is what you want in its place.

Escape sequence: \)

Interpolate a transparent dummy character—one that is transparent to end-of-sentence detection. It behaves as \&, except that \& is treated as letters and numerals normally are after ‘.’, ‘?’ and ‘!’; \& cancels end-of-sentence detection, and \) does not.

.de Suffix-&
.  nop \&\\$1
..
.
.de Suffix-)
.  nop \)\\$1
..
.
Here's a sentence.\c
.Suffix-& '
Another one.\c
.Suffix-) '
And a third.
    ⇒ Here's a sentence.' Another one.'  And a third.

Next: , Previous: , Up: GNU troff Reference   [Contents][Index]

5.20 Manipulating Type Size and Vertical Spacing

These concepts were introduced in Page Geometry. The height of a font’s tallest glyph is one em, which is equal to the type size in points.82 A vertical spacing of less than 120% of the type size can make a document hard to read. Larger proportions can be useful to spread the text for annotations or proofreader’s marks. By default, GNU troff uses 10 point type on 12 point spacing. Typographers call the difference between type size and vertical spacing leading.83


Next: , Previous: , Up: Manipulating Type Size and Vertical Spacing   [Contents][Index]

5.20.1 Changing the Type Size

Request: .ps [size]
Request: .ps +size
Request: .ps -size
Escape sequence: \ssize
Register: \n[.s]

Use the ps request or the \s escape sequence to change (increase, decrease) the type size (in scaled points). Specify size as either an absolute type size, or as a relative change from the current size. ps with no argument restores the previous size. The ps request’s default scaling unit is ‘z’. The requested size is rounded to the nearest valid size (with ties rounding down) within the limits supported by the device. If the requested size is non-positive, it is treated as 1u.

Type size alteration is incorrectly documented in the AT&T troff manual, which claims “if [the requested size] is invalid, the next larger valid size will result, with a maximum of 36”.84

The read-only string-valued register .s interpolates the type size in points as a decimal fraction; it is associated with the environment (see Environments). To obtain the type size in scaled points, interpolate the .ps register instead (see Using Fractional Type Sizes).

The \s escape sequence supports a variety of syntax forms.

\sn

Set the type size to n points. n must be a single digit. If n is 0, restore the previous size.

\s+n
\s-n

Increase or decrease the type size by n points. n must be exactly one digit.

\s(nn

Set the type size to nn points. nn must be exactly two digits.

\s+(nn
\s-(nn
\s(+nn
\s(-nn

Alter the type size in points by the two-digit value nn.

See Using Fractional Type Sizes, for further syntactical forms of the \s escape sequence that additionally accept decimal fractions.

snap, snap,
.ps +2
grin, grin,
.ps +2
wink, wink, \s+2nudge, nudge,\s+8 say no more!
.ps 10

The \s escape sequence affects the environment immediately and doesn’t produce an input token. Consequently, it can be used in requests like mc, which expects a single character as an argument, to change the type size on the fly.

.mc \s[20]x\s[0]
Request: .sizes s1 s2 … sn [0]

The DESC file specifies which type sizes are allowed by the output device; see DESC File Format. Use the sizes request to change this set of permissible sizes. Arguments are in scaled points; see Using Fractional Type Sizes. Each can be a single type size (such as ‘12000’), or a range of sizes (such as ‘4000-72000’). You can optionally end the list with a ‘0’.


Next: , Previous: , Up: Manipulating Type Size and Vertical Spacing   [Contents][Index]

5.20.2 Changing the Vertical Spacing

Request: .vs [space]
Request: .vs +space
Request: .vs -space
Register: \n[.v]

Set the vertical spacing to, or alter it by, space. The default scaling unit is ‘p’. If vs is called without an argument, the vertical spacing is reset to the previous value before the last call to vs. GNU troff emits a warning in category ‘range’ if space is negative; the vertical spacing is then set to the smallest possible positive value, the vertical motion quantum (as found in the .V register).

.vs 0 isn’t saved in a diversion since it doesn’t result in a vertical motion. You must explicitly issue this request before interpolating the diversion.

The read-only register .v contains the vertical spacing; it is associated with the environment (see Environments).

When a break occurs, GNU troff performs the following procedure.

Prefer vs or pvs over ls to produce double-spaced documents. vs and pvs have finer granularity than ls; moreover, some preprocessors assume single spacing. See Manipulating Spacing, regarding the \x escape sequence and the ls request.

Request: .pvs [space]
Request: .pvs +space
Request: .pvs -space
Register: \n[.pvs]

Set the post-vertical spacing to, or alter it by, space. The default scaling unit is ‘p’. If pvs is called without an argument, the post-vertical spacing is reset to the previous value before the last call to pvs. GNU troff emits a warning in category ‘range’ if space is negative; the post-vertical spacing is then set to zero.

The read-only register .pvs contains the post-vertical spacing; it is associated with the environment (see Environments).


Previous: , Up: Manipulating Type Size and Vertical Spacing   [Contents][Index]

5.20.3 Using Fractional Type Sizes

AT&T troff interpreted all type size measurements in points. Combined with integer arithmetic, this design choice made it impossible to support, for instance, ten and a half-point type. In GNU troff, an output device can select a scaling factor that subdivides a point into “scaled points”. A type size expressed in scaled points can thus represent a non-integral type size.

A scaled point is equal to 1/sizescale points, where sizescale is specified in the device description file DESC, and defaults to 1.85 Requests and escape sequences in GNU troff interpret arguments that represent a type size in scaled points, which the formatter multiplies by sizescale and converts to an integer. Arguments treated in this way comprise those to the escape sequences \H and \s, to the request ps, the third argument to the cs request, and the second and fourth arguments to the tkf request. Scaled points may be specified explicitly with the z scaling unit.

For example, if sizescale is 1000, then a scaled point is one thousandth of a point. The request ‘.ps 10.5’ is synonymous with ‘.ps 10.5z’ and sets the type size to 10,500 scaled points, or 10.5 points. Consequently, in GNU troff, the register .s can interpolate a non-integral type size.

Register: \n[.ps]

This read-only register interpolates the type size in scaled points; it is associated with the environment (see Environments).

It makes no sense to use the ‘z’ scaling unit in a numeric expression whose default scaling unit is neither ‘u’ nor ‘z’, so GNU troff disallows this. Similarly, it is nonsensical to use a scaling unit other than ‘z’ or ‘u’ in a numeric expression whose default scaling unit is ‘z’, and so GNU troff disallows this as well.

Another GNU troff scaling unit, ‘s’, multiplies by the number of basic units in a scaled point. Thus, ‘\n[.ps]s’ is equal to ‘1m’ by definition. Do not confuse the ‘s’ and ‘z’ scaling units.

Register: \n[.psr]
Register: \n[.sr]

Output devices may be limited in the type sizes they can employ. The .s and .ps registers represent the type size selected by the output driver as it understands a device’s capability. The last requested type size is interpolated in scaled points by the read-only register .psr and in points as a decimal fraction by the read-only string-valued register .sr. Both are associated with the environment (see Environments).

For example, if a type size of 10.95 points is requested, and the nearest size permitted by a sizes request (or by the sizes or sizescale directives in the device’s DESC file) is 11 points, the output driver uses the latter value.

The \s escape sequence offers the following syntax forms that work with fractional type sizes and accept scaling units. You may of course give them integral arguments. The delimited forms need not use the neutral apostrophe; see Delimiters.

\s[n]
\s'n'

Set the type size to n scaled points; n is a numeric expression with a default scaling unit of ‘z’.

\s[+n]
\s[-n]
\s+[n]
\s-[n]
\s'+n'
\s'-n'
\s+'n'
\s-'n'

Increase or decrease the type size by n scaled points; n is a numeric expression (which may start with a minus sign) with a default scaling unit of ‘z’.


Next: , Previous: , Up: GNU troff Reference   [Contents][Index]

5.21 Colors

GNU troff supports color output with a variety of color spaces and up to 16 bits per channel. Some devices, particularly terminals, may be more limited. When color support is enabled, two colors are current at any given time: the stroke color, with which glyphs, rules (lines), and geometric objects like circles and polygons are drawn, and the fill color, which can be used to paint the interior of a closed geometric figure.

Request: .color [n]
Register: \n[.color]

If n is missing or non-zero, enable the output of color-related device-independent output commands (this is the default); otherwise, disable them. This request sets a global flag; it does not produce an input token (see Gtroff Internals).

The read-only register .color is 1 if colors are enabled, 0 otherwise.

Color can also be disabled with the -c command-line option.

Request: .defcolor ident scheme color-component …

Define a color named ident. scheme selects a color space and determines the quantity of required color-components; it must be one of ‘rgb’ (three components), ‘cmy’ (three), ‘cmyk’ (four), or ‘gray’ (one). ‘grey’ is accepted as a synonym of ‘gray’. The color components can be encoded as a single hexadecimal value starting with ‘#’ or ‘##’. The former indicates that each component is in the range 0–255 (0–FF), the latter the range 0–65,535 (0–FFFF).

.defcolor half gray #7f
.defcolor pink rgb #FFC0CB
.defcolor magenta rgb  ##ffff0000ffff

Alternatively, each color component can be specified as a decimal fraction in the range 0–1, interpreted using a default scaling unit of f, which multiplies its value by 65,536 (but clamps it at 65,535).

.defcolor gray50 rgb 0.5 0.5 0.5
.defcolor darkgreen rgb 0.1f 0.5f 0.2f

Each output device has a color named ‘default’, which cannot be redefined. A device’s default stroke and fill colors are not necessarily the same. For the dvi, html, pdf, ps, and xhtml output devices, GNU troff automatically loads a macro file defining many color names at startup. By the same mechanism, the devices supported by grotty recognize the eight standard ISO 6429/EMCA-48 color names.86

Request: .gcolor [color]
Escape sequence: \mc
Escape sequence: \m(co
Escape sequence: \m[color]
Register: \n[.m]

Set the stroke color to color.

.gcolor red
The next words
.gcolor
\m[red]are in red\m[]
and these words are in the previous color.

The escape sequence \m[] restores the previous stroke color, as does a gcolor request without an argument.

The name of the current stroke color is available in the read-only string-valued register ‘.m’; it is associated with the environment (see Environments). It interpolates nothing when the stroke color is the default.

\m doesn’t produce an input token in GNU troff (see Gtroff Internals). It therefore can be used in requests like mc (which expects a single character as an argument) to change the color on the fly:

.mc \m[red]x\m[]
Request: .fcolor [color]
Escape sequence: \Mc
Escape sequence: \M(co
Escape sequence: \M[color]
Register: \n[.M]

Set the fill color for objects drawn with \D'…' escape sequences. The escape sequence \M[] restores the previous fill color, as does an fcolor request without an argument.

The name of the current fill color is available in the read-only string-valued register ‘.M’; it is associated with the environment (see Environments). It interpolates nothing when the fill color is the default. \M doesn’t produce an input token in GNU troff.

Create an ellipse with a red interior as follows.

\M[red]\h'0.5i'\D'E 2i 1i'\M[]

Next: , Previous: , Up: GNU troff Reference   [Contents][Index]

5.22 Strings

GNU troff supports strings primarily for user convenience. Conventionally, if one would define a macro only to interpolate a small amount of text, without invoking requests or calling any other macros, one defines a string instead. Only one string is predefined by the language.

String: \*[.T]

Contains the name of the output device (for example, ‘utf8’ or ‘pdf’).

The ds request creates a string with a specified name and contents and the \* escape sequence dereferences its name, interpolating its contents. If the string named by the \* escape sequence does not exist, it is defined as empty, nothing is interpolated, and a warning in category ‘mac’ is emitted. See Warnings, for information about the enablement and suppression of warnings.

Request: .ds name [contents]
Request: .ds1 name [contents]
Escape sequence: \*n
Escape sequence: \*(nm
Escape sequence: \*[name [arg1 arg2 …]]

Define a string called name with contents contents. If name already exists as an alias, the target of the alias is redefined; see als and rm below. If ds is called with only one argument, name is defined as an empty string. Otherwise, GNU troff stores contents in copy mode.87

The \* escape sequence interpolates a previously defined string variable name (one-character name n, two-character name nm). The bracketed interpolation form accepts arguments that are handled as macro arguments are; recall Calling Macros. In contrast to macro calls, however, if a closing bracket ‘]’ occurs in a string argument, that argument must be enclosed in double quotes. \* is interpreted even in copy mode. When defining strings, argument interpolations must be escaped if they are to reference parameters from the calling context; See Parameters.

.ds cite (\\$1, \\$2)
Gray codes are explored in \*[cite Morgan 1998].
    ⇒ Gray codes are explored in (Morgan, 1998).

Caution: Unlike other requests, the second argument to the ds request consumes the remainder of the input line, including trailing spaces. This means that comments on a line with such a request can introduce unwanted space into a string when they are set off from the material they annotate, as is conventional.

.ds H2O H\v'+.3m'\s'-2'2\v'-.3m'\s0O \" water

Instead, place the comment on another line or put the comment escape sequence immediately adjacent to the last character of the string.

.ds H2O H\v'+.3m'\s'-2'2\v'-.3m'\s0O\" water

Ending string definitions (and appendments) with a comment, even an empty one, prevents unwanted space from creeping into them during source document maintenance.

.ds author Alice Pleasance Liddell\"
.ds empty \" might be appended to later with .as

An initial neutral double quote " in contents is stripped to allow embedding of leading spaces. Any other " is interpreted literally, but it is wise to use the special character escape sequence \[dq] instead if the string might be interpolated as part of a macro argument; see Calling Macros.

.ds salutation "         Yours in a white wine sauce,\"
.ds c-var-defn "  char mydate[]=\[dq]2020-07-29\[dq];\"

Strings are not limited to a single input line of text. \RET works just as it does elsewhere. The resulting string is stored without the newlines. Care is therefore required when interpolating strings while filling is disabled.

.ds foo This string contains \
text on multiple lines \
of input.

It is not possible to embed a newline in a string that will be interpreted as such when the string is interpolated. To achieve that effect, use \* to interpolate a macro instead; see Punning Names.

Because strings are similar to macros, they too can be defined so as to suppress AT&T troff compatibility mode when used; see Writing Macros and Compatibility Mode. The ds1 request defines a string such that compatibility mode is off when the string is later interpolated. To be more precise, a compatibility save input token is inserted at the beginning of the string, and a compatibility restore input token at the end.

.nr xxx 12345
.ds aa The value of xxx is \\n[xxx].
.ds1 bb The value of xxx is \\n[xxx].
.
.cp 1
.
\*(aa
    error→ warning: register '[' not defined
    ⇒ The value of xxx is 0xxx].
\*(bb
    ⇒ The value of xxx is 12345.
Request: .as name [contents]
Request: .as1 name [contents]

The as request is similar to ds but appends contents to the string stored as name instead of redefining it. If name doesn’t exist yet, it is created. If as is called with only one argument, no operation is performed (beyond dereferencing the string).

.as salutation " with shallots, onions and garlic,\"

The as1 request is similar to as, but compatibility mode is switched off when the appended portion of the string is later interpolated. To be more precise, a compatibility save input token is inserted at the beginning of the appended string, and a compatibility restore input token at the end.

Several requests exist to perform rudimentary string operations. Strings can be queried (length) and modified (chop, substring, stringup, stringdown), and their names can be manipulated through renaming, removal, and aliasing (rn, rm, als).

Request: .length reg anything

Compute the number of characters of anything and store the count in the register reg. If reg doesn’t exist, it is created. anything is read in copy mode.

.ds xxx abcd\h'3i'efgh
.length yyy \*[xxx]
\n[yyy]
    ⇒ 14
Request: .chop object

Remove the last character from the macro, string, or diversion named object. This is useful for removing the newline from the end of a diversion that is to be interpolated as a string. This request can be used repeatedly on the same object; see Gtroff Internals, for details on nodes inserted additionally by GNU troff.

Request: .substring str start [end]

Replace the string named str with its substring bounded by the indices start and end, inclusively. The first character in the string has index 0. If end is omitted, it is implicitly set to the largest valid value (the string length minus one). Negative indices count backward from the end of the string: the last character has index -1, the character before the last has index -2, and so on.

.ds xxx abcdefgh
.substring xxx 1 -4
\*[xxx]
    ⇒ bcde
.substring xxx 2
\*[xxx]
    ⇒ de
Request: .stringdown str
Request: .stringup str

Alter the string named str by replacing each of its bytes with its lowercase (stringdown) or uppercase (stringup) version (if one exists). Special characters in the string will often transform in the expected way due to the regular naming convention for accented characters. When they do not, use substrings and/or catenation.

.ds resume R\['e]sum\['e]
\*[resume]
.stringdown resume
\*[resume]
.stringup resume
\*[resume]
    ⇒ Rsum rsum RSUM

(In practice, we would end the ds request with a comment escape \" to prevent space from creeping into the definition during source document maintenance.)

Request: .rn old new

Rename the request, macro, diversion, or string old to new.

Request: .rm name

Remove the request, macro, diversion, or string name. GNU troff treats subsequent invocations as if the name had never been defined.

Request: .als new old

Create an alias new for the existing request, string, macro, or diversion object named old, causing the names to refer to the same stored object. If old is undefined, a warning in category ‘mac’ is produced, and the request is ignored. See Warnings, for information about the enablement and suppression of warnings.

To understand how the als request works, consider two different storage pools: one for objects (macros, strings, etc.), and another for names. As soon as an object is defined, GNU troff adds it to the object pool, adds its name to the name pool, and creates a link between them. When als creates an alias, it adds a new name to the name pool that gets linked to the same object as the old name.

Now consider this example.

.de foo
..
.
.als bar foo
.
.de bar
.  foo
..
.
.bar
    error→ input stack limit exceeded (probable infinite
    error→ loop)

In the above, bar remains an alias—another name for—the object referred to by foo, which the second de request replaces. Alternatively, imagine that the de request dereferences its argument before replacing it. Either way, the result of calling bar is a recursive loop that finally leads to an error. See Writing Macros.

To remove an alias, call rm on its name. The object itself is not destroyed until it has no more names.

When a request, macro, string, or diversion is aliased, redefinitions and appendments “write through” alias names. To replace an alias with a separately defined object, you must use the rm request on its name first.


Next: , Previous: , Up: GNU troff Reference   [Contents][Index]

5.23 Conditionals and Loops

groff has if and while control structures like other languages. However, the syntax for grouping multiple input lines in the branches or bodies of these structures is unusual.


Next: , Previous: , Up: Conditionals and Loops   [Contents][Index]

5.23.1 Operators in Conditionals

In if, ie, and while requests, in addition to the numeric expressions described in Numeric Expressions, several Boolean operators are available; the members of this expanded class are termed conditional expressions.

c glyph

True if glyph is available, where glyph is an ordinary character, a special character ‘\(xx’ or ‘\[xxx]’, ‘\N'xxx'’, or has been defined by any of the char, fchar, fschar, or schar requests.

d name

True if a string, macro, diversion, or request called name exists.

e

True if the current page is even-numbered.

F font

True if font exists. font is handled as if it were opened with the ft request (that is, font translation and styles are applied), without actually mounting it.

m color

True if color is defined.

n

True if the document is being processed in nroff mode. See troff and nroff Modes.

o

True if the current page is odd-numbered.

r register

True if register exists.

S style

True if style is available for the current font family. Font translation is applied.

t

True if the document is being processed in troff mode. See troff and nroff Modes.

v

Always false. This condition is recognized only for compatibility with certain other troff implementations.88

If the first argument to an if, ie, or while request begins with a non-alphanumeric character apart from ! (see below); it performs an output comparison test. 89

'xxx'yyy'

True if formatting the comparands xxx and yyy produces the same output commands. The delimiter need not be a neutral apostrophe: the output comparison operator accepts the same delimiters as most escape sequences; see Delimiters. This output comparison operator formats xxx and yyy in separate environments; after the comparison, the resulting data are discarded.

.ie "|"\fR|\fP" true
.el false
    ⇒ true

The resulting glyph properties, including font family, style, size, and slant, must match, but not necessarily the requests and/or escape sequences used to obtain them. In the previous example, ‘|’ and ‘\fR|\fP’ result in ‘|’ glyphs in the same typefaces at the same positions, so the comparands are equal. If ‘.ft I’ had been added before the ‘.ie’, they would differ: the first ‘|’ would produce an italic ‘|’, not a roman one. Motions must match in orientation and magnitude to within the applicable horizontal and vertical motion quanta of the device, after rounding. ‘.if "\u\d"\v'0'"’ is false even though both comparands result in zero net motion, because motions are not interpreted or optimized but sent as-is to the output.90 On the other hand, ‘.if "\d"\v'0.5m'"’ is true, because \d is defined as a downward motion of one-half em.91

Surround the comparands with \? to avoid formatting them; this causes them to be compared character by character, as with string comparisons in other programming languages.

.ie "\?|\?"\?\fR|\fP\?" true
.el false
    ⇒ false

Since comparands protected with \? are read in copy mode (see Copy Mode), they need not even be valid groff syntax. The escape character is still lexically recognized, however, and consumes the next character.

.ds a \[
.ds b \[
.if '\?\*a\?'\?\*b\?' a and b equivalent
.if '\?\\?'\?\\?' backslashes equivalent
    ⇒ a and b equivalent

The above operators can’t be combined with most others, but a leading ‘!’, not followed immediately by spaces or tabs, complements an expression.

.nr x 1
.ie !r x register x is not defined
.el      register x is defined
    ⇒ register x is defined

Spaces and tabs are optional immediately after the ‘c’, ‘d’, ‘F’, ‘m’, ‘r’, and ‘S’ operators, but right after ‘!’, they end the predicate and the conditional evaluates true.92

.nr x 1
.ie ! r x register x is not defined
.el       register x is defined
    ⇒ r x register x is not defined

The unexpected ‘r x’ in the output is a clue that our conditional was not interpreted as we planned, but matters may not always be so obvious.


Next: , Previous: , Up: Conditionals and Loops   [Contents][Index]

5.23.2 if-then

Request: .if cond-expr anything

Evaluate the conditional expression cond-expr, and if it evaluates true (or to a positive value), interpret the remainder of the line anything as if it were an input line. Recall from Invoking Requests that any quantity of spaces between arguments to requests serves only to separate them; leading spaces in anything are thus not seen. anything effectively cannot be omitted; if cond-expr is true and anything is empty, the newline at the end of the control line is interpreted as a blank input line (and therefore a blank text line).

super\c
tanker
.nr force-word-break 1
super\c
.if ((\n[force-word-break] = 1) & \n[.int])
tanker
    ⇒ supertanker super tanker
Request: .nop anything

Interpret anything as if it were an input line. This is similar to ‘.if 1’. nop is not really “no operation”; its argument is processed—unconditionally. It can be used to cause text lines to share indentation with surrounding control lines.

.als real-MAC MAC
.de wrapped-MAC
.  tm MAC: called with arguments \\$@
.  nop \\*[real-MAC]\\
..
.als MAC wrapped-MAC
\# Later...
.als MAC real-MAC

In the above, we’ve used aliasing, nop, and the interpolation of a macro as a string to interpose a wrapper around the macro ‘MAC’ (perhaps to debug it).


Next: , Previous: , Up: Conditionals and Loops   [Contents][Index]

5.23.3 if-else

Request: .ie cond-expr anything
Request: .el anything

Use the ie and el requests to write an if-then-else. The first request is the “if” part and the latter is the “else” part. Unusually among programming languages, any number of non-conditional requests may be interposed between the ie branch and the el branch.

.nr a 0
.ie \na a is non-zero.
.nr a +1
.el a was not positive but is now \na.
    ⇒ a was not positive but is now 1.

Another way in which el is an ordinary request is that it does not lexically “bind” more tightly to its ie counterpart than it does to any other request. This fact can surprise C programmers.

.nr a 1
.nr z 0
.ie \nz \
.  ie \na a is true
.  el     a is false
.el z is false
    error→ warning: unbalanced 'el' request
    ⇒ a is false

To conveniently nest conditionals, keep reading.


Next: , Previous: , Up: Conditionals and Loops   [Contents][Index]

5.23.4 Conditional Blocks

It is frequently desirable for a control structure to govern more than one request, macro call, text line, or a combination of the foregoing. The opening and closing brace escape sequences \{ and \} define such groups. These conditional blocks can furthermore be nested.

Escape sequence: \{
Escape sequence: \}

\{ begins a conditional block; it must appear (after optional spaces and tabs) immediately subsequent to the conditional expression of an if, ie, or while request,93 or as the argument to an el request.

\} ends a condition block and should appear on a line with other occurrences of itself as necessary to match \{ sequences. It can be preceded by a control character, spaces, and tabs. Input after any quantity of \} sequences on the same line is processed only if all of the preceding conditions to which they correspond are true. Furthermore, a \} closing the body of a while request must be the last such escape sequence on an input line.

Brace escape sequences outside of control structures have no meaning and produce no output.

Caution: Input lines using \{ often end with \RET, especially in macros that consist primarily of control lines. Forgetting to use \RET on an input line after \{ is a common source of error.

We might write the following in a page header macro. If we delete \RET, the header will carry an unwanted extra empty line (except on page 1).

.if (\\n[%] != 1) \{\
.  ie ((\\n[%] % 2) = 0) .tl \\*[even-numbered-page-title]
.  el                    .tl \\*[odd-numbered-page-title]
.\}

Let us take a closer look at how conditional blocks nest.

A
.if 0 \{ B
C
D
\}E
F
    ⇒ A F
N
.if 1 \{ O
.  if 0 \{ P
Q
R\} S\} T
U
    ⇒ N O U

The above behavior may challenge the intuition; it was implemented to retain compatibility with AT&T troff. For clarity, it is idiomatic to end input lines with \{ (followed by \RET if appropriate), and to precede \} on an input line with nothing more than a control character, spaces, tabs, and other instances of itself.

We can use ie, el, and conditional blocks to simulate the multi-way “switch” or “case” control structures of other languages. The following example is adapted from the groff man package. Indentation is used to clarify the logic.

.\" Simulate switch/case in roff.
.      ie '\\$2'1' .ds title General Commands\"
.el \{.ie '\\$2'2' .ds title System Calls\"
.el \{.ie '\\$2'3' .ds title Library Functions\"
.el \{.ie '\\$2'4' .ds title Kernel Interfaces\"
.el \{.ie '\\$2'5' .ds title File Formats\"
.el \{.ie '\\$2'6' .ds title Games\"
.el \{.ie '\\$2'7' .ds title Miscellaneous Information\"
.el \{.ie '\\$2'8' .ds title System Management\"
.el \{.ie '\\$2'9' .ds title Kernel Development\"
.el                .ds title \" empty
.\}\}\}\}\}\}\}\}

Previous: , Up: Conditionals and Loops   [Contents][Index]

5.23.5 while

groff provides a looping construct: the while request. Its syntax matches the if request.

Request: .while cond-expr anything

Evaluate the conditional expression cond-expr, and repeatedly execute anything unless and until cond-expr evaluates false. anything, which is often a conditional block, is referred to as the while request’s body.

.nr a 0 1
.while (\na < 9) \{\
\n+a,
.\}
\n+a
    ⇒ 1, 2, 3, 4, 5, 6, 7, 8, 9, 10

GNU troff treats the body of a while request similarly to that of a de request (albeit one not read in copy mode94), but stores it under an internal name and deletes it when the loop finishes. The operation of a macro containing a while request can slow significantly if the while body is large. Each time the macro is executed, the while body is parsed and stored again.

.de xxx
.  nr num 10
.  while (\\n[num] > 0) \{\
.    \" many lines of code
.    nr num -1
.  \}
..

An often better solution—and one that is more portable, since AT&T troff lacked the while request—is to instead write a recursive macro. It will be parsed only once.95

.de yyy
.  if (\\n[num] > 0) \{\
.    \" many lines of code
.    nr num -1
.    yyy
.  \}
..
.
.de xxx
.  nr num 10
.  yyy
..

To prevent infinite loops, the default number of available recursion levels is 1,000 or somewhat less.96 You can disable this protective measure, or raise the limit, by setting the slimit register. See Debugging.

As noted above, if a while body begins with a conditional block, its closing brace must end an input line.

.if 1 \{\
.  nr a 0 1
.  while (\n[a] < 10) \{\
.    nop \n+[a]
.\}\}
    error→ unbalanced brace escape sequences
Request: .break

Exit a while loop. Do not confuse this request with a typographical break or the br request.

Request: .continue

Skip the remainder of a while loop’s body, immediately starting the next iteration.


Next: , Previous: , Up: GNU troff Reference   [Contents][Index]

5.24 Writing Macros

A macro is a stored collection of text and control lines that can be interpolated multiple times. Use macros to define common operations. Macros are called in the same way that requests are invoked. While requests exist for the purpose of creating macros, simply calling an undefined macro, or interpolating it as a string, will cause it to be defined as empty. See Identifiers.

Request: .de name [end]

Define a macro name, replacing the definition of any existing request, macro, string, or diversion called name. If name already exists as an alias, the target of the alias is redefined; recall Strings. GNU troff enters copy mode,97 storing subsequent input lines as the macro definition. If the optional second argument is not specified, the definition ends with the control line ‘..’ (two dots). Alternatively, end identifies a macro whose call syntax at the start of a control line ends the definition of name; end is then called normally. A macro definition must end in the same conditional block (if any) in which it began (see Conditional Blocks). Spaces or tabs are permitted after the control character in the line containing this ending token (either ‘.’ or ‘end’), but a tab immediately after the token prevents its recognition as the end of a macro definition. The macro end can be called with arguments.98

Here is a small example macro called ‘P’ that causes a break and inserts some vertical space. It could be used to separate paragraphs.

.de P
.  br
.  sp .8v
..

We can define one macro within another. Attempting to nest ‘..’ navely will end the outer definition because the inner definition isn’t interpreted as such until the outer macro is later interpolated. We can use an end macro instead. Each level of nesting should use a unique end macro.

An end macro need not be defined until it is called. This fact enables a nested macro definition to begin inside one macro and end inside another. Consider the following example.99

.de m1
.  de m2 m3
you
..
.de m3
Hello,
Joe.
..
.de m4
do
..
.m1
know?
.  m3
What
.m4
.m2
    ⇒ Hello, Joe.  What do you know?

A nested macro definition can be terminated with ‘..’ and nested macros can reuse end macros, but these control lines must be escaped multiple times for each level of nesting. The necessity of this escaping and the utility of nested macro definitions will become clearer when we employ macro parameters and consider the behavior of copy mode in detail.

de defines a macro that inherits the compatibility mode enablement status of its context (see Implementation Differences). Often it is desirable to make a macro that uses groff features callable from contexts where compatibility mode is on; for instance, when writing extensions to a historical macro package. To achieve this, compatibility mode needs to be switched off while such a macro is interpreted—without disturbing that state when it is finished.

Request: .de1 name [end]

The de1 request defines a macro to be interpreted with compatibility mode disabled. When name is called, compatibility mode enablement status is saved; it is restored when the call completes. Observe the extra backlash before the interpolation of register ‘xxx’; we’ll explore this subject in Copy Mode.

.nr xxx 12345
.de aa
The value of xxx is \\n[xxx].
.  br
..
.de1 bb
The value of xxx is \\n[xxx].
..
.cp 1
.aa
    error→ warning: register '[' not defined
    ⇒ The value of xxx is 0xxx].
.bb
    ⇒ The value of xxx is 12345.
Request: .dei name [end]
Request: .dei1 name [end]

The dei request defines a macro with its name and end macro indirected through strings. That is, it interpolates strings named name and end before performing the definition.

The following examples are equivalent.

.ds xx aa
.ds yy bb
.dei xx yy
.de aa bb

The dei1 request bears the same relationship to dei as de1 does to de; it temporarily turns compatibility mode off when name is called.

Request: .am name [end]
Request: .am1 name [end]
Request: .ami name [end]
Request: .ami1 name [end]

am appends subsequent input lines to macro name, extending its definition, and otherwise working as de does.

To make the previously defined ‘P’ macro set indented instead of block paragraphs, add the necessary code to the existing macro.

.am P
.ti +5n
..

The other requests are analogous to their ‘de’ counterparts. The am1 request turns off compatibility mode during interpretation of the appendment. The ami request appends indirectly, meaning that strings name and end are interpolated with the resulting names used before appending. The ami1 request is similar to ami, disabling compatibility mode during interpretation of the appended lines.

Using trace.tmac, you can trace calls to de, de1, am, and am1. You can also use the backtrace request at any point desired to troubleshoot tricky spots (see Debugging).

See Strings, for the als, rm, and rn requests to create an alias of, remove, and rename a macro, respectively.

Macro identifiers share their name space with requests, strings, and diversions; see Identifiers. The am, as, da, de, di, and ds requests (together with their variants) create a new object only if the name of the macro, diversion, or string is currently undefined or if it is defined as a request; normally, they modify the value of an existing object. See the description of the als request, for pitfalls when redefining a macro that is aliased.

Request: .return [anything]

Exit a macro, immediately returning to the caller. If called with an argument anything, exit twice—the current macro and the macro one level higher. This is used to define a wrapper macro for return in trace.tmac.


Next: , Previous: , Up: Writing Macros   [Contents][Index]

5.24.1 Parameters

Macro calls and string interpolations optionally accept a list of arguments; recall Calling Macros. At the time such an interpolation takes place, these parameters can be examined using a register and a variety of escape sequences starting with ‘\$’. All such escape sequences are interpreted even in copy mode, a fact we shall motivate and explain below (see Copy Mode).

Register: \n[.$]

The count of parameters available to a macro or string is kept in this read-only register. The shift request can change its value.

Any individual parameter can be accessed by its position in the list of arguments to the macro call, numbered from left to right starting at 1, with one of the following escape sequences.

Escape sequence: \$n
Escape sequence: \$(nn
Escape sequence: \$[nnn]

Interpolate the nth, nnth, or nnnth parameter. The first form expects only a single digit (1≤n≤9)), the second two digits (01≤nn≤99)), and the third any positive integer nnn. Macros and strings accept an unlimited number of parameters.

Request: .shift [n]

Shift the parameters n places (1 by default). This is a “left shift”: what was parameter i becomes parameter i-n. The parameters formerly in positions 1 to n are no longer available. Shifting by a non-positive amount performs no operation. The register .$ is adjusted accordingly.

In practice, parameter interpolations are usually seen prefixed with an extra escape character. This is because the \$ family of escape sequences is interpreted even in copy mode.100

Escape sequence: \$*
Escape sequence: \$@
Escape sequence: \$^

In some cases it is convenient to interpolate all of the parameters at once (to pass them to a request, for instance). The \$* escape concatenates the parameters, separating them with spaces. \$@ is similar, concatenating the parameters, surrounding each with double quotes and separating them with spaces. If not in compatibility mode, the interpolation depth of double quotes is preserved (see Calling Macros). \$^ interpolates all parameters as if they were arguments to the ds request.

.de foo
. tm $1='\\$1'
. tm $2='\\$2'
. tm $*='\\$*'
. tm $@='\\$@'
. tm $^='\\$^'
..
.foo " This is a "test"
    error→ $1=' This is a '
    error→ $2='test"'
    error→ $*=' This is a  test"'
    error→ $@='" This is a " "test""'
    error→ $^='" This is a "test"'

\$* is useful when writing a macro that doesn’t need to distinguish its arguments, or even to not interpret them; examples include macros that produce diagnostic messages by wrapping the tm or ab requests. Use \$@ when writing a macro that may need to shift its parameters and/or wrap a macro or request that finds the count significant. If in doubt, prefer \$@ to \$*. An application of \$^ is seen in trace.tmac, which redefines some requests and macros for debugging purposes.

Escape sequence: \$0

Interpolate the name by which the macro being interpreted was called. The als request can cause a macro to have more than one name. Applying string interpolation to a macro does not change this name.

.de foo
.  tm \\$0
..
.als bar foo
.
.de aaa
.  foo
..
.de bbb
.  bar
..
.de ccc
\\*[foo]\\
..
.de ddd
\\*[bar]\\
..
.
.aaa
    error→ foo
.bbb
    error→ bar
.ccc
    error→ ccc
.ddd
    error→ ddd

Previous: , Up: Writing Macros   [Contents][Index]

5.24.2 Copy Mode

When GNU troff processes certain requests, most importantly those which define or append to a macro or string, it does so in copy mode: it copies the characters of the definition into a dedicated storage region, interpolating the escape sequences \n, \g, \$, \*, \V, and \? normally; interpreting \RET immediately; discarding comments \" and \#; interpolating the current leader, escape, or tab character with \a, \e, and \t, respectively; and storing all other escape sequences in an encoded form.

The complement of copy mode—a roff formatter’s behavior when not defining or appending to a macro, string, or diversion—where all macros are interpolated, requests invoked, and valid escape sequences processed immediately upon recognition, can be termed interpretation mode.

Escape sequence: \\

The escape character, \ by default, can escape itself. This enables you to control whether a given \n, \g, \$, \*, \V, or \? escape sequence is interpreted at the time the macro containing it is defined, or later when the macro is called.101

.nr x 20
.de y
.nr x 10
\&\nx
\&\\nx
..
.y
    ⇒ 20 10

You can think of \\ as a “delayed” backslash; it is the escape character followed by a backslash from which the escape character has removed its special meaning. Consequently, ‘\\’ is not an escape sequence in the usual sense. In any escape sequence ‘\X’ that GNU troff does not recognize, the escape character is ignored and X is output. An unrecognized escape sequence causes a warning in category ‘escape’, with two exceptions—‘\\’ is the first.

Escape sequence: \.

\. escapes the control character. It is similar to \\ in that it isn’t a true escape sequence. It is used to permit nested macro definitions to end without a named macro call to conclude them. Without a syntax for escaping the control character, this would not be possible.

.de m1
foo
.
.  de m2
bar
\\..
.
..
.m1
.m2
    ⇒ foo bar

The first backslash is consumed while the macro is read, and the second is interpreted when macro m1 is called.

roff documents should not use the \\ or \. character sequences outside of copy mode; they serve only to obfuscate the input. Use \e to represent the escape character, \[rs] to obtain a backslash glyph, and \& before ‘.’ and ‘'’ where GNU troff expects them as control characters if you mean to use them literally (recall Requests and Macros).

Macro definitions can be nested to arbitrary depth. The mechanics of parsing the escape character have significant consequences for this practice.

.de M1
\\$1
.  de M2
\\\\$1
.    de M3
\\\\\\\\$1
\\\\..
.    M3 hand.
\\..
.  M2 of
..
This understeer is getting
.M1 out
    ⇒ This understeer is getting out of hand.

Each escape character is interpreted twice—once in copy mode, when the macro is defined, and once in interpretation mode, when the macro is called. As seen above, this fact leads to exponential growth in the quantity of escape characters required to delay interpolation of \n, \g, \$, \*, \V, and \? at each nesting level, which can be daunting. GNU troff offers a solution.

Escape sequence: \E

\E represents an escape character that is not interpreted in copy mode. You can use it to ease the writing of nested macro definitions.

.de M1
.  nop \E$1
.  de M2
.    nop \E$1
.    de M3
.      nop \E$1
\\\\..
.    M3 better.
\\..
.  M2 bit
..
This vehicle handles
.M1 a
    ⇒ This vehicle handles a bit better.

Observe that because \. is not a true escape sequence, we can’t use \E to keep ‘..’ from ending a macro definition prematurely. If the multiplicity of backslashes complicates maintenance, use end macros.

\E is also convenient to define strings containing escape sequences that need to work when used in copy mode (for example, as macro arguments), or which will be interpolated at varying macro nesting depths. We might define strings to begin and end superscripting as follows.102

.ds { \v'-.9m\s'\En[.s]*7u/10u'+.7m'
.ds } \v'-.7m\s0+.9m'

When the ec request is used to redefine the escape character, \E also makes it easier to distinguish the semantics of an escape character from the other meaning(s) its character might have. Consider the use of an unusual escape character, ‘-’.

.nr a 1
.ec -
.de xx
--na
..
.xx
    ⇒ -na

This result may surprise you; some people expect ‘1’ to be output since register ‘a’ has clearly been defined with that value. What has happened? The robotic replacement of ‘\’ with ‘-’ has led us astray. You might recognize the sequence ‘--’ more readily with the default escape character as ‘\-’, the special character escape sequence for the minus sign glyph.

.nr a 1
.ec -
.de xx
-Ena
..
.xx
    ⇒ 1

Next: , Previous: , Up: GNU troff Reference   [Contents][Index]

5.25 Page Motions

See Manipulating Spacing, for a discussion of the most commonly used request for vertical motion, sp, which spaces downward by one vee.

Request: .mk [reg]
Request: .rt [dist]

You can mark a location on a page for subsequent return. mk takes an argument, a register name in which to store the current page location. If given no argument, it stores the location in an internal register. This location can be used later by the rt or the sp requests (or the \v escape).

The rt request returns upward to the location marked with the last mk request. If used with an argument, it returns to a vertical position dist from the top of the page (no previous call to mk is necessary in this case). The default scaling unit is ‘v’.

If a page break occurs between a mk request and its matching rt request, the rt request is silently ignored.

A simple implementation of a macro to set text in two columns follows.

.nr column-length 1.5i
.nr column-gap 4m
.nr bottom-margin 1m
.
.de 2c
.  br
.  mk
.  ll \\n[column-length]u
.  wh -\\n[bottom-margin]u 2c-trap
.  nr right-side 0
..
.
.de 2c-trap
.  ie \\n[right-side] \{\
.    nr right-side 0
.    po -(\\n[column-length]u + \\n[column-gap]u)
.    \" remove trap
.    wh -\\n[bottom-margin]u
.  \}
.  el \{\
.    \" switch to right side
.    nr right-side 1
.    po +(\\n[column-length]u + \\n[column-gap]u)
.    rt
.  \}
..

Now let us apply our two-column macro.

.pl 1.5i
.ll 4i
This is a small test that shows how the
rt request works in combination with mk.

.2c
Starting here, text is typeset in two columns.
Note that this implementation isn't robust
and thus not suited for a real two-column
macro.
    ⇒ This is a small test that shows how the
    ⇒ rt request works in combination with mk.
    ⇒
    ⇒ Starting  here,    isn't    robust
    ⇒ text is typeset    and   thus  not
    ⇒ in two columns.    suited  for   a
    ⇒ Note that  this    real two-column
    ⇒ implementation     macro.

Several escape sequences enable fine control of movement about the page.

Escape sequence: \v'expr'

Vertically move the drawing position. expr indicates the magnitude of motion: positive is downward and and negative upward. The default scaling unit is ‘v’. The motion is relative to the current drawing position unless expr begins with the boundary-relative motion operator ‘|’. See Numeric Expressions.

Text processing continues at the new drawing position; usually, vertical motions should be in balanced pairs to avoid a confusing page layout.

\v will not spring a vertical position trap. This can be useful; for example, consider a page bottom trap macro that prints a marker in the margin to indicate continuation of a footnote. See Traps.

A few escape sequences that produce vertical motion are unusual. They are thought to originate early in AT&T nroff history to achieve super- and subscripting by half-line motions on line printers and teletypewriters before the phototypesetter made more precise positioning available. They are reckoned in ems—not vees—to maintain continuity with their original purpose of moving relative to the size of the type rather than the distance between text baselines (vees).103

Escape sequence: \r
Escape sequence: \u
Escape sequence: \d

Move upward 1m, upward .5m, and downward .5m, respectively.

Let us see these escape sequences in use.

Obtain 100 cm\u3\d of \ka\d\092\h'|\nau'\r233\dU.

In the foregoing we have paired \u and \d to typeset a superscript, and later a full em negative (“reverse”) motion to place a superscript above a subscript. A numeral-width horizontal motion escape sequence aligns the proton and nucleon numbers, while \k marks a horizontal position to which \h returns so that we could stack them. (We shall discuss these horizontal motion escape sequences presently.) In serious applications, we often want to alter the type size of the -scripts and to fine-tune the vertical motions, as the groff ms package does with its super- and subscripting string definitions.

Escape sequence: \h'expr'

Horizontally move the drawing position. expr indicates the magnitude of motion: positive is rightward and negative leftward. The default scaling unit is ‘m’. The motion is relative to the current drawing position unless expr begins with the boundary-relative motion operator ‘|’. See Numeric Expressions.

The following string definition sets the TeX logo.104

.ds TeX T\h'-.1667m'\v'.224m'E\v'-.224m'\h'-.125m'X\"

There are a number of special-case escape sequences for horizontal motion.

Escape sequence: \SP

Move right one word space. (The input is a backslash followed by a space.) This escape sequence can be thought of as a non-adjustable, unbreakable space. Usually you want \~ instead; see Manipulating Filling and Adjustment.

Escape sequence: \|

Move one-sixth em to the right on typesetting output devices. If a glyph named ‘\|’ is defined in the current font, its width is used instead, even on terminal output devices.

Escape sequence: \^

Move one-twelfth em to the right on typesetting output devices. If a glyph named ‘\^’ is defined in the current font, its width is used instead, even on terminal output devices.

Escape sequence: \0

Move right by the width of a numeral in the current font.

Horizontal motions are not discarded at the end of an output line as word spaces are. See Breaking.

Escape sequence: \w'anything'
Register: \n[st]
Register: \n[sb]
Register: \n[rst]
Register: \n[rsb]
Register: \n[ct]
Register: \n[ssc]
Register: \n[skw]

Interpolate the width of anything in basic units. This escape sequence allows several properties of formatted output to be measured without writing it out.

The length of the string 'abc' is \w'abc'u.
    ⇒ The length of the string 'abc' is 72u.

anything is processed in a dummy environment: this means that font and type size changes, for example, may occur within it without affecting subsequent output.

After each use, \w sets several registers.

st
sb

The maximum vertical displacements of the text baseline above and below, respectively. The sign convention is opposite that of relative vertical motions; that is, depth below the (original) baseline is negative. These registers are incorrectly documented in the AT&T troff manual as “the highest and lowest extent of [the argument to \w] relative to the baseline”.

rst
rsb

Like st and sb, but taking account of the heights and depths of glyphs. In other words, these registers store the highest and lowest vertical positions attained by anything, doing what AT&T troff documented st and sb as doing.

ct

Characterizes the geometry of glyphs occurring in anything.

0

only short glyphs, no descenders or tall glyphs

1

at least one descender

2

at least one tall glyph

3

at least one each of a descender and a tall glyph

ssc

The amount of horizontal space (possibly negative) that should be added to the last glyph before a subscript.

skw

How far to right of the center of the last glyph in the \w argument, the center of an accent from a roman font should be placed over that glyph.

Escape sequence: \kp
Escape sequence: \k(ps
Escape sequence: \k[position]

Store the current horizontal position in the input line in a register with the name position (one-character name p, two-character name ps). Use this, for example, to return to the beginning of a string for highlighting or other decoration.

Register: \n[hp]

The current horizontal position at the input line.

Register: \n[.k]

A read-only register containing the current horizontal output position (relative to the current indentation).

Escape sequence: \o'abc…'

Overstrike the glyphs of characters a, b, c, …; the glyphs are centered, written, and the drawing position advanced by the widest of the glyphs.

Escape sequence: \zc

Format the character c with zero width; that is, without advancing the drawing position. Use \z to overstrike glyphs aligned to their left edges, in contrast to \o’s centering.

Escape sequence: \Z'anything'

Save the drawing position, format anything, then restore it. Tabs and leaders in the argument are ignored with an error diagnostic.

We might implement a strike-through macro thus.

.de ST
.nr width \w'\\$1'
\Z@\v'-.25m'\l'\\n[width]u'@\\$1
..
.
This is
.ST "a test"
an actual emergency!

Next: , Previous: , Up: GNU troff Reference   [Contents][Index]

5.26 Drawing Geometric Objects

A few of the formatter’s escape sequences draw lines and other geometric objects. Combined with each other and with page motion commands (see Page Motions), a wide variety of figures is possible. For complex drawings, these operations can be cumbersome; the preprocessors gpic or ggrn are typically used instead.

The \l and \L escape sequences draw horizontal and vertical sequences of glyphs, respectively. Even the simplest of output devices supports them.

Escape sequence: \l'l'
Escape sequence: \l'lc'

Draw a horizontal line of length l from the drawing position. Rightward motion is positive. Afterward, the drawing position is at the right end of the line. The default scaling unit is ‘m’.

The optional second parameter c is a character with which to draw the line. The default is the baseline rule special character, \[ru].

If c is a valid scaling unit, put \& after l to disambiguate the input.

.de textbox
\[br]\\$*\[br]\l'|0\[rn]'\l'|0\[ul]'
..

The foregoing outputs a box rule (a vertical line), the text argument(s), and another box rule. We employ the boundary-relative motion operator ‘|’. Finally, the line-drawing escape sequences draw a radical extender (a form of overline) and an underline from the drawing position to the position coresponding to beginning of the input line. The drawing position returns to just after the right-hand box rule because the lengths of the drawn lines are negative, as noted above.

Escape sequence: \L'l'
Escape sequence: \L'lc'

Draw a vertical line of length l from the drawing position. Downward motion is positive. The default scaling unit is ‘v’. The default character is the box rule, \[br]. As with vertical motion escape sequences, text processing continues where the line ends. \L is otherwise similar to \l.

$ nroff <<EOF
This is a \L'3v'test.
EOF
    ⇒ This is a
    ⇒           |
    ⇒           |
    ⇒           |test.

When writing text, the drawing position is at the text baseline; recall Page Geometry.

The \D escape sequence provides drawing commands that direct the output device to render geometrical objects rather than glyphs. Specific devices may support only a subset, or may feature additional ones; consult the man page for the output driver in use. Terminal devices in particular implement almost none. See Graphics Commands.

Rendering starts at the drawing position; when finished, the drawing position is left at the rightmost point of the object, even for closed figures, except where noted. GNU troff draws stroked (outlined) objects with the stroke color, and shades filled ones with the fill color. See Colors. Coordinates h and v are horizontal and vertical motions relative to the drawing position or previous point in the command. The default scaling unit for horizontal measurements (and diameters of circles) is ‘m’; for vertical ones, ‘v’.

Circles, ellipses, and polygons can be drawn filled or stroked. These are independent properties; if you want a filled, stroked figure, you must draw the same figure twice using each drawing command. A filled figure is always smaller than an outlined one because the former is drawn only within its defined area, whereas strokes have a line thickness (set with ‘\D't'’).

\h'1i'\v'1i'\
\# increase line thickness
\Z'\D't 5p''\
\# draw stroked (unfilled) polygon
\Z'\D'p 3 3 -6 0''\
\# draw filled (solid) polygon
\Z'\D'P 3 3 -6 0''
Escape sequence: \D'command argument …'

Drawing command escape sequence parameters begin with an ordinary character, command, selecting the type of object to be drawn, followed by arguments whose meaning is determined by command.

\D'~ h1 v1hn vn'

Draw a B-spline to each point in sequence, leaving the drawing position at (hn, vn).

\D'a hc vc h v'

Draw a circular arc centered at (hc, vc) counterclockwise from the drawing position to a point (h, v) relative to the center. 105

\D'c d'

Draw a circle of diameter d with its leftmost point at the drawing position.

\D'C d'

As ‘\D'C '’, but the circle is filled.

\D'e h v'

Draw an ellipse of width h and height v with its leftmost point at the drawing position.

\D'E x y'

As ‘\D'e '’, but the ellipse is filled.

\D'l dx dy'

Draw line from the drawing position to (h, v).

The following is a macro for drawing a box around a text argument; for simplicity, the box margin is a fixed at 0.2m.

.de TEXTBOX
.  nr @wd \w'\\$1'
\h'.2m'\
\h'-.2m'\v'(.2m - \\n[rsb]u)'\
\D'l 0 -(\\n[rst]u - \\n[rsb]u + .4m)'\
\D'l (\\n[@wd]u + .4m) 0'\
\D'l 0 (\\n[rst]u - \\n[rsb]u + .4m)'\
\D'l -(\\n[@wd]u + .4m) 0'\
\h'.2m'\v'-(.2m - \\n[rsb]u)'\
\\$1\
\h'.2m'
..

The argument is measured with the \w escape sequence. Its width is stored in register @wd. \w also sets the registers rst and rsb; these contain its maximum vertical extents of the argument. Then, four lines are drawn to form a box, offset by the box margin.

\D'p h1 v1hn vn'

Draw polygon with vertices at drawing position and each point in sequence. GNU troff closes the polygon by drawing a line from (hn, vn) back to the initial drawing position. Afterward, the drawing position is left at (hn, vn).

\D'P dx1 dy1 dx2 dy2 …'

As ‘\D'P '’, but the polygon is filled.

The following macro is like the ‘\D'l'’ example, but shades the box. We draw the box before writing the text because colors in GNU troff have no transparency; in othe opposite order, the filled polygon would occlude the text.

.de TEXTBOX
.  nr @wd \w'\\$1'
\h'.2m'\
\h'-.2m'\v'(.2m - \\n[rsb]u)'\
\M[lightcyan]\
\D'P 0 -(\\n[rst]u - \\n[rsb]u + .4m) \
     (\\n[@wd]u + .4m) 0 \
     0 (\\n[rst]u - \\n[rsb]u + .4m) \
     -(\\n[@wd]u + .4m) 0'\
\h'.2m'\v'-(.2m - \\n[rsb]u)'\
\M[]\
\\$1\
\h'.2m'
..
\D't n'

Set the stroke thickness of geometric objects to n basic units. A zero n selects the minimal supported thickness. A negative n selects a thickness proportional to the type size; this is the default.

In a hazy penumbra between text rendering and drawing commands we locate the bracket-building escape sequence, \b. It can assemble apparently large glyphs by vertically stacking ordinary ones.

Escape sequence: \b'contents'

Pile and center a sequence of glyphs vertically on the output line. Piling stacks glyphs corresponding to each character in contents, read from left to right, and placed from top to bottom. GNU troff separates the glyphs vertically by 1m, and the pile itself is centered 0.5m above the text baseline. The horizontal drawing position is then advanced by the width of the widest glyph in the pile.

This rather inflexible positioning algorithm doesn’t work with the dvi output device since its bracket pieces vary in height. Instead, use the geqn preprocessor.

Manipulating Spacing describes how to adjust the vertical spacing of the output line with the \x escape sequence.

The application of \b that lends its name is construction of brackets, braces, and parentheses when typesetting mathematics. We might construct a large opening (left) brace as follows.

\b'\[lt]\[bv]\[lk]\[bv]\[lb]'

See groff_char(7) for a list of special character identifiers.


Next: , Previous: , Up: GNU troff Reference   [Contents][Index]

5.27 Deferring Output

A few roff language elements are generally not used in simple documents, but arise as page layouts become more sophisticated and demanding. Environments collect formatting parameters like line length and typeface. A diversion stores formatted output for later use. A trap is a condition on the input or output, tested automatically by the formatter, that is associated with a macro, causing it to be called when that condition is fulfilled.

Footnote support often exercises all three of the foregoing features. A simple implementation might work as follows. A pair of macros is defined: one starts a footnote and the other ends it. The author calls the first macro where a footnote marker is desired. The macro establishes a diversion so that the footnote text is collected at the place in the body text where its corresponding marker appears. An environment is created for the footnote so that it is set at a smaller typeface. The footnote text is formatted in the diversion using that environment, but it does not yet appear in the output. The document author calls the footnote end macro, which returns to the previous environment and ends the diversion. Later, after much more body text in the document, a trap, set a small distance above the page bottom, is sprung. The macro called by the trap draws a line across the page and emits the stored diversion. Thus, the footnote is rendered.

Diversions and traps make the text formatting process non-linear. Let us imagine a set of text lines or paragraphs labelled ‘A’, ‘B’, and so on. If we set up a trap that produces text ‘T’ (as a page footer, say), and we also use a diversion to store the formatted text ‘D’, then a document with input text in the order ‘A B C D E F’ might render as ‘A B C E T F’. The diversion ‘D’ will never be output if we do not call for it.

Environments of themselves are not a source of non-linearity in document formatting: environment switches have immediate effect. One could always write a macro to change as many formatting parameters as desired with a single convenient call. But because diversions can be nested and macros called by traps that are sprung by other trap-called macros, they may be called upon in varying contexts. For example, consider a page header that is always to be set in Helvetica. A document that uses Times for most of its body text, but Courier for displayed code examples, poses a challenge if a page break occurs in the middle of a code display; if the header trap assumes that the “previous font” is always Times, the rest of the example will be formatted in the wrong typeface. One could carefully save all formatting parameters upon entering the trap and restore them upon leaving it, but this is verbose, error-prone, and not future-proof as the groff language develops. Environments save us considerable effort.


Next: , Previous: , Up: GNU troff Reference   [Contents][Index]

5.28 Traps

Traps are locations in the output or conditions on the input that, when reached or fulfilled, call a specified macro. These traps can occur at a given location on the page, at a given location in the current diversion (together, these are known as vertical position traps), at a blank line, at a line with leading space characters, after a quantity of input lines, or at the end of input. Macros called by traps are passed no arguments. Setting a trap is also called planting one. It is said that a trap is sprung if its condition is fulfilled.


Next: , Previous: , Up: Traps   [Contents][Index]

5.28.1 Vertical Position Traps

A vertical position trap calls a macro when the formatter’s vertical drawing position reaches or passes, in the downward direction, a certain location on the output page or in a diversion. Its applications include setting page headers and footers, body text in multiple columns, and footnotes.

Request: .vpt [flag]
Register: \n[.vpt]

Enable vertical position traps if flag is non-zero or absent; disable them otherwise. Vertical position traps are those set by the wh request or by dt within a diversion. The parameter that controls whether vertical position traps are enabled is global. Initially, vertical position traps are enabled. The current value is stored in the .vpt read-only register.

A page can’t be ejected if vpt is set to zero; see The Implicit Page Trap.


Next: , Previous: , Up: Vertical Position Traps   [Contents][Index]

5.28.1.1 Page Location Traps

A page location trap is a vertical position trap that applies to the page; that is, to undiverted output. Many can be present; manage them with the wh and ch requests.

Request: .wh dist [name]

Plant macro name as page location trap at dist. The default scaling unit is ‘v’. Non-negative values for dist set the trap relative to the top of the page; negative values set the trap relative to the bottom of the page. It is not possible to plant a trap less than one basic unit from the page bottom: a dist of -0 is interpreted as 0, the top of the page.106 An existing visible trap (see below) at dist is removed; this is wh’s sole function if name is missing.

A trap is sprung only if it is visible, meaning that its location is reachable on the page107 and it is not hidden by another trap at the same location already planted there.

A macro package might set headers and footers as follows; this example configures vertical margins of one inch to the body text, and one half-inch to the titles. Observe the use of the no-break control character with sp request to position our text baselines, and the page number character ‘%’ used with the tl request.

.\" hdfo.roff
.de hd                  \" page header
'  sp .5i
'  tl '\\*(Ti''\\*(Da'  \" title and date strings
'  sp .5i
..
.de fo                  \" page footer
'  sp .5i
.  tl ''%''
.  bp
..
.wh 0   hd             \" trap at top of the page
.wh -1i fo             \" trap 1 inch from bottom

To use these traps, copy the above (or load it from a file with the so or mso requests), then set up the strings it uses.

.so hdfo.roff
.ds Ti Final Report\"
.ds Da 21 May 2023\"
.ti
On 5 August of last year,
this committee tasked me with the investigation of the
CFIT (controlled flight into terrain) incident of
.\" ...and so on...

A trap above the top or at or below the bottom of the page can be made visible by either moving it into the page area or increasing the page length so that the trap is on the page. Negative trap values always use the current page length; they are not converted to an absolute vertical position. We can use the ptr request to dump our page location traps to the standard error stream (see Debugging). Their positions are reported in basic units; an nroff device example follows.

.pl 5i
.wh -1i xx
.ptr
    error→ xx      -240
.pl 100i
.ptr
    error→ xx      -240

It is possible to have more than one trap at the same location (although only one at a time can be visible); to achieve this, the traps must be defined at different locations, then moved to the same place with the ch request. In the following example, the many empty lines caused by the bp request are not shown in the output.

.de a
.  nop a
..
.de b
.  nop b
..
.de c
.  nop c
..
.
.wh 1i a
.wh 2i b
.wh 3i c
.bp
    ⇒ a b c
.ch b 1i
.ch c 1i
.bp
    ⇒ a
.ch a 0.5i
.bp
    ⇒ a b
Register: \n[.t]

The read-only register .t holds the distance to the next vertical position trap. If there are no traps between the current position and the bottom of the page, it contains the distance to the page bottom. Within a diversion, in the absence of a diversion trap, this distance is the largest representable integer in basic units—effectively infinite.

Request: .ch name [dist]

Change the location of a trap by moving macro name to new location dist, or by unplanting it altogether if dist is absent. The default scaling unit is ‘v’. Parameters to ch are specified in the opposite order from wh. If name is the earliest planted macro of multiple traps at the same location, (re)moving it from that location exposes the macro next least recently planted at the same place.108

Changing a trap’s location is useful for building up footnotes in a diversion to allow more space at the bottom of the page for them.

The same macro can be installed simultaneously at multiple locations; however, only the earliest-planted instance—that has not yet been deleted with wh—will be moved by ch. The following example (using an nroff device) illustrates this behavior. Blank lines have been elided from the output.

.de T
Trap sprung at \\n(nlu.
.br
..
.wh 1i T
.wh 2i T
foo
.sp 11i
.bp
.ch T 4i
bar
.sp 11i
.bp
.ch T 5i
baz
.sp 11i
.bp
.wh 5i
.ch T 6i
qux
.sp 11i
    ⇒ foo
    ⇒ Trap sprung at 240u.
    ⇒ Trap sprung at 480u.
    ⇒ bar
    ⇒ Trap sprung at 480u.
    ⇒ Trap sprung at 960u.
    ⇒ baz
    ⇒ Trap sprung at 480u.
    ⇒ Trap sprung at 1200u.
    ⇒ qux
    ⇒ Trap sprung at 1440u.
Register: \n[.ne]

The read-only register .ne contains the amount of space that was needed in the last ne request that caused a trap to be sprung; it is useful in conjunction with the .trunc register. See Page Control. Since the .ne register is set only by traps, it doesn’t make sense to interpolate it outside of macros called by traps.

Register: \n[.trunc]

A read-only register containing the amount of vertical space truncated from an sp request by the most recently sprung vertical position trap, or, if the trap was sprung by an ne request, minus the amount of vertical motion produced by the ne request. In other words, at the point a trap is sprung, it represents the difference of what the vertical position would have been but for the trap, and what the vertical position actually is. Since the .trunc register is set only by traps, it doesn’t make sense to interpolate it outside of macros called by traps.

Register: \n[.pe]

This Boolean-valued, read-only register interpolates 1 while a page is being ejected, and 0 otherwise.

In the following example, we plant the same trap at the top and the bottom of the page. We also make the trap report its name and the vertical drawing position.

.de T
.tm \\$0: page \\n%, nl=\\n[nl] .pe=\\n[.pe]
..
.ll 46n
.wh 0 T
.wh -1v T
Those who can make you believe absurdities can make you
commit atrocities. \[em] Voltaire
    error→ T: page 1, nl=0 .pe=0
    error→ T: page 1, nl=2600 .pe=1
    ⇒ Those who can make you believe absurdities can
    ⇒ make you commit atrocities. -- Voltaire

When designing macros, keep in mind that diversions and traps do normally interact. For example, if a trap calls a header macro (while outputting a diversion) that tries to change the font on the current page, the effect is not visible before the diversion has completely been printed (except for input protected with \! or \?) since the data in the diversion is already formatted. In most cases, this is not the expected behaviour.


Next: , Previous: , Up: Vertical Position Traps   [Contents][Index]

5.28.1.2 The Implicit Page Trap

If, after starting GNU troff without loading a macro package, you use the ptr request to dump a list of the active traps to the standard error stream,109 nothing is reported. Yet the .t register will report a steadily decreasing value with every output line your document produces, and once the value of .t gets to within .V of zero, you will notice that something trap-like happens—the page is ejected, a new one begins, and the value of .t becomes large once more.

This implicit page trap always exists in the top-level diversion;110 it works like a trap in some ways but not others. Its purpose is to eject the current page and start the next one. It has no name, so it cannot be moved or deleted with wh or ch requests. You cannot hide it by placing another trap at its location, and can move it only by redefining the page length with pl. Its operation is suppressed when vertical page traps are disabled with GNU troff’s vpt request.


Next: , Previous: , Up: Vertical Position Traps   [Contents][Index]

5.28.1.3 Diversion Traps

A diversion is not formatted in the context of a page, so it lacks page location traps; instead it can have a diversion trap. There can exist at most one such vertical position trap per diversion.

Request: .dt [dist name]

Set a trap within a diversion at location dist, which is interpreted relative to diversion rather than page boundaries. If invoked with fewer than two arguments, any diversion trap in the current diversion is removed. The register .t works within diversions. It is an error to invoke dt in the top-level diversion. See Diversions.


Next: , Previous: , Up: Traps   [Contents][Index]

5.28.2 Input Line Traps

Request: .it [n name]
Request: .itc [n name]

Set an input line trap, calling macro name after processing the next n productive input lines (recall Manipulating Filling and Adjustment). Any existing input line trap in the environment is replaced. Without arguments, it and itc clear any input line trap that has not yet sprung.

Consider a macro ‘.ST s n’ which sets the next n input lines in the font style s.

.de ST \" Use style $1 for next $2 text lines.
.  it \\$2 ES
.  ft \\$1
..
.de ES \" end ST
.  ft R
..
.ST I 1
oblique
face
.ST I 1
oblique\c
face
    ⇒ oblique face obliqueface  (second “face” upright)

Unlike the ce and rj requests, it counts lines interrupted with the \c escape sequence separately (see Line Continuation); itc does not. To see the difference, let’s change the previous example to use itc instead.


.  itc \\$2 ES
oblique face obliqueface  (second “face” oblique)

You can think of the ce and rj requests as implicitly creating an input line trap with itc that schedules a break when the trap is sprung.

.de BR
.  br
.  internal: disable centering-without-filling
..
.
.de ce
.  if \\n[.br] .br
.  itc \\$1 BR
.  internal: enable centering-without-filling
..

Let us consider in more detail the sorts of input lines that are or are not “productive”.

.de Trap
TRAP SPRUNG
..
.de Mac
.if r a \l'5n'
..
.it 2 Trap
.
foo
.Mac
bar
baz
.it 1 Trap
.sp \" moves, but does not write or draw
qux
.itc 1 Trap
\h'5n'\c \" moves, but does not write or draw
jat

When ‘Trap’ gets called depends on whether the ‘a’ register is defined; the control line with the if request may or may not produce written output. We also see that the spacing request sp, while certainly affecting the output, does not spring the input line trap. Similarly, the horizontal motion escape sequence \h also affected the output, but was not “written”. Observe that we had to follow it with \c and use itc to prevent the newline at the end of the text line from causing a word break, which, like an ordinary space character, counts as written output.

$ groff -Tascii input-trap-example.groff
    ⇒ foo bar TRAP SPRUNG baz
    ⇒
    ⇒ qux TRAP SPRUNG      jat TRAP SPRUNG
$ groff -Tascii -ra1 input-trap-example.groff
    ⇒ foo _____ TRAP SPRUNG bar baz
    ⇒
    ⇒ qux TRAP SPRUNG      jat TRAP SPRUNG

Input line traps are associated with the environment (see Environments); switching to another environment suspends the current input line trap, and going back resumes it, restoring the count of qualifying lines enumerated in that environment.


Next: , Previous: , Up: Traps   [Contents][Index]

5.28.3 Blank Line Traps

Request: .blm [name]

Set a blank line trap, calling the macro name when GNU troff encounters a blank line in an input file, instead of the usual behavior (see Breaking). A line consisting only of spaces is also treated as blank and subject to this trap. If no argument is supplied, the default blank line behavior is (re-)established.


Next: , Previous: , Up: Traps   [Contents][Index]

5.28.4 Leading Space Traps

Request: .lsm [name]
Register: \n[lsn]
Register: \n[lss]

Set a leading space trap, calling the macro name when GNU troff encounters leading spaces in an input line; the implicit line break that normally happens in this case is suppressed. If no argument is supplied, the default leading space behavior is (re-)established (see Breaking).

The count of leading spaces on an input line is stored in register lsn, and the amount of corresponding horizontal motion in register lss, irrespective of whether a leading space trap is set. When it is, the leading spaces are removed from the input line, and no motion is produced before calling name.


Previous: , Up: Traps   [Contents][Index]

5.28.5 End-of-input Traps

Request: .em [name]

Set a trap at the end of input, calling macro name after the last line of the last input file has been processed. If no argument is given, any existing end-of-input trap is removed.

For example, if the document had to have a section at the bottom of the last page for someone to approve it, the em request could be used.

.de approval
\c
.  ne 3v
.  sp (\\n[.t]u - 3v)
.  in +4i
.  lc _
.  br
Approved:\t\a
.  sp
Date:\t\t\a
..
.
.em approval

The \c in the above example needs explanation. For historical reasons (compatibility with AT&T troff), the end-of-input macro exits as soon as it causes a page break if no partially collected line remains.111

Let us assume that there is no \c in the above approval macro, that the page is full, and last output line has been broken with, say, a br request. Because there is no more room, a ne request at this point causes a page ejection, which in turn makes troff exit immediately as just described. In most situations, this is not desired; people generally want to format the input after ne.

To force processing of the whole end-of-input macro independently of this behavior, it is thus advisable to (invisibly) ensure the existence of a partially collected line (\c) whenever there is a chance that a page break can happen. In the above example, invoking the ne request ensures that there is room for the subsequent formatted output on the same page, so we need insert \c only once.

The next example shows how to append three lines, then start a new page unconditionally. Since ‘.ne 1 doesn’t give the desired effect—there is always one line available or we are already at the beginning of the next page—we temporarily increase the page length by one line so that we can use ‘.ne 2.

.de EM
.pl +1v
\c
.ne 2
line one
.br
\c
.ne 2
line two
.br
\c
.ne 2
line three
.br
.pl -1v
\c
'bp
..
.em EM

This specific feature affects only the first potential page break caused by the end-of-input macro; further page breaks emitted by the macro are handled normally.

Another possible use of the em request is to make GNU troff emit a single large page instead of multiple pages. For example, one may want to produce a long plain text file for reading in a terminal or emulator without page footers and headers interrupting the body of the document. One approach is to set the page length at the beginning of the document to a very large value to hold all the text,112 and automatically adjust it to the exact height of the document after the text has been output.

.de adjust-page-length
.  br
.  pl \\n[nl]u \" \n[nl]: current vertical position
..
.
.de single-page-mode
.  pl 99999
.  em adjust-page-length
..
.
.\" Activate the above code if configured.
.if \n[do-continuous-rendering] \
.  single-page-mode

Since only one end-of-input trap exists and another macro package may already use it, care must be taken not to break the mechanism. A simple solution would be to append the above macro to the macro package’s end-of-input macro using the am request.


Next: , Previous: , Up: GNU troff Reference   [Contents][Index]

5.29 Diversions

In roff systems it is possible to format text as if for output, but instead of writing it immediately, one can divert the formatted text into a named storage area. It is retrieved later by specifying its name after a control character. The same name space is used for such diversions as for strings and macros; see Identifiers. Such text is sometimes said to be “stored in a macro”, but this coinage obscures the important distinction between macros and strings on one hand and diversions on the other; the former store unformatted input text, and the latter capture formatted output. Diversions also do not interpret arguments. Applications of diversions include “keeps” (preventing a page break from occurring at an inconvenient place by forcing a set of output lines to be set as a group), footnotes, tables of contents, and indices. For orthogonality it is said that GNU troff is in the top-level diversion if no diversion is active (that is, formatted output is being “diverted” immediately to the output device).

Dereferencing an undefined diversion will create an empty one of that name and cause a warning in category ‘mac’ to be emitted. See Warnings, for information about the enablement and suppression of warnings. A diversion does not exist for the purpose of testing with the d conditional operator until its initial definition ends (see Operators in Conditionals). The following requests are used to create and alter diversions.

Request: .di [name]
Request: .da [name]

Start collecting formatted output in a diversion called name. The da request appends to a diversion called name, creating it if necessary. If name already exists as an alias, the target of the alias is replaced or appended to; recall Strings. The pending output line is diverted as well. Switching to another environment (with the ev request) before invoking di or da avoids including any pending output line in the diversion; see Environments.

Invoking di or da without an argument stops diverting output to the diversion named by the most recent corresponding request. If di or da is called without an argument when there is no current diversion, a warning in category ‘di’ is produced. See Warnings, for information about the enablement and suppression of warnings.

Before the diversion.
.di yyy
In the diversion.
.br
.di
After the diversion.
.br
    ⇒ After the diversion.
.yyy
    ⇒ Before the diversion.  In the diversion.

GNU troff supports box requests to exclude a partially collected line from a diversion, as this is often desirable.

Request: .box [name]
Request: .boxa [name]

Divert (or append) output to name, similarly to the di and da requests, respectively. Any pending output line is not included in the diversion. Without an argument, stop diverting output; any pending output line inside the diversion is discarded.

Before the box.
.box xxx
In the box.
.br
Hidden treasure.
.box
After the box.
.br
    ⇒ Before the box.  After the box.
.xxx
    ⇒ In the box.

Apart from pending output line inclusion and the request names that populate them, boxes are handled exactly as diversions are. All of the following groff language elements can be used with them interchangeably.

Register: \n[.z]
Register: \n[.d]

Diversions may be nested. The read-only string-valued register .z contains the name of the current diversion. The read-only register .d contains the current vertical place in the diversion. If the input text is not being diverted, .d reports the same location as the register nl.

Register: \n[.h]

The read-only register .h stores the high-water mark on the current page or in the current diversion. It corresponds to the text baseline of the lowest line on the page.113

.tm .h==\n[.h], nl==\n[nl]
    ⇒ .h==0, nl==-1
This is a test.
.br
.sp 2
.tm .h==\n[.h], nl==\n[nl]
    ⇒ .h==40, nl==120

As implied by the example, vertical motion does not produce text baselines and thus does not increase the value interpolated by ‘\n[.h]’.

Register: \n[dn]
Register: \n[dl]

After completing a diversion, the writable registers dn and dl contain its vertical and horizontal sizes. Only the lines just processed are counted: for the computation of dn and dl, the requests da and boxa are handled as if di and box had been used, respectively—lines that have been already stored in the diversion (box) are not taken into account.

.\" Center text both horizontally and vertically.
.\" Macro .(c starts centering mode; .)c terminates it.
.
.\" Disable the escape character with .eo so that we
.\" don't have to double backslashes on the "\n"s.
.eo
.de (c
.  br
.  ev (c
.  evc 0
.  in 0
.  nf
.  di @c
..
.de )c
.  br
.  ev
.  di
.  nr @s (((\n[.t]u - \n[dn]u) / 2u) - 1v)
.  sp \n[@s]u
.  ce 1000
.  @c
.  ce 0
.  sp \n[@s]u
.  br
.  fi
.  rr @s
.  rm @c
..
.ec
Escape sequence: \!anything
Escape sequence: \?anything\?

Transparently embed anything into the current diversion, preventing requests, macro calls, and escape sequences from being interpreted when read into a diversion. This is useful for preventing them from taking effect until the diverted text is actually output. The \! escape sequence transparently embeds input up to and including the end of the line. The \? escape sequence transparently embeds input until its own next occurrence.

anything may not contain newlines; use \! by itself to embed newlines in a diversion. The escape sequence \? is also recognized in copy mode and turned into a single internal code; it is this code that terminates anything. Thus the following example prints 4.

.nr x 1
.nf
.di d
\?\\?\\\\?\\\\\\\\nx\\\\?\\?\?
.di
.nr x 2
.di e
.d
.di
.nr x 3
.di f
.e
.di
.nr x 4
.f

Both escape sequences read the data in copy mode.

If \! is used in the top-level diversion, its argument is directly embedded into GNU troff’s intermediate output. This can be used, for example, to control a postprocessor that processes the data before it is sent to an output driver.

The \? escape used in the top-level diversion produces no output at all; its argument is simply ignored.

Request: .output contents

Emit contents directly to GNU troff’s intermediate output (subject to copy mode interpretation); this is similar to \! used at the top level. An initial neutral double quote in contents is stripped to allow embedding of leading spaces.

This request can’t be used before the first page has started—if you get an error, simply insert .br before the output request.

Use with caution! It is normally only needed for mark-up used by a postprocessor that does something with the output before sending it to the output device, filtering out contents again.

Request: .asciify div

Unformat the diversion div in a way such that Unicode basic Latin (ASCII) characters, characters translated with the trin request, space characters, and some escape sequences, that were formatted and diverted into div are treated like ordinary input characters when div is reread. Doing so can be useful in conjunction with the writem request. asciify can be also used for gross hacks; for example, the following sets register n to 1.

.tr @.
.di x
@nr n 1
.br
.di
.tr @@
.asciify x
.x

asciify cannot return all items in a diversion to their source equivalent: nodes such as those produced by the \N escape sequence will remain nodes, so the result cannot be guaranteed to be a pure string. See Copy Mode. Glyph parameters such as the type face and size are not preserved; use unformat to achieve that.

Request: .unformat div

Like asciify, unformat the diversion div. However, unformat handles only tabs and spaces between words, the latter usually arising from spaces or newlines in the input. Tabs are treated as input tokens, and spaces become adjustable again. The vertical sizes of lines are not preserved, but glyph information (font, type size, space width, and so on) is retained.


Next: , Previous: , Up: GNU troff Reference   [Contents][Index]

5.30 Punning Names

Macros, strings, and diversions share a name space; recall Identifiers. Internally, the same mechanism is used to store them. You can thus call a macro with string interpolation syntax and vice versa.

.de subject
Typesetting
..
.de predicate
rewards attention to detail
..
\*[subject] \*[predicate].
Truly.
    ⇒ Typesetting
    ⇒  rewards attention to detail Truly.

What went wrong? Strings don’t contain newlines, but macros do. String interpolation placed a newline at the end of ‘\*[subject]’, and the next thing on the input was a space. Then when ‘\*[predicate]’ was interpolated, it was followed by the empty request ‘.’ on a line by itself. If we want to use macros as strings, we must take interpolation behavior into account.

.de subject
Typesetting\\
..
.de predicate
rewards attention to detail\\
..
\*[subject] \*[predicate].
Truly.
    ⇒ Typesetting rewards attention to detail.  Truly.

By ending each text line of the macros with an escaped \RET, we get the desired effect (see Line Continuation).114 What would have happened if we had used only one backslash at a time instead?

Interpolating a string does not hide existing macro arguments. We can also place the escaped newline outside the string interpolation instead of within the string definition. Thus, in a macro, a more efficient way of doing

.xx \\$@

is

\\*[xx]\\

The latter calling syntax doesn’t change the value of \$0, which is then inherited from the calling macro (see Parameters).

Diversions can be also called with string syntax. It is sometimes convenient to copy one-line diversions to a string.

.di xx
the
.ft I
interpolation system
.ft
.br
.di
.ds yy This is a test of \*(xx\c
\*(yy.
    ⇒ This is a test of the interpolation system.

As the previous example shows, it is possible to store formatted output in strings. The \c escape sequence prevents the subsequent newline from being interpreted as a break (again, see Line Continuation).

Copying multi-output line diversions produces unexpected results.

.di xxx
a funny
.br
test
.br
.di
.ds yyy This is \*[xxx]\c
\*[yyy].
    ⇒ test This is a funny.

Usually, it is not predictable whether a diversion contains one or more output lines, so this mechanism should be avoided. With AT&T troff, this was the only solution to strip off a final newline from a diversion. Another disadvantage is that the spaces in the copied string are already formatted, preventing their adjustment. This can cause ugly results.

A clean solution to this problem is available in GNU troff, using the requests chop to remove the final newline of a diversion, and unformat to make the horizontal spaces adjustable again.

.box xxx
a funny
.br
test
.br
.box
.chop xxx
.unformat xxx
This is \*[xxx].
    ⇒ This is a funny test.

See Gtroff Internals.


Next: , Previous: , Up: GNU troff Reference   [Contents][Index]

5.31 Environments

As discussed in Deferring Output, environments store most of the parameters that determine the appearance of text. A default environment named ‘0’ exists when GNU troff starts up; it is modified by formatting-related requests and escape sequences.

You can create new environments and switch among them. Only one is current at any given time. Active environments are managed using a stack, a data structure supporting “push” and “pop” operations. The current environment is at the top of the stack. The same environment name can be pushed onto the stack multiple times, possibly interleaved with others. Popping the environment stack does not destroy the current environment; it remains accessible by name and can be made current again by pushing it at any time. Environments cannot be renamed or deleted, and can only be modified when current. To inspect the environment stack, use the pev request; see Debugging.

Environments store the following information.

Request: .ev [ident]
Register: \n[.ev]

Enter the environment ident, which is created if it does not already exist, using the same parameters as for the default environment used at startup. With no argument, GNU troff switches to the previous environment.

Invoking ev with an argument puts environment ident onto the top of the environment stack. (If it isn’t already present in the stack, this is a proper push.) Without an argument, ev pops the environment stack, making the previous environment current. It is an error to pop the environment stack with no previous environment available. The read-only string-valued register .ev contains the name of the current environment—the one at the top of the stack.

.ev footnote-env
.fam N
.ps 6
.vs 8
.ll -.5i
.ev



.ev footnote-env
\[dg] Observe the smaller text and vertical spacing.
.ev

We can familiarize ourselves with stack behavior by wrapping the ev request with a macro that reports the contents of the .ev register to the standard error stream.

.de EV
.  ev \\$1
.  tm environment is now \\n[.ev]
..
.
.EV foo
.EV bar
.EV
.EV baz
.EV
.EV
.EV
    error→ environment is now foo
    error→ environment is now bar
    error→ environment is now foo
    error→ environment is now baz
    error→ environment is now foo
    error→ environment is now 0
    error→ error: environment stack underflow
    error→ environment is now 0
Request: .evc environment

Copy the contents of environment to the current environment.

The following environment data are not copied.

Register: \n[.w]
Register: \n[.cht]
Register: \n[.cdp]
Register: \n[.csk]

The \n[.w] register contains the width of the last glyph formatted in the environment.

The \n[.cht] register contains the height of the last glyph formatted in the environment.

The \n[.cdp] register contains the depth of the last glyph formatted in the environment. It is positive for glyphs extending below the baseline.

The \n[.csk] register contains the skew (how far to the right of the glyph’s center that GNU troff should place an accent) of the last glyph formatted in the environment.

Register: \n[.n]

The \n[.n] register contains the length of the previous output line emitted in the environment.


Next: , Previous: , Up: GNU troff Reference   [Contents][Index]

5.32 Suppressing Output

Escape sequence: \O[num]

Suppress GNU troff output of glyphs and geometric objects. The sequences \O2, \O3, \O4, and \O5 are intended for internal use by grohtml.

\O0

Disable the emission of glyphs and geometric objects to the output driver, provided that this sequence occurs at the outermost suppression level (see \O3 and \04 below). Horizontal motions corresponding to non-overstruck glyph widths still occur.

\O1

Enable the emission of glyphs and geometric objects to the output driver, provided that this sequence occurs at the outermost suppression level.

\O0 and \O1 also reset the four registers opminx, opminy, opmaxx, and opmaxy to -1. These four registers mark the top left and bottom right hand corners of a box encompassing all written or drawn output.

\O2

At the outermost suppression level, enable emission of glyphs and geometric objects, and write to the standard error stream the page number and values of the four aforementioned registers encompassing glyphs written since the last interpolation of a \O sequence, as well as the page offset, line length, image file name (if any), horizontal and vertical device motion quanta, and input file name. Numeric values are in basic units.

\O3

Begin a nested suppression level. grohtml uses this mechanism to create images of output preprocessed with gpic, geqn, and gtbl. At startup, GNU troff is at the outermost suppression level. pre-grohtml generates these sequences when processing the document, using GNU troff with the ps output device, Ghostscript, and the PNM tools to produce images in PNG format. They start a new page if the device is not html or xhtml, to reduce the number of images crossing a page boundary.

\O4

End a nested suppression level.

\O[5Pfile]

At the outermost suppression level, write the name file to the standard error stream at position P, which must be one of l, r, c, or i, corresponding to left, right, centered, and inline alignments within the document, respectively. file is a name associated with the production of the next image.

Register: \n[.O]

Output suppression nesting level applied by \O3 and \O4 escape sequences.


Next: , Previous: , Up: GNU troff Reference   [Contents][Index]

5.33 I/O

gtroff has several requests for including files:

Request: .so file
Request: .soquiet file

Replace the so request’s control line with the contents of the file named by the argument, “sourcing” it. file is sought in the directories specified by -I command-line option. If file does not exist, a warning in category ‘file’ is produced and the request has no further effect. See Warnings, for information about the enablement and suppression of warnings.

so can be useful for large documents; e.g., allowing each chapter of a book to be kept in a separate file. However, files interpolated with so are not preprocessed; to overcome this limitation, see the gsoelim(1) man page.

Since GNU troff replaces the entire control line with the contents of a file, it matters whether file is terminated with a newline or not. Assume that file xxx contains only the word ‘foo’ without a trailing newline.

$ printf 'foo' > xxx

The situation is
.so xxx
bar.
    ⇒ The situation is foobar.

soquiet works the same way, except that no warning diagnostic is issued if file does not exist.

Request: .pso command

Read the standard output from the specified command and include it in place of the pso request.

It is an error to use this request in safer mode, which is the default. Invoke GNU troff or a front end with the -U option to enable unsafe mode.

The comment regarding a final newline for the so request is valid for pso also.

Request: .mso file
Request: .msoquiet file

Identical to the so and soquiet requests, respectively, except that gtroff searches for the specified file in the same directories as macro files for the -m command-line option. If the file name to be included has the form name.tmac and it isn’t found, these requests try to include tmac.name and vice versa.

Request: .trf file
Request: .cf file

Transparently output the contents of file. Each line is output as if it were preceded by \!; however, the lines are not subject to copy mode interpretation. If the file does not end with a newline, trf adds one. Both requests cause a break.

When used in a diversion, these requests embed a node (see Gtroff Internals) in it that, when reread, causes the contents of file to be transparently copied to the output. In AT&T troff, the contents of file are immediately copied to the output regardless of whether there is a current diversion; this behaviour is so anomalous that it must be considered a bug.

While cf copies the contents of file completely unprocessed, trf disallows characters such as NUL that are not valid gtroff input characters (see Identifiers).

For cf, within a diversion, “completely unprocessed” means that each line of a file to be inserted is handled as if it were preceded by \!\\!.

To define a macro x containing the contents of file f, use

.ev 1
.di x
.trf f
.di
.ev

The calls to ev prevent the partially collected output line from becoming part of the diversion (see Diversions).

Request: .nx [file]

Force gtroff to continue processing of the file specified as an argument. If no argument is given, immediately jump to the end of file.

Request: .rd [prompt [arg1 arg2 …]]

Read from standard input, and include what is read as though it were part of the input file. Text is read until a blank line is encountered.

If standard input is a TTY input device (keyboard), write prompt to standard error, followed by a colon (or send BEL for a beep if no argument is given).

Arguments after prompt are available for the input. For example, the line

.rd data foo bar

with the input ‘This is \$2. prints

This is bar.

Using the nx and rd requests, it is easy to set up form letters. The form letter template is constructed like this, putting the following lines into a file called repeat.let:

.ce
\*(td
.sp 2
.nf
.rd
.sp
.rd
.fi
Body of letter.
.bp
.nx repeat.let

When this is run, a file containing the following lines should be redirected in. Requests included in this file are executed as though they were part of the form letter. The last block of input is the ex request, which tells GNU troff to stop processing. If this were not there, troff would not know when to stop.

Trent A. Fisher
708 NW 19th Av., #202
Portland, OR  97209

Dear Trent,

Len Adollar
4315 Sierra Vista
San Diego, CA  92103

Dear Mr. Adollar,

.ex
Request: .pi pipe

Pipe the output of gtroff to the shell command(s) specified by pipe. This request must occur before gtroff has a chance to print anything.

It is an error to use this request in safer mode, which is the default. Invoke GNU troff or a front end with the -U option to enable unsafe mode.

Multiple calls to pi are allowed, acting as a chain. For example,

.pi foo
.pi bar
...

is the same as ‘.pi foo | bar.

The intermediate output format of GNU troff is piped to the specified commands. Consequently, calling groff without the -Z option normally causes a fatal error.

Request: .sy cmds
Register: \n[systat]

Execute the shell command(s) specified by cmds. The output is not saved anywhere, so it is up to the user to do so.

It is an error to use this request in safer mode; this is the default. Give GNU troff or a front end program the -U option to enable unsafe mode.

The following code fragment introduces the current time into a document.

.sy perl -e 'printf ".nr H %d\\n.nr M %d\\n.nr S %d\\n",\
             (localtime(time))[2,1,0]' > /tmp/x\n[$$]
.so /tmp/x\n[$$]
.sy rm /tmp/x\n[$$]
\nH:\nM:\nS

This works by having the Perl script (run by sy) write nr requests that set the registers H, M, and S to a temporary file. The roff document then reads the temporary file using the so request.

The registers seconds, minutes, and hours, initialized at startup of GNU troff, should satisfy most requirements. Use the af request to format their values for output.

.af hours 00
.af minutes 00
.af seconds 00
\n[hours]:\n[minutes]:\n[seconds]
    ⇒ 02:17:54

The writable register systat contains the return value of the system() function executed by the last sy request.

Request: .open stream file
Request: .opena stream file

Open the specified file for writing and associates the specified stream with it.

The opena request is like open, but if the file exists, append to it instead of truncating it.

It is an error to use these requests in safer mode; this is the default. Give GNU troff or a front end program the -U option to enable unsafe mode.

Request: .write stream data
Request: .writec stream data

Write to the file associated with the specified stream. The stream must previously have been the subject of an open request. The remainder of the line is interpreted as the ds request reads its second argument: an initial neutral double quote in contents is stripped to allow embedding of leading spaces, and it is read in copy mode.

The writec request is like write, but only write appends a newline to the data.

Request: .writem stream xx

Write the contents of the macro or string xx to the file associated with the specified stream.

xx is read in copy mode, i.e., already formatted elements are ignored. Consequently, diversions must be unformatted with the asciify request before calling writem. Usually, this means a loss of information.

Request: .close stream

Close the specified stream; the stream is no longer an acceptable argument to the write request.

Here a simple macro to write an index entry.

.open idx test.idx
.
.de IX
.  write idx \\n[%] \\$*
..
.
.IX test entry
.
.close idx
Escape sequence: \Ve
Escape sequence: \V(ev
Escape sequence: \V[env]

Interpolate the contents of the specified environment variable env (one-character name e, two-character name ev) as returned by the function getenv(3). \V is interpreted even in copy mode (see Copy Mode).


Next: , Previous: , Up: GNU troff Reference   [Contents][Index]

5.34 Postprocessor Access

Two escape sequences and two requests enable documents to pass information directly to a postprocessor. These are useful for exercising device-specific capabilities that the groff language does not abstract or generalize; examples include the embedding of hyperlinks and image files. Device-specific functions are documented in each output driver’s man page, such as gropdf(1), grops(1), or grotty(1).

Request: .device xxx
Escape sequence: \X'xxx '

Embed all xxx arguments into GNU troff output as parameters to a device control command ‘x X. The meaning and interpretation of such parameters is determined by the output driver or other postprocessor.

The device request processes its arguments in copy mode (see Copy Mode). An initial neutral double quote in contents is stripped to allow embedding of leading spaces. By contrast, within \X arguments, the escape sequences \&, \), \%, and \: are ignored; \SP and \~ are converted to single space characters; and \\ has its escape character stripped. So that the basic Latin subset of the Unicode character set115 can be reliably encoded in device control commands, seven special character escape sequences (‘\-’, ‘\[aq]’, ‘\[dq]’, ‘\[ga]’, ‘\[ha]’, ‘\[rs]’, and ‘\[ti]’,) are mapped to basic Latin characters; see the groff_char(7) man page. For this transformation, character translations and special character definitions are ignored.116 The use of any other escape sequence in \X parameters is normally an error.

If the use_charnames_in_special directive appears in the output device’s DESC file, the use of special character escape sequences is not an error; they are simply output verbatim (with the exception of the seven mapped to Unicode basic Latin characters, discussed above). use_charnames_in_special is currently employed only by grohtml.

Request: .devicem name
Escape sequence: \Yn
Escape sequence: \Y(nm
Escape sequence: \Y[name]

This is approximately equivalent to ‘\X'\*[name]'’ (one-character name n, two-character name nm). However, the contents of the string or macro name are not interpreted; also it is permitted for name to have been defined as a macro and thus contain newlines (it is not permitted for the argument to \X to contain newlines). The inclusion of newlines requires an extension to the AT&T troff output format, and confuses drivers that do not know about this extension (see Device Control Commands).

Request: .tag name
Request: .taga name

Reserved for internal use.


Next: , Previous: , Up: GNU troff Reference   [Contents][Index]

5.35 Miscellaneous

We document here GNU troff features that fit poorly elsewhere.

Request: .nm [start [increment [space [indentation]]]]
Register: \n[ln]
Register: \n[.nm]

Begin (or, with no arguments, cease) numbering output lines. start assigns the number of the next output line. Only line numbers divisible by increment are marked (default: ‘1’). space configures the horizontal spacing between the number and the text (default: ‘1’). Any given indentation is applied to the numbers (default: ‘0’). The third and fourth arguments are reckoned in numeral widths (\0). start must be non-negative and increment positive.

The formatter aligns the number to the right in a width of three numeral spaces plus indentation, then catenates space and the output line. The line length is not reduced. Depending on the value of the page offset,117 numbers wider than the allocated space protrude into the left margin, or shift the output line to the right.

Line numbering parameters corresponding to missing arguments are not altered. After numbering is disabled, ‘.nm +0’ resumes it using the previously active parameters.

The parameters of nm are associated with the environment (see Environments).

While numbering is enabled, the output line number register ln is updated as each line is output, even if no line number is formatted with it because it is being skipped (it is not a multiple of increment) or because numbering is suppressed (see the nn request below).

The .nm register tracks the enablement status of numbering. Temporary suspension of numbering with the nn request does not alter its value.

.po 5n
.ll 44n
Programming,
when stripped of all its circumstantial irrelevancies,
.nm 999 1 1 -4
boils down to no more and no less than
.nm +0 3
very effective thinking so as to avoid unmastered
.nn 2
complexity,
to very vigorous separation of your many
different concerns.
.br
\(em Edsger Dijkstra
.sp
.nm 1 1 1
This guy's arrogance takes your breath away.
.br
\(em John Backus
    ⇒      Programming,  when  stripped of all its cir-
    ⇒  999 cumstantial irrelevancies, boils down to  no
    ⇒      more  and no less than very effective think-
    ⇒      ing so as to avoid unmastered complexity, to
    ⇒      very vigorous separation of your  many  dif-
    ⇒      ferent concerns.
    ⇒ 1002 -- Edsger Dijkstra
    ⇒
    ⇒    1 This guy’s arrogance takes your breath away.
    ⇒    2 -- John Backus
Request: .nn [skip]
Register: \n[.nn]

Suppress numbering of the next skip output lines that would otherwise be numbered. The default is 1. nn can be invoked when line numbering is not active; suppression of numbering will take effect for skip lines once nm enables it.

The .nn register stores the count of output lines still to have their numbering suppressed.

This count is associated with the environment (see Environments).

To test whether the current output line will be numbered, you must check both the .nm and .nn registers.

  .de is-numbered
  .  nop This line
  .  ie (\\n[.nm] & (1-\\n[.nn])) IS
  .  el                           ISN'T
  .  nop numbered.
  .  br
  ..
  Test line numbering.
  .is-numbered
  .nm 1
  .nn 1
  .is-numbered
  .is-numbered
  .nm
  .is-numbered
    ⇒ Test line numbering.  This line ISN’T numbered.
    ⇒ This line ISN’T numbered.
    ⇒   1 This line IS numbered.
    ⇒ This line ISN’T numbered.
Request: .mc [margin-character [distance]

Begin (or, with no arguments, cease) writing a margin-character to the right of each output line. The distance argument separates margin-character from the right margin. If absent, the most recent value is used; the default is 10 points. If an output line exceeds the line length, the margin character is appended to it. No margin character is written on lines produced by the tl request.

The margin character is a property of the output line; the margin character last configured when the line is output controls. If the margin character is disabled before an output line breaks, none is output (but see below).

The margin character is associated with the environment (see Environments).

.ll 5i
.nf
.mc \[br]
This paragraph is marked with a margin character.
.sp
As seen above, vertical space isn't thus marked.
\&
An output line that is present, but empty, is.
    ⇒ This paragraph is marked with a margin character.  |
    ⇒
    ⇒ As seen above, vertical space isn’t thus marked.   |
    ⇒                                                    |
    ⇒ An output line that is present, but empty, is.     |

For compatibility with AT&T troff, a call to mc to set the margin character can’t be undone immediately; at least one line gets a margin character.

.ll 10n
.nf
.mc |
.mc *
.mc
foo
bar
    ⇒ foo        *
    ⇒ bar

The margin character mechanism is commonly used to annotate changes in documents. The groff distribution ships a program, gdiffmk, to assist with this task.118

Request: .psbb file
Register: \n[llx]
Register: \n[lly]
Register: \n[urx]
Register: \n[ury]

Retrieve the bounding box of the PostScript image found in file, which must conform to Adobe’s Document Structuring Conventions (DSC), locate a %%BoundingBox comment, and store the (upper-, lower-, -left, -right) values into the registers llx, lly, urx, and ury. If an error occurs (for example, if no %%BoundingBox comment is present), the formatter sets these registers to 0.

The search path for file can be controlled with the -I command-line option.


Next: , Previous: , Up: GNU troff Reference   [Contents][Index]

5.36 gtroff Internals

gtroff processes input in three steps. One or more input characters are converted to an input token.119 Then, one or more input tokens are converted to an output node. Finally, output nodes are converted to the intermediate output language understood by all output devices.

Actually, before step one happens, gtroff converts certain escape sequences into reserved input characters (not accessible by the user); such reserved characters are used for other internal processing also – this is the very reason why not all characters are valid input. See Identifiers, for more on this topic.

For example, the input string ‘fi\[:u]’ is converted into a character token ‘f’, a character token ‘i’, and a special token ‘:u’ (representing u umlaut). Later on, the character tokens ‘f’ and ‘i’ are merged to a single output node representing the ligature glyph ‘fi’ (provided the current font has a glyph for this ligature); the same happens with ‘:u’. All output glyph nodes are ‘processed’, which means that they are invariably associated with a given font, font size, advance width, etc. During the formatting process, gtroff itself adds various nodes to control the data flow.

Macros, diversions, and strings collect elements in two chained lists: a list of input tokens that have been passed unprocessed, and a list of output nodes. Consider the following diversion.

.di xxx
a
\!b
c
.br
.di

It contains these elements.

node listtoken listelement number
line start node1
glyph node a2
word space node3
b4
\n5
glyph node c6
vertical size node7
vertical size node8
\n9

Elements 1, 7, and 8 are inserted by gtroff; the latter two (which are always present) specify the vertical extent of the last line, possibly modified by \x. The br request finishes the pending output line, inserting a newline input token, which is subsequently converted to a space when the diversion is reread. Note that the word space node has a fixed width that isn’t adjustable anymore. To convert horizontal space nodes back to input tokens, use the unformat request.

Macros only contain elements in the token list (and the node list is empty); diversions and strings can contain elements in both lists.

The chop request simply reduces the number of elements in a macro, string, or diversion by one. Exceptions are compatibility save and compatibility ignore input tokens, which are ignored. The substring request also ignores those input tokens.

Some requests like tr or cflags work on glyph identifiers only; this means that the associated glyph can be changed without destroying this association. This can be very helpful for substituting glyphs. In the following example, we assume that glyph ‘foo’ isn’t available by default, so we provide a substitution using the fchar request and map it to input character ‘x’.

.fchar \[foo] foo
.tr x \[foo]

Now let us assume that we install an additional special font ‘bar’ that has glyph ‘foo’.

.special bar
.rchar \[foo]

Since glyphs defined with fchar are searched before glyphs in special fonts, we must call rchar to remove the definition of the fallback glyph. Anyway, the translation is still active; ‘x’ now maps to the real glyph ‘foo’.

Macro and request arguments preserve compatibility mode enablement.

.cp 1     \" switch to compatibility mode
.de xx
\\$1
..
.cp 0     \" switch compatibility mode off
.xx caf\['e]
    ⇒ caf

Since compatibility mode is enabled while de is invoked, the macro xx enables compatibility mode when it is called. Argument $1 can still be handled properly because it inherits the compatibility mode enablement status that was active at the point where xx was called.

After interpolation of the parameters, the compatibility save and restore tokens are removed.


Next: , Previous: , Up: GNU troff Reference   [Contents][Index]

5.37 Debugging

Standard troff voodoo, just put a power of two backslashes in front of it until it works and if you still have problems add a \c. — Ron Natalie

GNU troff is not the easiest language to debug, in part thanks to its design features of recursive interpolation and the use of multi-stage pipeline processing in the surrounding system. Nevertheless there exist several features useful for troubleshooting.

Preprocessors use the lf request to preserve the identity of the line numbers and names of input files. GNU troff emits a variety of error diagnostics and supports several categories of warning; the output of these can be selectively suppressed. A trace of the formatter’s input processing stack can be emitted when errors or warnings occur by means of GNU troff’s -b option, or produced on demand with the backtrace request. The tm and related requests can be used to emit customized diagnostic messages or for instrumentation while troubleshooting. The ex and ab requests cause early termination with successful and error exit codes respectively, to halt further processing when continuing would be fruitless. Examine the state of the formatter with requests that write lists of defined names (macros, strings, and diversions), environments, registers, and page location traps to the standard error stream.

Request: .lf line [file]

Set the input line number (and, optionally, the file name) GNU troff shall use for error and warning messages. line is the input line number of the next line. Without an argument, the request is ignored.

lf’s primary purpose is to aid the debugging of documents that undergo preprocessing. Programs like tbl that transform input in their own languages into roff requests use it so that any diagnostic messages emitted by troff correspond to the source document.

Request: .tm message
Request: .tm1 message
Request: .tmc message

Send message, which consumes the remainder of the input line and cannot contain special characters, to the standard error stream, followed by a newline. Leading spaces in message are ignored.

tm1 is similar, but recognizes and strips a leading neutral double quote from message to allow the embedding of leading spaces.

tmc works as tm1, but does not append a newline.

Request: .ab [message]

Write any message to the standard error stream (like tm) and then abort GNU troff; that is, stop processing and terminate with a failure status.

Request: .ex

Exit GNU troff; that is, stop processing and terminate with a successful status. To stop processing only the current file, use the nx request; see I/O.

When doing something involved, it is useful to leave the debugging statements in the code and have them turned on by a command-line flag.

.if \n[DB] .tm debugging output

To activate such statements, use the -r option to set the register.

groff -rDB=1 file

If it is known in advance that there are many errors and no useful output, GNU troff can be forced to suppress formatted output with the -z option.

Request: .pev

Report the state of the current environment followed by that of all other environments to the standard error stream.

Request: .pm

Report, to the standard error stream, the names of all defined macros, strings, and diversions with their sizes in bytes.

Request: .pnr

Report the names and contents of all currently defined registers to the standard error stream.

Request: .ptr

Report the names and positions of all page location traps to the standard error stream. Empty slots in the list, where a trap has been planted but subsequently (re)moved, are printed as well.

Request: .fl

Instruct gtroff to flush its output immediately. The intent is for interactive use, but this behaviour is currently not implemented in gtroff. Contrary to Unix troff, TTY output is sent to a device driver also (grotty), making it non-trivial to communicate interactively.

This request causes a line break.

Request: .backtrace

Write the state of the input stack to the standard error stream.

Consider the following in a file test.

.de xxx
.  backtrace
..
.de yyy
.  xxx
..
.
.yyy
    error→ troff: backtrace: 'test':2: macro 'xxx'
    error→ troff: backtrace: 'test':5: macro 'yyy'
    error→ troff: backtrace: file 'test':8

The -b option of GNU troff causes a backtrace to be generated on each error or warning. Some warnings have to be enabled; See Warnings.

Register: \n[slimit]

If greater than 0, sets the maximum quantity of objects on GNU troff’s internal input stack. If less than or equal to 0, there is no limit: recursion can continue until program memory is exhausted. The default is 1,000.

Request: .warnscale su

Set the scaling unit used in certain warnings to su, which can take the values ‘u’, ‘i’, ‘c’, ‘p’, and ‘P’. The default is ‘i’.

Request: .spreadwarn [limit]

Emit a break warning if the additional space inserted for each space between words in an output line adjusted to both margins with ‘.ad b is larger than or equal to limit. A negative value is treated as zero; an absent argument toggles the warning on and off without changing limit. The default scaling unit is ‘m’. At startup, spreadwarn is inactive and limit is 3m.

For example,

.spreadwarn 0.2m

causes a warning if break warnings are not suppressed and gtroff must add 0.2m or more for each inter-word space in a line. See Warnings.

GNU troff has command-line options for reporting warnings (-w) and backtraces (-b) when a warning or an error occurs.

Request: .warn [n]
Register: \n[.warn]

Select the categories, or “types”, of reported warnings. n is the sum of the numeric codes associated with each warning category that is to be enabled; all other categories are disabled. The categories and their associated codes are listed in Warnings. For example, ‘.warn 0’ disables all warnings, and ‘.warn 1’ disables all warnings except those about missing glyphs. If no argument is given, all warning categories are enabled.

The read-only register .warn contains the sum of the numeric codes of enabled warning categories.


Previous: , Up: Debugging   [Contents][Index]

5.37.1 Warnings

Warning diagnostics emitted by GNU troff are divided into named, numbered categories. The name associated with each warning category is used by the -w and -W options. Each category is also assigned a power of two; the sum of enabled category values is used by the warn request and the .warn register.

Warnings of each category are produced under the following circumstances.

char
1

No mounted font defines a glyph for the requested character. This category is enabled by default.

number
2

An invalid numeric expression was encountered. This category is enabled by default. See Numeric Expressions.

break
4

A filled output line could not be broken such that its length was less than the output line length ‘\n[.l]’. This category is enabled by default.

delim
8

The closing delimiter in an escape sequence was missing or mismatched.

el
16

The el request was encountered with no prior corresponding ie request. See if-else.

scale
32

A scaling unit inappropriate to its context was used in a numeric expression.

range
64

A numeric expression was out of range for its context.

syntax
128

A self-contradictory hyphenation mode was requested; an empty or incomplete numeric expression was encountered; an operand to a numeric operator was missing; an attempt was made to define a recursive, empty, or nonsensical character class; or a groff extension conditional expression operator was used while in compatibility mode.

di
256

A di, da, box, or boxa request was invoked without an argument when there was no current diversion.

mac
512

An undefined string, macro, or diversion was used. When such an object is dereferenced, an empty one of that name is automatically created. So, unless it is later deleted, at most one warning is given for each.

This warning is also emitted upon an attempt to move an unplanted trap macro (see Page Location Traps). In such cases, the unplanted macro is not dereferenced, so it is not created if it does not exist.

reg
1024

An undefined register was used. When an undefined register is dereferenced, it is automatically defined with a value of 0. So, unless it is later deleted, at most one warning is given for each.

tab
2048

A tab character was encountered where a number was expected, or appeared in an unquoted macro argument.

right-brace
4096

A right brace escape sequence \} was encountered where a number was expected.

missing
8192

A request was invoked with a mandatory argument absent.

input
16384

An invalid character occurred on the input stream.

escape
32768

An unsupported escape sequence was encountered.

space
65536

A space was missing between a request or macro and its argument. This warning is produced when an undefined name longer than two characters is encountered and the first two characters of the name constitute a defined name. No request is invoked, no macro called, and an empty macro is not defined. This category is enabled by default. It never occurs in compatibility mode.

font
131072

A non-existent font was selected, or the selection was ignored because a font selection escape sequence was used after the output line continuation escape sequence on an input line. This category is enabled by default.

ig
262144

An invalid escape sequence occurred in input ignored using the ig request. This warning category diagnoses a condition that is an error when it occurs in non-ignored input.

color
524288

An undefined color was selected, an attempt was made to define a color using an unrecognized color space, an invalid component in a color definition was encountered, or an attempt was made to redefine a default color.

file
1048576

An attempt was made to load a file that does not exist. This category is enabled by default.

Two warning names group other warning categories for convenience.

all

All warning categories except ‘di’, ‘mac’ and ‘reg’. This shorthand is intended to produce all warnings that are useful with macro packages written for AT&T troff and its descendants, which have less fastidious diagnostics than GNU troff.

w

All warning categories. Authors of documents and macro packages targeting groff are encouraged to use this setting.


Next: , Previous: , Up: GNU troff Reference   [Contents][Index]

5.38 Implementation Differences

GNU troff has a number of features that cause incompatibilities with documents written for other versions of troff. Some GNU extensions to troff have become supported by other implementations.


Next: , Previous: , Up: Implementation Differences   [Contents][Index]

5.38.1 Safer Mode

The formatter operates in “safer” mode by default; to mitigate risks from untrusted input documents, the pi and sy requests are disabled. GNU troff’s -U option enables “unsafe mode”, restoring their function and enabling additional groff extension requests, open, opena, and pso. See I/O.


Next: , Previous: , Up: Implementation Differences   [Contents][Index]

5.38.2 Compatibility Mode

Long identifier names may be GNU troff’s most obvious innovation. AT&T troff interprets ‘.dsabcd’ as defining a string ‘ab’ with contents ‘cd’. Normally, GNU troff interprets this as a call of a macro named dsabcd. AT&T troff also interprets ‘\*[’ and ‘\n[’ as an interpolation of a string or register, respectively, named ‘[’. In GNU troff, however, the ‘[’ is normally interpreted as delimiting a long name. In compatibility mode, GNU troff interprets names in the traditional way; they thus can be two characters long at most.

Request: .cp [n]
Register: \n[.C]

If n is missing or non-zero, turn on compatibility mode; otherwise, turn it off.

The read-only register .C is 1 if compatibility mode is on, 0 otherwise.

Compatibility mode can be also turned on with the -C command-line option.

Request: .do name
Register: \n[.cp]

The do request interprets the string, request, diversion, or macro name (along with any further arguments) with compatibility mode disabled. Compatibility mode is restored (only if it was active) when the expansion of name is interpreted; that is, the restored compatibility state applies to the contents of the macro, string, or diversion name as well as data read from files or pipes if name is any of the so, soquiet, mso, msoquiet, or pso requests.

The following example illustrates several aspects of do behavior.

.de mac1
FOO
..
.de1 mac2
groff
.mac1
..
.de mac3
compatibility
.mac1
..
.de ma
\\$1
..
.cp 1
.do mac1
.do mac2 \" mac2, defined with .de1, calls "mac1"
.do mac3 \" mac3 calls "ma" with argument "c1"
.do mac3 \[ti] \" groff syntax accepted in .do arguments
    ⇒ FOO groff FOO compatibility c1 ~

The read-only register .cp, meaningful only when dereferenced from a do request, is 1 if compatibility mode was on when the do request was encountered, and 0 if it was not. This register is specialized and may require a statement of rationale.

When writing macro packages or documents that use GNU troff features and which may be mixed with other packages or documents that do not—common scenarios include serial processing of man pages or use of the so or mso requests—you may desire correct operation regardless of compatibility mode enablement in the surrounding context. It may occur to you to save the existing value of ‘\n(.C’ into a register, say, ‘_C’, at the beginning of your file, turn compatibility mode off with ‘.cp 0’, then restore it from that register at the end with ‘.cp \n(_C’. At the same time, a modular design of a document or macro package may lead you to multiple layers of inclusion. You cannot use the same register name everywhere lest you “clobber” the value from a preceding or enclosing context. The two-character register name space of AT&T troff is confining and mnemonically challenging; you may wish to use the more capacious name space of GNU troff. However, attempting ‘.nr _my_saved_C \n(.C’ will not work in compatibility mode; the register name is too long. “This is exactly what do is for,” you think, ‘.do nr _my_saved_C \n(.C’. The foregoing will always save zero to your register, because do turns compatibility mode off while it interprets its argument list.

To robustly save compatibility mode before switching it off, use

.do nr _my_saved_C \n[.cp]
.cp 0

at the beginning of your file, followed by

.cp \n[_my_saved_C]
.do rr _my_saved_C

at the end. As in the C language, we all have to share one big name space, so choose a register name that is unlikely to collide with other uses.

Normally, GNU troff preserves the interpolation depth in delimited arguments, but not in compatibility mode.

.ds xx '
\w'abc\*(xxdef'
    ⇒ 168 (normal mode on a terminal device)
    ⇒ 72def' (compatibility mode on a terminal device)

Furthermore, the escape sequences \f, \H, \m, \M, \R, \s, and \S are transparent for the purpose of recognizing a control character at the beginning of a line only in compatibility mode. For example, this code produces bold output in both cases, but the text differs.

.de xx
Hello!
..
\fB.xx\fP
    ⇒ .xx (normal mode)
    ⇒ Hello! (compatibility mode)

Normally, the syntax form \sn accepts only a single character (a digit) for n, consistently with other forms that originated in AT&T troff, like \*, \$, \f, \g, \k, \n, and \z. In compatibility mode only, a non-zero n must be in the range 4–39. Legacy documents relying upon this quirk of parsing120 should be migrated to another \s form.


Previous: , Up: Implementation Differences   [Contents][Index]

5.38.3 Other Differences

groff request names unrecognized by other troff implementations will likely be ignored by them; escape sequences that are groff extensions are liable to be interpreted as if the escape character were not present. For example, the adjustable, non-breaking escape sequence \~ is also supported by Heirloom Doctools troff 050915 (September 2005), mandoc 1.9.5 (2009-09-21), neatroff (commit 1c6ab0f6e, 2016-09-13), and Plan 9 from User Space troff (commit 93f8143600, 2022-08-12), but not by Solaris or Documenter’s Workbench troffs. See Manipulating Filling and Adjustment.

GNU troff does not allow the use of the escape sequences \|, \^, \&, \{, \}, \SP, \', \`, \-, \_, \!, \%, and \c in identifiers; AT&T troff does. The \A escape sequence (see Identifiers) may be helpful in avoiding use of these escape sequences in names.

When adjusting to both margins, AT&T troff at first adjusts spaces starting from the right; GNU troff begins from the left. Both implementations adjust spaces from opposite ends on alternating output lines in this adjustment mode to prevent “rivers” in the text.

GNU troff does not always hyphenate words as AT&T troff does. The AT&T implementation uses a set of hard-coded rules specific to English, while GNU troff uses language-specific hyphenation pattern files derived from TeX. Furthermore, in old versions of troff there was a limited amount of space to store hyphenation exceptions (arguments to the hw request); GNU troff has no such restriction.

GNU troff predefines a string .T containing the argument given to the -T command-line option, namely the current output device (for example, ‘pdf’ or ‘utf8’). The existence of this string is a common feature of post-CSTR #54 troffs121 but valid values are specific to each implementation.

AT&T troff ignored attempts to remove read-only registers; GNU troff honors such requests. See Built-in Registers.

The (read-only) register .T interpolates 1 if GNU troff is called with the -T command-line option, and 0 otherwise. This behavior differs from AT&T troff, which interpolated 1 only if nroff was the formatter and was called with -T.

AT&T troff and other implementations handle the lf request differently. For them, its line argument changes the line number of the current line.

AT&T troff had only environments named ‘0’, ‘1’, and ‘2’. In GNU troff, any number of environments may exist, using any valid identifiers for their names (see Identifiers.)

Fractional type sizes cause one noteworthy incompatibility. In AT&T troff the ps request ignores scaling units and thus ‘.ps 10u’ sets the type size to 10 points, whereas in GNU troff it sets the type size to 10 scaled points. See Using Fractional Type Sizes.

The ab request differs from AT&T troff: GNU troff writes no message to the standard error stream if no arguments are given, and it exits with a failure status instead of a successful one.

The bp request differs from AT&T troff: GNU troff does not accept a scaling unit on the argument, a page number; the former (somewhat uselessly) does.

The pm request differs from AT&T troff: GNU troff reports the sizes of macros, strings, and diversions in bytes and ignores an argument to report only the sum of the sizes.

Unlike AT&T troff, GNU troff does not ignore the ss request if the output is a terminal device; instead, the values of minimal inter-word and additional inter-sentence space are each rounded down to the nearest multiple of 12.

In GNU troff there is a fundamental difference between (unformatted) characters and (formatted) glyphs. Everything that affects how a glyph is output is stored with the glyph node; once a glyph node has been constructed, it is unaffected by any subsequent requests that are executed, including bd, cs, tkf, tr, or fp requests. Normally, glyphs are constructed from characters immediately before the glyph is added to an output line. Macros, diversions, and strings are all, in fact, the same type of object; they contain a sequence of intermixed character and glyph nodes. Special characters transform from one to the other: before being added to the output, they behave as characters; afterward, they are glyphs. A glyph node does not behave like a character node when it is processed by a macro: it does not inherit any of the special properties that the character from which it was constructed might have had. For example, the input

.di x
\\\\
.br
.di
.x

produces ‘\\’ in GNU troff. Each pair of backslashes becomes one backslash glyph; the resulting backslashes are thus not interpreted as escape characters when they are reread as the diversion is output. AT&T troff would interpret them as escape characters when rereading them and end up printing one ‘\’.

One correct way to obtain a printable backslash in most documents is to use the \e escape sequence; this always prints a single instance of the current escape character,122 regardless of whether or not it is used in a diversion; it also works in both GNU troff and AT&T troff.

The other correct way, appropriate in contexts independent of the backslash’s common use as a troff escape character—perhaps in discussion of character sets or other programming languages—is the character escape \(rs or \[rs], for “reverse solidus”, from its name in the ECMA-6 (ISO/IEC 646) standard.123

To store an escape sequence in a diversion that is interpreted when the diversion is reread, either use the traditional \! transparent output facility, or, if this is unsuitable, the new \? escape sequence. See Diversions and Gtroff Internals.

In the somewhat pathological case where a diversion exists containing a partially collected line and a partially collected line at the top-level diversion has never existed, AT&T troff will output the partially collected line at the end of input; GNU troff will not.


Next: , Previous: , Up: Top   [Contents][Index]

6 File Formats

All files read and written by gtroff are text files. The following two sections describe their format.


Next: , Previous: , Up: File Formats   [Contents][Index]

6.1 gtroff Output

This section describes the groff intermediate output format produced by GNU troff.

As groff is a wrapper program around GNU troff and automatically calls an output driver (or “postprocessor”), this output does not show up normally. This is why it is called intermediate. groff provides the option -Z to inhibit postprocessing such that the produced intermediate output is sent to standard output just as it is when calling GNU troff directly.

Here, the term troff output describes what is output by GNU troff, while intermediate output refers to the language that is accepted by the parser that prepares this output for the output drivers. This parser handles whitespace more flexibly than AT&T’s implementation and implements obsolete elements for compatibility; otherwise, both formats are the same.124

The main purpose of the intermediate output concept is to facilitate the development of postprocessors by providing a common programming interface for all devices. It has a language of its own that is completely different from the gtroff language. While the gtroff language is a high-level programming language for text processing, the intermediate output language is a kind of low-level assembler language by specifying all positions on the page for writing and drawing.

The intermediate output produced by gtroff is fairly readable, while output from AT&T troff is rather hard to understand because of strange habits that are still supported, but not used any longer by gtroff.


Next: , Previous: , Up: gtroff Output   [Contents][Index]

6.1.1 Language Concepts

The fundamental operation of the GNU troff formatter is the translation of the groff input language into a device-independent form primarily concerned with what has to be written or drawn at specific positions on the output device. This language is simple and imperative. In the following discussion, the term command always refers to this intermediate output language, and never to the groff language intended for direct use by document authors. Intermediate output commands comprise several categories: glyph output; font, color, and text size selection; motion of the printing position; page advancement; drawing of geometric objects; and device control commands, a catch-all for operations not easily classified as any of the foregoing, such as directives to start and stop output, identify the intended output device, or place URL hyperlinks in supported output formats.


Next: , Previous: , Up: Language Concepts   [Contents][Index]

6.1.1.1 Separation

AT&T troff output has strange requirements regarding whitespace. The gtroff output parser, however, is more tolerant, making whitespace maximally optional. Such characters, i.e., the tab, space, and newline, always have a syntactical meaning. They are never printable because spacing within the output is always done by positioning commands.

Any sequence of space or tab characters is treated as a single syntactical space. It separates commands and arguments, but is only required when there would occur a clashing between the command code and the arguments without the space. Most often, this happens when variable-length command names, arguments, argument lists, or command clusters meet. Commands and arguments with a known, fixed length need not be separated by syntactical space.

A line break is a syntactical element, too. Every command argument can be followed by whitespace, a comment, or a newline character. Thus a syntactical line break is defined to consist of optional syntactical space that is optionally followed by a comment, and a newline character.

The normal commands, those for positioning and text, consist of a single letter taking a fixed number of arguments. For historical reasons, the parser allows stacking of such commands on the same line, but fortunately, in gtroff’s intermediate output, every command with at least one argument is followed by a line break, thus providing excellent readability.

The other commands—those for drawing and device controlling—have a more complicated structure; some recognize long command names, and some take a variable number of arguments. So all ‘D’ and ‘x’ commands were designed to request a syntactical line break after their last argument. Only one command, ‘x X, has an argument that can span several input lines; all other commands must have all of their arguments on the same line as the command, i.e., the arguments may not be split by a line break.

Empty lines (these are lines containing only space and/or a comment), can occur everywhere. They are just ignored.


Next: , Previous: , Up: Language Concepts   [Contents][Index]

6.1.1.2 Argument Units

Some commands take integer arguments that are assumed to represent values in a measurement unit, but the letter for the corresponding scaling unit is not written with the output command arguments. Most commands assume the scaling unit ‘u’, the basic unit of the device, some use ‘z’, the scaled point unit of the device, while others, such as the color commands, expect plain integers.

Single characters can have the eighth bit set, as can the names of fonts and special characters. The names of characters and fonts can be of arbitrary length. A character that is to be printed is always in the current font.

A string argument is always terminated by the next whitespace character (space, tab, or newline); an embedded ‘#’ character is regarded as part of the argument, not as the beginning of a comment command. An integer argument is already terminated by the next non-digit character, which then is regarded as the first character of the next argument or command.


Previous: , Up: Language Concepts   [Contents][Index]

6.1.1.3 Document Parts

A correct intermediate output document consists of two parts, the prologue and the body.

The task of the prologue is to set the general device parameters using three exactly specified commands. gtroff’s prologue is guaranteed to consist of the following three lines (in that order):

x T device
x res n h v
x init

with the arguments set as outlined in Device Control Commands. The parser for the intermediate output format is able to interpret additional whitespace and comments as well even in the prologue.

The body is the main section for processing the document data. Syntactically, it is a sequence of any commands different from the ones used in the prologue. Processing is terminated as soon as the first ‘x stop command is encountered; the last line of any gtroff intermediate output always contains such a command.

Semantically, the body is page oriented. A new page is started by a ‘p’ command. Positioning, writing, and drawing commands are always done within the current page, so they cannot occur before the first ‘p’ command. Absolute positioning (by the ‘H’ and ‘V’ commands) is done relative to the current page; all other positioning is done relative to the current location within this page.


Next: , Previous: , Up: gtroff Output   [Contents][Index]

6.1.2 Command Reference

This section describes all intermediate output commands, both from AT&T troff as well as the gtroff extensions.


Next: , Previous: , Up: Command Reference   [Contents][Index]

6.1.2.1 Comment Command

#anythingend of line

A comment. Ignore any characters from the ‘#’ character up to the next newline character.

This command is the only possibility for commenting in the intermediate output. Each comment can be preceded by arbitrary syntactical space; every command can be terminated by a comment.


Next: , Previous: , Up: Command Reference   [Contents][Index]

6.1.2.2 Simple Commands

The commands in this subsection have a command code consisting of a single character, taking a fixed number of arguments. Most of them are commands for positioning and text writing. These commands are tolerant of whitespace. Optionally, syntactical space can be inserted before, after, and between the command letter and its arguments. All of these commands are stackable; i.e., they can be preceded by other simple commands or followed by arbitrary other commands on the same line. A separating syntactical space is necessary only when two integer arguments would clash or if the preceding argument ends with a string argument.

C idwhitespace

Typeset the glyph of the special character id. Trailing syntactical space is necessary to allow special character names of arbitrary length. The drawing position is not advanced.

c g

Typeset the glyph of the ordinary character c. The drawing position is not advanced.

f n

Select the font mounted at position n. n cannot be negative.

H n

Horizontally move the drawing position to n basic units from the left edge of the page. n cannot be negative.

h n

Move the drawing position right n basic units. AT&T troff allowed negative n; GNU troff does not produce such values, but groff’s output driver library handles them.

m color-scheme [component]

Select the stroke color using the components in the color space scheme. Each component is an integer between 0 and 65535. The quantity of components and their meanings vary with each scheme. This command is a groff extension.

mc cyan magenta yellow

Use the CMY color scheme with components cyan, magenta, and yellow.

md

Use the default color (no components; black in most cases).

mg gray

Use a grayscale color scheme with a component ranging between 0 (black) and 65535 (white).

mk cyan magenta yellow black

Use the CMYK color scheme with components cyan, magenta, yellow, and black.

mr red green blue

Use the RGB color scheme with components red, green, and blue.

N n

Typeset the glyph with index n in the current font. n is normally a non-negative integer. The drawing position is not advanced. The html and xhtml devices use this command with negative n to produce unbreakable space; the absolute value of n is taken and interpreted in basic units.

n b a

Indicate a break. No action is performed; the command is present to make the output more easily parsed. The integers b and a describe the vertical space amounts before and after the break, respectively. GNU troff issues this command but groff’s output driver library ignores it. See v and V below.

p n

Begin a new page, setting its number to n. Each page is independent, even from those using the same number. The vertical drawing position is set to 0. All positioning, writing, and drawing commands are interpreted in the context of a page, so a p command must precede them.

s n

Set type size to n scaled points (unit z in GNU troff. AT&T troff used unscaled points p instead; see Output Language Compatibility.

t xyzwhitespace
t xyz dummy-argwhitespace

Typeset a word xyz; that is, set a sequence of ordinary glyphs named x, y, z, …, terminated by a space character or a line break; an optional second integer argument is ignored (this allows the formatter to generate an even number of arguments). Each glyph is set at the current drawing position, and the position is then advanced horizontally by the glyph’s width. A glyph’s width is read from its metrics in the font description file, scaled to the current type size, and rounded to a multiple of the horizontal motion quantum. Use the C command to emplace glyphs of special characters. The t command is a groff extension and is output only for devices whose DESC file contains the tcommand directive; see DESC File Format.

u n xyzwhitespace

Typeset word xyz with track kerning. As t, but after placing each glyph, the drawing position is further advanced horizontally by n basic units (u). The u command is a groff extension and is output only for devices whose DESC file contains the tcommand directive; see DESC File Format.

V n

Vertically move the drawing position to n basic units from the top edge of the page. n cannot be negative.

v n

Move the drawing position down n basic units. AT&T troff allowed negative n; GNU troff does not produce such values, but groff’s output driver library handles them.

w

Indicate an inter-word space. No action is performed; the command is present to make the output more easily parsed. Only adjustable, breakable inter-word spaces are thus described; those resulting from \~ or horizontal motion escape sequences are not. GNU troff issues this command but groff’s output driver library ignores it. See h and H above.


Next: , Previous: , Up: Command Reference   [Contents][Index]

6.1.2.3 Graphics Commands

Each graphics or drawing command in the intermediate output starts with the letter ‘D’, followed by one or two characters that specify a subcommand; this is followed by a fixed or variable number of integer arguments that are separated by a single space character. A ‘D’ command may not be followed by another command on the same line (apart from a comment), so each ‘D’ command is terminated by a syntactical line break.

gtroff output follows the classical spacing rules (no space between command and subcommand, all arguments are preceded by a single space character), but the parser allows optional space between the command letters and makes the space before the first argument optional. As usual, each space can be any sequence of tab and space characters.

Some graphics commands can take a variable number of arguments. In this case, they are integers representing a size measured in basic units ‘u’. The arguments called h1, h2, …, hn stand for horizontal distances where positive means right, negative left. The arguments called v1, v2, …, vn stand for vertical distances where positive means down, negative up. All these distances are offsets relative to the current location.

Each graphics command directly corresponds to a similar gtroff \D escape sequence. See Drawing Geometric Objects.

Unknown ‘D’ commands are assumed to be device-specific. Its arguments are parsed as strings; the whole information is then sent to the postprocessor.

In the following command reference, the syntax element ‹line break› means a syntactical line break as defined above.

D~ h1 v1 h2 v2hn vnline break

Draw B-spline from current position to offset (h1,v1), then to offset (h2,v2), if given, etc., up to (hn,vn). This command takes a variable number of argument pairs; the current position is moved to the terminal point of the drawn curve.

Da h1 v1 h2 v2line break

Draw arc from current position to (h1,v1)+(h2,v2) with center at (h1,v1); then move the current position to the final point of the arc.

DC dline break
DC d dummy-argline break

Draw a solid circle using the current fill color with diameter d (integer in basic units ‘u’) with leftmost point at the current position; then move the current position to the rightmost point of the circle. An optional second integer argument is ignored (this allows the formatter to generate an even number of arguments). This command is a gtroff extension.

Dc dline break

Draw circle line with diameter d (integer in basic units ‘u’) with leftmost point at the current position; then move the current position to the rightmost point of the circle.

DE h vline break

Draw a solid ellipse in the current fill color with a horizontal diameter of h and a vertical diameter of v (both integers in basic units ‘u’) with the leftmost point at the current position; then move to the rightmost point of the ellipse. This command is a gtroff extension.

De h vline break

Draw an outlined ellipse with a horizontal diameter of h and a vertical diameter of v (both integers in basic units ‘u’) with the leftmost point at current position; then move to the rightmost point of the ellipse.

DF color-scheme [component]line break

Set fill color for solid drawing objects using different color schemes; the analogous command for setting the color of text, line graphics, and the outline of graphic objects is ‘m’. The color components are specified as integer arguments between 0 and 65535. The number of color components and their meaning vary for the different color schemes. These commands are generated by gtroff’s escape sequences ‘\D'F …'’ and \M (with no other corresponding graphics commands). No position changing. This command is a gtroff extension.

DFc cyan magenta yellowline break

Set fill color for solid drawing objects using the CMY color scheme, having the 3 color components cyan, magenta, and yellow.

DFd‹line break

Set fill color for solid drawing objects to the default fill color value (black in most cases). No component arguments.

DFg grayline break

Set fill color for solid drawing objects to the shade of gray given by the argument, an integer between 0 (black) and 65535 (white).

DFk cyan magenta yellow blackline break

Set fill color for solid drawing objects using the CMYK color scheme, having the 4 color components cyan, magenta, yellow, and black.

DFr red green blueline break

Set fill color for solid drawing objects using the RGB color scheme, having the 3 color components red, green, and blue.

Df nline break

The argument n must be an integer in the range -32767 to 32767.

0 ≤ n ≤ 1000

Set the color for filling solid drawing objects to a shade of gray, where 0 corresponds to solid white, 1000 (the default) to solid black, and values in between to intermediate shades of gray; this is obsoleted by command ‘DFg’.

n < 0 or n > 1000

Set the filling color to the color that is currently being used for the text and the outline, see command ‘m’. For example, the command sequence

mg 0 0 65535
Df -1

sets all colors to blue.

No position changing. This command is a gtroff extension.

Dl h vline break

Draw line from current position to offset (h,v) (integers in basic units ‘u’); then set current position to the end of the drawn line.

Dp h1 v1 h2 v2hn vnline break

Draw a polygon line from current position to offset (h1,v1), from there to offset (h2,v2), etc., up to offset (hn,vn), and from there back to the starting position. For historical reasons, the position is changed by adding the sum of all arguments with odd index to the actual horizontal position and the even ones to the vertical position. Although this doesn’t make sense it is kept for compatibility. This command is a gtroff extension.

DP h1 v1 h2 v2hn vnline break

Draw a solid polygon in the current fill color rather than an outlined polygon, using the same arguments and positioning as the corresponding ‘Dp’ command. This command is a gtroff extension.

Dt nline break

Set the current line thickness to n (an integer in basic units ‘u’) if n>0; if n=0 select the smallest available line thickness; if n<0 set the line thickness proportional to the type size (this is the default before the first ‘Dt’ command was specified). For historical reasons, the horizontal position is changed by adding the argument to the actual horizontal position, while the vertical position is not changed. Although this doesn’t make sense it is kept for compatibility. This command is a gtroff extension.


Next: , Previous: , Up: Command Reference   [Contents][Index]

6.1.2.4 Device Control Commands

Each device control command starts with the letter ‘x’, followed by a space character (optional or arbitrary space or tab in gtroff) and a subcommand letter or word; each argument (if any) must be preceded by a syntactical space. All ‘x’ commands are terminated by a syntactical line break; no device control command can be followed by another command on the same line (except a comment).

The subcommand is basically a single letter, but to increase readability, it can be written as a word, i.e., an arbitrary sequence of characters terminated by the next tab, space, or newline character. All characters of the subcommand word but the first are simply ignored. For example, gtroff outputs the initialization command ‘x i as ‘x init and the resolution command ‘x r as ‘x res.

In the following, the syntax element ‹line break› means a syntactical line break (see Separation).

xF nameline break

The ‘F’ stands for Filename.

Use name as the intended name for the current file in error reports. This is useful for remembering the original file name when gtroff uses an internal piping mechanism. The input file is not changed by this command. This command is a gtroff extension.

xf n sline break

The ‘f’ stands for font.

Mount font position n (a non-negative integer) with font named s (a text word). See Font Positions.

xH nline break

The ‘H’ stands for Height.

Set glyph height to n (a positive integer in scaled points ‘z’). AT&T troff uses the unit points (‘p’) instead. See Output Language Compatibility.

xi‹line break

The ‘i’ stands for init.

Initialize device. This is the third command of the prologue.

xp‹line break

The ‘p’ stands for pause.

Parsed but ignored. The AT&T troff manual documents this command as

pause device, can be restarted

but GNU troff output drivers do nothing with this command.

xr n h vline break

The ‘r’ stands for resolution.

Resolution is n, while h is the minimal horizontal motion, and v the minimal vertical motion possible with this device; all arguments are positive integers in basic units ‘u’ per inch. This is the second command of the prologue.

xS nline break

The ‘S’ stands for Slant.

Set slant to n (an integer in basic units ‘u’).

xs‹line break

The ‘s’ stands for stop.

Terminates the processing of the current file; issued as the last command of any intermediate troff output.

xt‹line break

The ‘t’ stands for trailer.

Generate trailer information, if any. In GNU troff, this is ignored.

xT xxxline break

The ‘T’ stands for Typesetter.

Set the name of the output driver to xxx, a sequence of non-whitespace characters terminated by whitespace. The possible names correspond to those of groff’s -T option. This is the first command of the prologue.

xu nline break

The ‘u’ stands for underline.

Configure underlining of spaces. If n is 1, start underlining of spaces; if n is 0, stop underlining of spaces. This is needed for the cu request in nroff mode and is ignored otherwise. This command is a gtroff extension.

xX anythingline break

The ‘x’ stands for X-escape.

Send string anything uninterpreted to the device. If the line following this command starts with a ‘+’ character this line is interpreted as a continuation line in the following sense. The ‘+’ is ignored, but a newline character is sent instead to the device, the rest of the line is sent uninterpreted. The same applies to all following lines until the first character of a line is not a ‘+’ character. This command is generated by the gtroff escape sequence \X. The line-continuing feature is a gtroff extension.


Previous: , Up: Command Reference   [Contents][Index]

6.1.2.5 Obsolete Command

In AT&T troff output, the writing of a single glyph is mostly done by a very strange command that combines a horizontal move and a single character giving the glyph name. It doesn’t have a command code, but is represented by a 3-character argument consisting of exactly 2 digits and a character.

ddg

Move right dd (exactly two decimal digits) basic units ‘u’, then print glyph g (represented as a single character).

In GNU troff, arbitrary syntactical space around and within this command is allowed. Only when a preceding command on the same line ends with an argument of variable length is a separating space obligatory. In AT&T troff, large clusters of these and other commands are used, mostly without spaces; this made such output almost unreadable.

For modern high-resolution devices, this command does not make sense because the width of the glyphs can become much larger than two decimal digits. In gtroff, this is only used for the devices X75, X75-12, X100, and X100-12. For other devices, the commands ‘t’ and ‘u’ provide a better functionality.


Next: , Previous: , Up: gtroff Output   [Contents][Index]

6.1.3 Intermediate Output Examples

This section presents the intermediate output generated from the same input for three different devices. The input is the sentence ‘hell world’ fed into gtroff on the command line.

High-resolution device ps

This is the standard output of gtroff if no -T option is given.

shell> echo "hell world" | groff -Z -T ps

x T ps
x res 72000 1 1
x init
p1
x font 5 TR
f5
s10000
V12000
H72000
thell
wh2500
tw
H96620
torld
n12000 0
x trailer
V792000
x stop

This output can be fed into grops to get its representation as a PostScript file.

Low-resolution device latin1

This is similar to the high-resolution device except that the positioning is done at a minor scale. Some comments (lines starting with ‘#’) were added for clarification; they were not generated by the formatter.

shell> echo "hell world" | groff -Z -T latin1

# prologue
x T latin1
x res 240 24 40
x init
# begin a new page
p1
# font setup
x font 1 R
f1
s10
# initial positioning on the page
V40
H0
# write text 'hell'
thell
# inform about space, and issue a horizontal jump
wh24
# write text 'world'
tworld
# announce line break, but do nothing because...
n40 0
# ...the end of the document has been reached
x trailer
V2640
x stop

This output can be fed into grotty to get a formatted text document.

AT&T troff output

Since a computer monitor has a much lower resolution than modern printers, the intermediate output for X11 devices can use the jump-and-write command with its 2-digit displacements.

shell> echo "hell world" | groff -Z -T X100

x T X100
x res 100 1 1
x init
p1
x font 5 TR
f5
s10
V16
H100
# write text with jump-and-write commands
ch07e07l03lw06w11o07r05l03dh7
n16 0
x trailer
V1100
x stop

This output can be fed into xditview or gxditview for displaying in X.

Due to the obsolete jump-and-write command, the text clusters in the AT&T troff output are almost unreadable.


Previous: , Up: gtroff Output   [Contents][Index]

6.1.4 Output Language Compatibility

The intermediate output language of AT&T troff was first documented in A Typesetter-independent TROFF, by Brian Kernighan, and by 1992 the AT&T troff manual was updated to incorprate a description of it.

The GNU troff intermediate output format is compatible with this specification except for the following features.


Previous: , Up: File Formats   [Contents][Index]

6.2 Device and Font Description Files

The groff font and output device description formats are slight extensions of those used by AT&T device-independent troff. In distinction to the AT&T implementation, groff lacks a binary format; all files are text files.125 The device and font description files for a device name are stored in a devname directory. The device description file is called DESC, and, for each font supported by the device, a font description file is called f, where f is usually an abbreviation of a font’s name and/or style. For example, the ps (PostScript) device has groff font description files for Times roman (TR) and Zapf Chancery Medium italic (ZCMI), among many others, while the utf8 device (for terminal emulators) has only font descriptions for the roman, italic, bold, and bold-italic styles (R, I, B, and BI, respectively).

Device and font description files are read both by the formatter, GNU troff, and by output drivers. The programs delegate these files’ processing to an internal library, libgroff, ensuring their consistent interpretation.


Next: , Previous: , Up: Device and Font Description Files   [Contents][Index]

6.2.1 DESC File Format

The DESC file contains a series of directives; each begins a line. Their order is not important, with two exceptions: (1) the res directive must precede any papersize directive; and (2) the charset directive must come last (if at all). If a directive name is repeated, later entries in the file override previous ones (except that the paper dimensions are computed based on the res directive last seen when papersize is encountered). Spaces and/or tabs separate words and are ignored at line boundaries. Comments start with the ‘#’ character and extend to the end of a line. Empty lines are ignored.

family fam

The default font family is fam.

fonts n F1 Fn

Fonts F1, …, Fn are mounted at font positions m+1, …, m+n where m is the number of styles (see below). This directive may extend over more than one line. A font name of 0 causes no font to be mounted at the corresponding position.

hor n

The horizontal motion quantum is n basic units. All horizontal quantities are rounded to multiples of n.

image_generator program

Use program to generate PNG images from PostScript input. Under GNU/Linux, this is usually gs, but under other systems (notably Cygwin) it might be set to another name. The grohtml driver uses this directive.

paperlength n

The vertical dimension of the output medium is n basic units (deprecated: use papersize instead).

papersize format-or-dimension-pair-or-file-name

The dimensions of the output medium are as according to the argument, which is either a standard paper format, a pair of dimensions, or the name of a plain text file containing either of the foregoing.

Recognized paper formats are the ISO and DIN formats A0A7, B0B7, C0C7, D0D7; the U.S. paper types letter, legal, tabloid, ledger, statement, and executive; and the envelope formats com10, monarch, and DL. Matching is performed without regard for lettercase.

Alternatively, the argument can be a custom paper format in the format length,width (with no spaces before or after the comma). Both length and width must have a unit appended; valid units are ‘i’ for inches, ‘c’ for centimeters, ‘p’ for points, and ‘P’ for picas. Example: ‘12c,235p’. An argument that starts with a digit is always treated as a custom paper format.

Finally, the argument can be a file name (e.g., /etc/papersize); if the file can be opened, the first line is read and a match attempted against each of the other forms. No comment syntax is supported.

More than one argument can be specified; each is scanned in turn and the first valid paper specification used.

paperwidth n

The horizontal dimension of the output medium is n basic units (deprecated: use papersize instead).

pass_filenames

Direct GNU troff to emit the name of the source file being processed. This is achieved with the intermediate output command ‘x F’, which grohtml interprets.

postpro program

Use program as the postprocessor.

prepro program

Use program as a preprocessor. The html and xhtml output devices use this directive.

print program

Use program as a spooler program for printing. If omitted, the -l and -L options of groff are ignored.

res n

The device resolution is n basic units per inch.

sizes s1 sn 0

The device has fonts at s1, …, sn scaled points (see below). The list of sizes must be terminated by 0. Each si can also be a range of sizes mn. The list can extend over more than one line.

sizescale n

A typographical point is subdivided into n scaled points. The default is 1. See Using Fractional Type Sizes.

styles S1 Sm

The first m mounting positions are associated with styles S1, …, Sm.

tcommand

The postprocessor can handle the ‘t’ and ‘u’ intermediate output commands.

unicode

The output device supports the complete Unicode repertoire. This directive is useful only for devices that produce character entities instead of glyphs.

If unicode is present, no charset section is required in the font description files since the Unicode handling built into groff is used. However, if there are entries in a font description file’s charset section, they either override the default mappings for those particular characters or add new mappings (normally for composite characters).

The utf8, html, and xhtml output devices use this directive.

unitwidth n

Quantities in the font description files are in basic units for fonts whose type size is n scaled points.

unscaled_charwidths

Make the font handling module always return unscaled character widths. The grohtml driver uses this directive.

use_charnames_in_special

GNU troff should encode special characters inside device control commands; see Postprocessor Access. The grohtml driver uses this directive.

vert n

The vertical motion quantum is n basic units. All vertical quantities are rounded to multiples of n.

charset

This line and everything following it in the file are ignored. It is recognized for compatibility with other troff implementations. In GNU troff, character set repertoire is described on a per-font basis.

GNU troff recognizes but ignores the directives spare1, spare2, and biggestfont.

The res, unitwidth, fonts, and sizes lines are mandatory. Directives not listed above are ignored by GNU troff but may be used by postprocessors to obtain further information about the device.


Previous: , Up: Device and Font Description Files   [Contents][Index]

6.2.2 Font Description File Format

On typesetting output devices, each font is typically available at multiple sizes. While paper measurements in the device description file are in absolute units, measurements applicable to fonts must be proportional to the type size. groff achieves this using the precedent set by AT&T device-independent troff: one font size is chosen as a norm, and all others are scaled linearly relative to that basis. The “unit width” is the number of basic units per point when the font is rendered at this nominal size.

For instance, groff’s lbp device uses a unitwidth of 800. Its Times roman font ‘TR’ has a spacewidth of 833; this is also the width of its comma, period, centered period, and mathematical asterisk, while its ‘M’ is 2,963 basic units. Thus, an ‘M’ on the lbp device is 2,963 basic units wide at a notional type size of 800 points.126

A font description file has two sections. The first is a sequence of directives, and is parsed similarly to the DESC file described above. Except for the directive names that begin the second section, their ordering is immaterial. Later directives of the same name override earlier ones, spaces and tabs are handled in the same way, and the same comment syntax is supported. Empty lines are ignored throughout.

name f

The name of the font is f. ‘DESC’ is an invalid font name. Simple integers are valid, but their use is discouraged.127

spacewidth n

The width of an unadjusted inter-word space is n basic units.

The directives above must appear in the first section; those below are optional.

slant n

The font’s glyphs have a slant of n degrees; a positive n slants in the direction of text flow.

ligatures lig1 lign [0]

Glyphs lig1, …, lign are ligatures; possible ligatures are ‘ff’, ‘fi’, ‘fl’, ‘ffi’ and ‘ffl’. For compatibility with other troff implementations, the list of ligatures may be terminated with a 0. The list of ligatures must not extend over more than one line.

special

The font is special: when a glyph is requested that is not present in the current font, it is sought in any mounted fonts that bear this property.

Other directives in this section are ignored by GNU troff, but may be used by postprocessors to obtain further information about the font.

The second section contains one or two subsections. These can appear in either order; the first one encountered commences the second section. Each starts with a directive on a line by itself. A charset subsection is mandatory unless the associated DESC file contains the unicode directive. Another subsection, kernpairs, is optional.

The directive charset starts the character set subsection.128 It precedes a series of glyph descriptions, one per line. Each such glyph description comprises a set of fields separated by spaces or tabs and organized as follows.

name metrics type code [entity-name] [-- comment]

name identifies the glyph: if name is a printable character c, it corresponds to the troff ordinary character c. If name is a multi-character sequence not beginning with \, it corresponds to the GNU troff special character escape sequence ‘\[name]’. A name consisting of three minus signs, ‘---’, is special and indicates that the glyph is unnamed: such glyphs can be accessed only by the \N escape sequence in troff. A special character named ‘---’ can still be defined using char and similar requests. The name\-’ defines the minus sign glyph. Finally, name can be the unbreakable one-sixth and one-twelfth space escape sequences, \| and \^ (“thin” and “hair” spaces, respectively), in which case only the width metric described below is interpreted; a font can thus customize the widths of these spaces.

The form of the metrics field is as follows.

width[,[height[,[depth[,[italic-correction
  [,[left-italic-correction[,[subscript-correction]]]]]]]]]]

There must not be any spaces, tabs, or newlines between these subfields (which have been split here into two lines only for better legibility). The subfields are in basic units expressed as decimal integers. Unspecified subfields default to 0. Since there is no associated binary format, these values are not required to fit into the C language data type ‘char’ as they are in AT&T device-independent troff.

The width subfield gives the width of the glyph. The height subfield gives the height of the glyph (upward is positive); if a glyph does not extend above the baseline, it should be given a zero height, rather than a negative height. The depth subfield gives the depth of the glyph, that is, the distance below the baseline to which the glyph extends (downward is positive); if a glyph does not extend below the baseline, it should be given a zero depth, rather than a negative depth. Italic corrections are relevant to glyphs in italic or oblique styles. The italic-correction is the amount of space that should be added after an oblique glyph to be followed immediately by an upright glyph. The left-italic-correction is the amount of space that should be added before an oblique glyph to be preceded immediately by an upright glyph. The subscript-correction is the amount of space that should be added after an oblique glyph to be followed by a subscript; it should be less than the italic correction.

For fonts used with typesetting devices, the type field gives a featural description of the glyph: it is a bit mask recording whether the glyph is an ascender, descender, both, or neither. When a \w escape sequence is interpolated, these values are bitwise or-ed together for each glyph and stored in the nr register. In font descriptions for terminal devices, all glyphs might have a type of zero, regardless of their appearance.

0

means the glyph lies entirely between the baseline and a horizontal line at the “x-height” of the font; typical examples are ‘a’, ‘c’, and ‘x’;

1

means the glyph descends below the baseline, like ‘p’;

2

means the glyph ascends above the font’s x-height, like ‘A’ or ‘b’; and

3

means the glyph is both an ascender and a descender—this is true of parentheses in some fonts.

The code field gives a numeric identifier that the postprocessor uses to render the glyph. The glyph can be specified to troff using this code by means of the \N escape sequence. code can be any integer.129

The entity-name field defines an identifier for the glyph that the postprocessor uses to print the GNU troff glyph name. This field is optional; it was introduced so that the grohtml output driver could encode its character set. For example, the glyph ‘\[Po]’ is represented by ‘&pound;’ in HTML 4.0. For efficiency, these data are now compiled directly into grohtml. grops uses the field to build sub-encoding arrays for PostScript fonts containing more than 256 glyphs. Anything on the line after the entity-name field or ‘--’ is ignored.

A line in the charset section can also have the form

name "

identifying name as another name for the glyph mentioned in the preceding line. Such aliases can be chained.

The directive kernpairs starts a list of kerning adjustments to be made to adjacent glyph pairs from this font. It contains a sequence of lines formatted as follows.

g1 g2 n

The foregoing means that when glyph g1 is typeset immediately before g2, the space between them should be increased by n. Most kerning pairs should have a negative value for n.


Next: , Previous: , Up: Top   [Contents][Index]

Appendix A Copying This Manual

Version 1.3, 3 November 2008
Copyright © 2000-2018 Free Software Foundation, Inc.
http://fsf.org/

Everyone is permitted to copy and distribute verbatim copies
of this license document, but changing it is not allowed.
  1. PREAMBLE

    The purpose of this License is to make a manual, textbook, or other functional and useful document free in the sense of freedom: to assure everyone the effective freedom to copy and redistribute it, with or without modifying it, either commercially or noncommercially. Secondarily, this License preserves for the author and publisher a way to get credit for their work, while not being considered responsible for modifications made by others.

    This License is a kind of “copyleft”, which means that derivative works of the document must themselves be free in the same sense. It complements the GNU General Public License, which is a copyleft license designed for free software.

    We have designed this License in order to use it for manuals for free software, because free software needs free documentation: a free program should come with manuals providing the same freedoms that the software does. But this License is not limited to software manuals; it can be used for any textual work, regardless of subject matter or whether it is published as a printed book. We recommend this License principally for works whose purpose is instruction or reference.

  2. APPLICABILITY AND DEFINITIONS

    This License applies to any manual or other work, in any medium, that contains a notice placed by the copyright holder saying it can be distributed under the terms of this License. Such a notice grants a world-wide, royalty-free license, unlimited in duration, to use that work under the conditions stated herein. The “Document”, below, refers to any such manual or work. Any member of the public is a licensee, and is addressed as “you”. You accept the license if you copy, modify or distribute the work in a way requiring permission under copyright law.

    A “Modified Version” of the Document means any work containing the Document or a portion of it, either copied verbatim, or with modifications and/or translated into another language.

    A “Secondary Section” is a named appendix or a front-matter section of the Document that deals exclusively with the relationship of the publishers or authors of the Document to the Document’s overall subject (or to related matters) and contains nothing that could fall directly within that overall subject. (Thus, if the Document is in part a textbook of mathematics, a Secondary Section may not explain any mathematics.) The relationship could be a matter of historical connection with the subject or with related matters, or of legal, commercial, philosophical, ethical or political position regarding them.

    The “Invariant Sections” are certain Secondary Sections whose titles are designated, as being those of Invariant Sections, in the notice that says that the Document is released under this License. If a section does not fit the above definition of Secondary then it is not allowed to be designated as Invariant. The Document may contain zero Invariant Sections. If the Document does not identify any Invariant Sections then there are none.

    The “Cover Texts” are certain short passages of text that are listed, as Front-Cover Texts or Back-Cover Texts, in the notice that says that the Document is released under this License. A Front-Cover Text may be at most 5 words, and a Back-Cover Text may be at most 25 words.

    A “Transparent” copy of the Document means a machine-readable copy, represented in a format whose specification is available to the general public, that is suitable for revising the document straightforwardly with generic text editors or (for images composed of pixels) generic paint programs or (for drawings) some widely available drawing editor, and that is suitable for input to text formatters or for automatic translation to a variety of formats suitable for input to text formatters. A copy made in an otherwise Transparent file format whose markup, or absence of markup, has been arranged to thwart or discourage subsequent modification by readers is not Transparent. An image format is not Transparent if used for any substantial amount of text. A copy that is not “Transparent” is called “Opaque”.

    Examples of suitable formats for Transparent copies include plain ASCII without markup, Texinfo input format, LaTeX input format, SGML or XML using a publicly available DTD, and standard-conforming simple HTML, PostScript or PDF designed for human modification. Examples of transparent image formats include PNG, XCF and JPG. Opaque formats include proprietary formats that can be read and edited only by proprietary word processors, SGML or XML for which the DTD and/or processing tools are not generally available, and the machine-generated HTML, PostScript or PDF produced by some word processors for output purposes only.

    The “Title Page” means, for a printed book, the title page itself, plus such following pages as are needed to hold, legibly, the material this License requires to appear in the title page. For works in formats which do not have any title page as such, “Title Page” means the text near the most prominent appearance of the work’s title, preceding the beginning of the body of the text.

    The “publisher” means any person or entity that distributes copies of the Document to the public.

    A section “Entitled XYZ” means a named subunit of the Document whose title either is precisely XYZ or contains XYZ in parentheses following text that translates XYZ in another language. (Here XYZ stands for a specific section name mentioned below, such as “Acknowledgements”, “Dedications”, “Endorsements”, or “History”.) To “Preserve the Title” of such a section when you modify the Document means that it remains a section “Entitled XYZ” according to this definition.

    The Document may include Warranty Disclaimers next to the notice which states that this License applies to the Document. These Warranty Disclaimers are considered to be included by reference in this License, but only as regards disclaiming warranties: any other implication that these Warranty Disclaimers may have is void and has no effect on the meaning of this License.

  3. VERBATIM COPYING

    You may copy and distribute the Document in any medium, either commercially or noncommercially, provided that this License, the copyright notices, and the license notice saying this License applies to the Document are reproduced in all copies, and that you add no other conditions whatsoever to those of this License. You may not use technical measures to obstruct or control the reading or further copying of the copies you make or distribute. However, you may accept compensation in exchange for copies. If you distribute a large enough number of copies you must also follow the conditions in section 3.

    You may also lend copies, under the same conditions stated above, and you may publicly display copies.

  4. COPYING IN QUANTITY

    If you publish printed copies (or copies in media that commonly have printed covers) of the Document, numbering more than 100, and the Document’s license notice requires Cover Texts, you must enclose the copies in covers that carry, clearly and legibly, all these Cover Texts: Front-Cover Texts on the front cover, and Back-Cover Texts on the back cover. Both covers must also clearly and legibly identify you as the publisher of these copies. The front cover must present the full title with all words of the title equally prominent and visible. You may add other material on the covers in addition. Copying with changes limited to the covers, as long as they preserve the title of the Document and satisfy these conditions, can be treated as verbatim copying in other respects.

    If the required texts for either cover are too voluminous to fit legibly, you should put the first ones listed (as many as fit reasonably) on the actual cover, and continue the rest onto adjacent pages.

    If you publish or distribute Opaque copies of the Document numbering more than 100, you must either include a machine-readable Transparent copy along with each Opaque copy, or state in or with each Opaque copy a computer-network location from which the general network-using public has access to download using public-standard network protocols a complete Transparent copy of the Document, free of added material. If you use the latter option, you must take reasonably prudent steps, when you begin distribution of Opaque copies in quantity, to ensure that this Transparent copy will remain thus accessible at the stated location until at least one year after the last time you distribute an Opaque copy (directly or through your agents or retailers) of that edition to the public.

    It is requested, but not required, that you contact the authors of the Document well before redistributing any large number of copies, to give them a chance to provide you with an updated version of the Document.

  5. MODIFICATIONS

    You may copy and distribute a Modified Version of the Document under the conditions of sections 2 and 3 above, provided that you release the Modified Version under precisely this License, with the Modified Version filling the role of the Document, thus licensing distribution and modification of the Modified Version to whoever possesses a copy of it. In addition, you must do these things in the Modified Version:

    1. Use in the Title Page (and on the covers, if any) a title distinct from that of the Document, and from those of previous versions (which should, if there were any, be listed in the History section of the Document). You may use the same title as a previous version if the original publisher of that version gives permission.
    2. List on the Title Page, as authors, one or more persons or entities responsible for authorship of the modifications in the Modified Version, together with at least five of the principal authors of the Document (all of its principal authors, if it has fewer than five), unless they release you from this requirement.
    3. State on the Title page the name of the publisher of the Modified Version, as the publisher.
    4. Preserve all the copyright notices of the Document.
    5. Add an appropriate copyright notice for your modifications adjacent to the other copyright notices.
    6. Include, immediately after the copyright notices, a license notice giving the public permission to use the Modified Version under the terms of this License, in the form shown in the Addendum below.
    7. Preserve in that license notice the full lists of Invariant Sections and required Cover Texts given in the Document’s license notice.
    8. Include an unaltered copy of this License.
    9. Preserve the section Entitled “History”, Preserve its Title, and add to it an item stating at least the title, year, new authors, and publisher of the Modified Version as given on the Title Page. If there is no section Entitled “History” in the Document, create one stating the title, year, authors, and publisher of the Document as given on its Title Page, then add an item describing the Modified Version as stated in the previous sentence.
    10. Preserve the network location, if any, given in the Document for public access to a Transparent copy of the Document, and likewise the network locations given in the Document for previous versions it was based on. These may be placed in the “History” section. You may omit a network location for a work that was published at least four years before the Document itself, or if the original publisher of the version it refers to gives permission.
    11. For any section Entitled “Acknowledgements” or “Dedications”, Preserve the Title of the section, and preserve in the section all the substance and tone of each of the contributor acknowledgements and/or dedications given therein.
    12. Preserve all the Invariant Sections of the Document, unaltered in their text and in their titles. Section numbers or the equivalent are not considered part of the section titles.
    13. Delete any section Entitled “Endorsements”. Such a section may not be included in the Modified Version.
    14. Do not retitle any existing section to be Entitled “Endorsements” or to conflict in title with any Invariant Section.
    15. Preserve any Warranty Disclaimers.

    If the Modified Version includes new front-matter sections or appendices that qualify as Secondary Sections and contain no material copied from the Document, you may at your option designate some or all of these sections as invariant. To do this, add their titles to the list of Invariant Sections in the Modified Version’s license notice. These titles must be distinct from any other section titles.

    You may add a section Entitled “Endorsements”, provided it contains nothing but endorsements of your Modified Version by various parties—for example, statements of peer review or that the text has been approved by an organization as the authoritative definition of a standard.

    You may add a passage of up to five words as a Front-Cover Text, and a passage of up to 25 words as a Back-Cover Text, to the end of the list of Cover Texts in the Modified Version. Only one passage of Front-Cover Text and one of Back-Cover Text may be added by (or through arrangements made by) any one entity. If the Document already includes a cover text for the same cover, previously added by you or by arrangement made by the same entity you are acting on behalf of, you may not add another; but you may replace the old one, on explicit permission from the previous publisher that added the old one.

    The author(s) and publisher(s) of the Document do not by this License give permission to use their names for publicity for or to assert or imply endorsement of any Modified Version.

  6. COMBINING DOCUMENTS

    You may combine the Document with other documents released under this License, under the terms defined in section 4 above for modified versions, provided that you include in the combination all of the Invariant Sections of all of the original documents, unmodified, and list them all as Invariant Sections of your combined work in its license notice, and that you preserve all their Warranty Disclaimers.

    The combined work need only contain one copy of this License, and multiple identical Invariant Sections may be replaced with a single copy. If there are multiple Invariant Sections with the same name but different contents, make the title of each such section unique by adding at the end of it, in parentheses, the name of the original author or publisher of that section if known, or else a unique number. Make the same adjustment to the section titles in the list of Invariant Sections in the license notice of the combined work.

    In the combination, you must combine any sections Entitled “History” in the various original documents, forming one section Entitled “History”; likewise combine any sections Entitled “Acknowledgements”, and any sections Entitled “Dedications”. You must delete all sections Entitled “Endorsements.”

  7. COLLECTIONS OF DOCUMENTS

    You may make a collection consisting of the Document and other documents released under this License, and replace the individual copies of this License in the various documents with a single copy that is included in the collection, provided that you follow the rules of this License for verbatim copying of each of the documents in all other respects.

    You may extract a single document from such a collection, and distribute it individually under this License, provided you insert a copy of this License into the extracted document, and follow this License in all other respects regarding verbatim copying of that document.

  8. AGGREGATION WITH INDEPENDENT WORKS

    A compilation of the Document or its derivatives with other separate and independent documents or works, in or on a volume of a storage or distribution medium, is called an “aggregate” if the copyright resulting from the compilation is not used to limit the legal rights of the compilation’s users beyond what the individual works permit. When the Document is included in an aggregate, this License does not apply to the other works in the aggregate which are not themselves derivative works of the Document.

    If the Cover Text requirement of section 3 is applicable to these copies of the Document, then if the Document is less than one half of the entire aggregate, the Document’s Cover Texts may be placed on covers that bracket the Document within the aggregate, or the electronic equivalent of covers if the Document is in electronic form. Otherwise they must appear on printed covers that bracket the whole aggregate.

  9. TRANSLATION

    Translation is considered a kind of modification, so you may distribute translations of the Document under the terms of section 4. Replacing Invariant Sections with translations requires special permission from their copyright holders, but you may include translations of some or all Invariant Sections in addition to the original versions of these Invariant Sections. You may include a translation of this License, and all the license notices in the Document, and any Warranty Disclaimers, provided that you also include the original English version of this License and the original versions of those notices and disclaimers. In case of a disagreement between the translation and the original version of this License or a notice or disclaimer, the original version will prevail.

    If a section in the Document is Entitled “Acknowledgements”, “Dedications”, or “History”, the requirement (section 4) to Preserve its Title (section 1) will typically require changing the actual title.

  10. TERMINATION

    You may not copy, modify, sublicense, or distribute the Document except as expressly provided under this License. Any attempt otherwise to copy, modify, sublicense, or distribute it is void, and will automatically terminate your rights under this License.

    However, if you cease all violation of this License, then your license from a particular copyright holder is reinstated (a) provisionally, unless and until the copyright holder explicitly and finally terminates your license, and (b) permanently, if the copyright holder fails to notify you of the violation by some reasonable means prior to 60 days after the cessation.

    Moreover, your license from a particular copyright holder is reinstated permanently if the copyright holder notifies you of the violation by some reasonable means, this is the first time you have received notice of violation of this License (for any work) from that copyright holder, and you cure the violation prior to 30 days after your receipt of the notice.

    Termination of your rights under this section does not terminate the licenses of parties who have received copies or rights from you under this License. If your rights have been terminated and not permanently reinstated, receipt of a copy of some or all of the same material does not give you any rights to use it.

  11. FUTURE REVISIONS OF THIS LICENSE

    The Free Software Foundation may publish new, revised versions of the GNU Free Documentation License from time to time. Such new versions will be similar in spirit to the present version, but may differ in detail to address new problems or concerns. See http://www.gnu.org/copyleft/.

    Each version of the License is given a distinguishing version number. If the Document specifies that a particular numbered version of this License “or any later version” applies to it, you have the option of following the terms and conditions either of that specified version or of any later version that has been published (not as a draft) by the Free Software Foundation. If the Document does not specify a version number of this License, you may choose any version ever published (not as a draft) by the Free Software Foundation. If the Document specifies that a proxy can decide which future versions of this License can be used, that proxy’s public statement of acceptance of a version permanently authorizes you to choose that version for the Document.

  12. RELICENSING

    “Massive Multiauthor Collaboration Site” (or “MMC Site”) means any World Wide Web server that publishes copyrightable works and also provides prominent facilities for anybody to edit those works. A public wiki that anybody can edit is an example of such a server. A “Massive Multiauthor Collaboration” (or “MMC”) contained in the site means any set of copyrightable works thus published on the MMC site.

    “CC-BY-SA” means the Creative Commons Attribution-Share Alike 3.0 license published by Creative Commons Corporation, a not-for-profit corporation with a principal place of business in San Francisco, California, as well as future copyleft versions of that license published by that same organization.

    “Incorporate” means to publish or republish a Document, in whole or in part, as part of another Document.

    An MMC is “eligible for relicensing” if it is licensed under this License, and if all works that were first published under this License somewhere other than this MMC, and subsequently incorporated in whole or in part into the MMC, (1) had no cover texts or invariant sections, and (2) were thus incorporated prior to November 1, 2008.

    The operator of an MMC Site may republish an MMC contained in the site under CC-BY-SA on the same site at any time before August 1, 2009, provided the MMC is eligible for relicensing.

ADDENDUM: How to use this License for your documents

To use this License in a document you have written, include a copy of the License in the document and put the following copyright and license notices just after the title page:

  Copyright (C)  year  your name.
  Permission is granted to copy, distribute and/or modify this document
  under the terms of the GNU Free Documentation License, Version 1.3
  or any later version published by the Free Software Foundation;
  with no Invariant Sections, no Front-Cover Texts, and no Back-Cover
  Texts.  A copy of the license is included in the section entitled ``GNU
  Free Documentation License''.

If you have Invariant Sections, Front-Cover Texts and Back-Cover Texts, replace the “with…Texts.” line with this:

    with the Invariant Sections being list their titles, with
    the Front-Cover Texts being list, and with the Back-Cover Texts
    being list.

If you have Invariant Sections without Cover Texts, or some other combination of the three, merge those two alternatives to suit the situation.

If your document contains nontrivial examples of program code, we recommend releasing these examples in parallel under your choice of free software license, such as the GNU General Public License, to permit their use in free software.


Next: , Previous: , Up: Top   [Contents][Index]

Appendix B Request Index

Request names appear without a leading control character; the defaults are . for the regular control character and ' for the no-break control character.

Jump to:   A   B   C   D   E   F   G   H   I   K   L   M   N   O   P   R   S   T   U   V   W  
Index Entry  Section

A
ab: Debugging
ad: Manipulating Filling and Adjustment
af: Assigning Register Formats
aln: Setting Registers
als: Strings
am: Writing Macros
am1: Writing Macros
ami: Writing Macros
ami1: Writing Macros
as: Strings
as1: Strings
asciify: Diversions

B
backtrace: Debugging
bd: Artificial Fonts
blm: Blank Line Traps
box: Diversions
boxa: Diversions
bp: Page Control
br: Manipulating Filling and Adjustment
break: while
brp: Manipulating Filling and Adjustment

C
c2: Control Characters
cc: Control Characters
ce: Manipulating Filling and Adjustment
cf: I/O
cflags: Using Symbols
ch: Page Location Traps
char: Using Symbols
chop: Strings
class: Character Classes
close: I/O
color: Colors
composite: Using Symbols
continue: while
cp: Compatibility Mode
cs: Artificial Fonts
cu: Artificial Fonts

D
da: Diversions
de: Writing Macros
de1: Writing Macros
defcolor: Colors
dei: Writing Macros
dei1: Writing Macros
device: Postprocessor Access
devicem: Postprocessor Access
di: Diversions
do: Compatibility Mode
ds: ms Document Control Settings
ds: Strings
ds1: Strings
dt: Diversion Traps

E
ec: Using Escape Sequences
ecr: Using Escape Sequences
ecs: Using Escape Sequences
el: if-else
em: End-of-input Traps
eo: Using Escape Sequences
ev: Environments
evc: Environments
ex: Debugging

F
fam: Font Families
fc: Fields
fchar: Using Symbols
fcolor: Colors
fi: Manipulating Filling and Adjustment
fl: Debugging
fp: Font Positions
fschar: Using Symbols
fspecial: Special Fonts
ft: Selecting Fonts
ftr: Selecting Fonts
fzoom: Selecting Fonts

G
gcolor: Colors

H
hc: Manipulating Hyphenation
hcode: Manipulating Hyphenation
hla: Manipulating Hyphenation
hlm: Manipulating Hyphenation
hpf: Manipulating Hyphenation
hpfa: Manipulating Hyphenation
hpfcode: Manipulating Hyphenation
hw: Manipulating Hyphenation
hy: Manipulating Hyphenation
hym: Manipulating Hyphenation
hys: Manipulating Hyphenation

I
ie: if-else
if: if-then
ig: Comments
in: Line Layout
it: Input Line Traps
itc: Input Line Traps

K
kern: Ligatures and Kerning

L
lc: Leaders
length: Strings
lf: Debugging
lg: Ligatures and Kerning
linetabs: Tabs and Fields
ll: Line Layout
ls: Manipulating Spacing
lsm: Leading Space Traps
lt: Page Layout

M
mc: Miscellaneous
mk: Page Motions
mso: I/O
msoquiet: I/O

N
na: Manipulating Filling and Adjustment
ne: Page Control
nf: Manipulating Filling and Adjustment
nh: Manipulating Hyphenation
nm: Miscellaneous
nn: Miscellaneous
nop: if-then
nr: ms Document Control Settings
nr: Setting Registers
nr: Setting Registers
nr: Auto-increment
nroff: troff and nroff Modes
ns: Manipulating Spacing
nx: I/O

O
open: I/O
opena: I/O
os: Page Control
output: Diversions

P
pc: Page Layout
pev: Debugging
pi: I/O
pl: Page Layout
pm: Debugging
pn: Page Layout
pnr: Debugging
po: Line Layout
ps: Changing the Type Size
psbb: Miscellaneous
pso: I/O
ptr: Debugging
pvs: Changing the Vertical Spacing

R
rchar: Using Symbols
rd: I/O
return: Writing Macros
rfschar: Using Symbols
rj: Manipulating Filling and Adjustment
rm: Strings
rn: Strings
rnn: Setting Registers
rr: Setting Registers
rs: Manipulating Spacing
rt: Page Motions

S
schar: Using Symbols
shc: Manipulating Hyphenation
shift: Parameters
sizes: Changing the Type Size
so: I/O
soquiet: I/O
sp: Manipulating Spacing
special: Special Fonts
spreadwarn: Debugging
ss: Manipulating Filling and Adjustment
stringdown: Strings
stringup: Strings
sty: Font Families
substring: Strings
sv: Page Control
sy: I/O

T
ta: Tabs and Fields
tag: Postprocessor Access
taga: Postprocessor Access
tc: Tabs and Fields
ti: Line Layout
tkf: Ligatures and Kerning
tl: Page Layout
tm: Debugging
tm1: Debugging
tmc: Debugging
tr: Character Translations
trf: I/O
trin: Character Translations
trnt: Character Translations
troff: troff and nroff Modes

U
uf: Artificial Fonts
ul: Artificial Fonts
unformat: Diversions

V
vpt: Vertical Position Traps
vs: Changing the Vertical Spacing

W
warn: Debugging
warnscale: Debugging
wh: Page Location Traps
while: while
write: I/O
writec: I/O
writem: I/O

Jump to:   A   B   C   D   E   F   G   H   I   K   L   M   N   O   P   R   S   T   U   V   W  

Next: , Previous: , Up: Top   [Contents][Index]

Appendix C Escape Sequence Index

The escape character, \ by default, is always followed by at least one more input character, making an escape sequence. Any input token \X with X not in the list below emits a warning and interpolates glyph X. Note the entries for \., which may be obscured by the leader dots, and for \RET and \SP, which are sorted alphabetically, not by code point order.

Jump to:   \
Index Entry  Section

\
\: Using Escape Sequences
\: Using Symbols
\!: Diversions
\": Comments
\#: Comments
\$: Parameters
\$*: Parameters
\$0: Parameters
\$@: Parameters
\$^: Parameters
\%: Manipulating Hyphenation
\&: Dummy Characters
\': Using Symbols
\(: Using Symbols
\): Dummy Characters
\*: Strings
\,: Italic Corrections
\-: Using Symbols
\.: Copy Mode
\/: Italic Corrections
\0: Page Motions
\:: Manipulating Hyphenation
\?: Diversions
\A: Identifiers
\a: Leaders
\B: Numeric Expressions
\b: Drawing Geometric Objects
\c: Line Continuation
\C: Using Symbols
\d: Page Motions
\D: Drawing Geometric Objects
<