path: root/doc/groff.info-1

This is groff.info, produced by makeinfo version 7.0.3 from groff.texi.

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".
INFO-DIR-SECTION Typesetting
START-INFO-DIR-ENTRY
* Groff: (groff).               The GNU roff document formatting system.
END-INFO-DIR-ENTRY


File: groff.info,  Node: Top,  Next: Introduction,  Prev: (dir),  Up: (dir)

GNU 'troff'
***********

* Menu:

* Introduction::
* Invoking groff::
* Tutorial for Macro Users::
* Major Macro Packages::
* GNU troff Reference::
* File Formats::
* Copying This Manual::
* Request Index::
* Escape Sequence Index::
* Operator Index::
* Register Index::
* Macro Index::
* String Index::
* File Keyword Index::
* Program and File Index::
* Concept Index::

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".


File: groff.info,  Node: Introduction,  Next: Invoking groff,  Prev: Top,  Up: Top

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.

* Menu:

* Background::
* What Is groff?::
* groff Capabilities::
* Macro Package Intro::
* Preprocessor Intro::
* Output Device Intro::
* Conventions Used in This Manual::
* Installation::
* Credits::


File: groff.info,  Node: Background,  Next: What Is groff?,  Prev: Introduction,  Up: Introduction

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.


File: groff.info,  Node: What Is groff?,  Next: groff Capabilities,  Prev: Introduction,  Up: Introduction

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.


File: groff.info,  Node: groff Capabilities,  Next: Macro Package Intro,  Prev: What Is groff?,  Up: Introduction

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.

   * text filling, breaking, alignment to the left or right margin;
     centering

   * adjustment of inter-word space size to justify text, and of
     inter-sentence space size to suit local style conventions

   * automatic and manual determination of hyphenation break points

   * pagination

   * selection of any font available to the output device

   * adjustment of type size and vertical spacing (or "leading")

   * configuration of line length and indentation amounts; columnation

   * drawing of geometric primitives (lines, arcs, polygons, circles,
     ...)

   * setup of stroke and fill colors (where supported by the output
     device)

   * embedding of hyperlinks, images, document metadata, and other
     inclusions (where supported by the output device)


File: groff.info,  Node: Macro Package Intro,  Next: Preprocessor Intro,  Prev: groff Capabilities,  Up: Introduction

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'.


File: groff.info,  Node: Preprocessor Intro,  Next: Output Device Intro,  Prev: Macro Package Intro,  Up: Introduction

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)  (*note Preprocessor
Intro-Footnote-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.


File: groff.info,  Node: Preprocessor Intro-Footnotes,  Up: Preprocessor Intro

   (1) The 'g' prefix is not used on all systems; see *note Invoking
groff::.


File: groff.info,  Node: Output Device Intro,  Next: Installation,  Prev: Preprocessor Intro,  Up: Introduction

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.


File: groff.info,  Node: Installation,  Next: Conventions Used in This Manual,  Prev: Output Device Intro,  Up: Introduction

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.


File: groff.info,  Node: Conventions Used in This Manual,  Next: Credits,  Prev: Installation,  Up: Introduction

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 *note 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.(1)  (*note Conventions Used
in This Manual-Footnote-1::)

     $ 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.(2)  (*note
Conventions Used in This Manual-Footnote-2::)


File: groff.info,  Node: Conventions Used in This Manual-Footnotes,  Up: Conventions Used in This Manual

   (1) Unix and related operating systems distinguish standard output
and standard error streams _because_ of 'troff':
<https://minnie.tuhs.org/pipermail/tuhs/2013-December/006113.html>.

   (2) *Note Line Layout::.


File: groff.info,  Node: Credits,  Prev: Conventions Used in This Manual,  Up: Introduction

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.


File: groff.info,  Node: Invoking groff,  Next: Tutorial for Macro Users,  Prev: Introduction,  Up: Top

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
(*note 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'
implementations(1) (*note Invoking groff-Footnote-1::) (with more or
less compatible changes).  Similarly, we say 'gpic', 'geqn', and so on.

* Menu:

* Groff Options::
* Environment::
* Macro Directories::
* Font Directories::
* Paper Format::
* Invocation Examples::


File: groff.info,  Node: Invoking groff-Footnotes,  Up: Invoking groff

   (1) Besides 'groff', 'neatroff' is an exception.


File: groff.info,  Node: Groff Options,  Next: Environment,  Prev: Invoking groff,  Up: Invoking groff

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.  *Note Built-in Registers::.
     This option produces a sort of abstract preview of the formatted
     output.

        * Page breaks are marked by a phrase in angle brackets; for
          example, '<beginning of page>'.

        * Lines are broken where they would be in the formatted output.

        * A horizontal motion of any size is represented as one space.
          Adjacent horizontal motions are not combined.  Inter-sentence
          space nodes (those arising from the second argument to the
          'ss' request) are not represented.

        * Vertical motions are not represented.

        * Special characters are rendered in angle brackets; for
          example, the default soft hyphen character appears as '<hy>'.

     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'.  *Note
     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.  *Note 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 *note 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'.

        * 'gsoelim' replaces 'so' requests with the contents of their
          file name arguments.

        * 'gtroff' searches for files named as operands in its command
          line and as arguments to 'psbb', 'so', and 'soquiet' requests.

        * Output drivers may search for files; for instance, 'grops'
          looks for files named in '\X'ps: import ...'', '\X'ps: file
          ...'', and '\X'pdf: pdfpic ...'' device control escape
          sequences.

     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 *note 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 *note 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 *note 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
     (*note 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').  *Note Built-in Registers::.

     The postprocessor to be used for a device is specified by the
     'postpro' command in the device description file.  (*Note 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 *note
     Warnings::.

'-WCATEGORY'
     Inhibit warnings in CATEGORY.  Categories are listed in *note
     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 *note
     gtroff Output:: for a description of this format.


File: groff.info,  Node: Environment,  Next: Macro Directories,  Prev: Groff Options,  Up: Invoking groff

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.  *Note
     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.  *Note 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 (*note 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.


File: groff.info,  Node: Macro Directories,  Next: Font Directories,  Prev: Environment,  Up: Invoking groff

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.(1)  (*note Macro Directories-Footnote-1::) 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.

   * Directories specified with GNU 'troff''s or 'groff''s '-M'
     command-line option.

   * Directories listed in the 'GROFF_TMAC_PATH' environment variable.

   * The current working directory (only if in unsafe mode using the
     '-U' command-line option).

   * The user's home directory, 'HOME'.

   * A platform-dependent directory, a site-local (platform-independent)
     directory, and the main tmac directory.  The locations
     corresponding to your installation are listed in section
     "Environment" of 'gtroff(1)'.  If not otherwise configured, they
     are as follows.

          /usr/local/lib/groff/site-tmac
          /usr/local/share/groff/site-tmac
          /usr/local/share/groff/1.23.0/tmac

     The foregoing assumes that the version of 'groff' is 1.23.0, and
     that the installation prefix was '/usr/local'.  It is possible to
     fine-tune these locations during the source configuration process.


File: groff.info,  Node: Macro Directories-Footnotes,  Up: Macro Directories

   (1) The 'mso' request does not have these limitations.  *Note I/O::.


File: groff.info,  Node: Font Directories,  Next: Paper Format,  Prev: Macro Directories,  Up: Invoking groff

2.4 Font Directories
====================

'groff' enforces few restrictions on how font description files are
named.  For its family/style mechanism to work (*note 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 'dev'NAME, 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.

   * Directories specified with GNU 'troff''s or 'groff''s '-f'
     command-line option.  All output drivers (and some preprocessors)
     support this option as well, because they require information about
     the glyphs to be rendered in the document.

   * Directories listed in the 'GROFF_FONT_PATH' environment variable.

   * A site-local directory and the main font description directory.
     The locations corresponding to your installation are listed in
     section "Environment" of 'gtroff(1)'.  If not otherwise configured,
     they are as follows.

          /usr/local/share/groff/site-font
          /usr/local/share/groff/1.23.0/font

     The foregoing assumes that the version of 'groff' is 1.23.0, and
     that the installation prefix was '/usr/local'.  It is possible to
     fine-tune these locations during the source configuration process.


File: groff.info,  Node: Paper Format,  Next: Invocation Examples,  Prev: Font Directories,  Up: Invoking groff

2.5 Paper Format
================

In 'groff', the page dimensions for the formatter GNU 'troff' and for
output devices are handled separately.  *Note Page Layout::, for
vertical manipulation of the page size, and *Note 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.  *Note 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


File: groff.info,  Node: Invocation Examples,  Prev: Paper Format,  Up: Invoking groff

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.


File: groff.info,  Node: Tutorial for Macro Users,  Next: Major Macro Packages,  Prev: Invoking groff,  Up: Top

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 *note GNU troff
Reference:: instead.

* Menu:

* Basics::
* Common Features::


File: groff.info,  Node: Basics,  Next: Common Features,  Prev: Tutorial for Macro Users,  Up: Tutorial for Macro Users

3.1 Basics
==========

Let us first survey some basic concepts necessary to use a macro package
fruitfully.(1)  (*note Basics-Footnote-1::) 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.  *Note 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).  *Note
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'.

   * First, keep the input lines short.  Short input lines are easier to
     edit, and GNU 'troff' packs words onto longer lines anyhow.

   * In keeping with this, it is helpful to begin a new line after every
     comma or phrase, since common corrections are to add or delete
     sentences or phrases.

   * End each sentence with two spaces--or better, start each sentence
     on a new line.  GNU 'troff' recognizes characters that usually end
     a sentence, and inserts inter-sentence space accordingly.

   * Do not hyphenate words at the end of lines--GNU 'troff' is smart
     enough to hyphenate words as needed, but is not smart enough to
     take hyphens out and join a word back together.  Also, words such
     as "mother-in-law" should not be broken over a line, since then a
     space can occur where not wanted, such as "mother- in-law".

   We offer further advice in *note 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 *note 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.  *Note
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.


File: groff.info,  Node: Basics-Footnotes,  Up: Basics

   (1) The remainder of this chapter is based on 'Writing Papers with
nroff using -me' by Eric P. Allman, which is distributed with 'groff' as
'meintro.me'.


File: groff.info,  Node: Common Features,  Prev: Basics,  Up: Tutorial for Macro Users

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 *note ms::.

* Menu:

* Paragraphs::
* Sections and Chapters::
* Headers and Footers::
* Page Layout Adjustment::
* Displays and Keeps::
* Footnotes and Endnotes::
* Table of Contents::
* Indexing::
* Document Formats::
* Columnation::
* Font and Size Changes::
* Predefined Text::
* Preprocessor Support::
* Configuration and Customization::


File: groff.info,  Node: Paragraphs,  Next: Sections and Chapters,  Prev: Common Features,  Up: Common Features

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.


File: groff.info,  Node: Sections and Chapters,  Next: Headers and Footers,  Prev: Paragraphs,  Up: Common Features

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.


File: groff.info,  Node: Headers and Footers,  Next: Page Layout Adjustment,  Prev: Sections and Chapters,  Up: Common Features

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.  *Note Page Layout::.


File: groff.info,  Node: Page Layout Adjustment,  Next: Displays and Keeps,  Prev: Headers and Footers,  Up: Common Features

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.  *Note Line Layout::.  Commonly, packages
support registers to tune these values.


File: groff.info,  Node: Displays and Keeps,  Next: Footnotes and Endnotes,  Prev: Page Layout Adjustment,  Up: Common Features

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.


File: groff.info,  Node: Footnotes and Endnotes,  Next: Table of Contents,  Prev: Displays and Keeps,  Up: Common Features

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.


File: groff.info,  Node: Table of Contents,  Next: Indexing,  Prev: Footnotes and Endnotes,  Up: Common Features

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.


File: groff.info,  Node: Indexing,  Next: Document Formats,  Prev: Table of Contents,  Up: Common Features

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.


File: groff.info,  Node: Document Formats,  Next: Columnation,  Prev: Indexing,  Up: Common Features

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.


File: groff.info,  Node: Columnation,  Next: Font and Size Changes,  Prev: Document Formats,  Up: Common Features

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.


File: groff.info,  Node: Font and Size Changes,  Next: Predefined Text,  Prev: Columnation,  Up: Common Features

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.  *Note Italic Corrections::.


File: groff.info,  Node: Predefined Text,  Next: Preprocessor Support,  Prev: Font and Size Changes,  Up: Common Features

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.


File: groff.info,  Node: Preprocessor Support,  Next: Configuration and Customization,  Prev: Predefined Text,  Up: Common Features

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'.


File: groff.info,  Node: Configuration and Customization,  Prev: Preprocessor Support,  Up: Common Features

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.


File: groff.info,  Node: Major Macro Packages,  Next: GNU troff Reference,  Prev: Tutorial for Macro Users,  Up: Top

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.

* Menu:

* man::
* mdoc::
* me::
* mm::
* mom::
* ms::


File: groff.info,  Node: man,  Next: mdoc,  Prev: Major Macro Packages,  Up: Major Macro Packages

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.

* Menu:

* Optional man extensions::


File: groff.info,  Node: Optional man extensions,  Up: man

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 \\$*
     ..


File: groff.info,  Node: mdoc,  Next: me,  Prev: man,  Up: Major Macro Packages

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.


File: groff.info,  Node: me,  Next: mm,  Prev: mdoc,  Up: Major Macro Packages

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.


File: groff.info,  Node: mm,  Next: mom,  Prev: me,  Up: Major Macro Packages

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)'.


File: groff.info,  Node: mom,  Next: ms,  Prev: mm,  Up: Major Macro Packages

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.

   * 'toc.html' Entry point to the full mom manual.

   * 'macrolist.html' Hyperlinked index of macros with brief
     descriptions, arranged by category.

   * 'mom-pdf.pdf' PDF features and usage.

   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.


File: groff.info,  Node: ms,  Prev: mom,  Up: Major Macro Packages

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.

* Menu:

* ms Introduction::
* ms Document Structure::
* ms Document Control Settings::
* ms Document Description Macros::
* ms Body Text::
* ms Page Layout::
* Differences from AT&T ms::
* ms Legacy Features::
* ms Naming Conventions::


File: groff.info,  Node: ms Introduction,  Next: ms Document Structure,  Prev: ms,  Up: ms

4.6.1 Introduction
------------------

The 'ms' macros are the oldest surviving package for 'roff' systems.(1)
(*note ms Introduction-Footnote-1::) 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.

* Menu:

* ms basic information::


File: groff.info,  Node: ms Introduction-Footnotes,  Up: ms Introduction

   (1) While manual _pages_ are older, early ones used macros supplanted
by the 'man' package of Seventh Edition Unix (1979).  'ms' shipped with
Sixth Edition (1975) and was documented by Mike Lesk in a Bell Labs
internal memorandum.


File: groff.info,  Node: ms basic information,  Next: ms Document Structure,  Prev: ms Introduction,  Up: ms Introduction

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 (*note 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 *note Request Index:: and *note 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.


File: groff.info,  Node: ms Document Structure,  Next: ms Document Control Settings,  Prev: ms Introduction,  Up: ms

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.  *Note 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.  *Note 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.  *Note 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.


File: groff.info,  Node: ms Document Control Settings,  Next: ms Document Description Macros,  Prev: ms Document Structure,  Up: ms

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.  *Note
     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 (*note 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 (*note 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' (*note 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.  *Note 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 number(1) (*note ms Document Control
          Settings-Footnote-1::) 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''


File: groff.info,  Node: ms Document Control Settings-Footnotes,  Up: ms Document Control Settings

   (1) defined in *note ms Footnotes::


File: groff.info,  Node: ms Document Description Macros,  Next: ms Body Text,  Prev: ms Document Control Settings,  Up: ms

4.6.4 Document Description Macros
---------------------------------

Only the simplest document lacks a title.(1)  (*note ms Document
Description Macros-Footnote-1::) 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...


File: groff.info,  Node: ms Document Description Macros-Footnotes,  Up: ms Document Description Macros

   (1) Distinguish a document title from "titles", which are what 'roff'
systems call headers and footers collectively.


File: groff.info,  Node: ms Body Text,  Next: ms Page Layout,  Prev: ms Document Description Macros,  Up: ms

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.

* Menu:

* Text settings in ms::
* Typographical symbols in ms::
* Paragraphs in ms::
* Headings in ms::
* Typeface and decoration::
* Lists in ms::
* Indented regions in ms::
* ms keeps and displays::
* ms Insertions::
* ms Footnotes::
* ms language and localization::


File: groff.info,  Node: Text settings in ms,  Next: Typographical symbols in ms,  Prev: ms Body Text,  Up: ms Body Text

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.


File: groff.info,  Node: Typographical symbols in ms,  Next: Paragraphs in ms,  Prev: Text settings in ms,  Up: ms Body Text

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.


File: groff.info,  Node: Paragraphs in ms,  Next: Headings in ms,  Prev: Typographical symbols in ms,  Up: ms Body Text

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 *note 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.


File: groff.info,  Node: Headings in ms,  Next: Typeface and decoration,  Prev: Paragraphs in ms,  Up: ms Body Text

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.  *Note ms keeps and displays:: and *note ms
Insertions::.


File: groff.info,  Node: Typeface and decoration,  Next: Lists in ms,  Prev: Headings in ms,  Up: ms Body Text

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.(1)  (*note
Typeface and decoration-Footnote-1::)

 -- 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 (*note 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'.


File: groff.info,  Node: Typeface and decoration-Footnotes,  Up: Typeface and decoration

   (1) This idiosyncrasy arose through feature accretion; for example,
the 'B' macro in Version 6 Unix 'ms' (1975) accepted only one argument,
the text to be set in boldface.  By Version 7 (1979) it recognized a
second argument; in 1990, 'groff' 'ms' added a "pre" argument, placing
it third to avoid breaking support for older documents.


File: groff.info,  Node: Lists in ms,  Next: Indented regions in ms,  Prev: Typeface and decoration,  Up: ms Body Text

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!


File: groff.info,  Node: Indented regions in ms,  Next: ms keeps and displays,  Prev: Lists in ms,  Up: ms Body Text

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


File: groff.info,  Node: ms keeps and displays,  Next: ms Insertions,  Prev: Indented regions in ms,  Up: ms Body Text

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 (*note 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 *note 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.


File: groff.info,  Node: ms Insertions,  Next: ms Footnotes,  Prev: ms keeps and displays,  Up: ms Body Text

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'.


File: groff.info,  Node: ms Footnotes,  Prev: ms Insertions,  Up: ms Body Text

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.(1)  (*note ms Footnotes-Footnote-1::)
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.
*Note 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,(2) (*note ms
Footnotes-Footnote-2::) 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.(3)  (*note ms Footnotes-Footnote-3::)

   '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.


File: groff.info,  Node: ms Footnotes-Footnotes,  Up: ms Footnotes

   (1) "Portable Document Format Publishing with GNU Troff",
'pdfmark.ms' in the 'groff' distribution, uses this technique.

   (2) Unix Version 7 'ms', its descendants, and GNU 'ms' prior to
'groff' version 1.23.0

   (3) You could reset it after each call to '.1C', '.2C', or '.MC'.


File: groff.info,  Node: ms language and localization,  Next: ms Page Layout,  Prev: ms Footnotes,  Up: ms Body Text

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.


File: groff.info,  Node: ms Page Layout,  Next: Differences from AT&T ms,  Prev: ms Body Text,  Up: ms

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.

* Menu:

* ms Headers and Footers::
* Tab Stops in ms::
* ms Margins::
* ms Multiple Columns::
* ms TOC::


File: groff.info,  Node: ms Headers and Footers,  Next: Tab Stops in ms,  Prev: ms Page Layout,  Up: ms Page Layout

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.


File: groff.info,  Node: Tab Stops in ms,  Next: ms Margins,  Prev: ms Headers and Footers,  Up: ms Page Layout

4.6.6.2 Tab stops
.................

Use the 'ta' request to define tab stops as needed.  *Note 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.


File: groff.info,  Node: ms Margins,  Next: ms Multiple Columns,  Prev: Tab Stops in ms,  Up: ms Page Layout

4.6.6.3 Margins
...............

Control margins using the registers summarized in "Margin settings" in
*note 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.


File: groff.info,  Node: ms Multiple Columns,  Next: ms TOC,  Prev: ms Margins,  Up: ms Page Layout

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.


File: groff.info,  Node: ms TOC,  Next: Differences from AT&T ms,  Prev: ms Multiple Columns,  Up: ms Page Layout

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.


File: groff.info,  Node: Differences from AT&T ms,  Next: ms Naming Conventions,  Prev: ms Page Layout,  Up: ms

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.

   * The internals of 'groff' 'ms' differ from the internals of AT&T
     'ms'.  Documents that depend upon implementation details of AT&T
     'ms' may not format properly with 'groff' 'ms'.  Such details
     include macros whose function was not documented in the AT&T 'ms'
     manual.(1)  (*note Differences from AT&T ms-Footnote-1::)

   * The error-handling policy of 'groff' 'ms' is to detect and report
     errors, rather than to ignore them silently.

   * Tenth Edition Research Unix supported 'P1'/'P2' macros to bracket
     code examples; 'groff' 'ms' does not.

   * 'groff' 'ms' does not work in GNU 'troff''s AT&T compatibility
     mode.  If loaded when that mode is enabled, it aborts processing
     with a diagnostic message.

   * Multiple line spacing is not supported.  Use a larger vertical
     spacing instead.

   * 'groff' 'ms' uses the same header and footer defaults in both
     'nroff' and 'troff' modes as AT&T 'ms' does in 'troff' mode; AT&T's
     default in 'nroff' mode is to put the date, in U.S. traditional
     format (e.g., "January 1, 2021"), in the center footer (the 'CF'
     string).

   * Many 'groff' 'ms' macros, including those for paragraphs, headings,
     and displays, cause a reset of paragraph rendering parameters, and
     may change the indentation; they do so not by incrementing or
     decrementing it, but by setting it absolutely.  This can cause
     problems for documents that define additional macros of their own
     that try to manipulate indentation.  Use the 'ms' 'RS' and 'RE'
     macros instead of the 'in' request.

   * AT&T 'ms' interpreted the values of the registers 'PS' and 'VS' in
     points, and did not support the use of scaling units with them.
     'groff' 'ms' interprets values of the registers 'PS', 'VS', 'FPS',
     and 'FVS' equal to or larger than 1,000 (one thousand) as decimal
     fractions multiplied by 1,000.(2)  (*note Differences from AT&T
     ms-Footnote-2::) This threshold makes use of a scaling unit with
     these parameters practical for high-resolution devices while
     preserving backward compatibility.  It also permits expression of
     non-integral type sizes.  For example, 'groff -rPS=10.5p' at the
     shell prompt is equivalent to placing '.nr PS 10.5p' at the
     beginning of the document.

   * AT&T 'ms''s 'AU' macro supported arguments used with some document
     types; 'groff' 'ms' does not.

   * Right-aligned displays are available.  The AT&T 'ms' manual
     observes that "it is tempting to assume that '.DS R' will right
     adjust lines, but it doesn't work".  In 'groff' 'ms', it does.

   * To make 'groff' 'ms' use the default page offset (which also
     specifies the left margin), the 'PO' register must stay undefined
     until the first 'ms' macro is called.

     This implies that '\n[PO]' should not be used early in the
     document, unless it is changed also: accessing an undefined
     register automatically defines it.

   * 'groff' 'ms' supports the 'PN' register, but it is not necessary;
     you can access the page number via the usual '%' register and
     invoke the 'af' request to assign a different format to it if
     desired.(3)  (*note Differences from AT&T ms-Footnote-3::)

   * The AT&T 'ms' manual documents registers 'CW' and 'GW' as setting
     the default column width and "intercolumn gap", respectively, and
     which applied when 'MC' was called with fewer than two arguments.
     'groff' 'ms' instead treats 'MC' without arguments as synonymous
     with '2C'; there is thus no occasion for a default column width
     register.  Further, the 'MINGW' register and the second argument to
     'MC' specify a _minimum_ space between columns, not the fixed
     gutter width of AT&T 'ms'.

   * The AT&T 'ms' manual did not document the 'QI' register; Berkeley
     and 'groff' 'ms' do.

 -- 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.

* Menu:

* Missing Unix Version 7 ms Macros::


File: groff.info,  Node: Differences from AT&T ms-Footnotes,  Up: Differences from AT&T ms

   (1) 'Typing Documents on the UNIX System: Using the -ms Macros with
Troff and Nroff', M. E. Lesk, Bell Laboratories, 1978

   (2) Register values are converted to and stored as basic units.
*Note Measurements::.

   (3) If you redefine the 'ms' 'PT' macro and desire special treatment
of certain page numbers (like '1'), you may need to handle a non-Arabic
page number format, as 'groff' 'ms''s 'PT' does; see the macro package
source.  'groff' 'ms' aliases the 'PN' register to '%'.


File: groff.info,  Node: Missing Unix Version 7 ms Macros,  Prev: Differences from AT&T ms,  Up: Differences from AT&T ms

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.(1)  (*note Missing Unix Version 7 ms
Macros-Footnote-1::) For informal purposes, a simple definition of 'UX'
should maintain the readability of the document's substance.

     .rm UX
     .ds UX Unix\"


File: groff.info,  Node: Missing Unix Version 7 ms Macros-Footnotes,  Up: Missing Unix Version 7 ms Macros

   (1) The removal beforehand is necessary because 'groff' 'ms' aliases
these macros to a diagnostic macro, and you want to redefine the aliased
name, not its target.


File: groff.info,  Node: ms Legacy Features,  Next: ms Naming Conventions,  Prev: Differences from AT&T ms,  Up: ms

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.


File: groff.info,  Node: ms Naming Conventions,  Prev: ms Legacy Features,  Up: ms

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.

   * Names used only within one module are of the form MODULE'*'NAME.

   * Names used outside the module in which they are defined are of the
     form MODULE'@'NAME.

   * Names associated with a particular environment are of the form
     ENVIRONMENT':'NAME; these are used only within the 'par' module.

   * NAME does not have a module prefix.

   * Constructed names used to implement arrays are of the form
     ARRAY'!'INDEX.

   Thus the 'groff' 'ms' macros reserve the following names.

   * Names containing the characters '*', '@', and ':'.

   * Names containing only uppercase letters and digits.


File: groff.info,  Node: GNU troff Reference,  Next: File Formats,  Prev: Major Macro Packages,  Up: Top

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.

* Menu:

* Text::
* Page Geometry::
* Measurements::
* Numeric Expressions::
* Identifiers::
* Formatter Instructions::
* Comments::
* Registers::
* Manipulating Filling and Adjustment::
* Manipulating Hyphenation::
* Manipulating Spacing::
* Tabs and Fields::
* Character Translations::
* troff and nroff Modes::
* Line Layout::
* Line Continuation::
* Page Layout::
* Page Control::
* Using Fonts::
* Manipulating Type Size and Vertical Spacing::
* Colors::
* Strings::
* Conditionals and Loops::
* Writing Macros::
* Page Motions::
* Drawing Geometric Objects::
* Deferring Output::
* Traps::
* Diversions::
* Punning Names::
* Environments::
* Suppressing Output::
* I/O::
* Postprocessor Access::
* Miscellaneous::
* Gtroff Internals::
* Debugging::
* Implementation Differences::


File: groff.info,  Node: Text,  Next: Measurements,  Prev: GNU troff Reference,  Up: GNU troff Reference

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.

* Menu:

* Filling::
* Hyphenation::
* Sentences::
* Breaking::
* Adjustment::
* Tabs and Leaders::
* Requests and Macros::
* Macro Packages::
* Input Encodings::
* Input Conventions::


File: groff.info,  Node: Filling,  Next: Sentences,  Prev: Text,  Up: Text

5.1.1 Filling
-------------

When GNU 'troff' starts up, it obtains information about the device for
which it is preparing output.(1)  (*note Filling-Footnote-1::) 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.(2)  (*note
Filling-Footnote-2::) To disable filling, see *note 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.


File: groff.info,  Node: Filling-Footnotes,  Up: Filling

   (1) *Note Device and Font Description Files::.

   (2) Tabs and leaders also separate words.  Escape sequences can
function as word characters, word separators, or neither--the last
simply have no effect on GNU 'troff''s idea of whether an input
character is within a word.  We'll discuss all of these in due course.


File: groff.info,  Node: Sentences,  Next: Hyphenation,  Prev: Filling,  Up: Text

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.(1)  (*note
Sentences-Footnote-1::) 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.(2)  (*note Sentences-Footnote-2::) 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 *note 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).(3)  (*note Sentences-Footnote-3::)

     \[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 *note Using Symbols::.  To change the additional
inter-sentence space amount--even to remove it entirely--see *note
Manipulating Filling and Adjustment::.


File: groff.info,  Node: Sentences-Footnotes,  Up: Sentences

   (1) A well-researched jeremiad appreciated by 'groff' contributors on
both sides of the sentence-spacing debate can be found at
<https://web.archive.org/web/20171217060354/http://www.heracliteanriver.com/?p=324>.

   (2) This statement oversimplifies; there are escape sequences whose
purpose is precisely to produce glyphs on the output device, and input
characters that _aren't_ part of escape sequences can undergo a great
deal of processing before getting to the output.

   (3) The mnemonics for the special characters shown here are "dagger",
"double dagger", "right (double) quote", and "closing (single) quote".
See the 'groff_char(7)' man page.


File: groff.info,  Node: Hyphenation,  Next: Breaking,  Prev: Sentences,  Up: Text

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.  *Note
Manipulating Hyphenation::.


File: groff.info,  Node: Breaking,  Next: Adjustment,  Prev: Hyphenation,  Up: Text

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 *note 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 *note 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 (*note
Leading Space Traps::).  Again, macro packages may provide other methods
of producing indented paragraphs.  Trailing spaces on text lines are
discarded.(1)  (*note Breaking-Footnote-1::)

   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 *note Manipulating Filling and Adjustment::.


File: groff.info,  Node: Breaking-Footnotes,  Up: Breaking

   (1) "Text lines" are defined in *note Requests and Macros::.


File: groff.info,  Node: Adjustment,  Next: Tabs and Leaders,  Prev: Breaking,  Up: Text

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 *note Manipulating Filling and
Adjustment::.


File: groff.info,  Node: Tabs and Leaders,  Next: Input Conventions,  Prev: Adjustment,  Up: Text

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 *note 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.(1)  (*note
Tabs and Leaders-Footnote-1::) The tab and leader glyphs can be
configured, and further facilities for sophisticated table composition
are available; see *note 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.


File: groff.info,  Node: Tabs and Leaders-Footnotes,  Up: Tabs and Leaders

   (1) "Tab" is short for "tabulation", revealing the term's origin as a
spacing mechanism for table arrangement.


File: groff.info,  Node: Requests and Macros,  Next: Macro Packages,  Prev: Tabs and Leaders,  Up: Text

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".(1)  (*note Requests and Macros-Footnote-1::)

   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.(2)  (*note Requests and
Macros-Footnote-2::)

   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".(3)  (*note Requests and Macros-Footnote-3::)
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.(4)  (*note
Requests and Macros-Footnote-4::)

     .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 (*note 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 (*note 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 (*note Conditionals and Loops::).  Maintainable
'roff' documents tend to arrange macro definitions to minimize forward
references.


File: groff.info,  Node: Requests and Macros-Footnotes,  Up: Requests and Macros

   (1) The '\<RET>' escape sequence can alter how an input line is
classified; see *note Line Continuation::.

   (2) Argument handling in macros is more flexible but also more
complex.  *Note Calling Macros::.

   (3) Some escape sequences undergo interpolation as well.

   (4) GNU 'troff' offers additional ones.  *Note Writing Macros::.


File: groff.info,  Node: Macro Packages,  Next: Input Encodings,  Prev: Requests and Macros,  Up: Text

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".(1)  (*note Macro Packages-Footnote-1::) 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.


File: groff.info,  Node: Macro Packages-Footnotes,  Up: Macro Packages

   (1) Macro files and packages frequently define registers and strings
as well.


File: groff.info,  Node: Input Encodings,  Next: Input Conventions,  Prev: Macro Packages,  Up: Text

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.(1)  (*note Input Encodings-Footnote-1::)

   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.  *Note 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.  (2)  (*note Input
Encodings-Footnote-2::)


File: groff.info,  Node: Input Encodings-Footnotes,  Up: Input Encodings

   (1) The _semantics_ of certain punctuation code points have gotten
stricter with the successive standards, a cause of some frustration
among man page writers; see the 'groff_char(7)' man page.

   (2) The DVI output device defaults to using the Computer Modern (CM)
fonts; 'ec.tmac' loads the EC fonts instead, which provide Euro '\[Eu]'
and per mille '\[%0]' glyphs.


File: groff.info,  Node: Input Conventions,  Prev: Input Encodings,  Up: Text

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.

   * Follow sentence endings in the input with newlines to ease their
     recognition (*note Sentences::).  It is frequently convenient to
     end text lines after colons and semicolons as well, as these
     typically precede independent clauses.  Consider doing so after
     commas; they often occur in lists that become easy to scan when
     itemized by line, or constitute supplements to the sentence that
     are added, deleted, or updated to clarify it.  Parenthetical and
     quoted phrases are also good candidates for placement on text lines
     by themselves.

   * Set your text editor's line length to 72 characters or fewer.(1)
     (*note Input Conventions-Footnote-1::) This limit, combined with
     the previous item of advice, makes it less common that an input
     line will wrap in your text editor, and thus will help you perceive
     excessively long constructions in your text.  Recall that natural
     languages originate in speech, not writing, and that punctuation is
     correlated with pauses for breathing and changes in prosody.

   * Use '\&' after '!', '?', and '.' if they are followed by space,
     tab, or newline characters and don't end a sentence.

   * In filled text lines, use '\&' before '.' and ''' if they are
     preceded by space, so that reflowing the input doesn't turn them
     into control lines.

   * Do not use spaces to perform indentation or align columns of a
     table.  Leading spaces are reliable when text is not being filled.

   * Comment your document.  It is never too soon to apply comments to
     record information of use to future document maintainers (including
     your future self).  We thus introduce another escape sequence,
     '\"', which causes GNU 'troff' to ignore the remainder of the input
     line.

   * Use the empty request--a control character followed immediately by
     a newline--to visually manage separation of material in input
     files.  Many of the 'groff' project's own documents use an empty
     request between sentences, after macro definitions, and where a
     break is expected, and two empty requests between paragraphs or
     other requests or macro calls that will introduce vertical space
     into the document.

     You can combine the empty request with the comment escape sequence
     to include whole-line comments in your document, and even "comment
     out" sections of it.

   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


File: groff.info,  Node: Input Conventions-Footnotes,  Up: Input Conventions

   (1) Emacs: 'fill-column: 72'; Vim: 'textwidth=72'


File: groff.info,  Node: Page Geometry,  Next: Measurements,  Prev: Text,  Up: GNU troff Reference

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 (*note 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 (*note 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 (*note 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.(1)  (*note Page Geometry-Footnote-1::) 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".(2)  (*note Page
Geometry-Footnote-2::) 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" (*note
Traps::); moreover, all but the simplest page layouts tend to have
headers and footers, or at least bear vertical margins larger than one
vee.


File: groff.info,  Node: Page Geometry-Footnotes,  Up: Page Geometry

   (1) 'groff' does not yet support right-to-left scripts.

   (2) 'groff''s terminal output devices have page offsets of zero.


File: groff.info,  Node: Measurements,  Next: Numeric Expressions,  Prev: Text,  Up: GNU troff Reference

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'
     *Note 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.  *Note 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 *note Page Geometry::.

'M'
     Hundredth of an em.

* Menu:

* Motion Quanta::
* Default Units::


File: groff.info,  Node: Motion Quanta,  Next: Default Units,  Prev: Measurements,  Up: Measurements

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.  *Note Drawing Geometric Objects::.

     .tm \n[.H]
         error-> 24
     .nf
     \l'36u' 36u
     \l'37u' 37u
         => _ 36u
         => __ 37u


File: groff.info,  Node: Default Units,  Prev: Motion Quanta,  Up: Measurements

5.3.2 Default Units
-------------------

A general-purpose register (one created or updated with the 'nr'
request; see *note 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 (*note 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.


File: groff.info,  Node: Numeric Expressions,  Next: Identifiers,  Prev: Measurements,  Up: GNU troff Reference

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.(1)  (*note Numeric Expressions-Footnote-1::) 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.(2)  (*note Numeric Expressions-Footnote-2::)

   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 (*note 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.(3)  (*note Numeric Expressions-Footnote-3::)

     .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.  *Note 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;(4) (*note Numeric Expressions-Footnote-4::)
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,(5) (*note
Numeric Expressions-Footnote-5::) 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 (*note
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
     *note Delimiters::.

   You might use '\B' along with the 'if' request to filter out invalid
macro or string arguments.  *Note 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.  *Note 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.  *Note Invoking Requests::, and *note
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 (*note 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.


File: groff.info,  Node: Numeric Expressions-Footnotes,  Up: Numeric Expressions

   (1) Provision is made for interpreting and reporting decimal
fractions in certain cases.

   (2) If that's not enough, see the 'groff_tmac(5)' man page for the
'62bit.tmac' macro package.

   (3) *Note Conditionals and Loops::.

   (4) Control structure syntax creates an exception to this rule, but
is designed to remain useful: recalling our example, '.if 1 .Underline
this' would underline only "this", precisely.  *Note Conditionals and
Loops::.

   (5) *Note Diversions::.


File: groff.info,  Node: Identifiers,  Next: Formatter Instructions,  Prev: Numeric Expressions,  Up: GNU troff Reference

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' (*note 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', '0x0D'-'0x1F', and
'0x80'-'0x9F'.  On an EBCDIC host, they are '0x00'-'0x01', '0x08',
'0x09', '0x0B', '0x0D'-'0x14', '0x17'-'0x1F', and '0x30'-'0x3F'.(1)
(*note Identifiers-Footnote-1::) 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.(2)  (*note Identifiers-Footnote-2::)

   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 *note
     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.  *Note 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.  *Note Warnings::, *note Interpolating
Registers::, and *note 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.


File: groff.info,  Node: Identifiers-Footnotes,  Up: Identifiers

   (1) Historically, control characters like ASCII STX, ETX, and BEL
(<Control+B>, <Control+C>, and <Control+G>) have been observed in 'roff'
documents, particularly in macro packages employing them as delimiters
with the output comparison operator to try to avoid collisions with the
content of arbitrary user-supplied parameters (*note Operators in
Conditionals::).  We discourage this expedient; in GNU 'troff' it is
unnecessary (outside of compatibility mode) because delimited arguments
are parsed at a different input level than the surrounding context.
*Note Implementation Differences::.

   (2) Consider what happens when a C1 control '0x80'-'0x9F' is
necessary as a continuation byte in a UTF-8 sequence.


File: groff.info,  Node: Formatter Instructions,  Next: Registers,  Prev: Identifiers,  Up: GNU troff Reference

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.

* Menu:

* Control Characters::
* Invoking Requests::
* Calling Macros::
* Using Escape Sequences::
* Delimiters::


File: groff.info,  Node: Control Characters,  Next: Invoking Requests,  Prev: Formatter Instructions,  Up: Formatter Instructions

5.6.1 Control Characters
------------------------

The mechanism of using 'roff''s control characters to invoke requests
and call macros was introduced in *note 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 *note
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.
*Note Manipulating Filling and Adjustment::.

   The control '.' and no-break control ''' characters can each be
changed to any ordinary character(1) (*note Control
Characters-Footnote-1::) 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 (*note 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 (*note
     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.  *Note 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.


File: groff.info,  Node: Control Characters-Footnotes,  Up: Control Characters

   (1) Recall *note Identifiers::.


File: groff.info,  Node: Invoking Requests,  Next: Calling Macros,  Prev: Control Characters,  Up: Formatter Instructions

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.  *Note 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.(1)  (*note Invoking
Requests-Footnote-1::)

   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 (*note Page Layout::).  Some requests, like 'ds',
interpret the remainder of the control line as a single argument.  *Note
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

   *Note Blank Line Traps::.


File: groff.info,  Node: Invoking Requests-Footnotes,  Up: Invoking Requests

   (1) In compatibility mode, a space is not necessary after a request
or macro name of two characters' length.  Also, Plan 9 'troff' allows
tabs to separate arguments.


File: groff.info,  Node: Calling Macros,  Next: Using Escape Sequences,  Prev: Invoking Requests,  Up: Formatter Instructions

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 (*note 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.(1)  (*note Calling Macros-Footnote-1::)

   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.  *Note 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.  *Note Using Escape
Sequences::.  Otherwise, you must escape the escape character repeatedly
to a context-dependent extent.  *Note 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.(2)  (*note Calling Macros-Footnote-2::) 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.  *Note
Implementation Differences::.


File: groff.info,  Node: Calling Macros-Footnotes,  Up: Calling Macros

   (1) '\~' is fairly portable; see *note Other Differences::.

   (2) Strictly, you can neglect to close the last quoted macro
argument, relying on the end of the control line to do so.  We consider
this lethargic practice poor style.


File: groff.info,  Node: Using Escape Sequences,  Next: Delimiters,  Prev: Calling Macros,  Up: Formatter Instructions

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.(1)
(*note Using Escape Sequences-Footnote-1::)

   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.  *Note 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.  *Note 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.(2)  (*note
Using Escape Sequences-Footnote-2::) When a macro is called, its
contents are interpreted literally.


File: groff.info,  Node: Using Escape Sequences-Footnotes,  Up: Using Escape Sequences

   (1) The omission of spaces before the comment escape sequences is
necessary; see *note Strings::.

   (2) TeX does have such a mechanism.


File: groff.info,  Node: Delimiters,  Prev: Using Escape Sequences,  Up: Formatter Instructions

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.

   * the numerals '0'-'9' and the decimal point '.'

   * the (single-character) operators '+-/*%<>=&:()'

   * the space and tab characters

   * any escape sequences other than '\%', '\:', '\{', '\}', '\'', '\`',
     '\-', '\_', '\!', '\/', '\c', '\e', and '\p'

   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.  *Note 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.


File: groff.info,  Node: Comments,  Next: Registers,  Prev: Formatter Instructions,  Up: GNU troff Reference

5.7 Comments
============

One of the most common forms of escape sequence is the comment.(1)
(*note Comments-Footnote-1::)

 -- 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),(2)
     (*note Comments-Footnote-2::) 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.(3)  (*note
     Comments-Footnote-3::)

          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


File: groff.info,  Node: Comments-Footnotes,  Up: Comments

   (1) This claim may be more aspirational than descriptive.

   (2) *Note Conditional Blocks::.

   (3) Exception: auto-incrementing registers defined outside the
ignored region _will_ be modified if interpolated with '\n�' inside it.
*Note Auto-increment::.


File: groff.info,  Node: Registers,  Next: Manipulating Filling and Adjustment,  Prev: Formatter Instructions,  Up: GNU troff Reference

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.  *Note
Identifiers::, for information on constructing a valid name for a
register.

* Menu:

* Setting Registers::
* Interpolating Registers::
* Auto-increment::
* Assigning Register Formats::
* Built-in Registers::


File: groff.info,  Node: Setting Registers,  Next: Interpolating Registers,  Prev: Registers,  Up: Registers

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 *note Delimiters::.  It also does not
     produce an input token in GNU 'troff'.  *Note 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.  *Note 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 *note
     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.  *Note
     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.


File: groff.info,  Node: Interpolating Registers,  Next: Auto-increment,  Prev: Setting Registers,  Up: Registers

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
     (*note 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.  *Note 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


File: groff.info,  Node: Auto-increment,  Next: Assigning Register Formats,  Prev: Interpolating Registers,  Up: Registers

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".(1)  (*note
Auto-increment-Footnote-1::)

 -- 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.


File: groff.info,  Node: Auto-increment-Footnotes,  Up: Auto-increment

   (1) A negative auto-increment can be considered an "auto-decrement".


File: groff.info,  Node: Assigning Register Formats,  Next: Built-in Registers,  Prev: Auto-increment,  Up: Registers

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'.  *Note
     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 (*note Copy Mode::).

   GNU 'troff' interprets only Arabic numerals.  The Roman numeral or
alphabetic formats cannot be used as operands to arithmetic operators in
expressions (*note 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


File: groff.info,  Node: Built-in Registers,  Prev: Assigning Register Formats,  Up: Registers

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 *note 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 (*note 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 (*note Groff Options::).

'\n[.R]'
     Count of available unused registers; always 10,000 in GNU
     'troff'.(1)  (*note Built-in Registers-Footnote-1::)

'\n[.T]'
     Indicator of output device selection (Boolean-valued); see 'groff'
     '-T' option (*note Groff Options::).

'\n[.U]'
     Unsafe mode enablement status (Boolean-valued); see 'groff' '-U'
     option (*note 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
*note 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.


File: groff.info,  Node: Built-in Registers-Footnotes,  Up: Built-in Registers

   (1) GNU 'troff' dynamically allocates memory for as many registers as
required.


File: groff.info,  Node: Manipulating Filling and Adjustment,  Next: Manipulating Hyphenation,  Prev: Registers,  Up: GNU troff Reference

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 *note 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.
*Note 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
re�nables 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 (*note Environments::).  *Note Line Continuation::, for
     interaction with the '\c' escape sequence.

 -- Request: .nf
     Disable filling of output lines: the output line length (*note 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 (*note
     Environments::).  See *note 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 (*note 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 (*note 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,(1)
(*note Manipulating Filling and Adjustment-Footnote-1::) 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.  (2)  (*note
Manipulating Filling and Adjustment-Footnote-2::)

     .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 (*note 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 (*note
     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 sentences(3) (*note
     Manipulating Filling and Adjustment-Footnote-3::) 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 (*note
     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.


File: groff.info,  Node: Manipulating Filling and Adjustment-Footnotes,  Up: Manipulating Filling and Adjustment

   (1) unless diverted; see *note Diversions::

   (2) *Note Line Continuation::.

   (3) Recall *note Filling:: and *note Sentences:: for the definitions
of word and sentence boundaries, respectively.


File: groff.info,  Node: Manipulating Hyphenation,  Next: Manipulating Spacing,  Prev: Manipulating Filling and Adjustment,  Up: GNU troff Reference

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,(1) (*note Manipulating Hyphenation-Footnote-1::) 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 (*note 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.  *Note 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.(2)  (*note Manipulating
     Hyphenation-Footnote-2::) Without an argument, 'hc' resets the
     hyphenation character to '\%' (the default).  The hyphenation
     character is associated with the environment (*note
     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.(3)  (*note Manipulating
     Hyphenation-Footnote-3::) 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 (*note 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.(4)  (*note Manipulating Hyphenation-Footnote-4::) 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,(5) (*note
          Manipulating Hyphenation-Footnote-5::) 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.

     language             pattern name   left min   right min
     -----------------------------------------------------------
     Czech                cs             2          2
     English              en             2          3
     French               fr             2          3
     German traditional   det            2          2
     German reformed      den            2          2
     Italian              it             2          2
     Swedish              sv             1          2

     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.

        * A percent sign starts a comment (up to the end of the line)
          even if preceded by a backslash.

        * "Digraphs" like '\$' are not supported.

        * '^^XX' (where each X is 0-9 or a-f) and '^^C' (character C in
          the code point range 0-127 decimal) are recognized; other uses
          of '^' cause an error.

        * No macro expansion is performed.

        * 'hpf' checks for the expression '\patterns{...}' (possibly
          with whitespace before or after the braces).  Everything
          between the braces is taken as hyphenation patterns.
          Consequently, '{' and '}' are not allowed in patterns.

        * Similarly, '\hyphenation{...}' gives a list of hyphenation
          exceptions.

        * '\endinput' is recognized also.

        * For backward compatibility, if '\patterns' is missing, the
          whole file is treated as a list of hyphenation patterns
          (except that the '%' character is recognized as the start of a
          comment).

     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.(6)
     (*note Manipulating Hyphenation-Footnote-6::)

     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
     'Kinderg�rten' (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-g�r-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 (*note 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 (*note 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 (*note
     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 (*note
     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.


File: groff.info,  Node: Manipulating Hyphenation-Footnotes,  Up: Manipulating Hyphenation

   (1) Whether a perfect algorithm for this application is even possible
is an unsolved problem in computer science:
<https://tug.org/docs/liang/liang-thesis.pdf>.

   (2) '\%' itself stops marking hyphenation points but still produces
no output glyph.

   (3) "Soft" because it appears in output only where a hyphenation
break is performed; a "hard" hyphen, as in "long-term", always appears.

   (4) The mode is a vector of Booleans encoded as an integer.  To a
programmer, this fact is easily deduced from the exclusive use of powers
of two for the configuration parameters; they are computationally easy
to "mask off" and compare to zero.  To almost everyone else, the
arrangement seems recondite and unfriendly.

   (5) Hyphenation is prevented if the next page location trap is closer
to the vertical drawing position than the next text baseline would be.
*Note Page Location Traps::.

   (6) For more on localization, see the 'groff_tmac(5)' man page.


File: groff.info,  Node: Manipulating Spacing,  Next: Tabs and Fields,  Prev: Manipulating Hyphenation,  Up: GNU troff Reference

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.(1)  (*note Manipulating
     Spacing-Footnote-1::) 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 *note 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 (*note Environments::).

   The 'ls' request is a coarse mechanism.  *Note 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 *note 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',(2) (*note Manipulating
     Spacing-Footnote-2::) 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 (*note Page Control::).  No-space mode ends automatically
     when text(3) (*note Manipulating Spacing-Footnote-3::) is formatted
     for output (4) (*note Manipulating Spacing-Footnote-4::) 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.


File: groff.info,  Node: Manipulating Spacing-Footnotes,  Up: Manipulating Spacing

   (1) *Note Page Location Traps::.

   (2) *Note Drawing Geometric Objects::.

   (3) or geometric objects; see *note Drawing Geometric Objects::

   (4) to the top-level diversion; see *note Diversions::


File: groff.info,  Node: Tabs and Fields,  Next: Character Translations,  Prev: Manipulating Spacing,  Up: GNU troff Reference

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 *note Copy Mode::.

 -- Request: .ta [[n1 n2 ... nn ]T r1 r2 ... rn]
 -- 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 *note Manipulating Filling and
     Adjustment:: and *note 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 (*note
     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.(1)  (*note Tabs and Fields-Footnote-1::)

          .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 (*note 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 (*note
     Environments::).  The read-only register '.linetabs' interpolates 1
     if line-tabs are active, and 0 otherwise.

* Menu:

* Leaders::
* Fields::


File: groff.info,  Node: Tabs and Fields-Footnotes,  Up: Tabs and Fields

   (1) Plan 9 'troff' uses the register '.S' for this purpose.


File: groff.info,  Node: Leaders,  Next: Fields,  Prev: Tabs and Fields,  Up: Tabs and Fields

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.(1)  (*note
Leaders-Footnote-1::)

   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 *note Copy Mode::.

 -- Request: .lc [c]
     Set the leader repetition character to the ordinary or special
     character C.  Recall *note 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 (*note 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


File: groff.info,  Node: Leaders-Footnotes,  Up: Leaders

   (1) This is pronounced to rhyme with "feeder", and refers to how the
glyphs "lead" the eye across the page to the corresponding page number
or other datum.


File: groff.info,  Node: Fields,  Prev: Leaders,  Up: Tabs and Fields

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 (*note Environments::).

          .fc # ^
          .ta T 3i
          #foo^bar^smurf#
          .br
          #foo^^bar^smurf#
              => foo         bar          smurf
              => foo            bar       smurf


File: groff.info,  Node: Character Translations,  Next: troff and nroff Modes,  Prev: Tabs and Fields,  Up: GNU troff Reference

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 (*note 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.  *Note
     Diversions::, for details about the 'asciify' request.

     Some notes:

        * Special characters ('\(XX', '\[XXX]', '\C'XXX'', '\'', '\`',
          '\-', '\_'), glyphs defined with the 'char' request, and
          numbered glyphs ('\N'XXX'') can be translated also.

        * The '\e' escape can be translated also.

        * Characters can be mapped onto the '\%' and '\~' escape
          sequences (but '\%' and '\~' can't be mapped onto another
          glyph).

        * The following characters can't be translated: space (with one
          exception, see below), backspace, newline, leader (and '\a'),
          tab (and '\t').

        * Translations are not considered for finding the soft hyphen
          character set with the 'shc' request.

        * The pair 'C\&' (an arbitrary character C followed by the dummy
          character) maps this character to "nothing".

               .tr a\&
               foo bar
                   => foo br

          Even the space character can be mapped to the dummy character.

               .tr aa \&
               foo bar
                   => foobar

          As shown in the example, the space character can't be the
          first character/glyph pair as an argument of 'tr'.
          Additionally, it is not possible to map the space character to
          any other glyph; requests like '.tr aa x' undo '.tr aa \&'
          instead.

          If justification is active, lines are justified in spite of
          the 'empty' space character (but there is no minimal distance,
          i.e., the space character, between words).

        * After an output glyph has been constructed (this happens at
          the moment immediately before the glyph is appended to an
          output glyph list, either by direct output, in a macro,
          diversion, or string), it is no longer affected by 'tr'.

        * Translating character to glyphs where one of them or both are
          undefined is possible also; 'tr' does not check whether the
          elements of its argument exist.

          *Note Gtroff Internals::.

        * Without an argument, the 'tr' request is ignored.

 -- 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 '\!'.  *Note 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'.


File: groff.info,  Node: troff and nroff Modes,  Next: Line Layout,  Prev: Character Translations,  Up: GNU troff Reference

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 executable(1) (*note troff and nroff
Modes-Footnote-1::) 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'.

   *Note Conditionals and Loops::, for more details on built-in
conditions.


File: groff.info,  Node: troff and nroff Modes-Footnotes,  Up: troff and nroff Modes

   (1) A GNU 'nroff' program is available for convenience; it calls GNU
'troff' to perform the formatting.


File: groff.info,  Node: Line Layout,  Next: Line Continuation,  Prev: troff and nroff Modes,  Up: GNU troff Reference

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 (*note 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 (*note 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 (*note 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 (*note
     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.


File: groff.info,  Node: Line Continuation,  Next: Page Layout,  Prev: Line Layout,  Up: GNU troff Reference

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 (*note
     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 on
              => Man-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 *note Manipulating Filling and Adjustment::.

        * If filling is enabled, a word interrupted with '\c' is
          continued with the text on the next input text line, without
          an intervening space.

               This is a te\c
               st.
                   => This is a test.

        * If filling is disabled, the next input text line after '\c' is
          handled as a continuation of the same input text line.

               .nf
               This is a \c
               test.
                   => This is a test.

     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 (*note Environments::).(1)  (*note Line
     Continuation-Footnote-1::)


File: groff.info,  Node: Line Continuation-Footnotes,  Up: Line Continuation

   (1) Historically, the '\c' escape sequence has proven challenging to
characterize.  Some sources say it "connects the next input text" (to
the input line on which it appears); others describe it as
"interrupting" text, on the grounds that a text line is interrupted
without breaking, perhaps to inject a request invocation or macro call.


File: groff.info,  Node: Page Layout,  Next: Page Control,  Prev: Line Continuation,  Up: GNU troff Reference

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,(1) (*note Page
     Layout-Footnote-1::) 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 *note 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 (*note 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.  *Note
Traps::.


File: groff.info,  Node: Page Layout-Footnotes,  Up: Page Layout

   (1) *Note Traps::.


File: groff.info,  Node: Page Control,  Next: Using Fonts,  Prev: Page Layout,  Up: GNU troff Reference

5.18 Page Control
=================

Discretionary page breaks can prevent the unwanted separation of
content.  A new page number takes effect during page ejection; see *note
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.  *Note Page Location Traps::.  'bp' has effect
     only if invoked within the top-level diversion.(1)  (*note Page
     Control-Footnote-1::) 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 *note 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 *note 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 *note 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 (*note 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.

     *Note Diversions::, for a comparison of 'nl' with the '.h' and '.d'
     registers.


File: groff.info,  Node: Page Control-Footnotes,  Up: Page Control

   (1) *Note Diversions::.


File: groff.info,  Node: Using Fonts,  Next: Manipulating Type Size and Vertical Spacing,  Prev: Page Control,  Up: GNU troff Reference

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.(1)  (*note Using Fonts-Footnote-1::) 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.(2)  (*note
Using Fonts-Footnote-2::) 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;(3) (*note Using Fonts-Footnote-3::) instead it works
with a "font description file" that characterizes it, including its
glyph repertoire and the "metrics" (dimensions) of each glyph.(4)
(*note Using Fonts-Footnote-4::) 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).

* Menu:

* Selecting Fonts::
* Font Families::
* Font Positions::
* Using Symbols::
* Character Classes::
* Special Fonts::
* Artificial Fonts::
* Ligatures and Kerning::
* Italic Corrections::
* Dummy Characters::


File: groff.info,  Node: Using Fonts-Footnotes,  Up: Using Fonts

   (1) Terminals and some output devices have fonts that render at only
one or two sizes.  As examples of the latter, take the 'groff' 'lj4'
device's Lineprinter, and 'lbp''s Courier and Elite faces.

   (2) Font designers prepare families such that the styles share
esthetic properties.

   (3) Historically, the fonts 'troff's dealt with were not Free
Software or, as with the Graphic Systems C/A/T, did not even exist in
the digital domain.

   (4) *Note Font Description File Format::.