groff

The GNU implementation of troff

Edition 1.16

Spring 2000

by Trent A. Fisher
and Werner Lemberg


GNU GENERAL PUBLIC LICENSE

Version 2, June 1991

Copyright (C) 1989, 1991 Free Software Foundation, Inc.
59 Temple Place, Suite 330, Boston, MA 02111, USA

Everyone is permitted to copy and distribute verbatim copies of this
license document, but changing it is not allowed.

Preamble

The licenses for most software are designed to take away your freedom to share and change it. By contrast, the GNU General Public License is intended to guarantee your freedom to share and change free software -- to make sure the software is free for all its users. This General Public License applies to most of the Free Software Foundation's software and to any other program whose authors commit to using it. (Some other Free Software Foundation software is covered by the GNU Library General Public License instead.) You can apply it to your programs, too.

When we speak of free software, we are referring to freedom, not price. Our General Public Licenses are designed to make sure that you have the freedom to distribute copies of free software (and charge for this service if you wish), that you receive source code or can get it if you want it, that you can change the software or use pieces of it in new free programs; and that you know you can do these things.

To protect your rights, we need to make restrictions that forbid anyone to deny you these rights or to ask you to surrender the rights. These restrictions translate to certain responsibilities for you if you distribute copies of the software, or if you modify it.

For example, if you distribute copies of such a program, whether gratis or for a fee, you must give the recipients all the rights that you have. You must make sure that they, too, receive or can get the source code. And you must show them these terms so they know their rights.

We protect your rights with two steps: (1) copyright the software, and (2) offer you this license which gives you legal permission to copy, distribute and/or modify the software.

Also, for each author's protection and ours, we want to make certain that everyone understands that there is no warranty for this free software. If the software is modified by someone else and passed on, we want its recipients to know that what they have is not the original, so that any problems introduced by others will not reflect on the original authors' reputations.

Finally, any free program is threatened constantly by software patents. We wish to avoid the danger that redistributors of a free program will individually obtain patent licenses, in effect making the program proprietary. To prevent this, we have made it clear that any patent must be licensed for everyone's free use or not licensed at all.

The precise terms and conditions for copying, distribution and modification follow.

TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION

  1. This License applies to any program or other work which contains a notice placed by the copyright holder saying it may be distributed under the terms of this General Public License. The "Program", below, refers to any such program or work, and a "work based on the Program" means either the Program or any derivative work under copyright law: that is to say, a work containing the Program or a portion of it, either verbatim or with modifications and/or translated into another language. (Hereinafter, translation is included without limitation in the term "modification".) Each licensee is addressed as "you". Activities other than copying, distribution and modification are not covered by this License; they are outside its scope. The act of running the Program is not restricted, and the output from the Program is covered only if its contents constitute a work based on the Program (independent of having been made by running the Program). Whether that is true depends on what the Program does.
  2. You may copy and distribute verbatim copies of the Program's source code as you receive it, in any medium, provided that you conspicuously and appropriately publish on each copy an appropriate copyright notice and disclaimer of warranty; keep intact all the notices that refer to this License and to the absence of any warranty; and give any other recipients of the Program a copy of this License along with the Program. You may charge a fee for the physical act of transferring a copy, and you may at your option offer warranty protection in exchange for a fee.
  3. You may modify your copy or copies of the Program or any portion of it, thus forming a work based on the Program, and copy and distribute such modifications or work under the terms of Section 1 above, provided that you also meet all of these conditions:
    1. You must cause the modified files to carry prominent notices stating that you changed the files and the date of any change.
    2. You must cause any work that you distribute or publish, that in whole or in part contains or is derived from the Program or any part thereof, to be licensed as a whole at no charge to all third parties under the terms of this License.
    3. If the modified program normally reads commands interactively when run, you must cause it, when started running for such interactive use in the most ordinary way, to print or display an announcement including an appropriate copyright notice and a notice that there is no warranty (or else, saying that you provide a warranty) and that users may redistribute the program under these conditions, and telling the user how to view a copy of this License. (Exception: if the Program itself is interactive but does not normally print such an announcement, your work based on the Program is not required to print an announcement.)
    These requirements apply to the modified work as a whole. If identifiable sections of that work are not derived from the Program, and can be reasonably considered independent and separate works in themselves, then this License, and its terms, do not apply to those sections when you distribute them as separate works. But when you distribute the same sections as part of a whole which is a work based on the Program, the distribution of the whole must be on the terms of this License, whose permissions for other licensees extend to the entire whole, and thus to each and every part regardless of who wrote it. Thus, it is not the intent of this section to claim rights or contest your rights to work written entirely by you; rather, the intent is to exercise the right to control the distribution of derivative or collective works based on the Program. In addition, mere aggregation of another work not based on the Program with the Program (or with a work based on the Program) on a volume of a storage or distribution medium does not bring the other work under the scope of this License.
  4. You may copy and distribute the Program (or a work based on it, under Section 2) in object code or executable form under the terms of Sections 1 and 2 above provided that you also do one of the following:
    1. Accompany it with the complete corresponding machine-readable source code, which must be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange; or,
    2. Accompany it with a written offer, valid for at least three years, to give any third party, for a charge no more than your cost of physically performing source distribution, a complete machine-readable copy of the corresponding source code, to be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange; or,
    3. Accompany it with the information you received as to the offer to distribute corresponding source code. (This alternative is allowed only for noncommercial distribution and only if you received the program in object code or executable form with such an offer, in accord with Subsection b above.)
    The source code for a work means the preferred form of the work for making modifications to it. For an executable work, complete source code means all the source code for all modules it contains, plus any associated interface definition files, plus the scripts used to control compilation and installation of the executable. However, as a special exception, the source code distributed need not include anything that is normally distributed (in either source or binary form) with the major components (compiler, kernel, and so on) of the operating system on which the executable runs, unless that component itself accompanies the executable. If distribution of executable or object code is made by offering access to copy from a designated place, then offering equivalent access to copy the source code from the same place counts as distribution of the source code, even though third parties are not compelled to copy the source along with the object code.
  5. You may not copy, modify, sublicense, or distribute the Program except as expressly provided under this License. Any attempt otherwise to copy, modify, sublicense or distribute the Program is void, and will automatically terminate your rights under this License. However, parties who have received copies, or rights, from you under this License will not have their licenses terminated so long as such parties remain in full compliance.
  6. You are not required to accept this License, since you have not signed it. However, nothing else grants you permission to modify or distribute the Program or its derivative works. These actions are prohibited by law if you do not accept this License. Therefore, by modifying or distributing the Program (or any work based on the Program), you indicate your acceptance of this License to do so, and all its terms and conditions for copying, distributing or modifying the Program or works based on it.
  7. Each time you redistribute the Program (or any work based on the Program), the recipient automatically receives a license from the original licensor to copy, distribute or modify the Program subject to these terms and conditions. You may not impose any further restrictions on the recipients' exercise of the rights granted herein. You are not responsible for enforcing compliance by third parties to this License.
  8. If, as a consequence of a court judgment or allegation of patent infringement or for any other reason (not limited to patent issues), conditions are imposed on you (whether by court order, agreement or otherwise) that contradict the conditions of this License, they do not excuse you from the conditions of this License. If you cannot distribute so as to satisfy simultaneously your obligations under this License and any other pertinent obligations, then as a consequence you may not distribute the Program at all. For example, if a patent license would not permit royalty-free redistribution of the Program by all those who receive copies directly or indirectly through you, then the only way you could satisfy both it and this License would be to refrain entirely from distribution of the Program. If any portion of this section is held invalid or unenforceable under any particular circumstance, the balance of the section is intended to apply and the section as a whole is intended to apply in other circumstances. It is not the purpose of this section to induce you to infringe any patents or other property right claims or to contest validity of any such claims; this section has the sole purpose of protecting the integrity of the free software distribution system, which is implemented by public license practices. Many people have made generous contributions to the wide range of software distributed through that system in reliance on consistent application of that system; it is up to the author/donor to decide if he or she is willing to distribute software through any other system and a licensee cannot impose that choice. This section is intended to make thoroughly clear what is believed to be a consequence of the rest of this License.
  9. If the distribution and/or use of the Program is restricted in certain countries either by patents or by copyrighted interfaces, the original copyright holder who places the Program under this License may add an explicit geographical distribution limitation excluding those countries, so that distribution is permitted only in or among countries not thus excluded. In such case, this License incorporates the limitation as if written in the body of this License.
  10. The Free Software Foundation may publish revised and/or new versions of the General Public License from time to time. Such new versions will be similar in spirit to the present version, but may differ in detail to address new problems or concerns. Each version is given a distinguishing version number. If the Program specifies a version number of this License which applies to it and "any later version", you have the option of following the terms and conditions either of that version or of any later version published by the Free Software Foundation. If the Program does not specify a version number of this License, you may choose any version ever published by the Free Software Foundation.
  11. If you wish to incorporate parts of the Program into other free programs whose distribution conditions are different, write to the author to ask for permission. For software which is copyrighted by the Free Software Foundation, write to the Free Software Foundation; we sometimes make exceptions for this. Our decision will be guided by the two goals of preserving the free status of all derivatives of our free software and of promoting the sharing and reuse of software generally.

    NO WARRANTY

  12. BECAUSE THE PROGRAM IS LICENSED FREE OF CHARGE, THERE IS NO WARRANTY FOR THE PROGRAM, TO THE EXTENT PERMITTED BY APPLICABLE LAW. EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR OTHER PARTIES PROVIDE THE PROGRAM "AS IS" WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE PROGRAM IS WITH YOU. SHOULD THE PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING, REPAIR OR CORRECTION.
  13. IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MAY MODIFY AND/OR REDISTRIBUTE THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED TO LOSS OF DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY YOU OR THIRD PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER PROGRAMS), EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES.

END OF TERMS AND CONDITIONS

How to Apply These Terms to Your New Programs

If you develop a new program, and you want it to be of the greatest possible use to the public, the best way to achieve this is to make it free software which everyone can redistribute and change under these terms.

To do so, attach the following notices to the program. It is safest to attach them to the start of each source file to most effectively convey the exclusion of warranty; and each file should have at least the "copyright" line and a pointer to where the full notice is found.

one line to give the program's name and an idea of what it does.
Copyright (C) 19yy name of author

This program is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 2 of the License, or (at
your option) any later version.

This program is distributed in the hope that it will be useful, but
WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
General Public License for more details.

You should have received a copy of the GNU General Public License
along with this program; if not, write to the Free Software
Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111, USA.

Also add information on how to contact you by electronic and paper mail.

If the program is interactive, make it output a short notice like this when it starts in an interactive mode:

Gnomovision version 69, Copyright (C) 19yy name of author
Gnomovision comes with ABSOLUTELY NO WARRANTY; for details type
`show w'.  This is free software, and you are welcome to redistribute
it under certain conditions; type `show c' for details.

The hypothetical commands `show w' and `show c' should show the appropriate parts of the General Public License. Of course, the commands you use may be called something other than `show 'w and `show c'; they could even be mouse-clicks or menu items -- whatever suits your program.

You should also get your employer (if you work as a programmer) or your school, if any, to sign a "copyright disclaimer" for the program, if necessary. Here is a sample; alter the names:

Yoyodyne, Inc., hereby disclaims all copyright interest
in the program `Gnomovision' (which makes passes at compilers)
written by James Hacker.

signature of Ty Coon, 1 April 1989
Ty Coon, President of Vice

This General Public License does not permit incorporating your program into proprietary programs. If your program is a subroutine library, you may consider it more useful to permit linking proprietary applications with the library. If this is what you want to do, use the GNU Library General Public License instead of this License.

Introduction

GNU troff (or groff) is a system for typesetting documents. troff is very flexible and has been in existence (and use) for about 3 decades. It is quite widespread and firmly entrenched in the @acronym{UNIX} community.

What Is groff?

groff belongs to an older generation of document preparation systems, which operate more like compilers than the more recent interactive @acronym{WYSIWYG}(1) systems. groff and its contemporary counterpart, TeX, both work using a batch paradigm: The input (or source) files are normal text files with embedded formatting commands. These files can then be processed by groff to produce a typeset document on a variety of devices.

Likewise, groff should not be confused with a word processor, since that term connotes an integrated system which includes an editor and a text formatter. Also, many word processors follow the @acronym{WYSIWYG} paradigm which was discussed earlier.

Although @acronym{WYSIWYG} systems may be easier to use, they have a number of disadvantages compared to troff:

"GUIs normally make it simple to accomplish simple actions and impossible to accomplish complex actions." --Doug Gwyn (22/Jun/91 in comp.unix.wizards)

History

troff can trace its origins back to a formatting program called runoff, written by J. E. Saltzer, which ran on MIT's CTSS operating system in the mid-sixties. This name came from the common phrase of the time "I'll run off a document." Bob Morris ported it to the 635 architecture and called the program roff (an abbreviation of runoff). It has then been rewritten as rf for the PDP-7 (before having @acronym{UNIX}), and at the same time (1969), Doug McIllroy rewrote an extended and simplified version of roff in the @acronym{BCPL} programming language.

The first version of @acronym{UNIX} was developed on a PDP-7 which was sitting around Bell Labs. In 1971 the developers wanted to get a PDP-11 for further work on the operating system. In order to justify the cost for this system, they proposed that they would implement a document formatting system for the AT&T patents division. This first formatting program was a reimplementation of McIllroy's roff, written by J. F. Ossanna.

When they needed a more flexible language, a new version of roff called nroff ("Newer roff") was written. It had a much more complicated syntax, but provided the basis for all future versions. When they got a Graphic Systems CAT Phototypesetter, Ossanna wrote a version of nroff which would drive it. It was dubbed troff, for "typesetter roff", although many people have speculated that it actually means "Times roff" because of the use of the Times font family in troff by default. As such, the name troff is pronounced `t-roff' rather than `trough'.

With troff came nroff (they were actually the same program except for some `#ifdef's), which was for producing output for line printers and character terminals. It understood everything troff did, and ignored the commands which were not applicable (e.g. font changes).

Since there are several things which cannot be done easily in troff, work on several preprocessors began. These programs would transform certain parts of a document into troff, which made a very natural use of pipes in @acronym{UNIX}.

The eqn preprocessor allowed mathematical formul@ae{} to be specified in a much simpler and more intuitive manner. tbl is a preprocessor for formatting tables. The refer preprocessor (and the similar program, bib) processes citations in a document according to a bibliographic database.

Unfortunately, Ossanna's troff was written in PDP-11 assembly language and produced output specifically for the CAT phototypesetter. He rewrote it in C, although it was now 7000 lines of uncommented code and still dependent on the CAT. As the CAT became less common, and was no longer supported by the manufacturer, the need to make it support other devices became a priority. However, before this could be done, Ossanna was killed in an auto accident.

So, Brian Kernighan took on the task of rewriting troff. The newly rewritten version produced a device independent code which was very easy for postprocessors to read and translate to the appropriate printer codes. Also, this new version of troff (called ditroff for "device independent troff") had several extensions, which included drawing functions.

Due to the additional abilities of the new version of troff, several new preprocessors appeared. The pic preprocessor provides a wide range of drawing functions. Likewise the ideal preprocessor did the same, although via a much different paradigm. The grap preprocessor took specifications for graphs, but, unlike other preprocessors, produced pic code.

James Clark began work on a GNU implementation of ditroff in early 1989. The first version, groff 0.3.1, was released June 1990. groff included:

Also, a front-end was included which could construct the, sometimes painfully long, pipelines required for all the post- and preprocessors.

Development of GNU troff progressed rapidly, and saw the additions of a replacement for refer, an implementation of the `ms' and `mm' macros, and a program to deduce how to format a document (grog).

It was declared a stable (i.e. non-beta) package with the release of version 1.04 around November 1991.

Beginning in 1999, groff has new maintainers (the package was an orphan for a few years). As a result, new features and programs like grn, a preprocessor for gremlin images, and grohtml, an output device to produce @acronym{HTML} output, have been added.

groff Capabilities

So what exactly is groff capable of doing? groff provides a wide range of low-level text formatting operations. Using these, it is possible to perform a wide range of formatting tasks, such as footnotes, table of contents, multiple columns, etc. Here's a list of the most important operations supported by groff:

Macro Packages

Since groff provides such low-level facilities, it can be quite difficult to use by itself. However, groff provides a macro facility to specify how certain routine operations (e.g. starting paragraphs, printing headers and footers, etc.) should be done. These macros can be collected together into a macro package. There are a number of macro packages available; the most common (and the ones described in this manual) are `man', `mdoc', `me', `ms', and `mm'.

Preprocessors

Although groff provides most functions needed to format a document, some operations would be unwieldy (e.g. to draw pictures). Therefore, programs called preprocessors were written which understand their own language and produce the necessary groff operations. These preprocessors are able to differentiate their own input from the rest of the document via markers.

To use a preprocessor, @acronym{UNIX} pipes are used to feed the output from the preprocessor into groff. Any number of preprocessors may be used on a given document; in this case, the preprocessors are linked together into one pipeline. However, in groff, the user does not need to construct the pipe, but only tell groff what preprocessors to use.

groff currently has preprocessors for producing tables (tbl), typesetting equations (eqn), drawing pictures (pic and grn), and for processing bibliographies (refer). An associated program which is useful when dealing with preprocessors is soelim.

A free implementation of grap, a preprocessor for drawing graphs, can be obtained as an extra package; groff can use grap also.

There are other preprocessors in existence, but, unfortunately, no free implementations are available. Among them are preprocessors for drawing mathematical pictures (ideal) and chemical structures (chem).

Output Devices

groff actually produces device independent code which may be fed into a postprocessor which will produce output for a particular device. Currently, groff has postprocessors for @acronym{PostScript} devices, character terminals, X Windows (for previewing), TeX DVI format, HP LaserJet 4 and Canon LBP printers (which use @acronym{CAPSL}), and @acronym{HTML}.

Credits

Large portions of this manual were taken from existing documents, most notably, the manual pages for the groff package by James Clark, and Eric Allman's papers on the `me' macro package.

The section on the `man' macro package is partly based on Susan G. Kleinmann's `groff_man' manual page written for the Debian GNU/Linux system.

Invoking groff

This section 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 it from its original counterparts provided by the host (see section Environment, for more details). Thus, for example, geqn is GNU eqn. On operating systems like Linux or the Hurd, which don't contain proprietary software, 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 used incarnation of troff. Exception: groff is never replaced by roff.

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, grefer, or gsoelim.

This section only documents options to the groff front end. Many of the arguments to groff are passed on to gtroff, therefore those are also included. Arguments to pre- or postprocessors can be found in section Invoking gpic, section Invoking geqn, section Invoking gtbl, section Invoking ggrn, section Invoking grefer, section Invoking gsoelim, section Invoking grotty, section Invoking grops, section Invoking grohtml, section Invoking grodvi, section Invoking grolj4, section Invoking grolbp, and section Invoking gxditview.

The command line format for groff is:

groff [ -abeghilpstvzCEGNRSUVXZ ] [ -Fdir ] [ -mname ]
      [ -Tdef ] [ -ffam ] [ -wname ] [ -Wname ]
      [ -Mdir ] [ -dcs ] [ -rcn ] [ -nnum ]
      [ -olist ] [ -Parg ] [ -Larg ] [ -Idir ]
      [ files... ]

The command line format for gtroff is as follows.

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

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

Options without an argument can be grouped behind a single @option{-}. A filename of `-' denotes the standard input. It is possible to have whitespace between an option and its parameter.

The grog command can be used to guess the correct groff command to format a file.

Here's the description of the command-line options:

`-h'
Print a help message.
`-e'
Preprocess with geqn.
`-t'
Preprocess with gtbl.
`-g'
Preprocess with ggrn.
`-G'
Preprocess with grap.
`-p'
Preprocess with gpic.
`-s'
Preprocess with gsoelim.
`-R'
Preprocess with grefer. No mechanism is provided for passing arguments to grefer because most grefer options have equivalent commands which can be included in the file. See section grefer, for more details. Note that gtroff also accepts a @option{-R} option, which is not accessible via groff. This option prevents the loading of the `troffrc' and `troffrc-end' files.
`-v'
Make programs run by groff print out their version number.
`-V'
Print the pipeline on stdout instead of executing it.
`-z'
Suppress output from gtroff. Only error messages will be printed.
`-Z'
Do not postprocess the output of gtroff. Normally groff will automatically run the appropriate postprocessor.
`-Parg'
Pass arg to the postprocessor. Each argument should be passed with a separate @option{-P} option. Note that groff does not prepend `-' to arg before passing it to the postprocessor.
`-l'
Send the output to a printer. The command used for this is specified by the print command in the device description file.
`-Larg'
Pass arg to the spooler. Each argument should be passed with a separate @option{-L} option. Note that groff does not prepend a `-' to arg before passing it to the postprocessor.
`-Tdev'
Prepare output for device dev. The default device is `ps', unless changed when groff was configured and built. The following are the output devices currently available:
ps
For @acronym{PostScript} printers and previewers.
dvi
For TeX DVI format.
X75
For a 75dpi X11 previewer.
X100
For a 100dpi X11 previewer.
ascii
For typewriter-like devices.
latin1
For typewriter-like devices that support the Latin-1 (ISO 8859-1) character set.
utf8
For typewriter-like devices which use the Unicode (ISO 10646) character set with UTF-8 encoding.
cp1047
For typewriter-like devices which use the @acronym{EBCDIC} encoding IBM cp1047.
lj4
For an HP LaserJet4-compatible (or other PCL5-compatible) printer.
lbp
For Canon @acronym{CAPSL} printers (LBP-4 and LBP-8 series laser printers).
html
To produce @acronym{HTML} output.
The predefined gtroff string register .T contains the current output device; the read-only number register .T is set to 1 if this option is used (which is always true if groff is used to call gtroff). See section Built-in Registers. The postprocessor to be used for a device is specified by the postpro command in the device description file. (See section Font Files, for more info.) This can be overridden with the @option{-X} option.
`-X'
Preview with gxditview instead of using the usual postprocessor. This is unlikely to produce good results except with @option{-Tps}. Note that this is not the same as using @option{-TX75} or @option{-TX100} to view a document with gxditview: The former will use the metrics of the specified device, whereas the latter will use X-specific fonts and metrics.
`-N'
Don't allow newlines with eqn delimiters. This is the same as the @option{-N} option in geqn.
`-S'
Safer mode. Pass the @option{-S} option to gpic and use the @option{-msafer} macros with gtroff (enabled by default).
`-U'
Unsafe mode. Reverts to the old unsafe behaviour.
`-a'
Generate an @acronym{ASCII} approximation of the typeset output. The read-only register .A is then set to 1. See section Built-in Registers.
`-b'
Print a backtrace with each warning or error message. This backtrace should help track down the cause of the error. The line numbers given in the backtrace may not always be correct: gtroff can get confused by as or am requests while counting line numbers.
`-i'
Read the standard input after all the named input files have been processed.
`-wname'
Enable warning name. Available warnings are described in section Debugging. Multiple @option{-w} options are allowed.
`-Wname'
Inhibit warning name. Multiple @option{-W} options are allowed.
`-E'
Inhibit all error messages.
`-C'
Enable compatibility mode. See section Implementation Differences, for the list of incompatibilites between groff and traditional Unix troff.
`-dcs'
`-dname=s'
Define c or name to be a string s. c must be a one-letter name; name can be of arbitrary length.
`-ffam'
Use fam as the default font family.
`-mname'
Read in the file `tmac.name'. Normally this will be searched for in the library directory of groff.
`-nnum'
Number the first page num.
`-olist'
Output only pages in list, which is a comma-separated list of page ranges; `n' means print page n, `m-n' means print every page between m and n, `-n' means print every page up to n, `n-' means print every page beginning with n. gtroff will exit after printing the last page in the list. All the ranges are inclusive on both ends. Within gtroff, this information can be extracted with the `.P' register. See section Built-in Registers.
`-rcn'
`-rname=n'
Set number register c or name to the value n. c must be a one-letter name; name can be of arbitrary length. n can be any gtroff numeric expression.
`-Fdir'
Search `dir' for subdirectories `devname' (name is the name of the device), for the `DESC' file, and for font files before looking in the standard directory.
`-Mdir'
Search directory `dir' for macro files before the standard directory.
`-Idir'
This option is as described in section gsoelim. It implies the @option{-s} option.

Environment

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

GROFF_COMMAND_PREFIX
If this is set to X, then groff will run Xtroff instead of gtroff. This also applies to tbl, pic, eqn, grn, refer, and soelim. It does not apply to grops, grodvi, grotty, grohtml, grolj4, and gxditview.
GROFF_TMAC_PATH
A colon-separated list of directories in which to search for macro files.
GROFF_TYPESETTER
The default output device.
GROFF_FONT_PATH
A colon-separated list of directories in which to search for the devname directory.
PATH
The search path for commands executed by groff.
GROFF_TMPDIR
The directory in which temporary files will be created. If this is not set and @env{TMPDIR} is set, temporary files will be created in that directory. Otherwise temporary files will be created in a system-dependent default directory (on Unix and GNU/Linux systems, this is usually `/tmp'). The grops and grefer commands can create temporary files in this directory.

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

Invocation Examples

This section will list several common uses of groff and the command line which will accomplish it.

groff file

This command processes `file' without a macro package or a preprocessor. The output device is the default, `ps', and the output is sent to stdout.

groff -t -mandoc -Tascii file | less

This is basically what a call to the man program does. The manual page `file' is processed with the `mandoc' macros (which in turn either calls the `man' or the `mdoc' macro package), using the tbl preprocessor and the @acronym{ASCII} output device. Finally, the result is displayed with the less pager.

groff -X -m me file

Preview `file' with gxditview, using the `me' macro package. Since no @option{-T} option is specified, use the default device (`ps'). Note that you can either say `-m me' or `-me'; the latter is an anachronism from the early days of @acronym{UNIX}.(2)

groff -man -rD1 -z file

Check `file' with the `man' macro package, forcing double-sided printing -- don't produce any output.

grog

grog reads files, guesses which of the groff preprocessors and/or macro packages are required for formatting them, and prints the groff command including those options on the standard output. The options generated are one of @option{-e}, @option{-man}, @option{-me}, @option{-mm}, @option{-ms}, @option{-p}, @option{-R}, @option{-g}, @option{-G}, @option{-s}, and @option{-t}.

A special file name `-' is taken to refer to the standard input. If no files are specified the standard input will be read. Any specified options will be included in the printed command. No space is allowed between options and their arguments. For example,

grog -Tdvi paper.ms

will guess the appropriate command to print `paper.ms' and then print it to the command line after adding the @option{-Tdvi} option. For direct execution, enclose the call to grog in backquotes at the @acronym{UNIX} shell prompt:

`grog -Tdvi paper.ms` > paper.dvi

As seen in the example, it is still necessary to redirect the output to something meaningful (i.e. either a file or a pager program like less).

Tutorial for Macro Users

Most users tend to use a macro package to format their papers. This means that the whole breadth of groff is not necessary for most people. This chapter covers the material needed to efficiently use a macro package.

Basics

This section covers some of the basic concepts necessary to understand how to use a macro package.(3) References are made throughout to more detailed information, if desired.

gtroff 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 escapes), which tell gtroff how to format the output. For more detail on this, see section Embedded Commands.

The word argument is used in this chapter to mean a word or number which 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. More details on this can be found in section Request Arguments.

The primary function of gtroff is to collect words from input lines, fill output lines with those words, justify the right-hand margin by inserting extra spaces in the line, 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,...

will be 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,...

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 cause a break, which starts a new output line. Some requests cause a break automatically, as do blank input lines and input lines beginning with a space.

Not all input lines are text to be formatted. Some of the input lines are requests which describe how to format the text. Requests always have a period (`.') or an apostrophe (`'') as the first character of the input line.

The text 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 gtroff. First, keep the input lines short. Short input lines are easier to edit, and gtroff will pack words onto longer lines anyhow. In keeping with this, it is helpful to begin a new line after every period, comma, or phrase, since common corrections are to add or delete sentences or phrases. Secondly, do not hyphenate words at the end of lines -- gtroff is smart enough to hyphenate words for the user 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".

gtroff will double space output text automatically if you use the request `.ls 2'. Single spaced mode can be reactivated by typing `.ls 1'.

A number of requests allow to change the way the output looks, sometimes called the layout of the output page. Most of these requests adjust the placing of white space (blank lines or spaces).

The `.bp' request starts a new page, causing a line break.

The request `.sp N' leaves N lines of blank space. N can be omitted (meaning skip 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.

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 groff to center zero more lines, in other words, stop centering.

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.

Common Features

gtroff provides very low level operations for formatting a document. There are many common routine operations which are done in all documents. These common operations are written into macros and collected into a macro package.

All macro packages provide certain common capabilities which fall into the following categories.

Paragraphs

One of the most common and most used capability is starting a paragraph. There are a number of different types of paragraphs, any of which can be initiated with macros supplied by the macro package. Normally, paragraphs start with a blank line and the first line indented, like the text in this manual. There are also block style paragraphs, which 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.

And there are also indented paragraphs which begin with a tag or label at the margin and the remaining text indented.

one   This is  the first paragraph.  Notice  how the first
      line of  the resulting  paragraph lines up  with the
      other lines in the paragraph.
longlabel
      This  paragraph   had  a  long   label.   The  first
      character of text on the first line will not line up
      with  the  text  on  second  and  subsequent  lines,
      although they will line up with each other.

A variation of this is a bulleted list.

Sections and Chapters

Most macro packages supply some form of section headers. The simplest kind is simply the heading on a line by itself in bold type. Others supply automatically numbered section heading or different heading styles at different levels. Some, more sophisticated, macro packages supply macros for starting chapters and appendices.

Headers and Footers

Every macro package gives some way to manipulate the headers and footers (or titles) on each page. Some packages will allow for different ones on the even and odd pages (for material printed in a book form).

The titles are called three-part titles, that is, there is a left-justified part, a centered part, and a right-justified part. An automatically generated page number may be put in any of these fields with the `%' character (see section Page Layout, for more details).

Page Layout

Most macro packages let the user specify top and bottom margins and other details about the appearance of the printed pages.

Displays

Displays are sections of text to be set off from the body of the paper. Major quotes, tables, and figures are types of displays, as are all the examples used in this document.

Major quotes are quotes which are several lines long, and hence are set in from the rest of the text without quote marks around them.

A list is an indented, single spaced, unfilled display. Lists should be used when the material to be printed should not be filled and justified like normal text, such as columns of figures or the examples used in this paper.

A keep is a display of lines which are kept on a single page if possible. An example for a keep might be a diagram. Keeps differ from lists in that lists may be broken over a page boundary whereas keeps will not.

Floating keeps move relative to the text. Hence, they are good for things which will be referred to by name, such as "See figure 3". A floating keep will appear at the bottom of the current page if it will fit; otherwise, it will appear at the top of the next page. Meanwhile, the surrounding text will `flow' around the keep, thus leaving now blank areas.

Footnotes and Annotations

There are a number of requests to save text for later printing.

Footnotes are printed at the bottom of the current page.

Delayed text is very similar to a footnote except that it is printed when called for explicitly. This allows a list of references to appear (for example) at the end of each chapter, as is the convention in some disciplines.

Most macro packages which supply this functionality also supply a means of automatically numbering either type of annotation.

Table of Contents

Tables of contents are a type of delayed text having a tag (usually the page number) attached to each entry after a row of dots. The table accumulates throughout the paper until printed, usually after the paper has ended. Many macro packages will provide the ability to have several tables of contents (i.e. one standard one, one for tables, etc).

Indices

While some macro packages will use the term index, none actually provide that functionality. The facilities they call indices are actually more appropriate for tables of contents.

Paper Formats

Some macro packages provide stock formats for various kinds of documents. Many of them provide a common format for the title and opening pages of a technical paper. The `mm' macros in particular provide formats for letters and memoranda.

Multiple Columns

Some macro packages (but not `man') provide the ability to have two or more columns on a page.

Font and Size Changes

The built-in font and size functions are not always intuitive, so all macro packages provide macros to make these operations simpler.

Predefined Strings

Most macro packages provide various predefined strings for a variety of uses; examples are sub- and superscripts, printable dates, quotes and various special characters.

Preprocessor Support

All macro packages provide support for the various preprocessors.

Configuration and Customization

Some macro packages provide means of customizing many of the details of how the package behaves. This ranges from setting the default type size to changing the appearance of section headers.

Macro Packages

This chapter documents the main macro packages that come with groff.

`man'

This is the most popular and probably the most important macro package of groff. It is easy to use, and a vast majority of manual pages are based on it.

Options

The command line format for using the `man' macros with groff is:

groff -m man [ -rC1 ] [ -rD1 ] [ -rPnnn ] [ -rSxx ]
      [ -rXnnn ] [ files... ] 

It is possible to use `-man' instead of `-m man'.

-rC1
If more than one manual page is given on the command line, number the pages continuously, rather than starting each at 1.
-rD1
Double-sided printing. Footers for even and odd pages are formatted differently.
-rPnnn
Enumeration of pages will start with nnn rather than with 1.
-rSxx
Use xx (which can be 10, 11, or 12pt) as the base document font size instead of the default value of 10pt.
-rXnnn
After page nnn, number pages as nnna, nnnb, nnnc, etc. For example, the option @option{-rX2} will produce the following page numbers: 1, 2, 2a, 2b, 2c, etc.

Usage

This section describes the available macros for manual pages. For further customization, put additional macros and requests into the file `man.local' which will be loaded immediately after `tmac.an'.

@Defmac{TH, title section [extra1] [extra2] [extra3]} Sets the title of the man page to title and the section to section, which must have a value between 1 and 8. The value of section may also have a string appended, e.g. `.pm', to indicate a specific subsection of the man pages.

Both title and section are positioned at the left and right in the header line (with section in parentheses immediately appended to title. extra1 will be positioned in the middle of the footer line. extra2 will be positioned at the left in the footer line (resp. at the left on even pages and at the right on odd pages if double-sided printing is active). extra3 is centered in the header line.

For @acronym{HTML} output, headers and footers are completely suppressed.

Additionally, this macro starts a new page; the new line number is 1 again (except if the @option{-rC1} option is given on the command line) -- this feature is intended only for formatting multiple man pages; a single man page should contain exactly one TH macro at the beginning of the file. @end_Defmac

@Defmac{SH, [heading]} Sets up an unnumbered section heading sticking out to the left. Prints out all the text following SH up to the end of the line (resp. the text in the next line if there is no argument to SH) in bold face, one size larger than the base document size. Additionally, the left margin for the following text is reset to its default value. @end_Defmac

@Defmac{SS, [heading]} Sets up an unnumbered section heading. Prints out all the text following SS up to the end of the line (resp. the text in the next line if there is no argument to SS) in bold face, at the same size as the base document size. Additionally, the left margin for the following text is reset to its default value. @end_Defmac

@Defmac{TP, [nnn]} Sets up an indented paragraph with label. The indentation is set to nnn if that argument is supplied (the default unit is `n' if omitted), otherwise it is set to the default indentation value.

The first line of text following this macro is interpreted as a string to be printed flush-left, as it is appropriate for a label. It is not interpreted as part of a paragraph, so there is no attempt to fill the first line with text from the following input lines. Nevertheless, if the label is not as wide as the indentation, then the paragraph starts at the same line (but indented), continuing on the following lines. If the label is wider than the indentation, then the descriptive part of the paragraph begins on the line following the label, entirely indented. Note that neither font shape nor font size of the label is set to a default value; on the other hand, the rest of the text will have default font settings. @end_Defmac

@Defmac{LP} @Defmacx{PP} @Defmacx{P} These macros are mutual aliases. Any of them causes a line break at the current position, followed by a vertical space downwards by the amount specified by the PD macro. The font size and shape are reset to the default value (10pt resp. Roman). Finally, the current left margin is restored. @end_Defmac

@Defmac{IP, [designator] [nnn]} Sets up an indented paragraph, using designator as a tag to mark its beginning. The indentation is set to nnn if that argument is supplied (default unit is `n'), otherwise the default indentation value is used. Font size and face of the paragraph (but not the designator) are reset to their default values. To start an indented paragraph with a particular indentation but without a designator, use `""' (two double quotes) as the first argument of IP.

For example, to start a paragraph with bullets as the designator and 4en indentation, write

.IP \(bu 4

@end_Defmac

@Defmac{HP, [nnn]} Sets up a paragraph with hanging left indentation. The indentation is set to nnn if that argument is supplied (default unit is `n'), otherwise the default indentation value is used. Font size and face are reset to their default values. @end_Defmac

@Defmac{RS, [nnn]} This macro moves the left margin to the right by the value nnn if specified (default unit is `n'); otherwise the default indentation value is used. Calls to the RS macro can be nested. @end_Defmac

@Defmac{RE, [nnn]} This macro moves the left margin back to level nnn; if no argument is given, it moves one level back. The first level (i.e., no call to RS yet) has number 1, and each call to RS increases the level by 1. @end_Defmac

To summarize, the following macros cause a line break with the insertion of vertical space (which amount can be changed with the PD macro): SH, SS, TP, LP (PP, P), IP, and HP.

The macros RS and RE also cause a break but do not insert vertical space.

Macros to set fonts

The standard font is Roman; the default text size is 10 point.

@Defmac{SM, [text]} Causes the text on the same line or the text on the next line to appear in a font that is one point size smaller than the default font. @end_Defmac

@Defmac{SB, [text]} Causes the text on the same line or the text on the next line to appear in boldface font, one point size smaller than the default font. @end_Defmac

@Defmac{BI, text} Causes text on the same line to appear alternately in bold face and italic. The text must be on the same line as the macro call. Thus,

.BI this "word and" that

would cause "this" and "that" to appear in bold face, while "word and" appears in italics. @end_Defmac

@Defmac{IB, text} Causes text to appear alternately in italic and bold face. The text must be on the same line as the macro call. @end_Defmac

@Defmac{RI, text} Causes text on the same line to appear alternately in roman and italic. The text must be on the same line as the macro call. @end_Defmac

@Defmac{IR, text} Causes text on the same line to appear alternately in italic and roman. The text must be on the same line as the macro call. @end_Defmac

@Defmac{BR, text} Causes text on the same line to appear alternately in bold face and roman. The text must be on the same line as the macro call. @end_Defmac

@Defmac{RB, text} Causes text on the same line to appear alternately in roman and bold face. The text must be on the same line as the macro call. @end_Defmac

@Defmac{R, [text]} Causes text to appear in roman font. If no text is present on the line where the macro is called, then the text of the next line appears in roman. This is the default font to which text is returned at the end of processing of the other macros. @end_Defmac

@Defmac{B, [text]} Causes text to appear in bold face. If no text is present on the line where the macro is called, then the text of the next line appears in bold face. @end_Defmac

@Defmac{I, [text]} Causes text to appear in italic. If no text is present on the line where the macro is called, then the text of the next line appears in italic. @end_Defmac

Miscellaneous

The default indentation is 7.2n for all output devices except for grohtml which uses 1.2i instead.

@Defmac{DT} Sets tabs every 0.5 inches. Since this macro is always called during a TH request, it makes sense to call it only if the tab positions have been changed. @end_Defmac

@Defmac{PD, [nnn]} Adjusts the empty space before a new paragraph (resp. section). The optional argument gives the amount of space (default units are `v'); without parameter, the value is reset to its default value (1 line for tty devices, 0.4v otherwise). @end_Defmac

This affects the macros SH, SS, TP, LP (resp. PP and P), IP, and HP.

The following strings are defined:

@Defmac{\\*S} Switch back to the default font size. @end_Defmac

@Defmac{\\*R} The `registered' sign. @end_Defmac

@Defmac{\\*(Tm} The `trademark' sign. @end_Defmac

@Defmac{\\*(lq} @Defmacx{\\*(rq} Left and right quote. This is equal to \(lq and \(rq, respectively. @end_Defmac

If a preprocessor like gtbl or geqn is needed, it has become common usage to make the first line of the man page look like this:

.\" word

Note the single space character after the double quote. word consists of letters for the needed preprocessors: `e' for geqn, `r' for grefer, `t' for gtbl. Modern implementations of the man program read this first line and automatically call the right preprocessor(s).

`mdoc'

`ms'

`me'

`mm'

Programming Tutorial

This chapter covers all of the facilities of gtroff. Users of macro packages may skip it if not interested in details.

Text

gtroff input files contain text with control commands interspersed throughout. But, even without control codes, gtroff will still do several things with the input text: filling and adjusting, adding additional space after sentences, hyphenating and inserting implicit line breaks.

Filling and Adjusting

When gtroff reads in text it collects words from input and fits as many of them together on one output line as it can. This is known as filling.

Once gtroff has a filled line it will try to adjust it. This means it will widen the spacing between words until the text reaches the right margin (in the default adjustment mode). Extra spaces between words are preserved, but spaces at the end of lines are ignored. Spaces at the front of a line will cause a break (breaks will be explained in section Implicit Line Breaks)

See section Manipulating Filling and Adjusting.

Hyphenation

Since the odds are not great for finding a set of words, for every output line, which will fit nicely on a line without inserting excessive amounts of space between words, gtroff will hyphenate words so that lines can be justified without there being too much space between words. It uses an internal hyphenation algorithm (a simplified version of the algorithm used within TeX) to indicate which words can be hyphenated and how to do so. When a word is hyphenated the first part of the word will be added to the current filled line being output (with an attached hyphen), and the other portion will be added to the next line to be filled.

See section Manipulating Hyphenation.

Sentences

Although it is often debated, some typesetting rules say there should be different amounts of space after various punctuation marks. For example, the Chicago typsetting manual says that a period at the end of a sentence should have twice as much space following it as would a comma or a period as part of an abbreviation.

gtroff does this by flagging certain characters (normally `!', `?' and `.') as end of sentence characters. When gtroff encounters one of these characters at the end of a line it will append two sentence spaces in the formatted output. (This justifies one of the conventions mentioned in section Input Conventions.)

In addition, the following characters resp. glyphs are treated transparently while handling end of sentence characters: `"', `'', `)', `]', `*', dg, and rq.

See the cflags request in section Using Symbols, for more details.

To prevent the insertion of extra space after an end of sentence character (at the end of a line), append \&.

Tab Stops

gtroff translates tabulator characters, also called tabs (normally code point @acronym{ASCII} 0x09 resp. @acronym{EBCDIC} 0x05), in the input into movements to the next tabulator stop. These tab stops are initially located every half inch across the page. Using this, simple tables can easily be made. However, it can often be deceptive as the appearance (and width) of the text on a terminal and the results from gtroff can vary greatly.

Also, a possible sticking point is that lines beginning with tab characters will still be filled, again producing unexpected results. For example, the following input

  • 1 2 3
  • 4 5 will produce
  • 1 2 3 4 5 See section Tabs and Fields.

    Implicit Line Breaks

    An important concept in gtroff is the break. When a break occurs, gtroff will output the partially filled line (unjustified), and resume collecting and filling text on the next output line.

    There are several ways to cause a break in gtroff. A blank line will not only cause a break, but it will also cause a one line vertical space (effectively a blank line) to be output.

    A line which begins with a space will cause a break and the space will be output at the beginning of the next line. Note that this space isn't adjusted, even in fill mode.

    The end of file will also cause a break -- otherwise the last line of the document may vanish!

    Certain requests also cause breaks, implicitly or explicitly. This will be discussed in section Manipulating Filling and Adjusting.

    Input Conventions

    Since gtroff does filling automatically, it is traditional in groff not to try and type things in as nicely formatted paragraphs. These are some conventions commonly used when typing gtroff text:

    Measurements

    gtroff (like any other programs) requires numeric parameters to specify various measurements. Most numeric parameters(4) may have a measurement unit attached. These units are specified as a single character which immediately follows the number or expression. Each of these units are understood, by gtroff, to be a multiple of its basic unit. So, whenever a different measurement unit is specified gtroff converts this into its basic units. This basic unit, represented by a `u', is a device dependent measurement which is quite small, ranging from 1/75th to 1/72000th of an inch. The values may be given as fractional numbers; however, fractional basic units are always rounded to integers.

    Some of the measurement units are completely independent of any of the current settings (e.g. type size) of gtroff.

    i
    Inches. An antiquated measurement unit still in use in certain backwards countries. One inch is equal to 2.54cm.
    c
    Centimeters. One centimeter is equal to 0.3937in.
    p
    Points. This is a typesetter's measurement used for measure type size. It is 72 points to an inch.
    P
    Pica. Another typesetting measurement. 6 Picas to an inch (and 12 points to a pica).
    s
    z
    See section Fractional Type Sizes, for a discussion of these units.

    The other measurements understood by gtroff are dependent on settings currently in effect in gtroff. These are very useful for specifying measurements which should look proper with any size of text.

    m
    Ems. This unit is equal to the current font size in points. So called because it is approximately the width of the letter `m' in the current font.
    n
    Ens. This is half of an em.
    v
    Vertical space. This is equivalent to the current line spacing. See section Sizes, for more information about this.
    M
    100ths of an em.

    Default Units

    Many requests take a default unit. While this can be helpful at times, it can cause strange errors in some expressions. For example, the line length request expects em units. Here are several attempts to get a line length of 3.5 inches and their results:

    3.5i      =>   3.5i
    7/2       =>   0i
    7/2i      =>   0i
    7i/2      =>   0.1i
    7i/2u     =>   3.5i
    

    Everything will be converted to basic units first. In the above example it is assumed that 1i equals 240u, and 1m equals 10p (thus 1m equals 33u). The value 7i/2 will be first handled as 7i/2m, then converted to 1680u/66u which is 25u, and this is approximately 0.1i.

    As a conclusion, the safest way to specify measurements is to always attach a scaling indicator. If you want to multiply or divide by a certain scalar value, use `u' as the unit for that value.

    Expressions

    gtroff has most of operators common to other languages:

    Parentheses may be used as in any other language. However, in gtroff they are necessary to ensure order of evaluation. gtroff has no operator precedence; expressions are evaluated left to right. This means that `3+5*4' is evaluated as if it were parenthesized like `(3+5)*4', not as `3+(5*4)', as might be expected.

    For many requests which cause a motion on the page, the unary operators work differently. The `+' and `-' operators then indicate a motion relative to the current position (down or up, respectively), and the `|' operator indicates an absolute position on the page or input line. `+' and `-' are also treated differently by the following requests and escapes: bp, in, ll, lt, nm, nr, pl, pn, po, ps, rt, ti, \R, and \s. Here the plus and minus signs indicate increments resp. decrements.

    See section Setting Registers.

    Due to the way arguments are parsed, spaces are not allowed in expressions, unless the entire expression is surrounded by parentheses.

    See section Request Arguments, and section Conditionals and Loops.

    Identifiers

    Like any other language, gtroff has rules for properly formed identifiers. In gtroff, an identifier can be made up of almost any printable character, with the exception of the following characters:

    For example, any of the following is valid.

    br
    PP
    (l
    end-list
    @_
    

    Note that identifiers longer than two characters with a closing bracket (`]') in its name can't be accessed with escape sequences which expect an identifier as a parameter. For example, `\[foo]]' will access the glyph `foo', followed by `]', whereas `\C'foo]'' really asks for glyph `foo]'.

    @Deffn{Escape, \\A, ident} Whether an identifier ident is valid in gtroff can be tested with the \A escape. It expands to the character 1 or 0 according to whether its argument (usually delimited by quotes) is or is not acceptable as the name of a string, macro, diversion, number register, environment, or font. It will return 0 if no argument is given. This is useful for looking up user input in some sort of associative table.

    \A'end-list'
        => 1
    

    @end_Deffn

    See section Escapes, for details on parameter delimiting characters.

    Identifiers in gtroff can be any length, but, in some contexts, gtroff needs to be told where identifiers end and text begins (and in different ways depending on their length):

    Unlike many other programming languages, undefined identifiers are silently ignored or expanded to nothing.

    See section Interpolating Registers, and section Strings.

    Embedded Commands

    Most documents need more functionality beyond filling, adjusting and implicit line breaking. In order to gain further functionality, gtroff allows commands to be embedded into the text, in two ways.

    The first is a request which takes up an entire line, and does some large scale operation (e.g. break lines, start new pages).

    The other is an escape which can be embedded anywhere in the text, or even as an argument to a request. Escapes generally do more minor operations like sub- and superscripts, print a symbol, etc.

    Requests

    A request line begins with a control character, which is either a single quote (`'', the no-break control character) or a period (`.', the normal control character). These can be changed; see section Character Translations, for details. After this there may be optional tabs or spaces followed by an identifier which is the name of the request. This may be followed by any number of space-separated arguments.

    To begin a line with a control character without it being interpreted, precede it with \&. This represents a zero width space, which means it will not affect the output.

    In most cases the period is used as a control character. Several requests will cause a break implicitly; using the single quote control character will prevent this.

    Request Arguments

    Arguments to requests (and macros) are processed much like the shell: The line is split into arguments according to spaces. An argument which is intended to contain spaces can either be enclosed in quotes (single or double), or have the spaces escaped with backslashes.

    Here are a few examples:

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

    The first line is the uh macro being called with 3 arguments, `The', `Mouse', and `Problem'. The latter two have the same effect or calling the uh macro with one argument, `The Mouse Problem'.(5)

    Note, however, that the ds request works differently. See section Strings, for more details.

    Macros

    gtroff has a macro facility for defining a series of lines which can be invoked by name. They are called in the same manner as requests -- arguments also may be passed in the same manner.

    See section Writing Macros, and section Request Arguments.

    Escapes

    Escapes may occur anywhere in the input to gtroff. They usually begin with a backslash and are followed by a single character which indicates the function to be performed. The escape character can be changed; see section Character Translations.

    Escape sequences which require an identifier as a parameter accept three possible syntax forms.

    Examples:

    \fB
    \n(XX
    \*[TeX]
    

    Other escapes may require several arguments and/or some special format. In such cases the argument is traditionally enclosed in single quotes (and quotes are always used in this manual for the definitions of escape sequences). The enclosed text is then processed according to what that escape expects. Example:

    \l'1.5i\(bu'
    

    Note that the quote character can be replaced with any other character which does not occur in the argument (even a newline or a space character) in the following escapes: \o, \b, and \X. This makes e.g.

    A caf
    \o
    e\'
    
    in Paris
      => A caf'e in Paris
    

    possible, but it is better not to use this feature to avoid confusion.

    The following escapes sequences (which are handled similarly to characters since they don't take a parameter) are also allowed as delimiters: \%, `\ ', \|, \^, \{, \}, \', \`, \-, \_, \!, \?, \@, \), \/, \,, \&, \~, \0, \a, \c, \d, \e, \E, \p, \r, \t, and \u. Again, don't use these if possible.

    No newline characters as delimiters are allowed in the following escapes: \A, \Z, \C, and \w.

    Finally, the escapes \D, \h, \H, \l, \L, \N, \R, \s, \S, \v, and \x can't use the following characters as delimiters:

    To have a backslash (resp. the current escape character) appear in the output several escapes are defined: \\, \e or \E. These are very similar, and only differ with respect to being used in macros or diversions. See section Copy-in Mode, and section Diversions, for more information.

    See section Identifiers, and section Character Translations.

    Comments

    Probably one of the most(6) common forms of escapes is the comment.

    @Deffn{Escape, \\"} Start a comment. Everything to the end of the input line is ignored.

    This may sound simple, but it can be tricky to keep the comments from interfering with the appearance of the final output.

    If the escape is to the right of some text or a request, that portion of the line will be ignored, but the space leading up to it will be noticed by gtroff. This only affects the .ds request.

    One possibly irritating idiosyncracy is that tabs must not be used to line up comments. Tabs are not treated as white space between the request and macro arguments.

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

    Test
    \" comment
    Test
    

    will produce

    Test
    
    Test
    

    As a consequence, it is common to start the line with .\" which will cause the line to be treated as an undefined request and thus ignored completely.

    Another commenting scheme seen sometimes is three consecutive single quotes ("') at the beginning of a line. This works, but gtroff will give a warning about an undefined macro (namely "), which is harmless, but irritating. @end_Deffn

    @Deffn{Escape, \\#} To avoid all this, gtroff has a new comment mechanism using the \# escape. This escape works the same as \" except that the newline is also ignored:

    Test
    \# comment
    Test
    

    will produce

    Test Test
    

    as expected. @end_Deffn

    For commenting away large blocks of text, the ig request may be useful.

    See section Strings.

    Registers

    Numeric variables in gtroff are called registers. There are a number of built-in registers, supplying anything from the date to details of formatting parameters.

    See section Identifiers, for details on register identifiers.

    Setting Registers

    Registers are defined resp. set via the nr request or the \R escape.

    @Deffn{Request, nr, ident value} @Deffnx{Escape, \\R, ident value} Set number register ident to value. If ident doesn't exist, it will be created.

    The argument to \R has to be enclosed in quotes usually. See section Escapes, for details on parameter delimiting characters. @end_Deffn

    For example, the following two lines are equivalent:

    .nr a 1
    \R'a 1'
    

    Both nr and \R have two additional special forms to increment resp. decrement a register.

    @Deffn{Request, nr, ident +value} @Deffnx{Request, nr, ident -value} @Deffnx{Escape, \\R, ident +value} @Deffnx{Escape, \\R, ident -value} Increment (decrement) register ident by value.

    .nr a 1
    .nr a +1
    \na
        => 2
    

    To assign the negated value of a register to another register, some care must be taken to get the desired result:

    .nr a 7
    .nr b 3
    .nr a -\nb
    \na
        => 4
    .nr a (-\nb)
    \na
        => -3
    

    The surrounding parentheses prevent the interpretation of the minus sign as a decrementing operator. An alternative is to start the assignment with a `0':

    .nr a 7
    .nr b -3
    .nr a \nb
    \na
        => 4
    .nr a 0\nb
    \na
        => -3
    

    @end_Deffn

    @Deffn{Request, rr, ident} Remove number register ident. If ident doesn't exist, the request is ignored. @end_Deffn

    @Deffn{Request, rnn, ident1 ident2} Rename number register ident1 to ident2. If either ident1 or ident2 doesn't exist, the request is ignored. @end_Deffn

    @Deffn{Request, aln, ident1 ident2} This request creates an alias ident1 for a number register ident2. The new name and the old name will be exactly equivalent. If ident1 is undefined, a warning of type `reg' will be generated, and the request will be ignored. See section Debugging, for information about warnings. @end_Deffn

    Interpolating Registers

    Numeric registers can be accessed via the \n escape.

    @Deffn{Escape, \\n, ident} Interpolate number register ident. This means that the value of the register is expanded in-place while gtroff is parsing the input line.

    .nr a 5
    .nr as \na+\na
    \n(as
        => 10
    

    @end_Deffn

    Auto-increment

    Number registers can also be auto-incremented and auto-decremented. The increment resp. decrement factor can be specified with a third argument to the nr request or \R escape.

    @Deffn{Request, nr, ident value incr} Set number register ident to value; the increment for auto-incrementing is set to incr. Note that the \R escape doesn't support this notation. @end_Deffn

    To activate auto-incrementing, the escape \n has a special syntax form.

    @Deffn{Escape, \\n, +ident} @Deffnx{Escape, \\n, -ident} Before interpolating, increment resp. decrement ident by the auto-increment value as specified with the nr request (or the \R escape). If no auto-increment value has been specified, both syntax forms are identical to \n. @end_Deffn

    For example,

    .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]
    

    produces

    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, the following can be used:

    .nr a \na 10
    

    Assigning Formats

    When a register is used in the text of an input file (as opposed to part of an expression), it is textually replaced (or interpolated) with a representation of that number. This output format can be changed to a variety of formats (numbers, Roman numerals, etc.). This is done using the af request.

    @Deffn{Request, af, ident format} Change the output format of a number register. The first argument ident is the name of the number register to be changed, and the second argument format is the output format. The following output formats are available:

    1
    Decimal arabic numbers. This is the default format: 1, 2, 3, @enddots{}
    0...0
    Decimal numbers with as many digits as specified. So, `00' would result in printing numbers as 01, 02, 03, @enddots{} In fact, any digit instead of zero will do; gtroff only counts how many digits are specified. As a consequence, af's default format `1' could be specified as `0' also (and exactly this is returned by the \g escape, see below).
    I
    Upper-case Roman numerals: 0, I, II, III, IV, @enddots{}
    i
    Lower-case Roman numerals: 0, i, ii, iii, iv, @enddots{}
    A
    Upper-case letters: A, B, C, ..., Z, AA, AB, @enddots{}
    a
    Lower-case letters: a, b, c, ..., z, aa, ab, @enddots{}

    Omitting the number register format will cause a warning of type `missing'. See section Debugging, for more details. Specifying a nonexistent format causes an error.

    The following example will produce `10, X, j, 010':

    .nr a 10
    .af a 1           \" the default format
    \na,
    .af a I
    \na,
    .af a a
    \na,
    .af a 001
    \na
    

    The largest number representable for the `i' and `I' formats is 39999 (resp. -39999); @acronym{UNIX} troff uses `z' and `w' to represent 10000 and 5000 in Roman numerals, and so does gtroff. Currently, the correct glyphs of Roman numeral five thousand and Roman numeral ten thousand (Unicode code points U+2182 and U+2181, respectively) are not available.

    If ident doesn't exist, it will be created.

    Changing the output format of a read-only register causes an error. It is necessary to first copy the register's value to a writeable register, then apply the af request to this other register. @end_Deffn

    @Deffn{Escape, \\g, ident} Return the current format of the specified register ident. For example, `\ga' after the previous example would produce the string `000'. If the register hasn't been defined yet, nothing is returned. @end_Deffn

    Built-in Registers

    The following lists some built-in registers which are not described elsewhere in this manual. Any register which begins with a `.' is read-only. A complete listing of all built-in registers can be found in section Register Index.

    .H
    Horizontal resolution in basic units.
    .V
    Vertical resolution in basic units.
    dw
    Day of the week (1-7).
    dy
    Day of the month (1-31).
    mo
    Current month (1-12).
    year
    The current year.
    yr
    The current year minus 1900. Unfortunately, the documentation of @acronym{UNIX} Version 7's troff had a year 2000 bug: It incorrectly claimed that yr contains the last two digits of the year. That claim has never been true of either traditional troff or GNU troff. Old troff input that looks like this:
    '\" The following line stopped working after 1999
    This document was formatted in 19\n(yr.
    
    can be corrected as follows:
    This document was formatted in \n[year].
    
    or, to be portable to older troff versions, as follows:
    .nr y4 1900+\n(yr
    This document was formatted in \n(y4.
    
    .c
    c.
    The current input line number. Register `.c' is read-only, whereas `c.' (a gtroff extension) is writable also, affecting both `.c' and `c.'.
    ln
    The current output line number after a call to the nm request to activate line numbering.
    .x
    The major version number. For example, if the version number is 1.03 then .x will contain `1'.
    .y
    The minor version number. For example, if the version number is 1.03 then .y will contain `03'.
    .Y
    The revision number of groff.
    .g
    Always 1. Macros should use this to determine whether they are running under GNU troff.
    .A
    If the command line option @option{-a} is used to produce an @acronym{ASCII} approximation of the output, this is set to 1, zero otherwise. See section Options.
    .P
    This register is set to 1 (and to 0 otherwise) if the current page is actually being printed, i.e., if the @option{-o} option is being used to only print selected pages. See section Options, for more information.
    .T
    If gtroff is called with the @option{-T} command line option, the number register .T is set to 1, and zero otherwise. See section Options. Additionally, gtroff predefines a single (read/write) string register .T which contains the current output device (for example, `latin1' or `ps').

    Manipulating Filling and Adjusting

    Various ways of causing breaks were given in section Implicit Line Breaks. The br request will likewise cause a break. Several other requests will also cause breaks, but implicitly. These are bp, ce, cf, fi, fl, in, nf, rj, sp, ti, and trf.

    @Deffn{Request, br, } Break the current line, i.e., the input collected so far will be emitted without adjustment.

    If the no-break control character is used, no break will happen:

    a
    'br
    b
        => a b
    

    @end_Deffn

    Initially, gtroff will fill and adjust text to both margins. Filling can be disabled via the nf request and re-enabled with the fi request.

    @Deffn{Request, fi, } Activate fill mode (which is the default). This request implicitly enables adjusting; it will also cause a break in the text currently being filled. The number register .u is set to 1.

    The fill mode status is associated with the current environment (see section Environments). @end_Deffn

    @Deffn{Request, nf, } Activate no-fill mode. Input lines are output as-is, retaining line breaks. The current line length will be ignored. This command implicitly disables adjusting; it also causes a break. The number register .u will be set to 0.

    The fill mode status is associated with the current environment (see section Environments). @end_Deffn

    @Deffn{Request, ad, [mode]} Set adjusting mode.

    Activation and deactivation of adjusting will be implicitly done with calls to the fi resp. nf requests.

    mode can have one of the following values:

    l
    Adjust text to the left margin. This produces what is traditionally called ragged-right text.
    r
    Adjust text to the right margin, producing ragged-left text.
    c
    Center filled text. This is different to the ce request which only centers text without filling.
    b
    n
    Justify to both margins. This is the default used by gtroff.

    With no argument, gtroff will adjust lines in the same way it did before adjusting has been deactivated (with a call to na, for example).

    text
    .ad r
    text
    .ad c
    text
    .na
    text
    .ad  \" back to centering
    text
    

    The current adjustment mode is available in the number register .j; it can be stored and subsequently used to set adjustment.

    The adjustment mode status is associated with the current environment (see section Environments). @end_Deffn

    @Deffn{Request, na, } Disable adjusting. This request won't change the current adjustment mode: A call to ad afterwards will use the previous adjustment setting.

    The adjustment mode status is associated with the current environment (see section Environments). @end_Deffn

    @Deffn{Escape, \\p, } Adjust the current line and cause a break.

    In most cases this will produce very ugly results, since gtroff doesn't have a sophisticated paragraph building algorithm (as TeX does, for example); instead, gtroff fills and adjusts a paragraph line by line:

      This is an uninteresting sentence.
      This is an uninteresting sentence.\p
      This is an uninteresting sentence.
    

    is formatted as

      This is  an uninteresting  sentence.   This  is an
      uninteresting                            sentence.
      This is an uninteresting sentence.
    

    @end_Deffn

    @Deffn{Request, ss, word_space_size [sentence_space_size]} Change the minimum size of a space between filled words. It takes its units as one twelfth of the space width parameter for the current font. Initially both the word_space_size and sentence_space_size are 12.

    If two arguments are given to the ss request, the second argument sets the sentence space size. If the second argument is not given, sentence space size will be set to word_space_size. The sentence space size is used in two circumstances: if the end of a sentence occurs at the end of a line in fill mode, then both an inter-word space and a sentence space will be added; if two spaces follow the end of a sentence in the middle of a line, then the second space will be a sentence space. Note that the behaviour of @acronym{UNIX} troff will be exactly that exhibited by GNU troff if a second argument is never given to the ss request. In GNU troff, as in @acronym{UNIX} troff, a sentence should always be followed by either a newline or two spaces.

    The number registers .ss and .sss hold the values of the parameters set by the first and second arguments of the ss request.

    The word space and sentence space values are associated with the current environment (see section Environments).

    This request is ignored in nroff mode; it is also ignored if there is no parameter. @end_Deffn

    @Deffn{Request, ce, [nnn]} Center text. While the `.ad c' request will also center text, it has the side effect of filling the text. ce will not fill the text it affects. This request causes a break.

    With no arguments, ce will center the next line of text. nnn specifies the number of lines to be centered. If the argument is zero or negative, centering is disabled.

    The basic length for centering text is the line length (as set with the ll request) minus the indentation (as set with the in request). Temporary indentation is ignored.

    A common idiom is to turn on centering for a large number of lines, and to turn off centering after text to be centered. This is useful for any request which takes a number of lines as an argument.

    .ce 1000
    replace this
    with
    something
    more interesting
    ...
    .ce 0
    

    The .ce number register contains the number of lines remaining to be centered, as set by the ce request. @end_Deffn

    @Deffn{Request, rj, [nnn]} Justify unfilled text to the right margin. Arguments are identical to the ce request. The .rj number register is the number of lines to be right-justified as set by the rj request. This request causes a line break. @end_Deffn

    Manipulating Hyphenation

    As discussed in section Hyphenation, gtroff will hyphenate words. There are a number of ways to influence hyphenation.

    @Deffn{Request, hy, [mode]} Enable hyphenation. The request has an optional numeric argument, mode, to restrict hyphenation if necessary:

    1
    The default argument if mode is omitted. Hyphenate without restrictions. This is also the start-up value of gtroff.
    2
    Do not hyphenate the last word on a page or column.
    4
    Do not hyphenate the last two characters of a word.
    8
    Do not hyphenate the first two characters of a word.

    Values in the previous table are additive. For example, the value 12 causes gtroff to neither hyphenate the last two nor the first two characters of a word.

    The current hyphenation restrictions can be found in the number register `.hy'.

    The hyphenation mode is associated with the current environment (see section Environments). @end_Deffn

    @Deffn{Request, nh, } Disable hyphenation (i.e., set the hyphenation mode to zero). Note that the hyphenation mode of the last call to hy is not remembered.

    The hyphenation mode is associated with the current environment (see section Environments). @end_Deffn

    @Deffn{Request, hlm, [nnn]} Set the maximum number of consecutive hyphenated lines to nnn. If this number is negative, there is no maximum. The default value is -1 if nnn is omitted. This value is associated with the current environment (see section Environments). Only lines output from a given environment count towards the maximum associated with that environment. Hyphens resulting from \% are counted; explicit hyphens are not.

    The current setting of hlm is available in the .hlm register. Also the number of immediately preceding consecutive hyphenated lines are available in the number register `.hlc'. @end_Deffn

    @Deffn{Request, hw, word1 word2 ...} Define how word1, word2, etc. are to be hyphenated. The words must be given with hyphens at the hyphenation points. For example:

    .hw in-sa-lub-rious
    

    Besides the space character, any character whose hyphenation code value is zero can be used to separate the arguments of hw (see the documentation for the hcode request below for more information). In addition, this request can be used more than once.

    Hyphenation exceptions specified with the hw request are associated with the current hyphenation language; it will cause an error if there is no current hyphenation language.

    This request is ignored if there is no parameter.

    In old versions of troff there was a limited amount of space to store such information; fortunately, with gtroff, this is no longer a restriction. @end_Deffn

    @Deffn{Escape, \\%, } To tell gtroff how to hyphenate words on the fly, the \% escape, also known as the hyphenation character, can be used. Preceding a word with this character will prevent it from being hyphenated, putting it in a word will indicate to gtroff that the word may be hyphenated at that point. Note that this mechanism will only affect that one occurrence of the word; to change the hyphenation of a word for the entire document, use the hw request. @end_Deffn

    @Deffn{Request, hc, [char]} Change the hyphenation character to char. This character will then work the same as the \% escape, and thus, no longer appear in the output. Without an argument, hc will reset the hyphenation character to be \% (the default) only.

    The hyphenation character is associated with the current environment (see section Environments). @end_Deffn

    @Deffn{Request, hpf, pattern_file} Read in a file of hyphenation patterns. This file will be searched for in the same way as `tmac.name' is searched for if the @option{-mname} option is specified.

    It should have the same format as the argument to the \patterns primitive in TeX (without using TeX's macro expansion); the letters appearing in this file are interpreted as hyphenation codes. A `%' character in the patterns file introduces a comment that continues