Manpages

Name

groff_diff - differences between GNU roff and AT&T troff

Description

The GNU roff text processing system, groff, is an extension of AT&T troff, the typesetting system originating in Unix systems of the 1970s. groff removes many arbitrary limitations and adds features, both to the input language and to the page description language output by the troff formatter. Differences arising from groff’s implementation of AT&T troff features are also noted. See roff(7) for background.

Language

GNU troff features identifiers of arbitrary length; supports color output, non-integral type sizes, and user-defined characters; adds more conditional expression operators; recognizes additional scaling units and numeric operators; enables general file I/O (in “unsafe mode” only); and exposes more formatter state.

Long names
GNU troff introduces many new requests; with three exceptions (cp, do, rj), they have names longer than two characters. The names of registers, fonts, strings/macros/diversions, environments, special characters, streams, and colors can be of any length. Anywhere AT&T troff supports a parameterized escape sequence that uses an opening parenthesis “(” to introduce a two-character argument, groff supports a square-bracketed form “[]” where the argument within can be of arbitrary length.

Font families, abstract styles, and translation
GNU troff can group text typefaces into families containing each of the styles “R”, “I”, “B”, and “BI”. So that a document need not be coupled to a specific font family, 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. A document can translate, or remap, fonts with the ftr request.

Applying the requests cs, bd, tkf, uf, or fspecial to an abstract style affects the member of the default family corresponding to that style. The default family can be set with the fam request or -f command-line option. The styles directive in the output device’s DESC file controls which mounting positions (if any) are initially associated with abstract styles rather than fonts, and the sty request can update this association.

Colors
groff supports color output with a variety of color spaces and up to 16 bits per channel. Some devices, particularly terminals, may be more limited. When color support is enabled, two colors are current at any given time: the stroke color, with which glyphs, rules (lines), and geometric figures are drawn, and the fill color, which paints the interior of filled geometric figures. The color, defcolor, gcolor, and fcolor requests; \m and \M escape sequences; and .color, .m, and .M registers exercise color support.

Fractional type sizes and new scaling units
AT&T troff interpreted all type size measurements in points. Combined with integer arithmetic, this design choice made it impossible to support, for instance, ten and a half-point type. In GNU troff, an output device can select a scaling factor that subdivides a point into “scaled points”. A type size expressed in scaled points can thus represent a non-integral type size.

A scaled point is equal to 1/sizescale points, where sizescale is specified in the device description file, DESC, and defaults to 1; see groff_font(5). Requests and escape sequences in GNU troff interpret arguments that represent a type size in points, which the formatter multiplies by sizescale and converts to an integer. Arguments treated in this way comprise those to the escape sequences \H and \s, to the request ps, the third argument to the cs request, and the second and fourth arguments to the tkf request. Scaled points may be specified explicitly with the z scaling unit. In GNU troff, the register \n[.s] can interpolate a non-integral type size. The register \n[.ps] interpolates the type size in scaled points.

For example, if sizescale is 1000, then a scaled point is one thousandth of a point. Consequently, “.ps 10.5” is synonymous with “.ps 10.5z”; both set the type size to 10,500 scaled points, or 10.5 points.

It makes no sense to use the “z” scaling unit in a numeric expression whose default scaling unit is neither “u” nor “z”, so GNU troff disallows this. Similarly, it is nonsensical to use a scaling unit other than “z” or “u” in a numeric expression whose default scaling unit is “z”, so GNU troff disallows this as well.

Another new scaling unit, “s”, multiplies by the number of basic units in a scaled point. Thus, “\n[.ps]s” is equal to “1m” by definition. Do not confuse the “s” and “z” scaling units.

Output devices may be limited in the type sizes they can employ. The .s and .ps registers represent the type size as selected by the output driver as it understands a device’s capability. The last requested type size is interpolated in scaled points by the read-only register .psr and in points as a decimal fraction by the read-only string-valued register .sr. Both are associated with the environment. For example, if a type size of 10.95 points is requested, and the nearest size permitted by a sizes request (or by the sizes or sizescale directives in the device’s DESC file) is 11 points, the output driver uses the latter value.

A further two new measurement units available in groff are “M”, which indicates hundredths of an em, and “f”, which multiplies by 65,536. The latter provides convenient fractions for color definitions with the defcolor request. For example, 0.5f equals 32768u.

Numeric expressions
GNU troff permits spaces in a numeric expression within parentheses, and offers three new operators.

Conditional expressions
More conditions can be tested with the “if” and ie requests, as well as the new “while” request.

Drawing commands
GNU troff offers drawing commands to create filled circles and ellipses, and polygons. Stroked (outlined) objects are drawn with the stroke color and filled (solid) ones shaded with the fill color. These are independent properties; if you want a filled, stroked figure, you must draw the same figure twice using each drawing command. A filled figure is always smaller than a stroked one because the former is drawn only within its defined area, whereas strokes have a line thickness (set with another new drawing command, \D't').

Escape sequences
groff introduces several new escape sequences and extends the syntax of a few AT&T troff escape sequences (namely, \D, \f, \k, \n, \s, \$, and \*). In the following list, escape sequences are collated alphabetically at first, and then by symbol roughly in Unicode code point order.
\A'anything'

Interpolate 1 if anything is a valid identifier, and 0 otherwise. Because invalid input characters are removed, 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.

\B'anything'

Interpolate 1 if anything is a valid numeric expression, and 0 otherwise. You might use \B along with the “if” request to filter out invalid macro arguments.

\D'C d'

Draw filled circle of diameter d with its leftmost point at the drawing position.

\D'E h v'

Draw filled ellipse with h and v as the axes and the leftmost point at the drawing position.

\D'p h1 v1 ... hn vn'

Draw polygon with vertices at drawing position and each point in sequence. GNU troff closes the polygon by drawing a line from (hn, vn) back to the initial drawing position; DWB and Heirloom troffs do not. Afterward, the drawing position is left at (hn, vn).

\D'P h1 v1 ... hn vn'

As \D'p', but the polygon is filled.

\D't n'

Set line thickness of geometric objects to to n basic units. A zero n selects the minimal supported thickness. A negative n selects a thickness proportional to the type size; this is the default.

\f[font]

Select font, which may be a mounting position, abstract style, or font name, to choose the typeface. \f[] and \fP are synonyms; we recommend the former.

\F[family]

Select default font family. \F[] makes the previous font family the default. \FP is unlike \fP; it selects font family “P” as the default. See the fam request below.

\k[reg]

Mark horizontal drawing position in two-character register name rg or arbitrary register name reg.

\m[col]

Set the stroke color. \m[] restores the previous stroke color, or the default if there is none.

\M[col]

Set the fill color. \M[] restores the previous fill color, or the default if there is none.

\n[reg]

Interpolate register reg.

\O0

\O5[Pfile]

At the outermost suppression level, write the name file to the standard error stream at position P, which must be one of l, r, c, or i, corresponding to left, right, centered, and inline alignments within the document, respectively. file is is a name associated with the production of the next image.

\R'name ±n'

Synonymous with “.nr name ±n”.

\V[env]

Interpolate contents of the environment variable env, as returned by getenv(3). \V is interpreted even in copy mode.

\X'anything'

Within \X arguments, the escape sequences \&, \), \%, and \: are ignored; \space and \~ are converted to single space characters; and \\ is reduced to \. So that the basic Latin subset of the Unicode character set (that is, ISO 646:1991-IRV or, popularly, “US-ASCII”) can be reliably encoded in anything, the special character escape sequences \-, \[aq], \[dq], \[ga], \[ha], \[rs], and \[ti] are mapped to basic Latin characters; see groff_char(7). For this transformation, character translations and definitions are ignored. Other escape sequences are not supported.

If the use_charnames_in_special directive appears in the output device’s DESC file, the use of special character escape sequences is not an error; they are simply output verbatim (with the exception of the seven mapped to Unicode basic Latin characters, discussed above). use_charnames_in_special is currently employed only by grohtml(1).

\Y[mac]

Interpolate a macro as a device control command. This is similar to \X'\*[mac]', except the contents of mac are not interpreted, and mac can be a macro and thus contain newlines, whereas the argument to \X cannot. This inclusion of newlines requires an extension to the AT&T troff output format, and will confuse postprocessors that do not know about it.

\Z'anything'

Save the drawing position, format anything, then restore it. Tabs and leaders in the argument are ignored with an error diagnostic.

\$[nnn]

In a macro or string definition, interpolate the nnth or nnnth argument. Macros and strings can have an unlimited number of arguments.

\*[string [arg ...]]

Interpolate string, passing it arg ... as arguments.

\?anything\?

When used in a diversion, this transparently embeds anything in the diversion. anything is read in copy mode. When the diversion is reread, anything is interpreted. anything may not contain newlines; use \! if you want to embed newlines in a diversion. The escape sequence \? is also recognized in copy mode and becomes an internal code; it is this code that terminates anything. Thus

.nr x 1
.nf
.di d
\?\\?\\\\?\\\\\\\\nx\\\\?\\?\?
.di
.nr x 2
.di e
.d
.di
.nr x 3
.di f
.e
.di
.nr x 4
.f

prints 4.

\[char]

Typeset the special character char.

\[base-char combining-component ...]

Typeset a composite glyph consisting of base-char overlaid with one or more combining-components. For example, “\[A ho]” is a capital letter “A” with a “hook accent” (ogonek). See the composite request below; Groff: The GNU Implementation of troff, the groff Texinfo manual, for details of composite glyph name construction; and groff_char(7) for a list of components used in composite glyph names.

Restricted requests
To mitigate risks from untrusted input documents, the pi and sy requests are disabled by default. troff(1)’s -U option enables the formatter’s “unsafe mode”, restoring their function (and enabling additional groff extension requests, open, opena, and pso).

New requests
.aln new old

Create alias new for existing register named old, causing the names to refer to the same stored value. If old is undefined, a warning in category “reg” is generated and the request is ignored. To remove a register alias, invoke rr on its name. A register’s contents do not become inaccessible until it has no more names.

.als new old

Create alias new for existing request, string, macro, or diversion named old, causing the names to refer to the same stored object. If old is undefined, a warning in category “mac” is produced, and the request is ignored. The “am”, “as”, da, de, di, and ds requests (together with their variants) create a new object only if the name of the macro, diversion, or string is currently undefined or if it is defined as a request; normally, they modify the value of an existing object. To remove an alias, invoke rm on its name. The object itself is not destroyed until it has no more names.

When a request, macro, string, or diversion is aliased, redefinitions and appendments “write through” alias names. To replace an alias with a separately defined object, you must use the rm request on its name first.

.am1 name [end-name]

As “am”, but compatibility mode is disabled while the appendment to name is interpreted: a “compatibility save” token is inserted at its beginning, and a “compatibility restore” token at its end. As a consequence, the requests “am”, am1, de, and de1 can be intermixed freely since the compatibility save/restore tokens affect only the parts of the macro populated by am1 and de1.

.ami name [end-name]

Append to macro indirectly. See dei below.

.ami1 name [end-name]

As ami, but compatibility mode is disabled during interpretation of the appendment.

.as1 name [contents]

As “as”, but compatibility mode is disabled while the appendment to name is interpreted: a “compatibility save” token is inserted at the beginning of contents, and a “compatibility restore” token after it. As a consequence, the requests “as”, as1, ds, and ds1 can be intermixed freely since the compatibility save/ restore tokens affect only the portions of the strings populated by as1 and ds1.

.asciify div

Unformat the diversion div in a way such that Unicode basic Latin (ASCII) characters, characters translated with the trin request, space characters, and some escape sequences, that were formatted in the diversion div are treated like ordinary input characters when div is reread. Doing so can be useful in conjunction with the writem request. asciify can be also used for gross hacks; for example, the following sets register n to 1.

.tr @.
.di x
@nr n 1
.br
.di
.tr @@
.asciify x
.x

asciify cannot return all items in a diversion to their source equivalent: nodes such as those produced by \N[...] will remain nodes, so the result cannot be guaranteed to be a pure string. See section “Copy mode” in groff(7). Glyph parameters such as the type face and size are not preserved; use unformat to achieve that.

.backtrace

Write backtrace of input stack to the standard error stream. See the -b option of troff(1).

.blm [name]

Set a blank line macro (trap). If a blank line macro is thus defined, groff executes macro when a blank line is encountered in the input file, instead of the usual behavior. A line consisting only of spaces is also treated as blank and subject to this trap. If no argument is supplied, the default blank line behavior is (re-)established.

.box [name]
.boxa [name]

Divert (or append) output to name, similarly to the di and da requests, respectively. Any pending output line is not included in the diversion. Without an argument, stop diverting output; any pending output line inside the diversion is discarded.

.cflags n c1 c2 ...

Assign properties encoded by the number n to characters c1, c2, and so on. Ordinary and special characters have certain associated properties. (Glyphs don’t: to GNU troff, like AT&T device-independent troff, a glyph is an identifier corresponding to a rectangle with some metrics; see groff_font(5).) The first argument is the sum of the desired flags and the remaining arguments are the characters to be assigned those properties. Spaces between the cn arguments are optional. Any argument cn can be a character class defined with the class request rather than an individual character.

The non-negative integer n is the sum of any of the following. Some combinations are nonsensical, such as “33” (1 + 32).

For example, if you need an automatic break point after the en-dash in numeric ranges like “3000–5000”, insert

.cflags 68 \[en]

into your document. However, this can lead to bad layout if done without thinking; in most situations, a better solution than changing the cflags value is inserting “\:” right after the hyphen at the places that really need a break point.

The remaining values were implemented for East Asian language support; those who use alphabetic scripts exclusively can disregard them.

In contrast to values 2 and 4, the values 128, 256, and 512 work pairwise. If, for example, the left character has value 512, and the right character 128, no break will be automatically inserted between them. If we use value 6 instead for the left character, a break after the character can’t be suppressed since the neighboring character on the right doesn’t get examined.

.char c contents

Define the ordinary or special character c as contents, which can be empty. More precisely, char defines a groff object (or redefines an existing one) that is accessed with the name c on input, and produces contents on output. Every time c is to be formatted, contents is processed in a temporary environment and the result is wrapped up into a single object. Compatibility mode is turned off and the escape character is set to \ while contents is processed. Any emboldening, constant spacing, or track kerning is applied to this object as a whole, not to each character in contents.

An object defined by this request can be used just like a glyph provided by the output device. In particular, other characters can be translated to it with the tr request; it can be made the tab or leader fill character with the tc and lc requests; sequences of it can be drawn with the \l and \L escape sequences; and, if the hcode request is used on c, it is subject to automatic hyphenation.

To prevent infinite recursion, occurrences of c within its own definition are treated normally (as if it were not being defined with char). The tr and trin requests take precedence if char both apply to c. A character definition can be removed with the rchar request.

.chop object

Remove the last character from the macro, string, or diversion object. This is useful for removing the newline from the end of a diversion that is to be interpolated as a string. This request can be used repeatedly on the same object; see section “gtroff Internals” in Groff: The GNU Implementation of troff, the groff Texinfo manual, for discussion of nodes inserted by groff.

.class name c1 c2 ...

Define a character class (or simply “class”) name comprising the characters or range expressions c1, c2, and so on.

A class thus defined can then be referred to in lieu of listing all the characters within it. Currently, only the cflags request can handle references to character classes.

In the request’s simplest form, each cn is a character (or special character).

.class [quotes] ' \[aq] \[dq] \[oq] \[cq] \[lq] \[rq]

Since class and special character names share the same name space, we recommend starting and ending the class name with “[” and “]”, respectively, to avoid collisions with existing character names defined by groff or the user (with char and related requests). This practice applies the presence of “]” in the class name to prevent the usage of the special character escape form “\[...]”, thus you must use the \C escape to access a class with such a name.

You can also use a character range expression consisting of a start character followed by “-” and then an end character. Internally, GNU troff converts these two character names to Unicode code points (according to the groff glyph list [GGL]), which determine the start and end values of the range. If that fails, the class definition is skipped. Furthermore, classes can be nested.

.class [prepunct] , : ; > }
.class [prepunctx] \C'[prepunct]' \[u2013]-\[u2016]

The class “[prepunctx]” thus contains the contents of the class “[prepunct]” and characters in the range U+2013–U+2016.

If you want to include “-” in a class, it must be the first character value in the argument list, otherwise it gets misinterpreted as part of the range syntax.

It is not possible to use class names as end points of range definitions.

A typical use of the class request is to control line-breaking and hyphenation rules as defined by the cflags request. For example, to inhibit line breaks before the characters belonging to the “[prepunctx]” class defined in the previous example, you can write the following.

.cflags 2 \C'[prepunctx]'

.close stream

Close the stream named stream, invalidating it as an argument to the write request. See open.

.composite c1 c2

Map character name c1 to character name c2 when c1 is a combining component in a composite glyph. Typically, this remaps a spacing glyph to a combining one.

.continue

Skip the remainder of a “while” loop’s body, immediately starting the next iteration. See break.

.color n

If n is non-zero or missing, enable colors (the default), otherwise disable them.

.defcolor ident scheme color-component ...

Define a color named ident. scheme identifies a color space and determines the number of required color-components; it must be one of “rgb” (three components), “cmy” (three components), “cmyk” (four components), or “gray” (one component). “grey” is accepted as a synonym of “gray”. The color components can be encoded as a hexadecimal value starting with # or ##. The former indicates that each component is in the range 0–255 (0–FF), the latter the range 0–65535 (0–FFFF). Alternatively, each color component can be specified as a decimal fraction in the range 0–1, interpreted using a default scaling unit of “f”, which multiplies its value by 65,536 (but clamps it at 65,535).

Each output device has a color named “default”, which cannot be redefined. A device’s default stroke and fill colors are not necessarily the same.

.de1 name [end-name]

Define a macro to be interpreted with compatibility mode disabled. When name is called, compatibility mode enablement status is saved; it is restored when the call completes.

.dei name [end-name]

Define macro indirectly, with the name of the macro to be defined in string name and the name of the end macro terminating its definition in string end-name.

.dei1 name [end-name]

As dei, but compatibility mode is disabled while the definition of the macro named in string name is interpreted.

.device anything

Write anything, read in copy mode, to troff output as a device control command. An initial neutral double quote is stripped to allow the embedding of leading spaces.

.devicem name

Write contents of macro or string name to troff output as a device control command.

.do name [arg ...]

Interpret the string, request, diversion, or macro name (along with any arguments) with compatibility mode disabled. Compatibility mode is restored (only if it was active) when the expansion of name is interpreted; that is, the restored compatibility state applies to the contents of the macro, string, or diversion name as well as data read from files or pipes if name is any of the so, soquiet, mso, msoquiet, or pso requests.

For example,

.de mac1
FOO
..
.de1 mac2
groff
.mac1
..
.de mac3
compatibility
.mac1
..
.de ma
\\$1
..
.cp 1
.do mac1
.do mac2 \" mac2, defined with .de1, calls "mac1"
.do mac3 \" mac3 calls "ma" with argument "c1"
.do mac3 \[ti] \" groff syntax accepted in .do arguments

results in

FOO groff FOO compatibility c1 ~

as output.

.ds1 name contents

As ds, but compatibility mode is disabled while name is interpreted: a “compatibility save” token is inserted at the beginning of contents, and a “compatibility restore” token after it.

.evc env

Copy the properties of environment env to the current environment, except for the following data.

.fam [family]

Set default font family to family. If no argument is given, the previous font family is selected, or the formatter’s default family if there is none. The formatter’s default font family is “T” (Times), but it can be overridden by the output device—see groff_font(5). The default font family is associated with the environment. See \F.

.fchar c contents

Define fallback character c as contents. The syntax of this request is the same as the char request; the difference is that a character defined with char hides a glyph with the same name in the selected font, whereas characters defined with fchar are checked only if c isn’t found in the selected font. This test happens before special fonts are searched.

.fcolor color

Set the fill color to color. Without an argument, the previous fill color is selected.

.fschar f c contents

Define fallback special character c for font f as contents. A character defined by fschar is located after the list of fonts declared with fspecial is searched but before those declared with the “special” request.

.fspecial f s1 s2 ...

When font f is selected, fonts s1, s2, ... are treated as special; that is, they are searched for glyphs not found in f. Any fonts specified in the “special” request are searched after s1, s2, and so on. Without s arguments, fspecial clears the list of fonts treated as special when f is selected.

.ftr f g

Translate font f to g. Whenever a font named f is referred to in an \f escape sequence, in the F and S conditional expression operators, or in the ft, ul, bd, cs, tkf, special, fspecial, fp, or sty requests, font g is used. If g is missing or identical to f, then font f is not translated.

.fzoom f zoom

Set zoom factor zoom for font f. zoom must a non-negative integer multiple of 1/1000th. If it is missing or is equal to zero, it means the same as 1000, namely no magnification. f must be a resolved font name, not an abstract style.

.gcolor color

Set the stroke color to color. Without an argument, the previous stroke color is selected.

.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. The request is ignored if given no arguments.

For hyphenation to work, hyphenation codes must be set up. At startup, groff 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 “ABBOT” and “Abbot” should be hyphenated exactly as “abbot” is. hcode 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.

.hla lang

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 in turn loaded by the troffrc or troffrc-end file; see the hpf request below. The hyphenation language is associated with the environment.

.hlm [n]

Set the maximum number 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. Only lines output from a given environment count towards the maximum associated with that environment. Hyphens resulting from \% are counted; explicit hyphens are not.

.hpf pattern-file

Read hyphenation patterns from pattern-file. This file is sought in the same way that macro files are with the mso request or the -mname command-line option to groff(1) and troff(1).

The pattern-file should have the same format as (simple) TeX pattern files. The following scanning rules are implemented.

Use the hpfcode request (see below) to map the encoding used in hyphenation pattern files to groff’s input encoding.

The set of hyphenation patterns is associated with the hyphenation language set by the hla request. The hpf request is usually invoked by a localization file loaded by the troffrc file. By default, troffrc loads the localization file for English. (As of groff 1.23.0, localization files for Czech (cs), German (de), English (en), French (fr), Japanese (ja), Swedish (sv), and Chinese (zh) exist.) For Western languages, the localization file sets the hyphenation mode and loads hyphenation patterns and exceptions.

A second call to hpf (for the same language) replaces the old patterns with the new ones.

Invoking hpf 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.

.hpfa pattern-file

As hpf, except that the hyphenation patterns and exceptions from pattern-file are appended to the patterns already applied to the hyphenation language of the environment.

.hpfcode a b [c d] ...

Define mapping values for character codes in pattern files. This is an older mechanism no longer used by groff’s own macro files; for its successor, see hcode above. hpf or hpfa apply the mapping after reading 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 groff 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”.

.hym [length]

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 default hyphenation margin is reset to its default value, 0. The default scaling unit is “m”. The hyphenation margin is associated with the environment. A negative argument resets the hyphenation margin to zero, emitting a warning in category “range”.

.hys [hyphenation-space]

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 current environment. A negative argument resets the hyphenation space adjustment threshold to zero, emitting a warning in category “range”.

.itc n name

As “it”, but lines interrupted with the \c escape sequence are not applied to the line count.

.kern n

If n is non-zero or missing, enable pairwise kerning (the default), otherwise disable it.

.length reg anything

Compute the number of characters in anything and return the count in the register reg. If reg doesn’t exist, it is created. anything is read in copy mode.

.ds xxx abcd\h'3i'efgh
.length yyy \*[xxx]
\n[yyy]
14

.linetabs n

If n is non-zero or missing, enable line-tabs mode, otherwise disable it (the default). In this mode, tab stops are computed relative to the start of the pending output line, instead of the drawing position corresponding to the start of the input line. Line-tabs mode is a property of the environment.

For example, the following

.ds x a\t\c
.ds y b\t\c
.ds z c
.ta 1i 3i
\*x
\*y
\*z

yields

a b c

whereas in line-tabs mode, the same input gives

a b c

instead.

.lsm [name]

Set the leading space macro (trap) to name. If there are leading space characters on an input line, name is invoked in lieu of the usual roff behavior; the leading spaces are removed. The count of leading spaces on an input line is stored in \n[lsn], and the amount of corresponding horizontal motion in \n[lss], irrespective of whether a leading space trap is set. When it is, the leading spaces are removed from the input line, and no motion is produced before calling name. If no argument is supplied, the default leading space behavior is (re-)established.

.mso file

As “so”, except that file is sought in the same directories as arguments to the groff(1) and troff(1) -m command-line option are (the “tmac path”). If the file name to be interpolated has the form name.tmac and it isn’t found, mso tries to include tmac.name instead and vice versa. If file does not exist, a warning in category “file” is emitted and the request has no other effect.

.msoquiet file

As mso, but no warning is emitted if file does not exist.

.nop anything

Interpret anything as if it were an input line. nop resembles “.if 1”; it puts a break on the output if anything is empty. Unlike “if”, it cannot govern conditional blocks. Its application is to maintain consistent indentation within macro definitions even when producing text lines.

.open stream file

Open file for writing and associate stream with it. See write and close.

.opena stream file

As open, but if file exists, append to it instead of truncating it.

.output contents

Emit contents, which are read in copy mode, to the formatter output; this is similar to \! used in the top-level diversion. An initial neutral double quote in contents is stripped to allow the embedding of leading spaces.

.psbb file

Get the bounding box of a PostScript image file. This file must conform to Adobe’s Document Structuring Conventions; the request attempts to extract the bounding box values from a %%BoundingBox comment. After invocation, the x and y coordinates (in PostScript units) of the lower left and upper right corners can be found in the registers \n[llx], \n[lly], \n[urx], and \n[ury], respectively. If an error occurs, these four registers are set to zero.

.pso command

As “so”, except that input comes from the standard output stream of command.

.pvs ±n

Set the post-vertical line spacing to n; default scaling unit is “p”. With no argument, the post-vertical line space is set to its previous value.

In GNU troff, the distance between text baselines consists of the extra pre-vertical line spacing set by the most negative \x argument on the pending output line, the vertical spacing (vs), the extra post-vertical line spacing set by the most positive \x argument on the pending output line, and the post-vertical line spacing set by this request.

.rchar c ...

Remove definition of each ordinary or special character c, undoing the effect of a char, fchar, or schar request. Glyphs, which are defined by font description files, cannot be removed. Spaces and tabs may separate c arguments.

.return

Within a macro, return immediately. If called with an argument, return twice, namely from the current macro and from the macro one level higher. No effect otherwise.

.rfschar f c ...

Remove each fallback special character c for font f. Spaces and tabs may separate c arguments. See fschar.

.rj [n]

Right-align the next n input lines. Without an argument, right-align the next input line. rj implies “.ce 0”, and ce implies “.rj 0”.

.rnn r1 r2

Rename register r1 to r2. If r1 doesn’t exist, the request is ignored.

.schar c contents

Define global fallback character c as contents. See char; the distinction is that a character defined with schar is located after the list of fonts declared with the special request but before any mounted special fonts.

.shc [c]

Set the soft hyphen character, inserted when a word is hyphenated automatically or at a hyphenation character, to c. If c is omitted, the soft hyphen character is set to the default, \[hy]. If the selected glyph does not exist in the font in use at a potential hyphenation point, then the line is not broken at that point. Neither character definitions (char and similar) nor translations (tr and similar) are considered when assigning the soft hyphen character.

.shift n

In a macro, shift the arguments by n positions: argument i becomes argument i-n; arguments 1 to n are no longer available. If n is missing, arguments are shifted by 1. No effect otherwise.

.sizes s1 s2 ... sn [0]

Set the available type sizes to s1, s2, ... sn scaled points. The list of sizes can be terminated by an optional “0”. Each si can also be a range m–n. In contrast to the device description file directive of the same name (see groff_font(5)), the argument list can’t extend over more than one line.

.soquiet file

As “so”, but no warning is emitted if file does not exist.

.special f ...

Declare each font f as special, searching it for glyphs not found in the selected font. Without arguments, this list of special fonts is made empty.

.spreadwarn [limit]

Emit a break warning if the additional space inserted for each space between words in an output line adjusted to both margins with “.ad b” is larger than or equal to limit. A negative value is treated as zero; an absent argument toggles the warning on and off without changing limit. The default scaling unit is m. At startup, spreadwarn is inactive and limit is 3 m.

For example, “.spreadwarn 0.2m” causes a warning if break warnings are not suppressed and troff must add 0.2 m or more for each inter-word space in a line.

.stringdown str
.stringup str

Alter the string named str by replacing each of its bytes with its lowercase (down) or uppercase (up) version (if one exists). Special characters (see groff_char(7)) will often transform in the expected way due to the regular naming convention for accented characters. When they do not, use substrings and/or catenation.

.ds resume R\['e]sum\['e]\"
\*[resume]
.stringdown resume
\*[resume]
.stringup resume
\*[resume]
Résumé résumé RÉSUMÉ

.sty n s

Associate abstract style s with font mounting position n.

.substring string start [end]

Replace the string named string with its substring bounded by the indices start and end, inclusively. The first character in the string has index 0. If end is omitted, it is implicitly set to the largest valid value (the string length minus one). Negative indices count backwards from the end of the string: the last character has index -1, the character before the last has index -2, and so on.

.ds xxx abcdefgh
.substring xxx 1 -4
\*[xxx]
bcde
.substring xxx 2
\*[xxx]
de

.tkf f s1 n1 s2 n2

Enable track kerning for font f. When the current font is f the width of every glyph is increased by an amount between n1 and n2; when the current type size is less than or equal to s1 the width is increased by n1; when it is greater than or equal to s2 the width is increased by n2; when the type size is greater than or equal to s1 and less than or equal to s2 the increase in width is a linear function of the type size.

.tm1 message

As tm request, but strips a leading neutral double quote from message to allow the embedding of leading spaces.

.tmc message

As tm1 request, but does not append a newline.

.trf file

Transparently output the contents of file file. Each line is output as if preceded by \!; however, the lines are not subject to copy-mode interpretation. If the file does not end with a newline, then a newline is added. Unlike cf, file cannot contain characters that are invalid as input to GNU troff.

For example, you can define a macro x containing the contents of file f, using

.di x
.trf f
.di

.trin abcd

This is the same as the tr request except that the asciify request uses the character code (if any) before the character translation. Example:

.trin ax
.di xxx
a
.br
.di
.xxx
.trin aa
.asciify xxx
.xxx

The result is “x a”. Using tr, the result would be “x x”.

.trnt abcd

This is the same as the tr request except that the translations do not apply to text that is transparently throughput into a diversion with \!. For example,

.tr ab
.di x
\!.tm a
.di
.x

prints b; if trnt is used instead of tr it prints a.

.unformat div

Unformat the diversion div. Unlike asciify, unformat handles only tabs and spaces between words, the latter usually arising from spaces or newlines in the input. Tabs are treated as input tokens, and spaces become adjustable again. The vertical sizes of lines are not preserved, but glyph information (font, type size, space width, and so on) is retained.

.warn [n]

Select the categories, or “types”, of reported warnings. n is the sum of the numeric codes associated with each warning category that is to be enabled; all other categories are disabled. The categories and their associated codes are listed in section “Warnings” of troff(1). For example, “.warn 0” disables all warnings, and “.warn 1” disables all warnings except those about missing glyphs. If no argument is given, all warning categories are enabled.

.warnscale si

Set the scaling unit used in warnings to si. Valid values for si are u, i (the default), c, p, and P.

.while cond-expr anything

Evaluate the conditional expression cond-expr, and repeatedly execute anything unless and until cond-expr evaluates false. anything, which is often a conditional block, is referred to as the while request’s body.

troff treats the body of a while request similarly to that of a de request (albeit one not read in copy mode), but stores it under an internal name and deletes it when the loop finishes. The operation of a macro containing a while request can slow significantly if the while body is large. Each time the macro is executed, the while body is parsed and stored again. An often better solution—and one that is more portable, since AT&T troff lacked the while request—is to instead write a recursive macro. It will be parsed only once (unless you redefine it). To prevent infinite loops, the default number of available recursion levels is 1,000 or somewhat less (because things other than macro calls can be on the input stack). You can disable this protective measure, or raise the limit, by setting the slimit register. See section “Debugging” below.

If a while body begins with a conditional block, its closing brace must end an input line.

The break and continue requests alter a while loop’s flow of control.

.write stream anything

Write anything to stream, which must previously have been the subject of an open request, followed by a newline. anything is read in copy mode. An initial neutral double quote in anything is stripped to allow the embedding of leading spaces.

.writec stream anything

As write, but without a trailing newline.

.writem stream name

Write the contents of the macro or string name to stream, which must previously have been the subject of an open request. name is read in copy mode.

Extended requests
.cf file

In a diversion, embed an object which, when reread, will cause the contents of file to be copied verbatim to the output. In AT&T troff, the contents of file are immediately copied to the output regardless of whether a diversion is being written to; this behavior is so anomalous that it must be considered a bug.

.de name [end-name]
.am name [end-name]
.ds name [contents]
.as name [contents]

In compatibility mode, these requests behave similarly to de1, am1, ds1, and as1, respectively: a “compatibility save” token is inserted at the beginning, and a “compatibility restore” token at the end, with compatibility mode switched on during execution.

.ss word-space-size [additional-sentence-space-size]

A 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. Both arguments are in twelfths of current font’s space width (typically one-fourth to one-third em for Western scripts; see groff_font(5)). The default for both parameters is 12. Negative values are erroneous.

.ta [[n1 n2 ... nn ]T r1 r2 ... rn]

groff supports an extended syntax to specify repeating tab stops after the “T” mark. These values are always taken as relative distances from the previous tab stop. This is the idiomatic way to specify tab stops at equal intervals in groff.

The syntax summary above instructs groff to set 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.

New registers
GNU troff exposes more formatter state via many new read-only registers. Their names often correspond to the requests that affect them.

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

\n[.linetabs]

Interpolate 1 if line-tabs mode is enabled, 0 otherwise.

The following (writable) registers are set by the psbb request.
\n[llx]
\n[lly]
\n[urx]
\n[ury]

Interpolate the (upper, lower, left, right) bounding box values (in PostScript units) of the most recently processed PostScript image.

The following (writable) registers are set by the \w escape sequence.

Other writable registers are as follows. Those relating to date and time are initialized using localtime(3) at formatter startup.

Miscellaneous
GNU troff predefines one string, .T, containing the argument given to the -T command-line option, namely the output device (for example, pdf or utf8). The (read-only) register .T interpolates 1 if GNU troff is run with the -T command-line option, and 0 otherwise.

A font not listed in the output device’s DESC file’s fonts directive is automatically mounted at the next available font position when it is selected. If you mount a font explicitly with the fp request, you should do so on the first unused position, which can be found in the .fp register.

Unparameterized string interpolation does not conceal the arguments to a macro being interpreted. Thus, in a macro definition, the call of another macro with the existing argument list,

.xx \\$@

is more efficiently done with

\\*[xx]\\

(that is, with string interpolation). The trailing backslashes prevent the final newline in the macro definition from being interpolated, potentially putting an unwanted blank line on the output. See section “Punning Names” in groff(7).

If a font description file contains pairwise kerning information, glyphs from that font are kerned. Kerning between two glyphs can be inhibited by placing a dummy character \& between them.

GNU troff keeps track of the nesting depth of escape sequence interpolations and other uses of delimiters, as in the tl request and the output comparison operator (that is, input like 'foo'bar' as a conditional expression), 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. Use visible characters as delimiters in GNU troff, not “ASCII” controls like BEL (Control+G). The implementation of \$@ ensures that the double quotes surrounding an argument appear at an interpolation depth different from that of the arguments themselves. Similarly, in bracket-form escape sequences like \f[ZCMI], a right bracket ] does not end the sequence unless it occurs at the same interpolation depth as the opening [, so input like

\f[\*[my-family]\*[my-style]]

works as desired. In compatibility mode, no attention is paid to the interpolation depth.

In GNU troff, the tr request can map characters to the unbreakable space escape sequence \~ as a special case (tr normally operates only on characters). This feature replaces the odd-parity tr mapping trick used in AT&T troff documents, where a character, often ~, was “sacrificed” by mapping it to “nothing”, drafting it into use as an unadjustable, unbreakable space. (This feature was gratuitous even in early AT&T troff, which supported the \space escape sequence by 1976.) Often, it makes more sense to use GNU troff’s \~ escape sequence instead, which has been adopted by every other active troff implementation except that of Illumos, as well as by the non-troff mandoc. Translation of a character to \~ is unnecessary.

GNU troff permits tabs and spaces after the first dot on a control line that ends a macro definition.

.if t \{\
. de bar
. nop Hello, I'm 'bar'.
. .
.\}

Formatter output

The page description language output by GNU troff is modeled after that used by AT&T troff once the latter adopted a device-independent approach in the early 1980s. Only the differences are documented here. For a fuller discussion, see groff_out(5).

Glyph and font names can be of arbitrary length; postprocessors should not assume that they are at most two characters. A glyph to be formatted is always drawn from the current font; in contrast to AT&T device-independent troff, drivers need not search special fonts to find a glyph.

Units
The argument to the s command is in scaled points (units of points/n, where n is the argument to the sizescale command in the DESC file). The argument to the “x H” command is also in scaled points.

Simple commands
If the tcommand directive is present in the output device’s DESC file, GNU troff employs the following two commands.
t xyz...

Typeset word xyz; that is, set a sequence of ordinary glyphs named x, y, z, ..., terminated by a space or newline; an optional second integer argument is ignored (this allows the formatter to generate an even number of arguments). Each glyph is set at the current drawing position, and the position is then advanced horizontally by the glyph’s width. A glyph’s width is read from its metrics in the font description file, scaled to the current type size, and rounded to a multiple of the horizontal motion quantum. Use the C command to emplace glyphs of special characters.

u n xyz...

Typeset word xyz with track kerning. As t, but after placing each glyph, the drawing position is further advanced horizontally by n basic units.

New commands implement color support.
mc cyan magenta yellow

mg gray
mk cyan magenta yellow black
mr red green blue

Set the components of the stroke color with respect to various color spaces. md resets the stroke color to the default value. The arguments are integers in the range 0 to 65535.

A new device control subcommand is available.

Extended drawing commands
GNU pic does not produce troff escape sequences employing these extensions if its -n option is given.

The corresponding \D'f' escape sequence should not be used: its argument is rounded to an integer multiple of the horizontal motion quantum, which can limit the precision of n.

Dp dx 1 dy 1 ...dx n dy n

Draw a polygon with, for i=1,...,n+1, its ith vertex at the drawing position + ij−1=1 Σ (dx j ,dy j ). groff output drivers automatically close polygons, drawing a line from (dx n ,dy n ) back to (dx 1 ,dy 1 ). The drawing position is left at the last specified vertex, but this may change in a future version of GNU troff. Heirloom Doctools troff, like DWB troff, by default does not close the polygon. In its groff compatibility mode, Heirloom closes the polygon but leaves the drawing position unchanged—that is, at the polygon’s initial drawing position.

At the moment, GNU pic uses this command only to generate triangles and rectangles.

DP dx 1 dy 1 ...dx n dy n

As Dp, but draw a filled rather than a stroked polygon.

A difficulty arises in how the drawing position should be changed after the execution of these commands. This has little importance to most users, since the output of GNU grn and pic does not depend on it. Given a drawing command of the form Dz x 1 y 1 ...x n y n , where z is not c or e, AT&T troff treats each x i as a horizontal motion, each y i as a vertical one, and therefore assumes that the width of the drawn object is i=1 Σ x i , and its height is i=1 Σ y i . (Verify its assumption about height by examining the st and sb registers after using such a drawing command in a \w escape sequence). For the sake of compatibility, GNU troff also follows this rule, even though it frustrates extensions to the D command that set drawing parameters rather than rendering objects, producing ugly results in the case of Dt and Df, or otherwise don’t parameterize objects as a series of vertices, as with GNU troff’s filled ellipse, DE. Thus after executing a D command of the form Dz x 1 y 1 ...x n y n , the drawing position should be increased by ( i=1 Σ x i , i=1 Σ y i ). In a future release, GNU troff and its output drivers may abandon the application of this assumption to drawing commands not explicitly specified in the AT&T “Troff User’s Manual”.

Fill color selection is implemented with another set of extensions.
DFc cyan magenta yellow

DFg gray
DFk cyan magenta yellow black
DFr red green blue

Set the components of the fill color as described under the \M escape sequence above. DFd restores the device’s default fill color. The drawing position is not updated, in contrast to Df.

Device control syntax extension
GNU troff introduces a line continuation convention, permitting the argument to the x X command to contain newlines. A newline in the input is transformed to the sequence “newline+”. When interpreting an x X command, a postprocessor should therefore be prepared for a plus sign after a newline; if it occurs, preserve the newline, discard the plus sign, and continue to collect the input into the argument of the x X command. A newline not followed by a plus sign terminates the x X command. An application of this feature is the embedding of PostScript or PDF language command streams into troff output.

GNU troff guarantees that the first three output commands it emits are as follows.

x T device
x res n h v
x init

Debugging

In addition to AT&T troff’s debugging features, GNU troff emits more error diagnostics when syntactical or semantic nonsense is encountered and supports several warning categories; the output of these can be selected with warn. Also see the -E, -w, and -W options of troff(1). Backtraces can be automatically produced when errors or warnings occur (the -b option of troff(1)) or generated on demand (backtrace).

groff also adds more flexible diagnostic output requests (tmc and tm1). More aspects of formatter state can be examined with requests that write lists of defined registers (pnr), environments (pev), and page location traps (ptr) to the standard error stream.

Implementation differences

GNU troff’s features sometimes cause incompatibilities with documents written assuming old implementations of troff. Some GNU extensions to troff are supported by other implementations.

When adjusting to both margins, AT&T troff at first adjusts spaces starting from the right; GNU troff begins from the left. Both implementations adjust spaces from opposite ends on alternating output lines to prevent “rivers” in the text.

GNU troff does not always hyphenate words as AT&T troff does. The AT&T implementation uses a set of hard-coded rules specific to U.S. English, while GNU troff uses language-specific hyphenation pattern files derived from TeX. In some versions of troff there was limited space to store hyphenation exceptions (arguments to the hw request); GNU troff has no such restriction.

Long names may be GNU troff’s most obvious innovation. AT&T troff interprets “.dsabcd” as defining a string “ab” with contents “cd”. Normally, GNU troff interprets this as a call of a macro named “dsabcd”. AT&T troff also interprets \*[ and \n[ as an interpolation of a string or register, respectively, called “[”. In GNU troff, however, the “[” is normally interpreted as beginning the enclosure of a long identifier. In compatibility mode, GNU troff interprets names in the traditional way, which means that they are limited to one or two characters. See the -C option in troff(1) and, above, the .C and .cp registers, and cp and “do” requests, for more on compatibility mode.

The register \n[.cp] is specialized and may require a statement of rationale. When writing macro packages or documents that use GNU troff features and which may be mixed with other packages or documents that do not—common scenarios include serial processing of man pages or use of the “so” or mso requests—you may desire correct operation regardless of compatibility mode enablement in the surrounding context. It may occur to you to save the existing value of \n(.C into a register, say, _C, at the beginning of your file, turn compatibility mode off with “.cp 0”, then restore it from that register at the end with “.cp \n(_C”. At the same time, a modular design of a document or macro package may lead you to multiple layers of inclusion. You cannot use the same register name everywhere lest you “clobber” the value from a preceding or enclosing context. The two-character register name space of AT&T troff is confining and mnemonically challenging; you may wish to use GNU troff’s more capacious name space. However, attempting “.nr _my_saved_C \n(.C” will not work in compatibility mode; the register name is too long. “This is exactly what .do is for,” you think, “.do nr _my_saved_C \n(.C”. The foregoing will always save zero to your register, because “do” turns compatibility mode off while it interprets its argument list. What you need is:

.do nr _my_saved_C \n[.cp]
.cp 0

at the beginning of your file, followed by

.cp \n[_my_saved_C]
.do rr _my_saved_C

at the end. As in the C language, we all have to share one big name space, so choose a register name that is unlikely to collide with other uses.

The existence of the .T string is a common feature of post-CSTR #54 troffs—DWB 3.3, Solaris, Heirloom Doctools, and Plan 9 troff all support it—but valid values are specific to each implementation. The behavior of the .T register in GNU troff differs from AT&T troff, which interpolated 1 only if nroff was the formatter and was called with -T.

The lf request sets the number of the current input line in AT&T troff, and the next in GNU troff.

AT&T troff had only environments named “0”, “1”, and “2”. In GNU troff, any number of environments may exist, using any valid identifiers for their names.

GNU troff normally tracks the interpolation depth of escape sequence parameters and other delimited structures, but not in compatibility mode. See section “Miscellaneous” above.

In compatibility mode, the escape sequences \f, \H, \m, \M, \R, \s, and \S are transparent at the beginning of an input line for the purpose of recognizing a control character, because they modify formatter state (\R) or properties of the environment (the rest) and therefore do not create output nodes. For example, this code produces bold output in both cases, but the text differs,

.de xx '
Hello!
..
\fB.xx\fP

formatting “.xx” normally and “Hello!” in compatibility mode.

GNU troff request names unrecognized by other troff implementations will likely be ignored; escape sequences that are GNU troff extensions are liable to format their function selector character. For example, the adjustable, non-breaking space escape sequence \~ is also supported by Heirloom Doctools troff 050915 (September 2005), mandoc 1.9.5 (2009-09-21), neatroff (commit 1c6ab0f6e, 2016-09-13), and Plan 9 from User Space troff (commit 93f8143600, 2022-08-12), but not by Solaris/Illumos troffs, which will render it as ~.

GNU troff does not allow the use of the escape sequences \|, \^, \&, \{, \}, \space, \', \`, \-, \_, \!, \%, or \c in identifiers; AT&T troff does. The \A escape sequence (see subsection “Escape sequences” above) may be helpful in avoiding their use.

Normally, the syntax form \sn accepts only a single character (a digit) for n, consistently with other forms that originated in AT&T troff, like \*, \$, \f, \g, \k, \n, and \z. In compatibility mode only, a non-zero n must be in the range 4–39. Legacy documents relying upon this quirk of parsing should be migrated to another \s form. [Background: The Graphic Systems C/A/T phototypesetter (the original device target for AT&T troff) supported only a few discrete type sizes in the range 6–36 points, so Ossanna contrived a special case in the parser to do what the user must have meant. Kernighan warned of this in the 1992 revision of CSTR #54 (§2.3), and more recently, McIlroy referred to it as a “living fossil”.]

Fractional type sizes cause one noteworthy incompatibility. In AT&T troff the ps request ignores scaling units and thus “.ps 10u” sets the type size to 10 points, whereas in GNU troff it sets the type size to 10 scaled points, which may be a much smaller measurement. See subsection “Fractional type sizes and new scaling units” above.

The ab request differs from AT&T troff: GNU troff writes no message to the standard error stream if no arguments are given, and it exits with a failure status instead of a successful one.

The bp request differs from AT&T troff: GNU troff does not accept a scaling unit on the argument, a page number; the former (somewhat uselessly) does.

In AT&T troff the pm request reports macro, string, and diversion sizes in units of 128-byte blocks, and an argument reduces the report to a sum of the above in the same units. GNU troff ignores any arguments and reports the sizes in bytes.

Unlike AT&T troff, GNU troff does not ignore the ss request if the output is a terminal device; instead, the values of minimum inter-word and additional inter-sentence space are each rounded down to the nearest multiple of 12.

In GNU troff there is a fundamental difference between (unformatted) characters and (formatted) glyphs. Everything that affects how a glyph is output is stored with the glyph node; once a glyph node has been constructed, it is unaffected by any subsequent requests that are executed, including bd, cs, tkf, tr, or fp requests. Normally, glyphs are constructed from characters immediately before the glyph is added to an output line. Macros, diversions, and strings are all, in fact, the same type of object; they contain a sequence of intermixed character and glyph nodes. Special characters transform from one to the other: before being added to the output, they behave as characters; afterward, they are glyphs. A glyph node does not behave like a character node when it is processed by a macro: it does not inherit any of the special properties that the character from which it was constructed might have had. For example, the input

.di x
\\\\
.br
.di
.x

produces “\\” in GNU troff. Each pair of backslashes becomes one backslash glyph; the resulting backslashes are thus not interpreted as escape characters when they are reread as the diversion is output. AT&T troff would interpret them as escape characters when rereading them and end up printing one “\”.

One way to format a backslash in most documents is with the \e escape sequence; this formats the glyph of the current escape character, regardless of whether it is used in a diversion; it also works in both GNU troff and AT&T troff. (Naturally, if you’ve changed the escape character, you need to prefix the “e” with whatever it is—and you’ll likely get something other than a backslash in the output.)

The other correct way, appropriate in contexts independent of the backslash’s common use as a roff escape character—perhaps in discussion of character sets or other programming languages—is the character escape \(rs or \[rs], for “reverse solidus”, from its name in the ECMA-6 (ISO/IEC 646) standard. [This escape sequence is not portable to AT&T troff, but is to its lineal descendant, Heirloom Doctools troff, as of its 060716 release (July 2006).]

To store an escape sequence in a diversion that is interpreted when the diversion is reread, either use the traditional \! transparent output facility, or, if this is unsuitable, the new \? escape sequence. See subsection “Escape sequences” above and sections “Diversions” and “gtroff Internals” in Groff: The GNU Implementation of troff, the groff Texinfo manual.

In the somewhat pathological case where a diversion exists containing a partially collected line and a partially collected line at the top-level diversion has never existed, AT&T troff will output the partially collected line at the end of input; GNU troff will not.

Formatter output incompatibilities
Its extensions notwithstanding, the groff intermediate output format has some incompatibilities with that of AT&T troff, but better compatibility is sought; problem reports and patches are welcome. The following incompatibilities are known.

Authors

This document was written by jjc [AT] jclark.com">James Clark, wl [AT] gnu.org">Werner Lemberg, groff-bernd.warken-72 [AT] web.de">Bernd Warken, and g.branden.robinson [AT] gmail.com">G. Branden Robinson.

See also

Groff: The GNU Implementation of troff, by Trent A. Fisher and Werner Lemberg, is the primary groff manual. You can browse it interactively with “info groff”.

“Troff User’s Manual” by Joseph F. Ossanna, 1976 (revised by Brian W. Kernighan, 1992), AT&T Bell Laboratories Computing Science Technical Report No. 54, widely called simply “CSTR #54”, documents the language, device and font description file formats, and output format referred to collectively in groff documentation as AT&T troff.

“A Typesetter-independent TROFF” by Brian W. Kernighan, 1982, AT&T Bell Laboratories Computing Science Technical Report No. 97, provides additional insights into the device and font description file formats and output format.

groff(1), groff(7), roff(7)