groff: Strings
5.19 Strings
============
'gtroff' has string variables, which are entirely for user convenience
(i.e. there are no built-in strings except '.T', but even this is a
read-write string variable).
Although the following requests can be used to create strings, simply
using an undefined string will cause it to be defined as empty. ⇒
Identifiers.
-- Request: .ds name [string]
-- Request: .ds1 name [string]
-- Escape: \*n
-- Escape: \*(nm
-- Escape: \*[name arg1 arg2 ...]
Define and access a string variable NAME (one-character name N,
two-character name NM). If NAME already exists, 'ds' overwrites
the previous definition. Only the syntax form using brackets can
take arguments that are handled identically to macro arguments; the
single exception is that a closing bracket as an argument must be
enclosed in double quotes. ⇒Request and Macro Arguments,
and ⇒Parameters.
Example:
.ds foo a \\$1 test
.
This is \*[foo nice].
=> This is a nice test.
The '\*' escape "interpolates" (expands in-place) a previously
defined string variable. To be more precise, the stored string is
pushed onto the input stack, which is then parsed by 'gtroff'.
Similar to number registers, it is possible to nest strings, i.e.,
string variables can be called within string variables.
If the string named by the '\*' escape does not exist, it is
defined as empty, and a warning of type 'mac' is emitted (see ⇒
Debugging, for more details).
*Caution:* Unlike other requests, the second argument to the 'ds'
request takes up the entire line including trailing spaces. This
means that comments on a line with such a request can introduce
unwanted space into a string.
.ds TeX T\h'-.2m'\v'.2m'E\v'-.2m'\h'-.1m'X \" Knuth's TeX
Instead the comment should be put on another line or have the
comment escape adjacent with the end of the string.
.ds TeX T\h'-.2m'\v'.2m'E\v'-.2m'\h'-.1m'X\" Knuth's TeX
To produce leading space the string can be started with a double
quote. No trailing quote is needed; in fact, any trailing quote is
included in your string.
.ds sign " Yours in a white wine sauce,
Strings are not limited to a single line of text. A string can
span several lines by escaping the newlines with a backslash. The
resulting string is stored _without_ the newlines.
.ds foo lots and lots \
of text are on these \
next several lines
It is not possible to have real newlines in a string. To put a
single double quote character into a string, use two consecutive
double quote characters.
The 'ds1' request turns off compatibility mode while interpreting a
string. To be more precise, a "compatibility save" input token is
inserted at the beginning of the string, and a "compatibility
restore" input token at the end.
.nr xxx 12345
.ds aa The value of xxx is \\n[xxx].
.ds1 bb The value of xxx is \\n[xxx].
.
.cp 1
.
\*(aa
=> warning: number register `[' not defined
=> The value of xxx is 0xxx].
\*(bb
=> The value of xxx is 12345.
Strings, macros, and diversions (and boxes) share the same name
space. Internally, even the same mechanism is used to store them.
This has some interesting consequences. For example, it is
possible to call a macro with string syntax and vice versa.
.de xxx
a funny test.
..
This is \*[xxx]
=> This is a funny test.
.ds yyy a funny test
This is
.yyy
=> This is a funny test.
In particular, interpolating a string does not hide existing macro
arguments. Thus in a macro, a more efficient way of doing
.xx \\$@
is
\\*[xx]\\
Note that the latter calling syntax doesn't change the value of
'\$0', which is then inherited from the calling macro.
Diversions and boxes can be also called with string syntax.
Another consequence is that you can copy one-line diversions or
boxes to a string.
.di xxx
a \fItest\fR
.br
.di
.ds yyy This is \*[xxx]\c
\*[yyy].
=> This is a test.
As the previous example shows, it is possible to store formatted
output in strings. The '\c' escape prevents the insertion of an
additional blank line in the output.
Copying diversions longer than a single output line produces
unexpected results.
.di xxx
a funny
.br
test
.br
.di
.ds yyy This is \*[xxx]\c
\*[yyy].
=> test This is a funny.
Usually, it is not predictable whether a diversion contains one or
more output lines, so this mechanism should be avoided. With Unix
'troff', this was the only solution to strip off a final newline
from a diversion. Another disadvantage is that the spaces in the
copied string are already formatted, making them unstretchable.
This can cause ugly results.
A clean solution to this problem is available in GNU 'troff', using
the requests 'chop' to remove the final newline of a diversion, and
'unformat' to make the horizontal spaces stretchable again.
.box xxx
a funny
.br
test
.br
.box
.chop xxx
.unformat xxx
This is \*[xxx].
=> This is a funny test.
⇒Gtroff Internals, for more information.
-- Request: .as name [string]
-- Request: .as1 name [string]
The 'as' request is similar to 'ds' but appends STRING to the
string stored as NAME instead of redefining it. If NAME doesn't
exist yet, it is created.
.as sign " with shallots, onions and garlic,
The 'as1' request is similar to 'as', but compatibility mode is
switched off while the appended string is interpreted. To be more
precise, a "compatibility save" input token is inserted at the
beginning of the appended string, and a "compatibility restore"
input token at the end.
Rudimentary string manipulation routines are given with the next two
requests.
-- Request: .substring str n1 [n2]
Replace the string named STR with the substring defined by the
indices N1 and N2. The first character in the string has index 0.
If N2 is omitted, it is implicitly set to the largest valid value
(the string length minus one). If the index value N1 or N2 is
negative, it is counted from the end of the string, going
backwards: The last character has index -1, the character before
the last character has index -2, etc.
.ds xxx abcdefgh
.substring xxx 1 -4
\*[xxx]
=> bcde
.substring xxx 2
\*[xxx]
=> de
-- Request: .length reg str
Compute the number of characters of STR and return it in the number
register REG. If REG doesn't exist, it is created. 'str' is read
in copy mode.
.ds xxx abcd\h'3i'efgh
.length yyy \*[xxx]
\n[yyy]
=> 14
-- Request: .rn xx yy
Rename the request, macro, diversion, or string XX to YY.
-- Request: .rm xx
Remove the request, macro, diversion, or string XX. 'gtroff'
treats subsequent invocations as if the object had never been
defined.
-- Request: .als new old
Create an alias named NEW for the request, string, macro, or
diversion object named OLD. The new name and the old name are
exactly equivalent (it is similar to a hard rather than a soft
link). If OLD is undefined, 'gtroff' generates a warning of type
'mac' and ignores the request.
To understand how the 'als' request works it is probably best to
think of two different pools: one pool for objects (macros,
strings, etc.), and another one for names. As soon as an object is
defined, 'gtroff' adds it to the object pool, adds its name to the
name pool, and creates a link between them. When 'als' creates an
alias, it adds a new name to the name pool that gets linked to the
same object as the old name.
Now consider this example.
.de foo
..
.
.als bar foo
.
.de bar
. foo
..
.
.bar
=> input stack limit exceeded
The definition of macro 'bar' replaces the old object this name is
linked to. However, the alias to 'foo' is still active! In other
words, 'foo' is still linked to the same object as 'bar', and the
result of calling 'bar' is an infinite, recursive loop that finally
leads to an error.
To undo an alias, simply call 'rm' on the aliased name. The object
itself is not destroyed until there are no more aliases.
-- Request: .chop xx
Remove (chop) the last character from the macro, string, or
diversion named XX. This is useful for removing the newline from
the end of diversions that are to be interpolated as strings. This
command can be used repeatedly; see ⇒Gtroff Internals, for
details on nodes inserted additionally by 'gtroff'.
⇒Identifiers, and ⇒Comments.