Initial import of binutils 2.22 on the new vendor branch

[dragonfly.git] / contrib / texinfo / doc / info-stnd.texi

vb\input texinfo.tex    @c -*-texinfo-*-
@comment $Id: info-stnd.texi,v 1.24 2008/08/29 17:27:18 karl Exp $
@c We must \input texinfo.tex instead of texinfo, otherwise make
@c distcheck in the Texinfo distribution fails, because the texinfo Info
@c file is made first, and texi2dvi must include . first in the path.
@comment %**start of header
@setfilename info-stnd.info
@include version-stnd.texi
@settitle GNU Info @value{VERSION}
@syncodeindex vr cp
@syncodeindex fn cp
@syncodeindex ky cp
@comment %**end of header

@copying
This manual is for GNU Info (version @value{VERSION}, @value{UPDATED}),
a program for viewing documents in Info format (usually created from
Texinfo source files).

Copyright @copyright{} 1992, 1993, 1996, 1997, 1998, 1999, 2001, 2002,
2003, 2004, 2005, 2006, 2007, 2008 Free Software Foundation, Inc.

@quotation
Permission is granted to copy, distribute and/or modify this document
under the terms of the GNU Free Documentation License, Version 1.2 or
any later version published by the Free Software Foundation; with no
Invariant Sections, with the Front-Cover texts being ``A GNU Manual'',
and with the Back-Cover Texts as in (a) below.  A copy of the
license is included in the section entitled ``GNU Free Documentation
License' in the Texinfo manual.

(a) The FSF's Back-Cover Text is: ``You have the freedom to
copy and modify this GNU manual.  Buying copies from the FSF
supports it in developing GNU and promoting software freedom.''
@end quotation

This document is part of a collection distributed under the GNU Free
Documentation License.  If you want to distribute this document
separately from the collection, you can do so by adding a copy of the
license to the document, as described in section 6 of the license.
@end copying

@dircategory Texinfo documentation system
@direntry
* info standalone: (info-stnd).            Read Info documents without Emacs.
* infokey: (info-stnd)Invoking infokey.    Compile Info customizations.
@end direntry

@titlepage
@title GNU Info
@subtitle for version @value{VERSION}, @value{UPDATED}
@author Brian J. Fox
@page
@vskip 0pt plus 1filll
@insertcopying
@end titlepage

@contents

@ifnottex
@node Top
@top GNU Info

If you do not know how to use Info, but have a working Info reader,
you should read the Info manual before this one (@pxref{Top, Getting
Started,,info, Info}), as it includes more background information and
a thorough tutorial.  This documentation describes the stand-alone Info
reader that is part of the Texinfo distribution, not the Info reader
that is part of GNU Emacs.
@end ifnottex

@menu
* Stand-alone Info::            What is Info?
* Invoking Info::               Options you can pass on the command line.
* Cursor Commands::             Commands which move the cursor within a node.
* Scrolling Commands::          Commands for reading the text within a node.
* Node Commands::               Commands for selecting a new node.
* Searching Commands::          Commands for searching an Info file.
* Xref Commands::               Commands for selecting cross references.
* Window Commands::             Commands which manipulate multiple windows.
* Printing Nodes::              How to print out the contents of a node.
* Miscellaneous Commands::      A few commands that defy categories.
* Variables::                   How to change the default behavior of Info.
* Custom Key Bindings::         How to define your own key-to-command bindings.
* Index::                       Global index with keystrokes, command names,
                                  variable names, and general concepts.
@end menu


@node Stand-alone Info
@chapter Stand-alone Info

The @dfn{Info} program is a stand-alone program, part of the Texinfo
distribution, which is used to view Info files on an ASCII terminal.
@dfn{Info files} are typically the result of processing Texinfo files
with the program @code{makeinfo} (also in the Texinfo distribution)

Texinfo itself is a documentation system that uses a single source
file to produce both on-line information and printed output.  You can
typeset and print the files that you read in Info.

GNU Emacs also provides an Info reader (just type @kbd{M-x info} in
Emacs).  Emacs Info and stand-alone Info have nearly identical user
interfaces, although customization and other details are different
(this manual explains the stand-alone Info reader).  The Emacs Info
reader supports the X Window System and other such bitmapped
interfaces, not just plain ASCII, so if you want the prettiest
display for Info files, you should try it.  You can use Emacs Info
without using Emacs for anything else.  (Type @kbd{C-x C-c} to exit;
this also works in the stand-alone Info reader.)


@node Invoking Info
@chapter Invoking Info

@cindex Info, invoking
@cindex invoking Info
@cindex command line options
@cindex options, command line
@cindex arguments, command line

GNU Info accepts several options to control the initial node being
viewed, and to specify which directories to search for Info files.  Here
is a template showing an invocation of GNU Info from the shell:

@example
info [@var{option}]@dots{} [@var{menu-item}@dots{}]
@end example

The program accepts the following options:

@table @code
@anchor{--apropos}
@item --apropos=@var{string}
@itemx -k @var{string}
@cindex Searching all indices
@cindex Info files@r{, searching all indices}
@cindex Apropos@r{, in Info files}
Specify a string to search in every index of every Info file installed
on your system.  Info looks up the named @var{string} in all the
indices it can find, prints the results to standard output, and then
exits.  If you are not sure which Info file explains certain issues,
this option is your friend.  (If your system has a lot of Info files
installed, searching all of them might take some time!)

You can invoke the apropos command from inside Info; see
@ref{Searching Commands}.

@cindex directory path
@item --directory @var{directory-path}
@itemx -d @var{directory-path}
Prepend @var{directory-path} to the list of directory paths searched
when Info needs to find a file.  You may issue @code{--directory}
multiple times; once for each directory which contains Info files.  The
list of directories searched by Info is constructed from the value of
the environment variable @code{INFOPATH}; @code{--directory} causes the
named @var{directory-path} to be prepended to that list.  The value of
@code{INFOPATH} is a list of directories usually separated by a colon;
on MS-DOS/MS-Windows systems, the semicolon is used.  If you do not
define @code{INFOPATH}, Info uses a default path defined when Info was
built as the initial list of directories.  If the value of
@code{INFOPATH} ends with a colon (or semicolon on MS-DOS/MS-Windows),
the initial list of directories is constructed by appending the
build-time default to the value of @code{INFOPATH}.

@cindex keystrokes, recording
@cindex remembering user keystrokes
@item --dribble=@var{dribble-file}
Specify a file where all user keystrokes will be recorded.  This file
can be used later to replay the same sequence of commands, see the
@samp{--restore} option below.

@item --file @var{filename}
@itemx -f @var{filename}
@cindex Info file, selecting
Specify a particular Info file to visit.  By default, Info visits
the file @code{dir}; if you use this option, Info will start with
@code{(@var{filename})Top} as the first file and node.

@cindex relative Info file names
@cindex file names, relative
@cindex Info files, relative
If @var{filename} is an absolute file name, or begins with @file{./} or
@file{../}, Info looks for @var{filename} only in the directory of the
specified @var{filename}, and adds the directory of @var{filename} to
the value of @code{INFOPATH}.  In contrast, if @var{filename} is in the
form of a relative file name, but without the @file{./} or @file{../}
prefix, Info will only look for it in the directories specified in
@code{INFOPATH}.  In other words, Info does @emph{not} treat file names
which lack @file{./} and @file{../} prefix as relative to the current
directory.

@cindex compressed Info files
@cindex files, compressed
@cindex Info files, compressed
In every directory Info tries, if @var{filename} is not found, Info
looks for it with a number of known extensions of Info files@footnote{
@file{.info}, @file{-info}, @file{/index}, and @file{.inf}.}.  For every
known extension, Info looks for a compressed file, if a regular file
isn't found.  Info supports files compressed with @code{gzip},
@code{bzip2}, @code{compress} and @code{yabba} programs; it calls
@code{gunzip}, @code{bunzip2}, @code{uncompress} and @code{unyabba},
accordingly, to decompress such files.  Compressed Info files are
assumed to have @file{.z}, @file{.gz}, @file{.bz2}, @file{.Z}, or
@file{.Y} extensions, possibly in addition to one of the known Info
files extensions@footnote{The MS-DOS version allows for the Info
extension, such as @code{.inf}, and the short compressed file
extensions, such as @file{.z} and @file{.gz}, to be merged into a single
extension, since DOS doesn't allow more than a single dot in the
basename of a file.  Thus, on MS-DOS, if Info looks for @file{bison},
file names like @file{bison.igz} and @file{bison.inz} will be found and
decompressed by @code{gunzip}.}.

@item --help
@itemx -h
Produces a relatively brief description of the available Info options.

@item --index-search @var{string}
@cindex index search, selecting from the command line
@cindex online help, using Info as
After processing all command-line arguments, go to the index in the Info
file and search for index entries which match @var{string}.  If such an
entry is found, the Info session begins with displaying the node pointed
to by the first matching index entry; press @kbd{,} to step through the
rest of the matching entries.  If no such entry exists, print @samp{no
entries found} and exit with nonzero status.  This can be used from
another program as a way to provide online help, or as a quick way of
starting to read an Info file at a certain node when you don't know the
exact name of that node.

This command can also be invoked from inside Info; see @ref{Searching
Commands}.

@item --node @var{nodename}
@itemx -n @var{nodename}
@cindex node, selecting from the command line
Specify a particular node to visit in the initial file that Info
loads.  This is especially useful in conjunction with
@code{--file}@footnote{Of course, you can specify both the file and node
in a @code{--node} command; but don't forget to escape the open and
close parentheses and whitespace from the shell as in: @code{info --node
"(emacs)Buffers"}.}.  You may specify @code{--node} multiple times; for
an interactive Info, each @var{nodename} is visited in its own window,
for a non-interactive Info (such as when @code{--output} is given) each
@var{nodename} is processed sequentially.

@item --output @var{filename}
@itemx -o @var{filename}
@cindex file, outputting to
@cindex outputting to a file
Specify @var{filename} as the name of a file to which to direct output.
Each node that Info visits will be output to @var{filename} instead of
interactively viewed.  A value of @code{-} for @var{filename} specifies
the standard output.

@cindex colors in documents
@cindex ANSI escape sequences in documents
@item --raw-escapes
@itemx --no-raw-escapes
@itemx -R
Do not remove ANSI escape sequences from documents.  Some versions of
Groff, the GNU document formatter, produce man pages with ANSI escape
sequences for bold, italics, and underlined characters, and for
colorized text.  By default, Info lets those escape sequences pass
through directly to the terminal.  If your terminal does not support
these escapes, use @code{--no-raw-escapes} to make Info remove them.

@cindex replaying recorded keystrokes
@item --restore=@var{dribble-file}
Read keystrokes from @var{dribble-file}, presumably recorded during
previous Info session (see the description of the @samp{--dribble}
option above).  When the keystrokes in the files are all read, Info
reverts its input to the usual interactive operation.

@item --show-malformed-multibytes
@itemx --no-show-malformed-multibytes
  Show malformed multibyte sequences in the output.  By default, such
sequences are dropped.

@anchor{--show-options}
@cindex command-line options, how to find
@cindex invocation description, how to find
@item --show-options
@itemx --usage
@itemx -O
This option causes Info to look for the node that describes how to
invoke the program and its command-line options, and begin the session
by displaying that node.  It is provided to make it easier to find the
most important usage information in a manual without the need to wade
through complex menu hierarchies.  The effect is similar to the
@code{M-x goto-invocation} command (@pxref{goto-invocation}) from inside
Info.

@cindex speech synthesizers
@item --speech-friendly
@itemx -b
On MS-DOS/MS-Windows only, this option causes Info to use standard file
I/O functions for screen writes.  (By default, Info uses direct writes
to the video memory on these systems, for faster operation and colored
display support.)  This allows the speech synthesizers used by blind
persons to catch the output and convert it to audible speech.

@item --subnodes
@cindex @code{--subnodes}, command line option
This option only has meaning when given in conjunction with
@code{--output}.  It means to recursively output the nodes appearing in
the menus of each node being output.  Menu items which resolve to
external Info files are not output, and neither are menu items which are
members of an index.  Each node is only output once.

@item --version
@cindex version information
Prints the version information of Info and exits.

@anchor{--vi-keys}
@cindex vi-like key bindings
@cindex Less-like key bindings
@item --vi-keys
This option binds functions to keys differently, to emulate the key
bindings of @code{vi} and Less.  The default key bindings are generally
modeled after Emacs.
(@xref{Custom Key Bindings},
for a more general way of altering GNU Info's key bindings.)

@cindex Info manual location
@cindex Where is an Info manual?
@item --where
@itemx --location
@itemx -w
Show the filename that would be read and exit, instead of actually
reading it and starting Info.

@item @var{menu-item}
@cindex menu, following
@anchor{command-line menu items}
Info treats its remaining arguments as the names of menu items.  The
first argument is a menu item in the initial node visited (generally
@code{dir}), the second argument is a menu item in the first argument's
node, etc.  You can easily move to the node of your choice by specifying
the menu names which describe the path to that node.  For example,

@example
info emacs buffers
@end example

@noindent
first selects the menu item @samp{Emacs} in the node @samp{(dir)Top},
and then selects the menu item @samp{Buffers} in the node
@samp{(emacs)Top}.
@end table

To avoid searching the @file{dir} files and just show some arbitrary
file, use @samp{-f} and the filename, as in @samp{info -f ./foo.info}.

The index search and the search for the node which describes program
invocation and command-line options begins @emph{after} processing all
the command-line menu items.  Therefore, the Info file searched for the
index or the invocation node is the file where Info finds itself after
following all the menu items given on the command line.  This is so
@samp{info emacs --show-options} does what you'd expect.

Finally, Info defines many default key bindings and variables.
@xref{Custom Key Bindings}, for information on how to customize these
settings.

@c FIXME: the feature with lowercasing the file name isn't documented


@node Cursor Commands
@chapter Moving the Cursor
@cindex cursor, moving
@cindex moving the cursor

Many people find that reading screens of text page by page is made
easier when one is able to indicate particular pieces of text with
some kind of pointing device.  Since this is the case, GNU Info (both
the Emacs and stand-alone versions) have several commands which allow
you to move the cursor about the screen.  The notation used in this
manual to describe keystrokes is identical to the notation used within
the Emacs manual, and the GNU Readline manual.  @xref{User Input,,,
emacs, The GNU Emacs Manual}, if you are unfamiliar with the
notation.@footnote{Here's a short summary.  @kbd{C-@var{x}} means
press the @kbd{CTRL} key and the key @var{x}.  @kbd{M-@var{x}} means
press the @kbd{META} key and the key @var{x}.  On many terminals th
@kbd{META} key is known as the @kbd{ALT} key.  @kbd{SPC} is the space
bar.  The other keys are usually called by the names imprinted on
them.}

The following table lists the basic cursor movement commands in Info.
Each entry consists of the key sequence you should type to execute the
cursor movement, the @code{M-x}@footnote{@code{M-x} is also a command; it
invokes @code{execute-extended-command}.  @xref{M-x, , Executing an
extended command, emacs, The GNU Emacs Manual}, for more detailed
information.} command name (displayed in parentheses), and a short
description of what the command does.  All of the cursor motion commands
can take a @dfn{numeric} argument (see @ref{Miscellaneous Commands,
@code{universal-argument}, to find out how to supply them}.  With a
numeric argument, the motion commands are simply executed that
many times; for example, a numeric argument of 4 given to
@code{next-line} causes the cursor to move down 4 lines.  With a
negative numeric argument, the motion is reversed; an argument of -4
given to the @code{next-line} command would cause the cursor to move
@emph{up} 4 lines.

@table @asis
@item @kbd{C-n} (@code{next-line})
@itemx @key{DOWN} (an arrow key)
@kindex C-n
@kindex DOWN (an arrow key)
@findex next-line
Move the cursor down to the next line.

@item @kbd{C-p} (@code{prev-line})
@itemx @key{UP} (an arrow key)
@kindex C-p
@kindex UP (an arrow key)
@findex prev-line
Move the cursor up to the previous line.

@item @kbd{C-a} (@code{beginning-of-line})
@itemx @key{Home} (on DOS/Windows only)
@kindex C-a, in Info windows
@kindex Home
@findex beginning-of-line
Move the cursor to the start of the current line.

@item @kbd{C-e} (@code{end-of-line})
@itemx @key{End} (on DOS/Windows only)
@kindex C-e, in Info windows
@kindex End
@findex end-of-line
Move the cursor to the end of the current line.

@item @kbd{C-f} (@code{forward-char})
@itemx @key{RIGHT} (an arrow key)
@kindex C-f, in Info windows
@kindex RIGHT (an arrow key)
@findex forward-char
Move the cursor forward a character.

@item @kbd{C-b} (@code{backward-char})
@itemx @key{LEFT} (an arrow key)
@kindex C-b, in Info windows
@kindex LEFT (an arrow key)
@findex backward-char
Move the cursor backward a character.

@item @kbd{M-f} (@code{forward-word})
@itemx @kbd{C-@key{RIGHT}} (on DOS/Windows only)
@kindex M-f, in Info windows
@kindex C-RIGHT
@findex forward-word
Move the cursor forward a word.

@item @kbd{M-b} (@code{backward-word})
@itemx @kbd{C-@key{LEFT}} (on DOS/Windows only)
@kindex M-b, in Info windows
@kindex C-LEFT
@findex backward-word
Move the cursor backward a word.

@item @kbd{M-<} (@code{beginning-of-node})
@itemx @kbd{C-@key{Home}} (on DOS/Windows only)
@itemx @kbd{b}
@itemx @kbd{M-b}, vi-like operation
@kindex b, in Info windows
@kindex M-<
@kindex C-Home
@kindex M-b, vi-like operation
@findex beginning-of-node
Move the cursor to the start of the current node.

@item @kbd{M->} (@code{end-of-node})
@itemx @kbd{C-@key{End}} (on DOS/Windows only)
@itemx @kbd{e}
@kindex M->
@kindex e, in Info windows
@kindex C-End
@findex end-of-node
Move the cursor to the end of the current node.

@item @kbd{M-r} (@code{move-to-window-line})
@kindex M-r
@findex move-to-window-line
Move the cursor to a specific line of the window.  Without a numeric
argument, @code{M-r} moves the cursor to the start of the line in the
center of the window.  With a numeric argument of @var{n}, @code{M-r}
moves the cursor to the start of the @var{n}th line in the window.
@end table


@node Scrolling Commands
@chapter Moving Text Within a Window
@cindex scrolling

Sometimes you are looking at a screenful of text, and only part of the
current paragraph you are reading is visible on the screen.  The
commands detailed in this section are used to shift which part of the
current node is visible on the screen.

Scrolling commands are bound differently when @samp{--vi-keys}
operation is in effect (@pxref{--vi-keys}).  These key bindings are
designated with ``vi-like operation''.  @xref{Custom Key Bindings},
for information on arbitrarily customizing key bindings and variable
settings.

@table @asis
@item @key{SPC} (@code{scroll-forward})
@kindex SPC, in Info windows
@findex scroll-forward
Shift the text in this window up.  That is, show more of the node which
is currently below the bottom of the window.  With a numeric argument,
show that many more lines at the bottom of the window; a numeric
argument of 4 would shift all of the text in the window up 4 lines
(discarding the top 4 lines), and show you four new lines at the bottom
of the window.  Without a numeric argument, @key{SPC} takes the bottom
two lines of the window and places them at the top of the window,
redisplaying almost a completely new screenful of lines.  If you are at
the end of a node, @key{SPC} takes you to the ``next'' node, so that you can
read an entire manual from start to finish by repeating @key{SPC}.

The default scroll size is one screen-full, but it can be changed by
invoking the (@code{scroll-forward-page-only-set-window}) command,
@samp{z} under @samp{--vi-keys}, with a numeric argument.

@item @key{NEXT} (an arrow key) (@code{scroll-forward-page-only})
@itemx @kbd{C-v}
@itemx @kbd{C-f}, vi-like operation
@itemx @kbd{f}, vi-like operation
@itemx @kbd{M-SPC}, vi-like operation
@kindex NEXT
@kindex C-v
@kindex C-f, vi-like operation
@kindex f, vi-like operation
@kindex M-SPC, vi-like operation
@findex scroll-forward-page-only
Shift the text in this window up.  This is identical to the @key{SPC}
operation above, except that it never scrolls beyond the end of the
current node.

@kindex PageDown
The @key{NEXT} key is known as the @key{PageDown} key on some
keyboards.

@item @kbd{z} (@code{scroll-forward-page-only-set-window}, vi-like operation)
@kindex z, vi-like operation
@findex scroll-forward-page-only-set-window
Scroll forward, like with @key{NEXT}, but if a numeric argument is
specified, it becomes the default scroll size for subsequent
@code{scroll-forward} and @code{scroll-backward} commands and their
ilk.

@item @key{DEL} (@code{scroll-backward})
@kindex DEL, in Info windows
@findex scroll-backward
Shift the text in this window down.  The inverse of
@code{scroll-forward}.
If you are at the start of a node, @key{DEL} takes you to the
``previous'' node, so that you can read an entire manual from finish to
start by repeating @key{DEL}.  The default scroll size can be changed by
invoking the (@code{scroll-backward-page-only-set-window}) command,
@samp{w} under @samp{--vi-keys}, with a numeric argument.

@itemx @key{PREVIOUS} (arrow key) (@code{scroll-backward-page-only})
@itemx @key{PRIOR} (arrow key)
@itemx @kbd{M-v}
@itemx @kbd{b}, vi-like operation
@itemx @kbd{C-b}, vi-like operation
@kindex PREVIOUS
@kindex M-v
@kindex b, vi-like operation
@kindex C-b, vi-like operation
@findex scroll-backward-page-only
Shift the text in this window down.  The inverse of
@code{scroll-forward-page-only}.  Does not scroll beyond the start of
the current node.  The default scroll size can be changed by invoking
the(@code{scroll-backward-page-only-set-window}) command, @samp{w} under
@samp{--vi-keys}, with a numeric argument.

@item @kbd{w} (@code{scroll-backward-page-only-set-window}, vi-like operation)
@kindex w, vi-like operation
@findex scroll-backward-page-only-set-window
Scroll backward, like with @key{PREVIOUS}, but if a numeric argument is
specified, it becomes the default scroll size for subsequent
@code{scroll-forward} and @code{scroll-backward} commands.

@item @kbd{C-n} (@code{down-line}, vi-like operation)
@itemx @kbd{C-e}, vi-like operation
@itemx @key{RET}, vi-like operation
@itemx @key{LFD}, vi-like operation
@itemx @key{DOWN}, vi-like operation
@kindex C-n, vi-like operation
@kindex C-e, vi-like operation
@kindex RET, vi-like operation
@kindex LFD, vi-like operation
@kindex DOWN, vi-like operation
@findex down-line
Scroll forward by one line.  With a numeric argument, scroll forward
that many lines.

@item @kbd{C-p} (@code{up-line}, vi-like operation)
@itemx @key{UP}, vi-like operation
@itemx @kbd{y}, vi-like operation
@itemx @kbd{k}, vi-like operation
@itemx @kbd{C-k}, vi-like operation
@itemx @kbd{C-y}, vi-like operation
@kindex C-p, vi-like operation
@kindex UP, vi-like operation
@kindex y, vi-like operation
@kindex k, vi-like operation
@kindex C-k, vi-like operation
@kindex C-y, vi-like operation
@findex up-line
Scroll backward one line.  With a numeric argument, scroll backward that
many lines.

@item @kbd{d} (@code{scroll-half-screen-down}, vi-like operation)
@itemx @kbd{C-d}, vi-like operation
@kindex d, vi-like operation
@kindex C-d, vi-like operation
@findex scroll-half-screen-down
Scroll forward by half of the screen size.  With a numeric argument,
scroll that many lines.  If an argument is specified, it becomes the new
default number of lines to scroll for subsequent @samp{d} and @samp{u}
commands.

@item @kbd{u} (@code{scroll-half-screen-up}, vi-like operation)
@itemx @kbd{C-u}, vi-like operation
@kindex u, vi-like operation
@kindex C-u, vi-like operation
@findex scroll-half-screen-up
Scroll back by half of the screen size.  With a numeric argument,
scroll that many lines.  If an argument is specified, it becomes the new
default number of lines to scroll for subsequent @samp{u} and @samp{d}
commands.
@end table

@cindex scrolling through node structure
The @code{scroll-forward} and @code{scroll-backward} commands can also
move forward and backward through the node structure of the file.  If
you press @key{SPC} while viewing the end of a node, or @key{DEL} while
viewing the beginning of a node, what happens is controlled by the
variable @code{scroll-behavior} (@pxref{scroll-behavior}).

The @code{scroll-forward-page-only} and @code{scroll-backward-page-only}
commands never scroll beyond the current node.

@kindex PageUp
The @key{PREVIOUS} key is the @key{PageUp} key on many keyboards.  Emacs
refers to it by the name @key{PRIOR}.  When you use @key{PRIOR} or
@key{PageUp} to scroll, Info never scrolls beyond the beginning of the
current node.

@kindex BS (backspace)
If your keyboard lacks the @key{DEL} key, look for a key called
@key{BS}, or @samp{Backspace}, sometimes designated with an arrow which
points to the left, which should perform the same function.

@table @asis
@item @kbd{C-l} (@code{redraw-display})
@kindex C-l
@findex redraw-display
Redraw the display from scratch, or shift the line containing the cursor
to a specified location.  With no numeric argument, @samp{C-l} clears
the screen, and then redraws its entire contents.  Given a numeric
argument of @var{n}, the line containing the cursor is shifted so that
it is on the @var{n}th line of the window.

@item @kbd{C-x @kbd{w}} (@code{toggle-wrap})
@kindex C-w
@findex toggle-wrap
Toggles the state of line wrapping in the current window.  Normally,
lines which are longer than the screen width @dfn{wrap}, i.e., they are
continued on the next line.  Lines which wrap have a @samp{\} appearing
in the rightmost column of the screen.  You can cause such lines to be
terminated at the rightmost column by changing the state of line
wrapping in the window with @code{C-x w}.  When a line which needs more
space than one screen width to display is displayed, a @samp{$} appears
in the rightmost column of the screen, and the remainder of the line is
invisible.  When long lines are truncated, the modeline displays the
@samp{$} character near its left edge.
@end table


@node Node Commands
@chapter Selecting a Node
@cindex nodes, selection of

This section details the numerous Info commands which select a new node
to view in the current window.

The most basic node commands are @samp{n}, @samp{p}, @samp{u}, and
@samp{l}.  Note that the commands to select nodes are mapped differently
when @samp{--vi-keys} is in effect; these keybindings are designated
below as ``vi-like operation''.

When you are viewing a node, the top line of the node contains some Info
@dfn{pointers} which describe where the next, previous, and up nodes
are.  Info uses this line to move about the node structure of the file
when you use the following commands:

@table @asis
@item @kbd{n} (@code{next-node})
@itemx @kbd{C-@key{NEXT}} (on DOS/Windows only)
@itemx @kbd{C-x @kbd{n}}, vi-like operation
@kindex n
@kindex C-NEXT
@kindex C-x n, vi-like operation
@findex next-node
Select the `Next' node.

@kindex C-PgDn
The @key{NEXT} key is known as the @key{PgDn} key on some
keyboards.

@item @kbd{p} (@code{prev-node})
@itemx @kbd{C-@key{PREVIOUS}} (on DOS/Windows only)
@kindex p
@kindex C-PREVIOUS
@findex prev-node
Select the `Prev' node.

@kindex C-PgUp
The @key{PREVIOUS} key is known as the @key{PgUp} key on some
keyboards.

@item @kbd{u} (@code{up-node})
@itemx @kbd{C-@key{UP}} (an arrow key on DOS/Windows only)
@itemx @kbd{C-x @kbd{u}}, vi-like operation
@kindex u
@kindex C-UP
@kindex C-x u, vi-like operation
@findex up-node
Select the `Up' node.
@end table

You can easily select a node that you have already viewed in this window
by using the @samp{l} command---this name stands for ``last'', and
actually moves backwards through the history of visited nodes for this
window.  This is handy when you followed a reference to another node,
possibly to read about a related issue, and would like then to resume
reading at the same place where you started the excursion.

Each node where you press @samp{l} is discarded from the history.  Thus,
by the time you get to the first node you visited in a window, the
entire history of that window is discarded.

@table @asis
@item @kbd{l} (@code{history-node})
@itemx @kbd{C-@key{CENTER}} (on DOS/Windows only)
@itemx @kbd{'}, vi-like operation
@kindex l
@kindex C-CENTER
@kindex ', vi-like operation
@findex history-node
Pop the most recently selected node in this window from the node
history.
@end table

Two additional commands make it easy to select the most commonly
selected nodes; they are @samp{t} and @samp{d}.

@table @asis
@item @kbd{t} (@code{top-node})
@itemx @kbd{M-t}, vi-like operation
@kindex t
@kindex M-t, vi-like operation
@findex top-node
Select the node @samp{Top} in the current Info file.

@item @kbd{d} (@code{dir-node})
@itemx @kbd{M-d}, vi-like operation
@kindex d
@kindex M-d, vi-like operation
@findex dir-node
Select the directory node (i.e., the node @samp{(dir)}).
@end table

Here are some other commands which immediately result in the selection
of a different node in the current window:

@table @asis
@item @kbd{<} (@code{first-node})
@itemx @kbd{g}, vi-like operation
@kindex <
@kindex g, vi-like operation
@findex first-node
Selects the first node which appears in this file.  This node is most
often @samp{Top}, but it does not have to be.  With a numeric argument
@var{N}, select the @var{N}th node (the first node is node 1).  An
argument of zero is the same as the argument of 1.

@item @kbd{>} (@code{last-node})
@itemx @kbd{G}, vi-like operation
@kindex >
@kindex G, vi-like operation
@findex last-node
Select the last node which appears in this file.  With a numeric argument
@var{N}, select the @var{N}th node (the first node is node 1).  An
argument of zero is the same as no argument, i.e., it selects the last
node.

@item @kbd{]} (@code{global-next-node})
@kindex ]
@findex global-next-node
Move forward or down through node structure.  If the node that you are
currently viewing has a @samp{Next} pointer, that node is selected.
Otherwise, if this node has a menu, the first menu item is selected.  If
there is no @samp{Next} and no menu, the same process is tried with the
@samp{Up} node of this node.

@item @kbd{[} (@code{global-prev-node})
@kindex [
@findex global-prev-node
Move backward or up through node structure.  If the node that you are
currently viewing has a @samp{Prev} pointer, that node is selected.
Otherwise, if the node has an @samp{Up} pointer, that node is selected,
and if it has a menu, the last item in the menu is selected.
@end table

You can get the same behavior as @code{global-next-node} and
@code{global-prev-node} while simply scrolling through the file with
@key{SPC} and @key{DEL} (@pxref{scroll-behavior}).

@table @asis
@anchor{goto-node}
@item @kbd{g} (@code{goto-node})
@itemx @kbd{C-x @kbd{g}}, vi-like operation
@kindex g
@kindex C-x g, vi-like operation
@findex goto-node
Read the name of a node and select it.  While reading the node name,
completion (@pxref{The Echo Area, completion}) is only done for the
nodes which reside in one of the Info files that were loaded in the
current Info session; if the desired node resides in some other file,
you must type the node exactly as it appears in that Info file, and you
must include the Info file of the other file.  For example,

@example
@code{g(emacs)Buffers}
@end example

finds the node @samp{Buffers} in the Info file @file{emacs}.

@anchor{goto-invocation}
@item @kbd{O} (@code{goto-invocation}
@itemx @kbd{I}
@kindex O
@kindex I
@findex goto-invocation
@cindex finding the Invocation node
Read the name of a program and look for a node in the current Info file
which describes the invocation and the command-line options for that
program.  The default program name is derived from the name of the
current Info file.  This command does the same as the
@samp{--show-options} command-line option (@pxref{--show-options}), but
it also allows to specify the program name; this is important for those
manuals which describe several programs.

If you need to find the Invocation node of a program that is documented
in another Info file, you need to visit that file before invoking
@samp{I}.  For example, if you are reading the Emacs manual and want to
see the command-line options of the @code{makeinfo} program, type @kbd{g
(texinfo) @key{RET}} and then @kbd{I makeinfo @key{RET}}.  If you don't
know what Info file documents the command, or if invoking @samp{I}
doesn't display the right node, go to the @samp{(dir)} node (using the
@samp{d} command) and invoke @samp{I} from there.

@item @kbd{G} (@code{menu-sequence})
@kindex G
@findex menu-sequence
@cindex menu, following, from inside Info
Read a sequence of menu entries and follow it.  Info prompts for a
sequence of menu items separated by commas.  (Since commas are not
allowed in a node name, they are a natural choice for a delimiter in a
list of menu items.)  Info then looks up the first item in the menu of
the node @samp{(dir)} (if the @samp{(dir)} node cannot be found, Info
uses @samp{Top}).  If such an entry is found, Info goes to the node it
points to and looks up the second item in the menu of that node, etc.
In other words, you can specify a complete path which descends through
the menu hierarchy of a particular Info file starting at the
@samp{(dir)} node.  This has the same effect as if you typed the menu
item sequence on Info's command line, see @ref{command-line menu items,,
Info command-line arguments processing}.  For example,

@example
 @kbd{G Texinfo,Overview,Reporting Bugs @key{RET}}
@end example

@noindent
displays the node @samp{Reporting Bugs} in the Texinfo manual.  (You
don't actually need to type the menu items in their full length, or in
their exact letter-case.  However, if you do type the menu items
exactly, Info will find it faster.)

If any of the menu items you type are not found, Info stops at the last
entry it did find and reports an error.

@item @kbd{C-x @kbd{k}} (@code{kill-node})
@kindex C-x k
@findex kill-node
Kill a node.  The node name is prompted for in the echo area, with a
default of the current node.  @dfn{Killing} a node means that Info tries
hard to forget about it, removing it from the list of history nodes kept
for the window where that node is found.  Another node is selected in
the window which contained the killed node.

@item @kbd{C-x C-f} (@code{view-file})
@kindex C-x C-f
@findex view-file
Read the name of a file and selects the entire file.  The command
@example
@code{C-x C-f @var{filename}}
@end example
is equivalent to typing
@example
@code{g(@var{filename})*}
@end example

@item @kbd{C-x C-b} (@code{list-visited-nodes})
@kindex C-x C-b
@findex list-visited-nodes
Make a window containing a menu of all of the currently visited nodes.
This window becomes the selected window, and you may use the standard
Info commands within it.

@item @kbd{C-x @kbd{b}} (@code{select-visited-node})
@kindex C-x b
@findex select-visited-node
Select a node which has been previously visited in a visible window.
This is similar to @samp{C-x C-b} followed by @samp{m}, but no window is
created.
@end table


@node Searching Commands
@chapter Searching an Info File
@cindex searching

GNU Info allows you to search for a sequence of characters throughout an
entire Info file, search through the indices of an Info file, or find
areas within an Info file which discuss a particular topic.

@table @asis
@item @kbd{s} (@code{search})
@itemx @kbd{/}
@kindex s
@kindex /
@findex search
@cindex regular expression search
Read a string in the echo area and search for it, either as a regular
expression (by default) or a literal string.  If the string includes
upper-case characters, the Info file is searched case-sensitively;
otherwise Info ignores the letter case.  With a numeric argument of
@var{N}, search for @var{N}th occurrence of the string.  Negative
arguments search backwards.

@item @kbd{?} (@code{search-backward}, vi-like operation)
@kindex ?, vi-like operation
@findex search-backward
Read a string in the echo area and search backward through the Info file
for that string.  If the string includes upper-case characters, the Info
file is searched case-sensitively; otherwise Info ignores the letter
case.  With a numeric argument of @var{N}, search for @var{N}th
occurrence of the string.  Negative arguments search forward.

@item @kbd{R} (@code{toggle-regexp})
@kindex R
@findex toggle-regexp
Toggle between using regular expressions and literal strings for
searching.  Info uses so-called `extended' regular expression syntax,
similar to Emacs (@pxref{Regexps, , Using Regular Expressions, emacs,
The GNU Emacs Manual}).

@item @kbd{S} (@code{search-case-sensitively}
@kindex S
@findex search-case-sensitively
@cindex search, case-sensitive
@cindex case-sensitive search
Read a string in the echo area and search for it case-sensitively, even
if the string includes only lower-case letters.  With a numeric argument
of @var{N}, search for @var{N}th occurrence of the string.  Negative
arguments search backwards.

@item @kbd{C-x @kbd{n}} (@code{search-next})
@itemx @kbd{n}, vi-like operation
@kindex C-x n
@kindex n, vi-like operation
@findex search-next
@cindex repeated search
Search for the same string used in the last search command, in the same
direction, and with the same case-sensitivity option.  With a numeric
argument of @var{N}, search for @var{N}th next occurrence.

@item @kbd{C-x @kbd{N}} (@code{search-previous})
@itemx @kbd{N}, vi-like operation
@kindex C-x N
@kindex n, vi-like operation
@findex search-previous
Search for the same string used in the last search command, and with the
same case-sensitivity option, but in the reverse direction.  With a
numeric argument of @var{N}, search for @var{N}th previous occurrence.

@item @kbd{C-s} (@code{isearch-forward})
@kindex C-s
@findex isearch-forward
@cindex incremental search
Interactively search forward through the Info file for a string as you
type it.  If the string includes upper-case characters, the search is
case-sensitive; otherwise Info ignores the letter case.

@item @kbd{C-r} (@code{isearch-backward})
@kindex C-r
@findex isearch-backward
Interactively search backward through the Info file for a string as
you type it.  If the string includes upper-case characters, the search
is case-sensitive; otherwise Info ignores the letter case.

@item @kbd{i} (@code{index-search})
@kindex i
@findex index-search
@cindex index, searching
@cindex searching, in the indices
Look up a string in the indices for this Info file, and select a node
to which the found index entry points.

@item @kbd{,} (@code{next-index-match})
@kindex ,
@findex next-index-match
Move to the node containing the next matching index item from the last
@samp{i} command.

@item @kbd{M-x index-apropos}
@findex index-apropos
Grovel the indices of all the known Info files on your system for a
string, and build a menu of the possible matches.
@end table

The most basic searching command is @samp{s} or @samp{/}
(@code{search}).  The @samp{s} command prompts you for a string in the
echo area, and then searches the remainder of the Info file for an
occurrence of that string.  If the string is found, the node containing
it is selected, and the cursor is left positioned at the start of the
found string.  Subsequent @samp{s} commands show you the default search
string within @samp{[} and @samp{]}; pressing @key{RET} instead of
typing a new string will use the default search string.  Under
@samp{--vi-keys} (@pxref{--vi-keys}), using the @samp{n} or @samp{N}
commands is a faster way of searching for the same string.

@dfn{Incremental searching} is similar to basic searching, but the
string is looked up while you are typing it, instead of waiting until
the entire search string has been specified.

@cindex search, and case-sensitivity
@cindex case-sensitivity, and search
Both incremental and non-incremental search by default ignore the case
of letters when comparing the Info file text with the search string.
However, an uppercase letter in the search string makes the search
case-sensitive.  You can force a case-sensitive non-incremental search,
even for a string that includes only lower-case letters, by using the
@samp{S} command (@code{search-case-sensitively}).  The @samp{n} and
@samp{N} commands operate case-sensitively if the last search command
was @samp{S}.

The most efficient means of finding something quickly in a manual is
the @samp{i} command (@code{index-search}).  This command prompts for
a string, and then looks for that string in all the indices of the
current Info manual.  If it finds a matching index entry, it displays
the node to which that entry refers and prints the full text of the
entry in the echo area.  You can press @samp{,}
(@code{next-index-match}) to find more matches.  A good Info manual
has all of its important concepts indexed, so the @samp{i} command
lets you use a manual as a reference.

If you don't know what manual documents something, try the @kbd{M-x
index-apropos} command.  It prompts for a string and then looks up
that string in all the indices of all the Info documents installed on
your system.  It can also be invoked from the command line; see
@ref{--apropos}.


@node Xref Commands
@chapter Selecting Cross References

We have already discussed the @samp{Next}, @samp{Prev}, and @samp{Up}
pointers which appear at the top of a node.  In addition to these
pointers, a node may contain other pointers which refer you to a
different node, perhaps in another Info file.  Such pointers are called
@dfn{cross references}, or @dfn{xrefs} for short.

@menu
* Parts of an Xref::            What a cross reference is made of.
* Selecting Xrefs::             Commands for selecting menu or note items.
@end menu

@node Parts of an Xref
@section Parts of an Xref

Cross references have two major parts: the first part is called the
@dfn{label}; it is the name that you can use to refer to the cross
reference, and the second is the @dfn{target}; it is the full name of
the node that the cross reference points to.

The target is separated from the label by a colon @samp{:}; first the
label appears, and then the target.  For example, in the sample menu
cross reference below, the single colon separates the label from the
target.

@example
* Foo Label: Foo Target.        More information about Foo.
@end example

Note the @samp{.} which ends the name of the target.  The @samp{.} is
not part of the target; it serves only to let Info know where the target
name ends.

A shorthand way of specifying references allows two adjacent colons to
stand for a target name which is the same as the label name:

@example
* Foo Commands::                Commands pertaining to Foo.
@end example

In the above example, the name of the target is the same as the name of
the label, in this case @code{Foo Commands}.

You will normally see two types of cross reference while viewing nodes:
@dfn{menu} references, and @dfn{note} references.  Menu references
appear within a node's menu; they begin with a @samp{*} at the beginning
of a line, and continue with a label, a target, and a comment which
describes what the contents of the node pointed to contains.

Note references appear within the body of the node text; they begin with
@code{*Note}, and continue with a label and a target.

Like @samp{Next}, @samp{Prev}, and @samp{Up} pointers, cross references
can point to any valid node.  They are used to refer you to a place
where more detailed information can be found on a particular subject.
Here is a cross reference which points to a node within the Texinfo
documentation:  @xref{xref, , Writing an Xref, texinfo, the Texinfo
Manual}, for more information on creating your own texinfo cross
references.

@node Selecting Xrefs
@section Selecting Xrefs

The following table lists the Info commands which operate on menu items.

@table @asis
@item @kbd{1} (@code{menu-digit})
@itemx @kbd{2} @dots{} @kbd{9}
@itemx @kbd{M-1}, vi-like operation
@itemx @kbd{M-2} @dots{} @kbd{M-9}, vi-like operation
@cindex 1 @dots{} 9, in Info windows
@cindex M-1 @dots{} M-9, vi-like operation
@kindex 1 @dots{} 9, in Info windows
@kindex M-1 @dots{} M-9, vi-like operation
@findex menu-digit
Within an Info window, pressing a single digit, (such as @samp{1}),
selects that menu item, and places its node in the current window.
For convenience, there is one exception; pressing @samp{0} selects the
@emph{last} item in the node's menu.  When @samp{--vi-keys} is in
effect, digits set the numeric argument, so these commands are remapped
to their @samp{M-} varieties.  For example, to select the last menu
item, press @kbd{M-0}.

@item @kbd{0} (@code{last-menu-item})
@itemx @kbd{M-0}, vi-like operation
@kindex 0, in Info windows
@kindex M-0, vi-like operation
@findex last-menu-item
Select the last item in the current node's menu.

@item @kbd{m} (@code{menu-item})
@kindex m
@findex menu-item
Reads the name of a menu item in the echo area and selects its node.
Completion is available while reading the menu label.  @xref{The Echo
Area, completion}.

@item @kbd{M-x find-menu}
@findex find-menu
Move the cursor to the start of this node's menu.
@end table

This table lists the Info commands which operate on cross references.

@table @asis
@item @kbd{f} (@code{xref-item})
@itemx @kbd{r}
@item @kbd{M-f}, vi-like operation
@itemx @kbd{C-x @kbd{r}}, vi-like operation
@kindex f
@kindex r
@kindex M-f, vi-like operation
@kindex C-x r, vi-like operation
@findex xref-item
Reads the name of a note cross reference in the echo area and selects
its node.  Completion is available while reading the cross reference
label.  @xref{The Echo Area, completion}.
@end table

Finally, the next few commands operate on menu or note references alike:

@table @asis
@item @key{TAB} (@code{move-to-next-xref})
@kindex TAB, in Info windows
@findex move-to-next-xref
Move the cursor to the start of the next nearest menu item or note
reference in this node.  You can then use @key{RET}
(@code{select-reference-this-line}) to select the menu or note reference.

@item @kbd{M-TAB} (@code{move-to-prev-xref})
@itemx @kbd{Shift-@key{TAB}} (on DOS/Windows only)
@kindex M-TAB, in Info windows
@findex move-to-prev-xref
Move the cursor the start of the nearest previous menu item or note
reference in this node.

@kindex Shift-TAB, in Info windows
@kindex BackTab, in Info windows
On DOS/Windows only, the @kbd{Shift-@key{TAB}} key is an alias for
@kbd{M-@key{TAB}}.  This key is sometimes called @samp{BackTab}.

@item @key{RET} (@code{select-reference-this-line})
@itemx @kbd{M-g}, vi-like operation
@kindex RET, in Info windows
@kindex M-g, vi-like operation
@findex select-reference-this-line
Select the menu item or note reference appearing on this line.
@end table


@node Window Commands
@chapter Manipulating Multiple Windows
@cindex windows, manipulating

A @dfn{window} is a place to show the text of a node.  Windows have a
view area where the text of the node is displayed, and an associated
@dfn{mode line}, which briefly describes the node being viewed.

GNU Info supports multiple windows appearing in a single screen; each
window is separated from the next by its modeline.  At any time, there
is only one @dfn{active} window, that is, the window in which the cursor
appears.  There are commands available for creating windows, changing
the size of windows, selecting which window is active, and for deleting
windows.

@menu
* The Mode Line::               What appears in the mode line?
* Basic Windows::               Manipulating windows in Info.
* The Echo Area::               Used for displaying errors and reading input.
@end menu

@node The Mode Line
@section The Mode Line

A @dfn{mode line} is a line of inverse video which appears at the bottom
of an Info window.  It describes the contents of the window just above
it; this information includes the name of the file and node appearing in
that window, the number of screen lines it takes to display the node,
and the percentage of text that is above the top of the window.  It can
also tell you if the indirect tags table for this Info file needs to be
updated, and whether or not the Info file was compressed when stored on
disk.

Here is a sample mode line for a window containing an uncompressed file
named @file{dir}, showing the node @samp{Top}.

@example
@group
-----Info: (dir)Top, 40 lines --Top-------------------------------------
            ^^   ^   ^^^        ^^
          (file)Node #lines    where
@end group
@end example

When a node comes from a file which is compressed on disk, this is
indicated in the mode line with two small @samp{z}'s.  In addition, if
the Info file containing the node has been split into subfiles, the name
of the subfile containing the node appears in the modeline as well:

@example
--zz-Info: (emacs)Top, 291 lines --Top-- Subfile: emacs-1.Z-------------
@end example

Truncation of long lines (as opposed to wrapping them to the next
display line, @pxref{Scrolling Commands, toggle-wrap}) is indicated by a
@samp{$} at the left edge of the mode line:

@example
--$--Info: (texinfo)Top, 480 lines --Top-- Subfile: texinfo-1-----------
@end example

When Info makes a node internally, such that there is no corresponding
info file on disk, the name of the node is surrounded by asterisks
(@samp{*}).  The name itself tells you what the contents of the window
are; the sample mode line below shows an internally constructed node
showing possible completions:

@example
-----Info: *Completions*, 7 lines --All---------------------------------
@end example

@node Basic Windows
@section Window Commands

It can be convenient to view more than one node at a time.  To allow
this, Info can display more than one @dfn{window}.  Each window has its
own mode line (@pxref{The Mode Line}) and history of nodes viewed in that
window (@pxref{Node Commands, , @code{history-node}}).

@table @asis
@item @kbd{C-x @kbd{o}} (@code{next-window})
@cindex windows, selecting
@kindex C-x o
@findex next-window
Select the next window on the screen.  Note that the echo area can only be
selected if it is already in use, and you have left it temporarily.
Normally, @samp{C-x o} simply moves the cursor into the next window on
the screen, or if you are already within the last window, into the first
window on the screen.  Given a numeric argument, @samp{C-x o} moves over
that many windows.  A negative argument causes @samp{C-x o} to select
the previous window on the screen.

@item @kbd{M-x prev-window}
@findex prev-window
Select the previous window on the screen.  This is identical to
@samp{C-x o} with a negative argument.

@item @kbd{C-x @kbd{2}} (@code{split-window})
@cindex windows, creating
@kindex C-x 2
@findex split-window
Split the current window into two windows, both showing the same node.
Each window is one half the size of the original window, and the cursor
remains in the original window.  The variable @code{automatic-tiling}
can cause all of the windows on the screen to be resized for you
automatically (@pxref{Variables, , automatic-tiling}).

@item @kbd{C-x @kbd{0}} (@code{delete-window})
@cindex windows, deleting
@kindex C-x 0
@findex delete-window
Delete the current window from the screen.  If you have made too many
windows and your screen appears cluttered, this is the way to get rid of
some of them.

@item @kbd{C-x @kbd{1}} (@code{keep-one-window})
@kindex C-x 1
@findex keep-one-window
Delete all of the windows excepting the current one.

@item @kbd{ESC @kbd{C-v}} (@code{scroll-other-window})
@kindex ESC C-v, in Info windows
@findex scroll-other-window
Scroll the other window, in the same fashion that @samp{C-v} might
scroll the current window.  Given a negative argument, scroll the
``other'' window backward.

@item @kbd{C-x @kbd{^}} (@code{grow-window})
@kindex C-x ^
@findex grow-window
Grow (or shrink) the current window.  Given a numeric argument, grow
the current window that many lines; with a negative numeric argument,
shrink the window instead.

@item @kbd{C-x @kbd{t}} (@code{tile-windows})
@cindex tiling
@kindex C-x t
@findex tile-windows
Divide the available screen space among all of the visible windows.
Each window is given an equal portion of the screen in which to display
its contents.  The variable @code{automatic-tiling} can cause
@code{tile-windows} to be called when a window is created or deleted.
@xref{Variables, , @code{automatic-tiling}}.
@end table

@node The Echo Area
@section The Echo Area
@cindex echo area

The @dfn{echo area} is a one line window which appears at the bottom of
the screen.  It is used to display informative or error messages, and to
read lines of input from you when that is necessary.  Almost all of the
commands available in the echo area are identical to their Emacs
counterparts, so please refer to that documentation for greater depth of
discussion on the concepts of editing a line of text.  The following
table briefly lists the commands that are available while input is being
read in the echo area:

@table @asis
@item @kbd{C-f} (@code{echo-area-forward})
@itemx @key{RIGHT} (an arrow key)
@itemx @kbd{M-h}, vi-like operation
@kindex C-f, in the echo area
@kindex RIGHT, in the echo area
@kindex M-h, in the echo area, vi-like operation
@findex echo-area-forward
Move forward a character.

@item @kbd{C-b} (@code{echo-area-backward})
@itemx @key{LEFT} (an arrow key)
@itemx @kbd{M-l}, vi-like operation
@kindex LEFT, in the echo area
@kindex C-b, in the echo area
@kindex M-l, in the echo area, vi-like operation
@findex echo-area-backward
Move backward a character.

@item @kbd{C-a} (@code{echo-area-beg-of-line})
@itemx @kbd{M-0}, vi-like operation
@kindex C-a, in the echo area
@kindex M-0, in the echo area, vi-like operation
@findex echo-area-beg-of-line
Move to the start of the input line.

@item @kbd{C-e} (@code{echo-area-end-of-line})
@itemx @kbd{M-$}, vi-like operation
@kindex C-e, in the echo area
@kindex M-$, vi-like operation
@findex echo-area-end-of-line
Move to the end of the input line.

@item @kbd{M-f} (@code{echo-area-forward-word})
@itemx @kbd{C-@key{RIGHT}} (DOS/Windows only)
@itemx @kbd{M-w}, vi-like operation
@kindex M-f, in the echo area
@kindex M-w, in the echo area, vi-like operation
@findex echo-area-forward-word
Move forward a word.

@kindex C-RIGHT, in the echo area
On DOS/Windows, @kbd{C-@key{RIGHT}} moves forward by words.

@item @kbd{M-b} (@code{echo-area-backward-word})
@itemx @kbd{C-@key{LEFT}} (DOS/Windows only)
@kindex M-b, in the echo area
@findex echo-area-backward-word
Move backward a word.

@kindex C-LEFT, in the echo area
On DOS/Windows, @kbd{C-@key{LEFT}} moves backward by words.

@item @kbd{C-d} (@code{echo-area-delete})
@itemx @kbd{M-x}, vi-like operation
@kindex C-d, in the echo area
@kindex M-x, in the echo area, vi-like operation
@findex echo-area-delete
Delete the character under the cursor.

@item @key{DEL} (@code{echo-area-rubout})
@kindex DEL, in the echo area
@findex echo-area-rubout
Delete the character behind the cursor.

On some keyboards, this key is designated @key{BS}, for
@samp{Backspace}.  Those keyboards will usually bind @key{DEL} in the
echo area to @code{echo-area-delete}.

@item @kbd{C-g} (@code{echo-area-abort})
@itemx @kbd{C-u}, vi-like operation
@kindex C-g, in the echo area
@kindex C-u, in the echo area, vi-like operation
@findex echo-area-abort
Cancel or quit the current operation.  If completion is being read, this
command discards the text of the input line which does not match any
completion.  If the input line is empty, it aborts the calling function.

@item @key{RET} (@code{echo-area-newline})
@kindex RET, in the echo area
@findex echo-area-newline
Accept (or forces completion of) the current input line.

@item @kbd{C-q} (@code{echo-area-quoted-insert})
@itemx @kbd{C-v}, vi-like operation
@kindex C-q, in the echo area
@kindex C-v, in the echo area, vi-like operation
@findex echo-area-quoted-insert
Insert the next character verbatim.  This is how you can insert control
characters into a search string, for example, or the @samp{?} character
when Info prompts with completion.

@item @var{printing character} (@code{echo-area-insert})
@kindex printing characters, in the echo area
@findex echo-area-insert
Insert the character.  Characters that have their 8th bit set, and not
bound to @samp{M-} commands, are also inserted verbatim; this is useful
for terminals which support Latin scripts.

@item @kbd{M-TAB} (@code{echo-area-tab-insert})
@itemx @kbd{Shift-@key{TAB}} (on DOS/Windows only)
@kindex M-TAB, in the echo area
@kindex Shift-TAB, in the echo area
@findex echo-area-tab-insert
Insert a TAB character.

@kindex Shift-TAB, in the echo area
@kindex BackTab, in the echo area
On DOS/Windows only, the @kbd{Shift-@key{TAB}} key is an alias for
@kbd{M-@key{TAB}}.  This key is sometimes called @samp{BackTab}.

@item @kbd{C-t} (@code{echo-area-transpose-chars})
@kindex C-t, in the echo area
@findex echo-area-transpose-chars
Transpose the characters at the cursor.
@end table

The next group of commands deal with @dfn{killing}, and @dfn{yanking}
text@footnote{
Some people are used to calling these operations @dfn{cut} and
@dfn{paste}, respectively.}.   For an in-depth discussion of killing and
yanking, see @ref{Killing, , Killing and Deleting, emacs, the GNU Emacs
Manual}.

@table @asis
@item @kbd{M-d} (@code{echo-area-kill-word})
@itemx @kbd{M-X}, vi-like operation
@kindex M-d, in the echo area
@kindex M-X, in the echo area, vi-like operation
@findex echo-area-kill-word
Kill the word following the cursor.

@item @kbd{M-@key{DEL}} (@code{echo-area-backward-kill-word})
@itemx @kbd{M-@key{BS}}
@kindex M-DEL, in the echo area
@findex echo-area-backward-kill-word
Kill the word preceding the cursor.

@kindex M-BS, in the echo area
On some keyboards, the @samp{Backspace} key is used instead of
@code{DEL}, so @code{M-@key{Backspace}} has the same effect as
@code{M-@key{DEL}}.

@item @kbd{C-k} (@code{echo-area-kill-line})
@kindex C-k, in the echo area
@findex echo-area-kill-line
Kill the text from the cursor to the end of the line.

@item @kbd{C-x @key{DEL}} (@code{echo-area-backward-kill-line})
@kindex C-x DEL, in the echo area
@findex echo-area-backward-kill-line
Kill the text from the cursor to the beginning of the line.

@item @kbd{C-y} (@code{echo-area-yank})
@kindex C-y, in the echo area
@findex echo-area-yank
Yank back the contents of the last kill.

@item @kbd{M-y} (@code{echo-area-yank-pop})
@kindex M-y, in the echo area
@findex echo-area-yank-pop
Yank back a previous kill, removing the last yanked text first.
@end table

@cindex completion
Sometimes when reading input in the echo area, the command that needed
input will only accept one of a list of several choices.  The choices
represent the @dfn{possible completions}, and you must respond with one
of them.  Since there are a limited number of responses you can make,
Info allows you to abbreviate what you type, only typing as much of the
response as is necessary to uniquely identify it.  In addition, you can
request Info to fill in as much of the response as is possible; this
is called @dfn{completion}.

The following commands are available when completing in the echo area:

@table @asis
@item @key{TAB} (@code{echo-area-complete})
@itemx @key{SPC}
@kindex TAB, in the echo area
@kindex SPC, in the echo area
@findex echo-area-complete
Insert as much of a completion as is possible.

@item @kbd{?} (@code{echo-area-possible-completions})
@kindex ?, in the echo area
@findex echo-area-possible-completions
Display a window containing a list of the possible completions of what
you have typed so far.  For example, if the available choices are:

@example
@group
bar
foliate
food
forget
@end group
@end example

@noindent
and you have typed an @samp{f}, followed by @samp{?}, Info will pop up a
window showing a node called @samp{*Completions*} which lists the
possible completions like this:

@example
@group
3 completions:
foliate         food
forget
@end group
@end example

@noindent
i.e., all of the choices which begin with @samp{f}.  Pressing @key{SPC}
or @key{TAB} would result in @samp{fo} appearing in the echo area, since
all of the choices which begin with @samp{f} continue with @samp{o}.
Now, typing @samp{l} followed by @samp{TAB} results in @samp{foliate}
appearing in the echo area, since that is the only choice which begins
with @samp{fol}.

@item @key{ESC C-v} (@code{echo-area-scroll-completions-window})
@kindex ESC C-v, in the echo area
@findex echo-area-scroll-completions-window
Scroll the completions window, if that is visible, or the ``other''
window if not.
@end table


@node Printing Nodes
@chapter Printing Nodes
@cindex printing

In general, we recommend that you use @TeX{} to format the document and
print sections of it, by running @code{tex} on the Texinfo source file.
However, you may wish to print out the contents of a node as a quick
reference document for later use, or if you don't have @TeX{} installed.
Info provides you with a command for doing this.

@table @asis
@item @kbd{M-x print-node}
@findex print-node
@cindex INFO_PRINT_COMMAND, environment variable
Pipe the contents of the current node through the command in the
environment variable @code{INFO_PRINT_COMMAND}.  If the variable does not
exist, the node is simply piped to @code{lpr} (on DOS/Windows, the
default is to print the node to the local printer device, @file{PRN}).

@cindex printing nodes to the local printer
@cindex local printer device
The value of @code{INFO_PRINT_COMMAND} may begin with the @samp{>}
character, as in @samp{>/dev/printer}, in which case Info treats the
rest as the name of a file or a device.  Instead of piping to a command,
Info opens the file, writes the node contents, and closes the file,
under the assumption that text written to that file will be printed by
the underlying OS.
@end table


@node Miscellaneous Commands
@chapter Miscellaneous Commands

GNU Info contains several commands which self-document GNU Info:

@table @asis
@item @kbd{M-x describe-command}
@cindex functions, describing
@cindex commands, describing
@findex describe-command
Read the name of an Info command in the echo area and then display a
brief description of what that command does.

@item @kbd{M-x describe-key}
@cindex keys, describing
@findex describe-key
Read a key sequence in the echo area, and then display the name and
documentation of the Info command that the key sequence invokes.

@item @kbd{M-x describe-variable}
Read the name of a variable in the echo area and then display a brief
description of what the variable affects.

@item @kbd{M-x where-is}
@findex where-is
Read the name of an Info command in the echo area, and then display
a key sequence which can be typed in order to invoke that command.

@item @kbd{C-h} (@code{get-help-window})
@itemx @kbd{?}
@itemx @key{F1} (on DOS/Windows only)
@itemx h, vi-like operation
@kindex C-h
@kindex ?, in Info windows
@kindex F1
@kindex h, vi-like operation
@findex get-help-window
Create (or Move into) the window displaying @code{*Help*}, and place
a node containing a quick reference card into it.  This window displays
the most concise information about GNU Info available.

@item @kbd{h} (@code{get-info-help-node})
@itemx @kbd{M-h}, vi-like operation
@kindex h
@kindex M-h, vi-like operation
@findex get-info-help-node
Try hard to visit the node @code{(info)Help}.  The Info file
@file{info.texi} distributed with GNU Info contains this node.  Of
course, the file must first be processed with @code{makeinfo}, and then
placed into the location of your Info directory.
@end table

Here are the commands for creating a numeric argument:

@table @asis
@item @kbd{C-u} (@code{universal-argument})
@cindex numeric arguments
@kindex C-u
@findex universal-argument
Start (or multiply by 4) the current numeric argument.  @samp{C-u} is
a good way to give a small numeric argument to cursor movement or
scrolling commands; @samp{C-u C-v} scrolls the screen 4 lines, while
@samp{C-u C-u C-n} moves the cursor down 16 lines.  @samp{C-u} followed
by digit keys sets the numeric argument to the number thus typed:
@kbd{C-u 1 2 0} sets the argument to 120.

@item @kbd{M-1} (@code{add-digit-to-numeric-arg})
@itemx @kbd{1}, vi-like operation
@itemx @kbd{M-2} @dots{} @kbd{M-9}
@itemx @kbd{2} @dots{} @kbd{9}, vi-like operation
@itemx @kbd{M-0}
@itemx @kbd{0}, vi-like operation
@kindex M-0 @dots{} M-9
@kindex 0 @dots{} 9, vi-like operation
@findex add-digit-to-numeric-arg
Add the digit value of the invoking key to the current numeric
argument.  Once Info is reading a numeric argument, you may just type
the digits of the argument, without the Meta prefix.  For example, you
might give @samp{C-l} a numeric argument of 32 by typing:

@example
@kbd{C-u 3 2 C-l}
@end example

@noindent
or

@example
@kbd{M-3 2 C-l}
@end example

@item @kbd{M--} (@code{add-digit-to-numeric-arg}
@itemx @kbd{-}
@kindex M--
@kindex -
@cindex negative arguments
@cindex arguments, negative
@cindex numeric arguments, negative
To make a negative argument, type @kbd{-}.  Typing @kbd{-} alone makes a
negative argument with a value of -1.  If you continue to type digit or
Meta-digit keys after @kbd{-}, the result is a negative number produced
by those digits.

@kbd{-} doesn't work when you type in the echo area, because you need to
be able to insert the @samp{-} character itself; use @kbd{M--} instead,
if you need to specify negative arguments in the echo area.
@end table

@samp{C-g} is used to abort the reading of a multi-character key
sequence, to cancel lengthy operations (such as multi-file searches) and
to cancel reading input in the echo area.

@table @asis
@item @kbd{C-g} (@code{abort-key})
@itemx @kbd{C-u}, vi-like operation
@cindex cancelling typeahead
@cindex cancelling the current operation
@kindex C-g, in Info windows
@kindex C-u cancels typeahead, vi-like operation
@findex abort-key
Cancel current operation.
@end table

The @samp{q} command of Info simply quits running Info.  Under
@samp{--vi-keys} (@pxref{--vi-keys}), you can also exit with @samp{:q}
or @samp{ZZ}.

@table @asis
@item @kbd{q} (@code{quit})
@itemx @kbd{C-x C-c}
@itemx @kbd{:q}, vi-like operation
@itemx @kbd{ZZ}, vi-like operation
@cindex quitting
@kindex q
@kindex C-x C-c
@kindex ZZ, vi-like operation
@findex quit
Exit GNU Info.
@end table

If the operating system tells GNU Info that the screen is 60 lines tall,
and it is actually only 40 lines tall, here is a way to tell Info that
the operating system is correct.

@table @asis
@item @kbd{M-x set-screen-height}
@findex set-screen-height
@cindex screen, changing the height of
Read a height value in the echo area and set the height of the
displayed screen to that value.
@end table

On MS-DOS/MS-Windows, this command actually tries to change the
dimensions of the visible screen to the value you type in the echo
area.

Finally, Info provides a convenient way to display footnotes which might
be associated with the current node that you are viewing:

@table @asis
@item @kbd{ESC C-f} (@code{show-footnotes})
@kindex ESC C-f
@findex show-footnotes
@cindex footnotes, displaying
Show the footnotes (if any) associated with the current node in another
window.  You can have Info automatically display the footnotes
associated with a node when the node is selected by setting the variable
@code{automatic-footnotes}.  @xref{Variables, , @code{automatic-footnotes}}.
@end table


@node Variables
@chapter Manipulating Variables

GNU Info contains several @dfn{variables} whose values are looked at by
various Info commands.  You can change the values of these variables,
and thus change the behavior of Info to more closely match your
environment and Info file reading manner.

There are two ways to set the value of a variable: interactively, using
the @code{set-variable} command described below, or in the @code{#var}
section of the @code{.infokey} file.  @xref{Custom Key Bindings}.

@table @asis
@item @kbd{M-x set-variable}
@cindex variables, setting
@findex set-variable
Read the name of a variable, and the value for it, in the echo area and
then set the variable to that value.  Completion is available when
reading the variable name (@pxref{The Echo Area, completion}); often,
completion is available when reading the value to give to the variable,
but that depends on the variable itself.  If a variable does @emph{not}
supply multiple choices to complete over, it expects a numeric value.

@item @kbd{M-x describe-variable}
@cindex variables, describing
@findex describe-variable
Read the name of a variable in the echo area and then display a brief
description of what the variable affects.
@end table

Here is a list of the variables that you can set in Info.

@table @code
@item automatic-footnotes
@vindex automatic-footnotes
When set to @code{On}, footnotes appear and disappear automatically;
else, they appear at the bottom of the node text.  This variable is
@code{Off} by default.  When a node is selected, a window containing
the footnotes which appear in that node is created, and the footnotes
are displayed within the new window.  The window that Info creates to
contain the footnotes is called @samp{*Footnotes*}.  If a node is
selected which contains no footnotes, and a @samp{*Footnotes*} window
is on the screen, the @samp{*Footnotes*} window is deleted.  Footnote
windows created in this fashion are not automatically tiled so that
they can use as little of the display as is possible.

@item automatic-tiling
@vindex automatic-tiling
When set to @code{On}, creating or deleting a window resizes other
windows.  This variable is @code{Off} by default.  Normally, typing
@samp{C-x 2} divides the current window into two equal parts.  When
@code{automatic-tiling} is set to @code{On}, all of the windows are
resized automatically, keeping an equal number of lines visible in each
window.  There are exceptions to the automatic tiling; specifically, the
windows @samp{*Completions*} and @samp{*Footnotes*} are @emph{not}
resized through automatic tiling; they remain their original size.

@anchor{cursor-movement-scrolls}
@item cursor-movement-scrolls
Normally, cursor movement commands (@pxref{Cursor Commands}) stop when
top or bottom of a node is reached.  When this variable is set to
@code{On}, cursor movement commands act as scrolling ones and their
behavior is controlled by the @code{scroll-behavior} variable (see below).

@item errors-ring-bell
@vindex errors-ring-bell
When set to @code{On}, errors cause the bell to ring.  The default
setting of this variable is @code{On}.

@item gc-compressed-files
@vindex gc-compressed-files
When set to @code{On}, Info garbage collects files which had to be
uncompressed.  The default value of this variable is @code{Off}.
Whenever a node is visited in Info, the Info file containing that node
is read into core, and Info reads information about the tags and nodes
contained in that file.  Once the tags information is read by Info, it
is never forgotten.  However, the actual text of the nodes does not need
to remain in core unless a particular Info window needs it.  For
non-compressed files, the text of the nodes does not remain in core when
it is no longer in use.  But de-compressing a file can be a time
consuming operation, and so Info tries hard not to do it twice.
@code{gc-compressed-files} tells Info it is okay to garbage collect the
text of the nodes of a file which was compressed on disk.

@item ISO-Latin
@cindex ISO Latin characters
@vindex ISO-Latin
When set to @code{On}, Info accepts and displays ISO Latin characters.
By default, Info assumes an ASCII character set.  @code{ISO-Latin} tells
Info that it is running in an environment where the European standard
character set is in use, and allows you to input such characters to
Info, as well as display them.

@anchor{scroll-behavior}
@item scroll-behavior
@itemx scroll-behaviour
@vindex scroll-behavior
@vindex scroll-behaviour
Control what happens when forward scrolling is requested at the end of
a node, or when backward scrolling is requested at the beginning of a
node.  The default value for this variable is @code{Continuous}.
There are three possible values for this variable:

@table @code
@item Continuous
Try to get the first item in this node's menu, or failing that, the
@samp{Next} node, or failing that, the @samp{Next} of the @samp{Up}.
This behavior is identical to using the @samp{]}
(@code{global-next-node}) and @samp{[} (@code{global-prev-node})
commands.

@item scroll-last-node
@vindex scroll-last-node
Control what happens when a scrolling command is issued at the end of
the last node. Possible values are:

@table @code
@item Stop
Do not scroll. Display the @samp{No more nodes within this document.}
message. This is the default.

@item Scroll
Scroll as usual. Since the last node is usually an index, this means
that the very first node from the menu will be selected.

@item Top
Go to the top node of this document.
@end table

This variable is in effect only if @code{scroll-behaviour} is set to
@code{Continuous}.

Notice that the default behavior has changed in version 4.12. Previous
versions behaved as if @code{scroll-last-node=Scroll} was set. This
behavior was counter-intuitive, therefore since version 4.12 the
default is to stop at the last node.

@item Next Only
Only try to get the @samp{Next} node.

@item Page Only
Simply give up, changing nothing.  If @code{scroll-behavior} is
@code{Page Only}, no scrolling command can change the node that is being
viewed.
@end table

This variable normally affects only scrolling commands.
@xref{cursor-movement-scrolls}, for information on how to widen its
scope.

The two names, @code{scroll-behavior} and @code{scroll-behaviour}, are
a historical accident.  They are merely synonyms.

@item scroll-step
@vindex scroll-step
The number of lines to scroll when the cursor moves out of the window.
Scrolling happens automatically if the cursor has moved out of the
visible portion of the node text when it is time to display.  Usually
the scrolling is done so as to put the cursor on the center line of the
current window.  However, if the variable @code{scroll-step} has a
nonzero value, Info attempts to scroll the node text by that many lines;
if that is enough to bring the cursor back into the window, that is what
is done.  The default value of this variable is 0, thus placing the
cursor (and the text it is attached to) in the center of the window.
Setting this variable to 1 causes a kind of ``smooth scrolling'' which
some people prefer.

@item show-index-match
@vindex show-index-match
When set to @code{On}, the portion of the matched search string is
highlighted in the message which explains where the matched search
string was found.  The default value of this variable is @code{On}.
When Info displays the location where an index match was found,
(@pxref{Searching Commands, , @code{next-index-match}}), the portion of the
string that you had typed is highlighted by displaying it in the inverse
case from its surrounding characters.

@item visible-bell
@vindex visible-bell
When set to @code{On}, GNU Info attempts to flash the screen instead of
ringing the bell.  This variable is @code{Off} by default.  Of course,
Info can only flash the screen if the terminal allows it; in the case
that the terminal does not allow it, the setting of this variable has no
effect.  However, you can make Info perform quietly by setting the
@code{errors-ring-bell} variable to @code{Off}.

@end table


@node Custom Key Bindings
@chapter Customizing Key Bindings and Variables

@cindex default key bindings, overriding
@cindex overriding default key bindings
@cindex customizing key bindings
@cindex key bindings, customizing
@cindex infokey
@cindex .info
@cindex .infokey
@cindex _info file (MS-DOS)

GNU Info provides a way to define arbitrary key-to-command bindings
and variable settings, overriding the defaults described in this
document.  (The @option{--vi-keys} option rebinds many keys at once;
@pxref{--vi-keys}.)

On startup, GNU Info looks for a configuration file in the invoker's
HOME directory called @file{.info}@footnote{Due to the limitations of
DOS filesystems, the MS-DOS version of Info looks for a file
@file{_info} instead.  If the @env{HOME} variable is not defined, Info
additionally looks in the current directory.}.  If it is present, and
appears to contain Info configuration data, and was created with the
current version of the @code{infokey} command, then Info adopts the
key bindings and variable settings contained therein.

The @file{.info} file contains compact, non-textual data for reasons of
efficiency and because its design was lifted wholesale from the GNU Less
program, which also does it that way.  It must be created by compiling a
textual source file using the @code{infokey} command.

@menu
* Invoking infokey::
* infokey source format::
@end menu


@node Invoking infokey
@section Invoking @command{infokey}

@cindex invoking infokey
@cindex infokey, invoking
@cindex _infokey file (MS-DOS)

@command{infokey} compiles a source file
(@file{$HOME/.infokey}@footnote{This file is named @file{_infokey} in
the MS-DOS version, and is looked for in the current directory if
@env{HOME} is undefined.} by default) containing Info customizations
into a binary format (@file{$HOME/.info} by default).  GNU Info reads
the binary file at startup to override the default key bindings and
variable definitions.  Synopsis:

@example
infokey [@var{option}@dots{}] [@var{input-file}]
@end example

Besides the standard @option{--help} and @option{--version}, the only
option is @option{--output @var{file}}.  This tells @command{infokey} to
write the binary data to @var{file} instead of @file{$HOME/.info}.


@node infokey source format
@section @command{infokey} source format

@cindex infokey source format
@cindex .infokey source format
@cindex format of .infokey source

The format of the source file read by @command{infokey} is most easily
illustrated by example.  For instance, here is a sample @file{.infokey}
source file suitable for aficionados of @command{vi} or @command{less}:

@example
#info
j       next-line
k       prev-line
l       forward-char
h       backward-char
\kd     next-line
\ku     prev-line
\kr     forward-char
\kl     backward-char
\       scroll-forward
\kD     scroll-forward-page-only
b       scroll-backward
\kU     scroll-backward-page-only
g       beginning-of-node
\kh     beginning-of-node
G       end-of-node
\ke     end-of-node
\t      select-reference-this-line
-       history-node
n       next-node
p       prev-node
u       up-node
t       top-node
d       dir-node
#var
scroll-step=1
@end example

The source file consists of one or more @dfn{sections}.
Each section starts with a line that identifies the type of section.
Possible sections are:

@table @code
@item #info
Key bindings for Info windows.
The start of this section is indicated by a line containing just
@code{#info} by itself.  If this is the first section in the source
file, the @code{#info} line can be omitted.  The rest of this section
consists of lines of the form:

@example
@var{string} whitespace @var{action} [ whitespace [ # comment ] ] newline
@end example

Whitespace is any sequence of one or more spaces and/or tabs.  Comment
is any sequence of any characters, excluding newline.  @var{string} is
the key sequence which invokes the action.  @var{action} is the name of
an Info command.  The characters in @var{string} are interpreted
literally or prefixed by a caret (@code{^}) to indicate a control
character.  A backslash followed by certain characters specifies input
keystrokes as follows:

@table @code
@item \b
Backspace
@item \e
Escape (ESC)
@item \n
Newline
@item \r
Return
@item \t
Tab
@item \ku
Up arrow
@item \kd
Down arrow
@item \kl
Left arrow
@item \kr
Right arrow
@item \kU
Page Up
@item \kD
Page Down
@item \kh
HOME
@item \ke
END
@item \kx
Delete (DEL)
@item \m@var{x}
Meta-@var{x} where @var{x} is any character as described above.
@end table

Backslash followed by any other character indicates that character is to
be taken literally.  Characters which must be preceded by a backslash
include caret, space, tab, and backslash itself.

@item #echo-area
Key bindings for the echo area.
The start of this section is indicated by a line containing just
@code{#echo-area} by itself.  The rest of this section has a syntax
identical to that for the key definitions for the Info area, described
above.

@item #var
Variable initializations.
The start of this section is indicated by a line containing just
@code{#var} by itself.  Following this line is a list of variable
assignments, one per line.  Each line consists of a variable name
(@xref{Variables},) followed by @code{=} followed by a value.
There may be no white space between the variable name and the @code{=},
and all characters following the @code{=}, including white space,
are included in the value.
@end table

Blank lines and lines starting with @code{#} are ignored, except for
the special section header lines.

Key bindings defined in the @file{.info} file take precedence over GNU
Info's default key bindings, whether or not @samp{--vi-keys} is used.  A
default key binding may be disabled by overriding it in the @file{.info}
file with the action @code{invalid}.  In addition, @emph{all} default
key bindings can be disabled by adding this line @emph{anywhere} in the
relevant section:

@example
#stop
@end example

This will cause GNU Info to ignore all the default key commands for that
section.

Beware: @code{#stop} can be dangerous.  Since it disables all default
key bindings, you must supply enough new key bindings to enable all
necessary actions.  Failure to bind any key to the @code{quit} command,
for example, can lead to frustration.

The order in which key bindings are defined in the @file{.info} file is
not important, except that the command summary produced by the
@code{get-help-window} command only displays the @emph{first} key that
is bound to each command.


@c the following is incomplete
@ignore
@c node Info for Sys Admins
@c chapter Info for System Administrators

This text describes some common ways of setting up an Info hierarchy
from scratch, and details the various options that are available when
installing Info.  This text is designed for the person who is installing
GNU Info on the system; although users may find the information present
in this section interesting, none of it is vital to understanding how to
use GNU Info.

@menu
* Setting the INFOPATH::        Where are my Info files kept?
* Editing the DIR node::        What goes in `DIR', and why?
* Storing Info files::          Alternate formats allow flexibility in setups.
* Using `localdir'::            Building DIR on the fly.
* Example setups::              Some common ways to organize Info files.
@end menu

@c node Setting the INFOPATH
@c section Setting the INFOPATH

Where are my Info files kept?

@c node Editing the DIR node
@c section Editing the DIR node

What goes in `DIR', and why?

@c node Storing Info files
@c section Storing Info files

Alternate formats allow flexibility in setups.

@c node Using `localdir'
@c section Using `localdir'

Building DIR on the fly.

@c node Example setups
@c section Example setups

Some common ways to organize Info files.
@end ignore


@node Index
@appendix Index

@printindex cp

@bye