\input texinfo @c -*-texinfo-*-
@c %**start of header
@setfilename ed.info
@settitle GNU @command{ed} Manual
@finalout
@c %**end of header

@set UPDATED 18 July 2007
@set VERSION 0.7

@dircategory Basics
@direntry
* Ed: (ed).                     The GNU line editor.
@end direntry

@copying
Copyright @copyright{} 1993, 2006, 2007 Free Software Foundation, Inc.

Permission is granted to copy, distribute and/or modify this document
under the terms of the GNU Free Documentation License, Version 1.2 or
any later version published by the Free Software Foundation; with no
Invariant Sections, no Front-Cover Texts, and no Back-Cover Texts.
@end copying

@titlepage
@title GNU ed
@subtitle The GNU line editor
@subtitle for GNU ed version @value{VERSION}, @value{UPDATED}
@author by Andrew L. Moore and Antonio Diaz Diaz
@page
@vskip 0pt plus 1filll
@insertcopying
@end titlepage

@contents

@ifnottex
@node Top
@top The GNU ed line editor

This manual is for GNU ed (version @value{VERSION}, @value{UPDATED}).

@sp 1
GNU ed is an 8-bit clean, more or less POSIX-compliant implementation of
the standard Unix line editor. These days, full-screen editors have
rendered @command{ed} mostly of historical interest. Nonetheless, it
appeals to a handful of aging programmers who still believe that ``Small
is Beautiful''.
@end ifnottex

@menu
* Overview::                      Overview of the @command{ed} command
* Introduction to Line Editing::  Getting started with GNU @command{ed}
* Invoking Ed::                   Command line interface
* Line Addressing::               Specifying lines/ranges in the buffer
* Regular Expressions::           Patterns for selecting text
* Commands::                      Commands recognized by GNU @command{ed}
* Limitations::                   Intrinsic limits of GNU @command{ed}
* Diagnostics::                   GNU @command{ed} error handling

@end menu

@sp 1
@insertcopying


@node Overview
@chapter Overview

@command{ed} is a line-oriented text editor. It is used to create,
display, modify and otherwise manipulate text files. @command{red} is a
restricted @command{ed}: it can only edit files in the current directory
and cannot execute shell commands.

If invoked with a @var{file} argument, then a copy of @var{file} is read
into the editor's buffer. Changes are made to this copy and not directly
to @var{file} itself. Upon quitting @command{ed}, any changes not
explicitly saved with a @samp{w} command are lost.

Editing is done in two distinct modes: @dfn{command} and @dfn{input}.
When first invoked, @command{ed} is in command mode. In this mode
commands are read from the standard input and executed to manipulate the
contents of the editor buffer. A typical command might look like:

@example
,s/@var{old}/@var{new}/g
@end example

which replaces all occurences of the string @var{old} with @var{new}.

When an input command, such as @samp{a} (append), @samp{i} (insert) or
@samp{c} (change), is given, @command{ed} enters input mode. This is the
primary means of adding text to a file. In this mode, no commands are
available; instead, the standard input is written directly to the editor
buffer. A @dfn{line} consists of the text up to and including a
@key{newline} character. Input mode is terminated by entering a single
period (@samp{.}) on a line.

All @command{ed} commands operate on whole lines or ranges of lines;
e.g., the @samp{d} command deletes lines; the @samp{m} command moves
lines, and so on. It is possible to modify only a portion of a line by
means of replacement, as in the example above. However even here, the
@samp{s} command is applied to whole lines at a time.

In general, @command{ed} commands consist of zero or more line
addresses, followed by a single character command and possibly
additional parameters; i.e., commands have the structure:

@example
[@var{address} [,@var{address}]]@var{command}[@var{parameters}]
@end example

The @var{address}es indicate the line or range of lines to be affected
by the command. If fewer addresses are given than the command accepts,
then default addresses are supplied.


@node Introduction to Line Editing
@chapter Introduction to Line Editing

@command{ed} was created, along with the Unix operating system, by Ken
Thompson and Dennis Ritchie. It is the refinement of its more complex,
programmable predecessor, @cite{QED}, to which Thompson and Ritchie had
already added pattern matching capabilities (@pxref{Regular
Expressions}).

For the purposes of this tutorial, a working knowledge of the Unix shell
@command{sh} (@pxref{Bash,,, bash, The GNU Bash Reference Manual}) and
the Unix file system is recommended, since @command{ed} is designed to
interact closely with them.

The principal difference between line editors and display editors is
that display editors provide instant feedback to user commands, whereas
line editors require sometimes lengthy input before any effects are
seen. The advantage of instant feedback, of course, is that if a mistake
is made, it can be corrected immediately, before more damage is done.
Editing in @command{ed} requires more strategy and forethought; but if
you are up to the task, it can be quite efficient.

Much of the @command{ed} command syntax is shared with other Unix
utilities.

As with the shell, @key{RETURN} (the carriage-return key) enters a line
of input. So when we speak of ``entering'' a command or some text in
@command{ed}, @key{RETURN} is implied at the end of each line. Prior to
typing @key{RETURN}, corrections to the line may be made by typing
either @key{BACKSPACE} (sometimes labeled @key{DELETE} or @key{DEL}) to
erase characters backwards, or @key{CONTROL}-u (i.e., hold the CONTROL
key and type u) to erase the whole line.

When @command{ed} first opens, it expects to be told what to do but
doesn't prompt us like the shell. So let's begin by telling @command{ed}
to do so with the @key{P} (@dfn{prompt}) command:

@example
$ ed
P
*
@end example

By default, @command{ed} uses asterisk (@samp{*}) as command prompt to
avoid confusion with the shell command prompt (@samp{$}).

We can run Unix shell (@command{sh}) commands from inside @command{ed}
by prefixing them with @key{!} (exclamation mark, aka ``bang''). For
example:

@example
*!date
Mon Jun 26 10:08:41 PDT 2006
!
*!for s in hello world; do echo $s; done
hello
world
!
*
@end example

So far, this is no different from running commands in the Unix shell.
But let's say we want to edit the output of a command, or save it to a
file. First we must capture the command output to a temporary location
called a @dfn{buffer} where @command{ed} can access it. This is done
with @command{ed}'s @key{r} command (mnemonic: @dfn{read}):

@example
*r !cal
143
*
@end example

Here @command{ed} is telling us that it has just read 139 characters
into the editor buffer - i.e., the output of the @command{cal} command,
which prints a simple ASCII calendar. To display the buffer contents we
issue the @key{p} (@dfn{print}) command (not to be confused with the
prompt command, which is uppercase!). To indicate the range of lines in
the buffer that should be printed, we prefix the command with @key{,}
(comma) which is shorthand for ``the whole buffer'':

@example
*,p
   September 2006
Mo Tu We Th Fr Sa Su 
             1  2  3
 4  5  6  7  8  9 10
11 12 13 14 15 16 17
18 19 20 21 22 23 24
25 26 27 28 29 30

*
@end example

Now let's write the buffer contents to a file named @code{junk} with the
@key{w} (@dfn{write}) command. Again, we use the @key{,} prefix to
indicate that it's the whole buffer we want:

@example
*,w junk
143
*
@end example

Need we say? It's good practice to frequently write the buffer contents,
since unwritten changes to the buffer will be lost when we exit
@command{ed}.

The sample sessions below illustrate some basic concepts of line editing
with @command{ed}. We begin by creating a file, @samp{sonnet}, with some
help from Shakespeare. As with the shell, all input to @command{ed} must
be followed by a @key{newline} character. Comments begin with a @samp{#}.

@example
$ ed
# The `a' command is for appending text to the editor buffer.
a
No more be grieved at that which thou hast done.
Roses have thorns, and filvers foutians mud.
Clouds and eclipses stain both moon and sun,
And loathsome canker lives in sweetest bud.
.
# Entering a single period on a line returns @command{ed} to command mode.
# Now write the buffer to the file @samp{sonnet} and quit:
w sonnet
183
# @command{ed} reports the number of characters written.
q
$ ls -l
total 2
-rw-rw-r--    1 alm           183 Nov 10 01:16 sonnet
$
@end example

In the next example, some typos are corrected in the file @samp{sonnet}.

@example
$ ed sonnet
183
# Begin by printing the buffer to the terminal with the @samp{p} command.
# The `,' means ``all lines.''
,p
No more be grieved at that which thou hast done.
Roses have thorns, and filvers foutians mud.
Clouds and eclipses stain both moon and sun,
And loathsome canker lives in sweetest bud.
# Select line 2 for editing.
2
Roses have thorns, and filvers foutians mud.
# Use the substitute command, @samp{s}, to replace `filvers' with `silver',
# and print the result.
s/filvers/silver/p
Roses have thorns, and silver foutians mud.
# And correct the spelling of `fountains'.
s/utia/untai/p
Roses have thorns, and silver fountains mud.
w sonnet
183
q
$
@end example

Since @command{ed} is line-oriented, we have to tell it which line, or
range of lines we want to edit. In the above example, we do this by
specifying the line's number, or sequence in the buffer. Alternatively,
we could have specified a unique string in the line, e.g.,
@samp{/filvers/}, where the @samp{/}s delimit the string in question.
Subsequent commands affect only the selected line, a.k.a. the
@dfn{current} line. Portions of that line are then replaced with the
substitute command, whose syntax is @samp{s/@var{old}/@var{new}/}.

Although @command{ed} accepts only one command per line, the print
command @samp{p} is an exception, and may be appended to the end of most
commands.

In the next example, a title is added to our sonnet.

@example
$ ed sonnet
183
a
 Sonnet #50
.
,p
No more be grieved at that which thou hast done.
Roses have thorns, and silver fountains mud.
Clouds and eclipses stain both moon and sun,
And loathsome canker lives in sweetest bud.
 Sonnet #50
# The title got appended to the end; we should have used `0a'
# to append ``before the first line.''
# Move the title to its proper place.
5m0p
 Sonnet #50
# The title is now the first line, and the current line has been
# set to this line as well.
,p
 Sonnet #50
No more be grieved at that which thou hast done.
Roses have thorns, and silver fountains mud.
Clouds and eclipses stain both moon and sun,
And loathsome canker lives in sweetest bud.
wq sonnet
195
$
@end example

When @command{ed} opens a file, the current line is initially set to the
last line of that file. Similarly, the move command @samp{m} sets the
current line to the last line moved.

In summary:

Structurally,
Related programs or routines are @command{vi (1)}, @command{sed (1)},
@command{regex (3)},  @command{sh (1)}. Relevant documents
are:

@quotation
Unix User's Manual Supplementary Documents: 12 --- 13
@end quotation

@quotation
B. W. Kernighan and P. J. Plauger: ``Software Tools in Pascal'',
Addison-Wesley, 1981.
@end quotation


@node Invoking Ed
@chapter Invoking Ed

The format for running @command{ed} is:

@example
ed [@var{options}] [@var{file}]
red [@var{options}] [@var{file}]
@end example

@var{file} specifies the name of a file to read. If @var{file} is
prefixed with a bang (!), then it is interpreted as a shell command. In
this case, what is read is the standard output of @var{file} executed
via @command{sh (1)}. To read a file whose name begins with a bang,
prefix the name with a backslash (@kbd{\}). The default filename is set
to @var{file} only if it is not prefixed with a bang.

@command{ed} supports the following options:

@table @code
@item --help
@itemx -h
Print an informative help message describing the options and exit.

@item --version
@itemx -V
Print the version number of @command{ed} on the standard output and exit.

@item --loose-exit-status
@itemx -l
Do not exit with bad status if a command happens to "fail" (for example
if a substitution command finds nothing to replace). This can be useful
when @command{ed} is invoked as the editor for crontab.

@item --prompt=@var{string}
@itemx -p @var{string}
Specifies a command prompt. This may be toggled on and off with the
@samp{P} command.

@item --quiet
@itemx --silent
@itemx -s
Suppresses diagnostics. This should be used if @command{ed}'s standard
input is from a script.

@item --traditional
@itemx -G
Forces backwards compatibility. This affects the behavior of the
@command{ed} commands @samp{G}, @samp{V}, @samp{f}, @samp{l}, @samp{m},
@samp{t} and @samp{!!}. If the default behavior of these commands does
not seem familiar, then try invoking @command{ed} with this switch.

@item --verbose
@itemx -v
Verbose mode. This may be toggled on and off with the @samp{H} command.

@end table


@node Line Addressing
@chapter Line Addressing

An address represents the number of a line in the buffer. @command{ed}
maintains a @dfn{current address} which is typically supplied to
commands as the default address when none is specified. When a file is
first read, the current address is set to the last line of the file. In
general, the current address is set to the last line affected by a
command.

A line address is constructed from one of the bases in the list below,
optionally followed by a numeric offset. The offset may include any
combination of digits, operators (i.e., @samp{+} and @samp{-}) and
whitespace. Addresses are read from left to right, and their values may
be absolute or relative to the current address.

One exception to the rule that addresses represent line numbers is the
address @samp{0} (zero). This means ``before the first line,'' and is
valid wherever it makes sense.

An address range is two addresses separated either by a comma or
semicolon. The value of the first address in a range cannot exceed the
value of the the second. If only one address is given in a range, then
the second address is set to the given address. If an @var{n}-tuple of
addresses is given where @var{n} > 2, then the corresponding range is
determined by the last two addresses in the @var{n}-tuple. If only one
address is expected, then the last address is used.

Each address in a comma-delimited range is interpreted relative to the
current address. In a semicolon-delimited range, the first address is
used to set the current address, and the second address is interpreted
relative to the first.

The following address symbols are recognized.

@table @code
@item .
The current line (address) in the buffer.

@item $
The last line in the buffer.

@item @var{n}
The @var{n}th, line in the buffer where @var{n} is a number in the range
@samp{0,$}.

@item +
The next line. This is equivalent to @samp{+1} and may be repeated with
cumulative effect.

@item -
The previous line. This is equivalent to @samp{-1} and may be repeated
with cumulative effect.

@item +@var{n}
@itemx @var{whitespace} @var{n}
The @var{n}th next line, where @var{n} is a non-negative number.
Whitespace followed by a number @var{n} is interpreted as
@samp{+@var{n}}.

@item -@var{n}
The @var{n}th previous line, where @var{n} is a non-negative number.

@item ,
The first through last lines in the buffer. This is equivalent to the
address range @samp{1,$}.

@item ;
The current through last lines in the buffer. This is equivalent to the
address range @samp{.,$}.

@item /@var{re}/
The next line containing the regular expression @var{re}. The search
wraps to the beginning of the buffer and continues down to the current
line, if necessary. @samp{//} repeats the last search.

@item ?@var{re}?
The previous line containing the regular expression @var{re}. The search
wraps to the end of the buffer and continues up to the current line, if
necessary. @samp{??} repeats the last search.

@item 'x
The apostrophe-x character pair addresses the line previously marked by
a @samp{k} (mark) command, where @samp{x} is a lower case letter from
the portable character set.

@end table


@node Regular Expressions
@chapter Regular Expressions

Regular expressions are patterns used in selecting text. For example,
the @command{ed} command

@example
g/@var{string}/
@end example

@noindent
prints all lines containing @var{string}. Regular expressions are also
used by the @samp{s} command for selecting old text to be replaced with
new text.

In addition to a specifying string literals, regular expressions can
represent classes of strings. Strings thus represented are said to be
matched by the corresponding regular expression. If it is possible for a
regular expression to match several strings in a line, then the
left-most longest match is the one selected.

The following symbols are used in constructing regular expressions:

@table @code

@item @var{c}
Any character @var{c} not listed below, including @samp{@{}, @samp{@}},
@samp{(}, @samp{)}, @samp{<} and @samp{>}, matches itself.

@item \@var{c}
Any backslash-escaped character @var{c}, other than @samp{@{},
`@samp{@}}, @samp{(}, @samp{)}, @samp{<}, @samp{>}, @samp{b}, @samp{B},
@samp{w}, @samp{W}, @samp{+} and @samp{?}, matches itself.

@item .
Matches any single character.

@item [@var{char-class}]
Matches any single character in @var{char-class}. To include a @samp{]}
in @var{char-class}, it must be the first character. A range of
characters may be specified by separating the end characters of the
range with a @samp{-}, e.g., @samp{a-z} specifies the lower case
characters. The following literal expressions can also be used in
@var{char-class} to specify sets of characters:

@example
[:alnum:] [:cntrl:] [:lower:] [:space:]
[:alpha:] [:digit:] [:print:] [:upper:]
[:blank:] [:graph:] [:punct:] [:xdigit:]
@end example

If @samp{-} appears as the first or last character of @var{char-class},
then it matches itself. All other characters in @var{char-class} match
themselves.

Patterns in
@var{char-class}
of the form:
@example
[.@var{col-elm}.]
[=@var{col-elm}=]
@end example

@noindent
where @var{col-elm} is a @dfn{collating element} are interpreted
according to @code{locale (5)}. See
@code{regex (3)} for an explanation of these constructs.

@item [^@var{char-class}]
Matches any single character, other than newline, not in
@var{char-class}.  @var{char-class} is defined as above.

@item ^
If @samp{^} is the first character of a regular expression, then it
anchors the regular expression to the beginning of a line. Otherwise,
it matches itself.

@item $
If @samp{$} is the last character of a regular expression, it anchors
the regular expression to the end of a line. Otherwise, it matches
itself.

@item \(@var{re}\)
Defines a (possibly null) subexpression @var{re}. Subexpressions may be
nested. A subsequent backreference of the form @samp{\@var{n}}, where
@var{n} is a number in the range [1,9], expands to the text matched by
the @var{n}th subexpression. For example, the regular expression
@samp{\(a.c\)\1} matches the string @samp{abcabc}, but not
@samp{abcadc}. Subexpressions are ordered relative to their left
delimiter.

@item *
Matches the single character regular expression or subexpression
immediately preceding it zero or more times. If @samp{*} is the first
character of a regular expression or subexpression, then it matches
itself. The @samp{*} operator sometimes yields unexpected results. For
example, the regular expression @samp{b*} matches the beginning of the
string @samp{abbb}, as opposed to the substring @samp{bbb}, since a null
match is the only left-most match.

@item \@{@var{n},@var{m}\@}
@itemx \@{@var{n},\@}
@itemx \@{@var{n}\@}
Matches the single character regular expression or subexpression
immediately preceding it at least @var{n} and at most @var{m} times. If
@var{m} is omitted, then it matches at least @var{n} times. If the comma
is also omitted, then it matches exactly @var{n} times. If any of these
forms occurs first in a regular expression or subexpression, then it is
interpreted literally (i.e., the regular expression @samp{\@{2\@}}
matches the string @samp{@{2@}}, and so on).

@item \<
@itemx \>
Anchors the single character regular expression or subexpression
immediately following it to the beginning (in the case of @samp{\<}) or
ending (in the case of @samp{\>}) of a @dfn{word}, i.e., in ASCII, a
maximal string of alphanumeric characters, including the underscore (_).

@end table

The following extended operators are preceded by a backslash @samp{\} to
distinguish them from traditional @command{ed} syntax.

@table @code

@item  \`
@itemx \'
Unconditionally matches the beginning @samp{\`} or ending @samp{\'} of a line.

@item \?
Optionally matches the single character regular expression or
subexpression immediately preceding it. For example, the regular
expression @samp{a[bd]\?c} matches the strings @samp{abc}, @samp{adc}
and @samp{ac}. If @samp{\?} occurs at the beginning of a regular
expressions or subexpression, then it matches a literal @samp{?}.

@item \+
Matches the single character regular expression or subexpression
immediately preceding it one or more times. So the regular expression
@samp{a+} is shorthand for @samp{aa*}. If @samp{\+} occurs at the
beginning of a regular expression or subexpression, then it matches a
literal @samp{+}.

@item \b
Matches the beginning or ending (null string) of a word. Thus the
regular expression @samp{\bhello\b} is equivalent to @samp{\<hello\>}.
However, @samp{\b\b} is a valid regular expression whereas @samp{\<\>}
is not.

@item \B
Matches (a null string) inside a word.

@item \w
Matches any character in a word.

@item \W
Matches any character not in a word.

@end table


@node Commands
@chapter Commands

All @command{ed} commands are single characters, though some require
additonal parameters. If a command's parameters extend over several
lines, then each line except for the last must be terminated with a
backslash (@samp{\}).

In general, at most one command is allowed per line. However, most
commands accept a print suffix, which is any of @samp{p} (print),
@samp{l} (list), or @samp{n} (enumerate), to print the last line
affected by the command.

An interrupt (typically @key{Control-C}) has the effect of aborting the
current command and returning the editor to command mode.

@command{ed} recognizes the following commands. The commands are shown
together with the default address or address range supplied if none is
specified (in parenthesis).

@table @code

@item (.)a
Appends text to the buffer after the addressed line, which may be the
address @samp{0} (zero). Text is entered in input mode. The current
address is set to last line entered.

@item (.,.)c
Changes lines in the buffer. The addressed lines are deleted from the
buffer, and text is appended in their place. Text is entered in input
mode. The current address is set to last line entered.

@item (.,.)d
Deletes the addressed lines from the buffer. If there is a line after
the deleted range, then the current address is set to this line.
Otherwise the current address is set to the line before the deleted
range.

@item e @var{file}
Edits @var{file}, and sets the default filename. If @var{file} is not
specified, then the default filename is used. Any lines in the buffer
are deleted before the new file is read. The current address is set to
the last line read.

@item e !@var{command}
Edits the standard output of @samp{!@var{command}}, (see the @samp{!}
command below). The default filename is unchanged. Any lines in the
buffer are deleted before the output of @var{command} is read. The
current address is set to the last line read.

@item E @var{file}
Edits @var{file} unconditionally. This is similar to the @samp{e}
command, except that unwritten changes are discarded without warning.
The current address is set to the last line read.

@item f @var{file}
Sets the default filename to @var{file}. If @var{file} is not specified,
then the default unescaped filename is printed.

@item (1,$)g /@var{re}/@var{command-list}
Global command. Applies @var{command-list} to each of the addressed
lines matching a regular expression @var{re}. The current address is set
to the line currently matched before @var{command-list} is executed. At
the end of the @samp{g} command, the current address is set to the last
line affected by @var{command-list}.

At least the first command of @var{command-list} must appear on the same
line as the @samp{g} command. All lines of a multi-line
@var{command-list} except the last line must be terminated with a
backslash (@samp{\}). Any commands are allowed, except for @samp{g},
@samp{G}, @samp{v}, and @samp{V}. By default, a newline alone in
@var{command-list} is equivalent to a @samp{p} command. If @command{ed}
is invoked with the command-line option @samp{-G}, then a newline in
@var{command-list} is equivalent to a @samp{.+1p} command.

@item (1,$)G /@var{re}/
Interactive global command. Interactively edits the addressed lines
matching a regular expression @var{re}. For each matching line, the line
is printed, the current address is set, and the user is prompted to
enter a @var{command-list}. At the end of the @samp{G} command, the
current address is set to the last line affected by (the last)
@var{command-list}.

The format of @var{command-list} is the same as that of the @samp{g}
command. A newline alone acts as a null command list. A single @samp{&}
repeats the last non-null command list.

@item H
Toggles the printing of error explanations. By default, explanations are
not printed. It is recommended that ed scripts begin with this command
to aid in debugging.

@item h
Prints an explanation of the last error.

@item (.)i
Inserts text in the buffer before the current line. The address @samp{0}
(zero) is valid for this command; it is equivalent to address @samp{1}.
Text is entered in input mode. The current address is set to the last
line entered.

@item (.,.+1)j
Joins the addressed lines. The addressed lines are deleted from the
buffer and replaced by a single line containing their joined text. The
current address is set to the resultant line.

@item (.)kx
Marks a line with a lower case letter @samp{x}. The line can then be
addressed as @samp{'x} (i.e., a single quote followed by @samp{x}) in
subsequent commands. The mark is not cleared until the line is deleted
or otherwise modified.

@item (.,.)l
Prints the addressed lines unambiguously. The end of each line is marked
with a @samp{$}, and every @samp{$} character within the text is printed
with a preceding backslash. The current address is set to the last line
printed.

@item (.,.)m(.)
Moves lines in the buffer. The addressed lines are moved to after the
right-hand destination address, which may be the address @samp{0}
(zero). The current address is set to the last line moved.

@item (.,.)n
Prints the addressed lines, preceding each line by its line number and a
@key{tab}. The current address is set to the last line printed.

@item (.,.)p
Prints the addressed lines. The current address is set to the last line
printed.

@item P
Toggles the command prompt on and off. Unless a prompt is specified with
command-line option @samp{-p}, the command prompt is by default turned
off.

@item q
Quits @command{ed}.

@item Q
Quits @command{ed} unconditionally. This is similar to the @code{q}
command, except that unwritten changes are discarded without warning.

@item ($)r @var{file}
Reads @var{file} to after the addressed line. If @var{file} is not
specified, then the default filename is used. If there is no default
filename prior to the command, then the default filename is set to
@var{file}. Otherwise, the default filename is unchanged. The current
address is set to the last line read.

@item ($)r !@var{command}
Reads to after the addressed line the standard output of
@samp{!command}, (see the @samp{!} command below). The default filename
is unchanged. The current address is set to the last line read.

@item (.,.)s /@var{re}/@var{replacement}/
@itemx (.,.)s /@var{re}/@var{replacement}/g
@itemx (.,.)s /@var{re}/@var{replacement}/@var{n}
Replaces text in the addressed lines matching a regular expression
@var{re} with @var{replacement}. By default, only the first match in
each line is replaced. If the @samp{g} (global) suffix is given, then
every match is replaced. The @var{n} suffix, where @var{n} is a postive
number, causes only the @var{n}th match to be replaced. It is an error
if no substitutions are performed on any of the addressed lines. The
current address is set the last line affected.

@var{re} and @var{replacement} may be delimited by any character other
than @key{space} and @key{newline} (see the @samp{s} command below). If
one or two of the last delimiters is omitted, then the last line
affected is printed as if the print suffix @samp{p} were specified.

An unescaped @samp{&} in @var{replacement} is replaced by the currently
matched text. The character sequence @samp{\@var{m}} where @var{m} is a
number in the range [1,9], is replaced by the @var{m}th backreference
expression of the matched text. If @var{replacement} consists of a
single @samp{%}, then @var{replacement} from the last substitution is
used. Newlines may be embedded in @var{replacement} if they are escaped
with a backslash (@samp{\}).

@item (.,.)s
Repeats the last substitution. This form of the @samp{s} command accepts
a count suffix @var{n}, or any combination of the characters @samp{r},
@samp{g}, and @samp{p}. If a count suffix @var{n} is given, then only
the @var{n}th match is replaced. The @samp{r} suffix causes the regular
expression of the last search to be used instead of the that of the last
substitution. The @samp{g} suffix toggles the global suffix of the last
substitution. The @samp{p} suffix toggles the print suffix of the last
substitution The current address is set to the last line affected.

@item (.,.)t(.)
Copies (i.e., transfers) the addressed lines to after the right-hand
destination address, which may be the address @samp{0} (zero). The
current address is set to the last line copied.

@item u
Undoes the last command and restores the current address to what it was
before the command. The global commands @samp{g}, @samp{G}, @samp{v},
and @samp{V} are treated as a single command by undo. @samp{u} is its
own inverse.

@item (1,$)v /@var{re}/@var{command-list}
This is similar to the @samp{g} command except that it applies
@var{command-list} to each of the addressed lines not matching the
regular expression @var{re}.

@item (1,$)V /@var{re}/
This is similar to the @samp{G} command except that it interactively
edits the addressed lines not matching the regular expression @var{re}.

@item (1,$)w @var{file}
Writes the addressed lines to @var{file}. Any previous contents of
@var{file} is lost without warning. If there is no default filename,
then the default filename is set to @var{file}, otherwise it is
unchanged. If no filename is specified, then the default filename is
used. The current address is unchanged.

@item (1,$)w !@var{command}
Writes the addressed lines to the standard input of
@samp{!@var{command}}, (see the @samp{!} command below). The default
filename and current address are unchanged.

@item (1,$)wq @var{file}
Writes the addressed lines to @var{file}, and then executes a @samp{q}
command.

@item (1,$)W @var{file}
Appends the addressed lines to the end of @var{file}. This is similar to
the @samp{w} command, expect that the previous contents of file is not
clobbered. The current address is unchanged.

@item (.)x
Copies (puts) the contents of the cut buffer to after the addressed
line. The current address is set to the last line copied.

@item (.,.)y
Copies (yanks) the addressed lines to the cut buffer. The cut buffer is
overwritten by subsequent @samp{y}, @samp{s}, @samp{j}, @samp{d}, or
@samp{c} commands. The current address is unchanged.

@item (.+1)z @var{n}
Scrolls @var{n} lines at a time starting at addressed line. If @var{n}
is not specified, then the current window size is used. The current
address is set to the last line printed.

@item !@var{command}
Executes @var{command} via @command{sh (1)}. If the first character of
@var{command} is @samp{!}, then it is replaced by text of the previous
@samp{!@var{command}}. @command{ed} does not process @var{command} for
backslash (@samp{\}) escapes. However, an unescaped @samp{%} is replaced
by the default filename. When the shell returns from execution, a
@samp{!} is printed to the standard output. The current line is
unchanged.

@item (.,.)#
Begins a comment; the rest of the line, up to a newline, is ignored. If
a line address followed by a semicolon is given, then the current
address is set to that address. Otherwise, the current address is
unchanged.

@item ($)=
Prints the line number of the addressed line.

@item (.+1)@key{newline}
An address alone prints the addressed line. A @key{newline} alone is
equivalent to @samp{+1p}. the current address is set to the address of
the printed line.

@end table


@node Limitations
@chapter Limitations

If the terminal hangs up, @command{ed} attempts to write the buffer to
file @file{ed.hup}.

@command{ed} processes @var{file} arguments for backslash escapes, i.e.,
in a filename, any characters preceded by a backslash (@samp{\}) are
interpreted literally.

If a text (non-binary) file is not terminated by a newline character,
then @command{ed} appends one on reading/writing it. In the case of a
binary file, @command{ed} does not append a newline on reading/writing.

Per line overhead: 4 @code{int}s.


@node Diagnostics
@chapter Diagnostics

When an error occurs, if @command{ed}'s input is from a regular file or
here document, then it exits, otherwise it prints a @samp{?} and returns
to command mode. An explanation of the last error can be printed with
the @samp{h} (help) command.

If the @samp{u} (undo) command occurs in a global command list, then the
command list is executed only once.

Attempting to quit @command{ed} or edit another file before writing a
modified buffer results in an error. If the command is entered a second
time, it succeeds, but any changes to the buffer are lost.

@command{ed} exits with 0 if no errors occurred; otherwise >0.

@bye