.Dd August 11, 2007
.Dt ED 1
.Os
.Sh NAME
.Nm ed ,
.Nm red
.Nd text editor
.Sh SYNOPSIS
.Nm
.Op Fl
.Op Fl EGhrsvV
.Op Fl f Ar script
.Op Fl p Ar string
.Op Ar
.Nm red
.Op Fl
.Op Fl EGhrsvV
.Op Fl f Ar script
.Op Fl p Ar string
.Op Ar
.Sh DESCRIPTION
The
.Nm
utility is a line-oriented text editor.
It is used to create, display, modify and
otherwise manipulate text files.
When invoked as
.Nm red ,
the editor is
.Qq restricted
to editing only regular files in the working directory.
.Pp
If invoked with a
.Ar file
argument, then a copy of
.Ar file
is read into the editor's buffer.
Changes are made to this copy and not directly to
.Ar file
itself.
Upon quitting
.Nm ,
any changes not explicitly saved with a
.Cm w
command are lost.
.Pp
Editing is done in two distinct modes:
.Em command
and
.Em input .
When first invoked,
.Nm
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
.Pp
.Sm off
.D1 Cm ,s/ Ar old Cm \&/ Ar new Cm \&/g
.Sm on
.Pp
which replaces all occurrences of the string
.Ar old
with
.Ar new .
.Pp
When an input command, such as
.Cm a
(append),
.Cm i
(insert) or
.Cm c
(change), is given,
.Nm
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
.Em line
consists of text up to and including a
.Em newline
(aka end-of-line) character.
Input mode is terminated by entering a single period
.Pq Cm \&.
on a line.
.Pp
All
.Nm
commands operate on whole lines or ranges of lines; e.g.\&,
the
.Cm d
command deletes lines; the
.Cm m
command moves lines, and so on.
It is possible to modify only part of a line
by means of replacement, as in the example above.
However even here, the
.Cm s
command is applied to whole lines at a time.
.Pp
In general,
.Nm
commands consist of zero or more line addresses,
followed by a single character command and
possibly additional parameters; i.e.\&,
commands have the structure:
.Pp
.Sm off
.D1 Oo Ar address Oo , Ar address Oc Oc Ar command Op Ar parameters
.Sm on
.Pp
The
.Ar address Ns No (es)
indicate the line or range of lines
to be affected by
.Ar command .
If fewer addresses are given than
.Ar command
accepts, then default addresses are supplied.
.Sh OPTIONS
The following command-line options are recognized:
.Bl -tag -width indent
.  It Fl E
Enable extended regular expression syntax.
By default,
.Nm
interprets patterns using basic regular expression syntax.
.  It Fl f Ar script
Read editing commands from file
.Ar script
instead of standard input.
Implies
.Fl s .
.It Fl G
Enable backward compatibility.
.It Fl h
Display help, then exit.
.It Fl p Ar string
Prompt for commands with
.Ar string .
The prompt can be toggled on and off with the
.Cm P
command.
.It Fl r
Same as
.Fl E .
.It Fl s
Suppress interactive diagnostics.
Use this option if
.Nm Ns \&'s
standard input is from a script.
.It Fl v
Enable verbose error diagnostics.
Verbose mode can be toggled on and off with the
.Cm H
command.
.It Fl V
Print version information, then exit.
.It Ar file
If
.Ar file
is given, open it for editing.
If
.Ar file
is prefixed with
.Cm \&!
(exclamation mark), then execute the (string)
.Ar file
as a shell command,
and read its standard output for editing.
See
.Xr sh 1 .
To open a file whose name begins with
.Cm \&! ,
prefix the name with a path separator.
.El
.Sh LINE ADDRESSING
A
.Em line address
is an expression which
represents a line number in the buffer.
.Nm
maintains a
.Em 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.
.Pp
A line address is constructed from one of the expressions in the list below,
optionally followed by a numeric offset.
The offset may include any combination of digits, operators (i.e.\&,
.Cm \&+ ,
.Cm \&-
and
.Cm ^ )
and whitespace.
Addresses are read from left to right, and their values are computed relative
to the current address.
.Pp
One exception to the rule that addresses represent line numbers is the address
.Cm 0
(zero).
This means
.Qq before the first line ,
and is legal wherever it makes sense.
.Pp
An
.Em address range
is two addresses separated either by a
.Cm \&,
(comma) or
.Cm \&;
(semi-colon).
The value of the first address in a range cannot exceed the value of the second.
Each address in a comma-delimited range is interpreted relative to the current
address.
In a semicolon-delimited range, the semicolon acts as an operator which sets the
current address to the address immediately preceding the semicolon.
Subsequent addresses are interpreted relative to the new current address.
.Pp
If only the first address in a range is given,
then the second address is set to the given address.
If only the second address in a comma-delimited range is given,
then the first address is set to
.Cm 1 .
If only the second address in a semicolon-delimited range is given,
then the first address is set to the current address.
.Pp
If an
.Ar n Ns \&-tuple
of addresses is given where
.Ar n  No \&> 2 ,
then the corresponding (two-address) range is
set to the last two addresses
after evaluating each address of the
.Ar n Ns \&-tuple
in sequence.
If only one address is expected, then the last address is used.
Intermediate addresses can be either negative or greater than the last address
in the buffer as long as the resulting address range is valid.
.Pp
The following expressions can be used to form an address:
.Bl -tag -width indent
.It Cm \&.
The current line (address) in the buffer.
.It Cm $
The last line in the buffer.
.It Ar n
As an absolute address, the
.Ar n Ns th
line in the buffer, where
.Ar n
is a number in the interval
.Sm off
.No \&[0 , Cm $ No \&] .
.Sm on
As an offset, the
.Ar n Ns No th
next line.
For example,
.Cm \&.5
represents the fifth line after the current address.
.It Cm \&- No or Xo
.     Cm ^
.  Xc
The previous line.
This is equivalent to
.Cm \&-1 .
.
.Sm off
.It Cm \&- Ar n No \ or\ \& Xo
.Cm ^ Ar n
.Xc
.Sm on
The
.Ar n Ns th
previous line, where
.Ar n
is a non-negative number.
.It Cm \&+
The next line.
This is equivalent to
.Cm \&+1 .
.It Cm \&+ Ns Ar n No or Xo
.Em whitespace Ar n
.Xc
The
.Ar n Ns th
next line, where
.Ar n
is a non-negative number.
.Em Whitespace
followed by a number
.Ar n
is interpreted as
.Cm \&+ Ns Ar n .
.It Cm \&, No or Xo
.Cm \&%
.Xc
The first through last lines in the buffer.
This is equivalent to the address range
.Cm 1,$ .
.It Cm \&;
The current through last lines in the buffer.
This is equivalent to the address range
.Cm \&.,$ .
.Sm off
.It Cm \&/ Ar re Cm \&/
.Sm on
The next line containing the regular expression
.Ar re .
The search wraps to the beginning of the buffer and continues down to
the current line, if necessary.
.Cm \&/\&/
repeats the last search.
.Sm off
.It Cm \&? Ar re Cm \&?
.Sm on
The previous line containing the regular expression
.Ar re .
The search wraps to the end of the buffer and continues up to the
current line, if necessary.
.Cm ??
repeats the last search.
.It Cm \&' Ns Ar lc
The line previously marked by a
.Cm k
(mark) command, where
.Ar lc
is a lower case letter.
.El
.Sh BASIC REGULAR EXPRESSIONS
.Em Regular expressions
are patterns used in selecting text.
For example, the
.Nm
command:
.Bd -literal
.Sm off
.Cm g\&/ Ar string Cm \&/
.Sm on
.Ed
.Pp
prints all lines containing
.Ar string .
Regular expressions are also
used by the
.Cm s
command for selecting old text to be replaced with new.
.Pp
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.
.Pp
The following patterns can be used in basic regular expressions
.Pq Tn BRE Ns s :
.Bl -tag -width indent
.It Cm \&.
Matches any single character except for
.Em newline .
Compare to definition of
.Cm \&.
in
.Sx "LINE ADDRESSING"
(above).
.It Ar c
Any character
.Ar c ,
except for
.Cm \&. ,
.Cm \&* ,
.Cm \&[ ,
.Cm \&^ ,
.Cm \&$ ,
and
.Cm \e ,
matches itself.
.It Cm \e Ns Ar c
Any
.Cm \e
(backslash) escaped character,
.Ar c ,
except for
.Cm \&{ ,
.Cm \&} ,
.Cm \&(
and
.Cm \&)
matches itself.
.Sm off
.It Cm \&[ Ar char-class Cm \&]
.Sm on
Matches any single character in the string
.Em char-class ,
except for
.Em newline .
A pattern of the form
.Sm off
.Em x Cm \&- Em y
.Sm on
within
.Em char-class
such that
.Bk -words
.Em x No \&<= Em y
.Ek
represents the range of characters between the two endpoints, inclusive.
If
.Cm \&-
appears as either the first or last character of
.Ar char-class ,
then it matches itself.
To include a
.Cm \&]
(right bracket), it must be the first character.
All other characters in
.Ar char-class
match themselves.
.Pp
The following literal expressions can also be used in
.Ar char-class
to match sets of characters:
.Pp
.Bl -column "[:alnum:]" "[:cntrl:]" "[:lower:]" "[:xdigit:]" -offset indent -compact
.It Cm [:alnum:] Ta Cm [:cntrl:] Ta Cm [:lower:] Ta Cm [:space:]
.It Cm [:alpha:] Ta Cm [:digit:] Ta Cm [:print:] Ta Cm [:upper:]
.It Cm [:blank:] Ta Cm [:graph:] Ta Cm [:punct:] Ta Cm [:xdigit:]
.  El
.Pp
Patterns in
.Ar char-class
of the form:
.Pp
.Sm off
.Cm [\&. Ar col-elm Cm \&.\&]
.Sm on
.Pp
where
.Ar col-elm
is a
.Em "collating element"
.Pq e.g.\&, a multibyte character
are interpreted according to the current locale settings and match a single
localized character, which may be used as the endpoint of a character range.
.Pp
Patterns in
.Ar char-class
of the form:
.Pp
.Sm off
.Cm [\&= Ns Ar col-elm Ns Cm =\&]
.Sm on
.Pp
represent an
.Em  equivalence class
of similar characters.
For example, the equivalence class
.Cm [=a=]
in a Latin locale matches all accented variations of the letter
.Cm a .
.Pp
See
.Xr regex 3
for further explanation of these constructs.
.Sm off
.It Cm \&[^ Ar char-class Cm \&]
.Sm on
Matches any single character, except for
.Em newline ,
not in
.Ar char-class
(as defined above).
.It Cm ^
If
.Cm ^
is the first character of a regular expression, then it anchors the regular
expression to the beginning of a line.
Otherwise, it matches itself.
Compare to definition of
.Cm ^
in
.Sx "LINE ADDRESSING"
(above).
.It Cm $
If
.Cm $
is the last character of a regular expression, then it anchors the regular
expression to the end of a line.
Otherwise, it matches itself.
Compare to definition of
.Cm $
in
.Sx "LINE ADDRESSING"
(above).
.Sm off
.It Cm \e( Ar re Cm \e)
.Sm on
Defines a (possibly null) subexpression
.Ar re .
A subsequent backreference of the form
.Cm \e Ns Ar n ,
where
.Ar n
is a number in the interval [1,9], expands to the text matched by the
.Ar n Ns th
subexpression.
For example, the regular expression
.Cm \e(a.c\e)\e1
matches the string
.Cm abcabc ,
but not
.Cm abcadc .
Subexpressions may be nested and
are ordered relative to their left delimiter.
.  It Cm \&*
Matches the single character regular expression or subexpression
immediately preceding it zero or more times.
To match at least once, repeat the pattern before applying
.Cm \&* .
For example, the regular expression
.Pp
.D1 Cm [[:digit:]][[:digit:]]*\e(\e.[[:digit:]]*\e)*
.Pp
matches a decimal number with at least one integral digit.
.Sm off
.It Cm \e{ Ar n Cm \&, Ar m Cm \e} No \ or \ \& Xo
.Cm \e{ Ar n Cm ,\e} No \ or\ \&
.Cm \e{ Ar n Cm \e}
.Xc
.Sm on
Matches the single character regular expression or subexpression
immediately preceding it at least
.Ar n
and at most
.Ar m
times.
If
.Ar m
is omitted, then it matches at least
.Ar n
times.
If the comma is omitted as well, then it matches exactly
.Ar n
times.
The previous example for matching a decimal number can now be written more
precisely as:
.Pp
.D1 Cm [[:digit:]]\e{1,\e}\e(\e.[[:digit:]]\e{0,\e}\e)\e{0,1\e}
.Pp
It is illegal for any of these forms to occur first in
a regular expression or subexpression.
.El
.Sh EXTENDED REGULAR EXPRESSIONS
If
.Nm
is invoked with either the
.Xo
.Fl E No or
.Fl r
.Xc
option, then patterns are expected in
.Em extended regular expression
.Pq Tn ERE
format, which adds three new operators to
.Tn BRE Ns s:
.Pp
.Cm \&? \&+ No and Cm \&|
.Pp
Only the
.Cm \&|
(vertical bar) operator has no
.Tn BRE
equivalent.
It allows searching for one of multiple patterns concurrently.
For example, the
.Nm
command
.Pp
.Sm off
.Cm g/ Ar string1 Xo
.Cm \&| Ar string2
.Cm \&/
.Xc
.Sm on
.Pp
prints all lines matching either
.Ar string1
or
.Ar string2 .
.Pp
The major difference in syntax between
.Tn BRE Ns s
and
.Tn ERE Ns s
is that
.Tn ERE
operators for subexpressions and repetitions are not prefixed with
.Cm \e .
For example, the
.Tn BRE
.Cm ed\e(it \e)\e{2\e}
is equivalent to the
.Tn ERE
.Bk -words
.Cm ed(it ){2} .
.Ek
.Pp
.Tn ERE Ns s
are easy to understand in terms of their differences with
.Tn BRE Ns s
(as defined in
.Sx "BASIC REGULAR EXPRESSIONS" ,
above) as follows.
.Bl -tag -width indent
.It Cm \&.
See
.Tn  BRE
definition.
.It Ar c
See
.Tn  BRE
definition.
Additional
.Tn Pf ERE \&-specific
operators are:
.Pp
.Cm \&? \&+ \&( \&{ \&|
.It Cm \e Ns Ar c
Any
.Cm \e
(backslash) escaped character,
.Ar c ,
matches itself.
Compare with
.Tn BRE
definition.
.Sm off
.It Cm \&[ Ar char-class Cm \&]
.Sm on
See
.Tn BRE
definition.
.Sm off
.It Cm \&[^ Ar char-class Cm \&]
.Sm on
See
.Tn BRE
definition.
.It Cm ^
See
.Tn BRE
definition.
.It Cm $
See
.Tn BRE
definition.
.Sm off
.It Cm \&( Ar re Cm \&)
.Sm on
See corresponding
.Sm off
.Cm \e( Ar re Cm \e)
.Sm on
.Tn BRE
definition.
.It Cm \&*
See
.Tn BRE
definition.
.Sm off
.It Cm { Ar n , Ar m Cm } No \ or\ \& Xo
.Cm { Ar n Cm ,} No \ or\ \&
.Cm { Ar n Cm }
.Xc
.Sm on
See corresponding
.Sm off
.Cm \e{ Ar n Cm \&, Ar m Cm \e}
.Sm on
.Tn BRE
definition.
.It Cm \&?
Equivalent to
.Cm {0,1} .
.It Cm \&+
Equivalent to
.Cm {1,} .
.It Cm \&|
Separates alternative patterns within a regular expression or subexpression.
The left-most pattern producing the left-most longest match is the one selected.
For the purposes of backreferencing, all subexpressions are counted, whether or
not they occur in the currently selected alternative.
.El
.Sh COMMANDS
All
.Nm
commands are single characters, though some require additional parameters.
If a command's parameters extend over several lines, then each line except for
the last must be terminated with a
.Cm \e
(baskslash).
.Pp
In general, only one command per line is allowed.
However, most commands accept a print suffix, which is any of
.Cm p
(print),
.Cm l
(list),
or
.Cm n
(enumerate),
to print the last line affected by the command.
.Pp
An interrupt (typically
.Sm off
.No \&< Cm CTL \&>\&+ Cm C )
.Sm on
has the effect of aborting the current command and returning the editor to
command mode..
.Pp
.Nm
recognizes the following commands.
The commands are shown together with the default address or address range
supplied if none is specified (in parenthesis).
.Bl -tag -width indent
.It (.) Ns Ic a
Appends text to the buffer after the addressed line.
Text is entered in input mode.
The current address is set to last line entered.
.It (.,.) Ns Ic 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.
.It (.,.) Ns Ic 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.
.It Cm e Ar file
Edits
.Ar file ,
and sets the default filename.
If
.Ar 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.
.It Cm e\ \&! Ns Ar command
Edits the standard output of
.Cm \&! Ns Ar command ,
(see
.Cm \&! Ns Ar command
below).
The default filename is unchanged.
Any lines in the buffer are deleted before the output of
.Ar command
is read.
The current address is set to the last line read.
.It Cm E Ar file
Edits
.Ar file
unconditionally.
This is similar to the
.Cm e
command, except that unwritten changes are discarded without warning.
The current address is set to the last line read.
.It Cm f Ar file
Sets the default filename to
.Ar file .
If
.Ar file
is not specified, then the default unescaped filename is printed.
.Sm off
.It Xo
.Pf ( Cm 1,$ ) Ic g No /
.Ar re No / Ar command-list
.Xc
.Sm on
Applies
.Ar command-list
to each of the addressed lines matching a regular expression
.Ar re .
The current address is set to the line currently matched before
.Ar command-list
is executed.
At the end of the
.Cm g
command, the current address is set to the last line affected by
.Ar command-list .
.Pp
Each command in
.Ar command-list
must be on a separate line, and every line except for the last must be
terminated by a backslash (\e). Any commands are allowed, except for
.Cm G ,
.Cm g ,
.Cm V ,
and
.Cm v .
A null command in
.Ar command-list
is equivalent to a
.Cm p
command.
.Sm off
.It ( Cm 1,$ ) Ic G No / Ar re No /
.Sm on
Interactively edits the addressed lines matching a regular expression
.Ar re .
For each matching line, the line is printed, the current address is
set, and the user is prompted to enter a
.Ar command-list .
At the end of the
.Cm G
command, the current address is set to the last line affected by (the
last)
.Ar command-list .
.Pp
The format of
.Ar command-list
is the same as that of the
.Cm g
command.
A single
.Cm \&&
repeats the last non-null command list.
.It Cm 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.
.It Cm h
Prints an explanation of the last error.
.Sm off
.It No \&( Cm \&. No \&) Xo
.Cm i
.Xc
.Sm on
Inserts text in the buffer before the current line.
Text is entered in input mode.
The current address is set to the last line entered.
.Sm off
.It No \&( Cm \&. , Cm +1 ) Xo
.Cm j
.Xc
.Sm on
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.
.Sm off
.It No \&( Cm \&. No \&) Xo
.Cm k Ar lc
.Xc
.Sm on
Marks the addressed line with a lower case letter
.Ar lc
without changing the current address.
The line can then be addressed as
.Cm \&' Ns Ar lc
(i.e.\&, a single quote followed by
.Ar lc )
in subsequent commands.
The mark is not cleared until the line is deleted or otherwise modified.
.Sm off
.It No \&( Cm \&. , Cm \&. No \&) Xo
.Cm l
.Xc
.Sm on
Prints the addressed lines unambiguously. The current address is set
to the last line printed.
.Sm off
.It No \&( Cm \&. , Cm \&. No \&) Xo
.Cm m No \&( Cm \&. No \&)
.Xc
.Sm on
Moves lines in the buffer.
The addressed lines are moved to after the right-hand destination address,
which may be the address
.Cm 0
(zero). The current address is set to the last line moved.
.Sm off
.It No \&( Cm \&. , Cm \&. No \&) Xo
.Cm n
.Xc
.Sm on
Prints the addressed lines along with their line numbers.
The current address is set to the last line printed.
.  It Cm P
Toggles the command prompt on and off.
Unless a prompt was specified by with command-line option
.Fl p Ar string ,
the command prompt is by default turned off.
.Sm off
.It No \&( Cm \&. , Cm \&. No \&) Xo
.Cm p
.Xc
.Sm on
Prints the addressed lines.
The current address is set to the last line printed.
.  It Cm Q
Quits
.Nm
unconditionally.
This is similar to the
.Cm q
command, except that unwritten changes are discarded without warning.
.It Cm q
Quits
.Nm
if there are no unwritten changes.
Otherwise, a warning diagnostic is printed to the standard output.
.Sm off
.It No \&( Cm $ No \&) Xo
.Cm r\  Ar file
.Xc
.Sm on
Reads
.Ar file
to after the addressed line.
If
.Ar file
is not specified, then the default filename is used.
If there was no default filename prior to the command, then the default
filename is set to
.Ar file .
Otherwise, the default filename is unchanged.
The current address is set to the last line read.
.Sm off
.It No \&( Cm $ No \&) Xo
.Cm r\ \&! Ar command
.Xc
.Sm on
Reads to after the addressed line the standard output of
.Cm \&! Ns Ar command ,
(see the
.Cm \&! Ns Ar command
below). The default filename is unchanged.
The current address is set to the last line read.
.Sm off
.It No \&( Cm \&. , Cm \&. No \&) Xo
.Cm s/ Ar re Cm / Ar replacement Cm / Op Cm g Ar n
.Xc
.Sm on
.\".
.\". It (.,.)s/re/replacement/g
.\".
.\".  It (.,.)s/re/replacement/n
Replaces text in the addressed lines matching a regular expression
.Ar re
with
.Ar replacement .
By default, only the first match in each line is replaced.
If the
.Cm g
(global) suffix is given, then every match to be replaced.
The
.Ar n
suffix, where
.Ar n
is a positive number, causes only the
.Ar n Ns 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.
.Pp
.Ar re
and
.Ar replacement
may be delimited by any character other than space and
.Em newline
(see the
.Cm s
command below).
If one or two of the last delimiters is omitted,
then the last line affected is printed as though the print suffix
.Cm p
were specified.
.Pp
An unescaped
.Cm \&&
in
.Ar replacement
is replaced by the currently matched text.
The character sequence
.Cm \e Ns Ar m ,
where
.Ar m
is a number in the interval [1,9], is replaced by the
.Em m Ns No th
backreference expression of the matched text.
If
.Ar replacement
consists of a single
.Cm \&% ,
then
.Ar replacement
from the last substitution is used.
A
.Em newline
can be embedded in
.Ar replacement
if it is escaped with a backslash (\e).
.Sm off
.It No \&( Cm \&. , Cm \&. No \&) Xo
.Cm s Op Cm rgp Ar n
.Xc
.Sm on
Repeat the last substitution.
This form of the
.Cm s
command accepts any combination of the parameters
.Cm r ,
.Cm g ,
.Cm p ,
and/or a count suffix
.Ar n .
If a count suffix
.Ar n
is given, then only the
.Ar n Ns th
match is replaced.
The
.Cm r
suffix substitutes the regular expression of the last search in place of
that of the last substitution.
The
.Cm g
suffix toggles the global suffix of the last substitution.
The
.Cm p
suffix toggles the print suffix of the last substitution.
Within a non-interactive global command, the
.Cm g
and
.Cm p
suffixes are evaluated only once.
The current address is set to the last line affected.
.Sm off
.It No \&( Cm \&. , Cm \&. No \&) Xo
.Cm t No \&( Cm \&. No \&)
.Xc
.Sm on
Copies (or transfers) the addressed lines to after the right-hand
destination address, which may be the address
.Cm 0
(zero).
The current address is set to the last line copied.
.It Cm u
Undoes the last command and restores the current address to what it was
before the command.
The global commands
.Cm g ,
.Cm G ,
.Cm v ,
and
.Cm V .
are treated as a single command by undo.
.Cm u
is its own inverse.
.Sm off
.It No \&( Cm 1 , Cm $ No \&) Xo
.Cm v\&/ Ar re Cm / Ar command-list
.Xc
.Sm on
Applies
.Ar command-list
to each of the addressed lines not matching a regular expression
.Ar re .
This is similar to the
.Cm g
command.
.Sm off
.It No \&( Cm 1 , Cm $ No \&) Xo
.Cm V\&/ Ar re Cm \&/
.Xc
.Sm on
Interactively edits the addressed lines not matching a regular expression
.Ar re .
This is similar to the
.Cm G
command.
.Sm off
.It No \&( Cm 1 , Cm $ No \&) Xo
.Cm W\  Ar file
.Xc
.Sm on
Appends the addressed lines to the end of
.Ar file .
This is similar to the
.Cm w
command, expect that the previous contents of file is not clobbered.
The current address is unchanged.
.Sm off
.It No \&( Cm 1 , Cm $ \&) Xo
.Cm w\  Ar file
.Xc
.Sm on
Writes the addressed lines to
.Ar file .
Any previous contents of
.Ar file
are lost without warning.
If there is no default filename, then the default filename is set to
.Ar file ,
otherwise it is unchanged.
If no filename is specified, then the default filename is used.
The current address is unchanged.
.Sm off
.It \&( Cm 1 , Cm $ \&) Xo
.Cm wq\  Ar file
.Xc
.Sm on
Write the addressed lines to
.Ar file ,
and then executes a
.Em q
command.
.Sm off
.It ( Cm 1 , Cm $ ) Xo
.Cm w\ \&! Ar command
.Xc
.Sm on
Writes the addressed lines to the standard input of
.Cm \&! Ns Ar command ,
(see the
.Cm \&! Ns Ar command
below). The default filename and current address are unchanged.
.Sm off
.It No \&( Cm \&.\&+1 No \&) Xo
.Cm z Ar n
.Xc
.Sm on
Scrolls
.Ar n
lines at a time starting at addressed line.
If
.Ar n
is not specified, then the current window size is used.
The current address is set to the last line printed.
.It Cm \&! Ns Ar command
Executes
.Ar command
via
.Xr sh 1 .
If the first character of
.Ar command
is
.Ql \&! ,
then it is replaced by text of the previous
.Cm \&! Ns Ar command .
The
.Nm
utility does not process
.Ar command
for backslash (\e) escapes.
However, an unescaped
.Em \&%
is replaced by the default filename.
When the shell returns from execution,
.Ql \&!
is printed to the standard output.
The current line is unchanged.
.Sm off
.It No \&( Cm $ No \&) Cm =
.Sm on
Prints the line number of the addressed line.
.Sm off
.It No \&( Cm .+1 ) Xo
.No < Cm RET No >
.Xc
.Sm on
Prints the addressed line, and sets the current address to
that line.
.El
.Sh FILES
.Bl -tag -width /tmp/ed.* -compact
.It \&/tmp/ed.*
buffer file
.It ed.hup
the file to which
.Nm
attempts to write the buffer if the terminal hangs up
.El
.Sh DIAGNOSTICS
When an error occurs,
.Nm
prints a
.Ql \&?
and either returns to command mode or exits if its input is from a script.
An explanation of the last error can be printed with the
.Em h
(help) command.
.Pp
Since the
.Em g
(global) command masks any errors from failed searches and substitutions,
it can be used to perform conditional operations in scripts; e.g.\&,
.Pp
.Sm off
.Cm g No \&/ Em old Xo
.No \&/ Cm s
.No \&// Em new
.No \&/
.Xc
.Sm on
.Pp
replaces any occurrences of
.Em old
with
.Em new .
If the
.Em u
(undo) command occurs in a global command list, then the command list
is executed only once.
.Pp
If diagnostics are not disabled, attempting to quit
.Nm
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.
.Sh SEE ALSO
.Xr bdes 1 ,
.Xr sed 1 ,
.Xr sh 1 ,
.Xr vi 1 ,
.Xr regex 3
.Pp
USD:12-13
.Rs
.%A B. W. Kernighan
.%A P. J. Plauger
.%B Software Tools in Pascal
.%O Addison-Wesley
.%D 1981
.Re
.Sh LIMITATIONS
When processing a file name, backslash escape processing is
limited to protecting an initial exclamation mark (!).
Pp
If a text (non-binary) file is not terminated by an
.Em end-of-line
character, then
.Nm
appends one on reading/writing it.
In the case of a binary file,
.Nm
does not append an
.Em end-of-line
on reading/writing.
.Pp
per line overhead: 4 ints
.Sh HISTORY
An
.Nm
command appeared in
Version 1 AT&T UNIX.
.Sh BUGS
The
.Nm
utility does not recognize multibyte characters.