sed.1

.de L
\&\\f(CW\\$1\\fP
..
.de LR
\&\\f(CW\\$1\\fR\\$2\\f(CW\\$3\\fR\\$4\\f(CW\\$5\\fR\\$6\\fR\\$7\\f(CW\\$8\\fR\\$9
..
.TH SED 1
.CT 1 files editor
.SH NAME
sed \- stream editor
.SH SYNOPSIS
.B sed
[
.B -nb
]
.I script
[
.I file ...
]
.PP
.B sed
[
option ...
] [
.I file ...
]
.SH DESCRIPTION
.I Sed
copies the named
.I files
(standard input default) to the standard output,
edited according to a 
.I script
of commands.
The script may be accumulated from
one or more
.B -e 
and
.B -f
options.
The options are
.TP
.BI  -e " ascript
Append
.I ascript
to the script.
.TP
.BI -f " sfile
Append
.I sfile
to the script.
.TP
.B -n
Suppress default output, giving output only as directed by
.BR p 
in the script.
.TP
.B -b
Strip leading blanks from
.I text
in commands (nonstandard option for interpreting some old scripts).
.PP
A script consists of commands, one per line (with semicolon
equivalent to newline, a common but nonstandard convention).
The form of a command is
.IP
[\fIaddress\fR [\f5,\fI address\fR] ] \fIfunction\fR [\fIargument\fR ...]
.PP
In normal operation
.I sed
cyclically copies a line of input into a
.I pattern space
(unless there is something left after a 
.L D
command),
applies in sequence
all commands whose
.I addresses
select that pattern space,
and at the end of the script copies the pattern space
to the standard output (except under
.BR -n )
and deletes the pattern space.
.PP
An
.I address
is either a decimal number that counts
input lines cumulatively across files, a 
.L $
that
addresses the last line of input, or a context address,
.BI / regular-expression / ,
as in
.IR ed (1)
with the added conventions that
.IP
.L \en
matches an embedded newline and
.BI \e c
for any character
.I c
matches
.I c.
.IP
.BI \e "c ... c",
for any character
.I c,
can play the delimiting role of
.BR / ... / .
.PP
A command line with no addresses selects every pattern space.
.PP
A command line with
one address selects each pattern space that matches the address.
(Address
.L 0
is never matched.)
.PP
A command line with
two addresses selects the inclusive range from the first
pattern space that matches the first address through
the next pattern space that matches
the second, or the end of input.
(If the second address is a number less than or equal
to the line number first selected, only one
line is selected.)
Thereafter the process is repeated, looking again for the
first address.
.PP
The negation function
.L !
complements the set of selected pattern spaces.
.PP
In the following list of functions the
maximum number of permissible addresses
for each function is indicated in parentheses.
.PP
A
.I text
argument consists of one or more lines,
all but the last of which end with 
.L \e
to hide the
newline.
Backslashes in text are treated like backslashes
in the replacement string of an 
.L s
command,
and may be used to protect initial blanks and tabs
against the stripping that is done on
every script line.
.PP
An
.I rfile
or
.I wfile
argument must terminate the command
line and must be preceded by exactly one blank.
Each
.I wfile
is created before processing begins.
.TP
.RB (1) \|a\e
.br
.ns
.TP
.I text
Append.
Place
.I text
on the output before
reading the next input line.
.HP
.RB (2) \|b
.I label
.br
Branch to the 
.B :
command bearing the
.I label.
If
.I label
is empty, branch to the end of the script.
.TP
.RB (2) \|c\e
.br
.ns
.TP
.I text
Change.
Delete the pattern space.
With 0 or 1 address or at the end of a 2-address range, place
.I text
on the output.
Start the next cycle.
.TP
.RB (2) \|d
Delete the pattern space.
Start the next cycle.
.TP
.RB (2) \|D
Delete the initial segment of the
pattern space through the first newline.
Start the next cycle.
.TP
.RB (2) \|g
Replace the contents of the pattern space
by the contents of the hold space.
.TP
.RB (2) \|G
Append the contents of the hold space to the pattern space.
.TP
.RB (2) \|h
Replace the contents of the hold space by the contents of the pattern space.
.TP
.RB (2) \|H
Append the contents of the pattern space to the hold space.
.TP
.RB (1) \|i\e
.br
.ns
.TP
.I text
Insert.
Place
.I text
on the standard output.
.TP
.RB (2) \|l
Literal.  
Place an unambiguous image of the pattern
space on the standard output,
using C escape sequences.
Break long lines, indicating the breakpoint by
a single backslash.
Append
.B \en
if pattern space ends with space or newline.
.TP
.RB (2) \|n
Copy the pattern space to the standard output.
Replace the pattern space with the next line of input.
.TP
.RB (2) \|N
Append the next line of input to the pattern space
with an embedded newline.
(The current line number changes.)
.TP
.RB (2) \|p
Print.
Copy the pattern space to the standard output.
.TP
.RB (2) \|P
Copy the initial segment of the pattern space through
the first newline to the standard output.
.TP
.RB (1) \|q
Quit.
Branch to the end of the script.
Do not start a new cycle.
.HP
.RB (2) \|r 
.I rfile
.br
Read the contents of
.IR rfile .
Place them on the output before reading
the next input line.
.TP
.RB (2) \|s/\fIregular-expression\fP/\fIreplacement\fP/\fIflags
Substitute the
.I replacement
string for instances of the
.I regular-expression
in the pattern space.
Any character may be used as a delimiter instead of 
.LR / .
In the replacement text, 
.L &
stands for a copy of the matched part
of the input text;
.BI \e n,
where 
.I n
is a digit (1-9)
stands for a copy of the last match for the
.IR n th
parenthesized (with
.BR \e( ... \e) )
subexpression; and
.L \e
quotes the following nondigit (even a newline).
.I Flags
is zero or more of
.RS
.TP
.I n
Substitute for the
.IR n th
occurrence of the regular expression;
.IR n =1
by default.
.TP
.B g
Global.
Substitute for all non-overlapping instances of the
regular expression.
.TP
.B p
Print the pattern space if a substitution was made.
.TP
.BI w " wfile"
Write.
Append the pattern space to
.I wfile
if a substitution
was made.
.RE
.HP
.RB (2) \|t
.I label
.br
Test.
Branch to the 
.L :
command bearing the
.I label
if any
substitutions have been made since the most recent
reading of an input line or execution of a 
.LR t .
If
.I label
is empty, branch to the end of the script.
.TP
.RB (2) \|w
.I wfile
.br
Write.
Append the pattern space to
.I wfile.
.TP
.RB (2) \|x
Exchange the contents of the pattern and hold spaces.
.TP
.RB (2) \|y/\fIstring1\fP/\fIstring2\fP/
Transform.
Replace all occurrences of characters in
.I string1
with the corresponding character in
.I string2.
The lengths of
.I
string1
and
.I string2
must be equal.
.HP
.RB (2) !
.I function
.br
Don't.
Apply the
.I function
(or group, if
.I function
is 
.LR { )
only to lines
.I not
selected by the address(es).
.HP
.RB (0) \|:
.I label
.br
This command does nothing; it bears a
.I label
for 
.B b
and 
.B t
commands to branch to.
.TP
.RB (1) \|=
Place the current line number on the standard output as a line.
.TP
.RB (2) \|{
Execute the following commands through a matching 
.L }
function only when the pattern space is selected.
.TP
.RB (0) \|
Ignore this empty line.
.TP
.RB (0) \|#
Comment.  Ignore this line (and any semicolons in it).
.ne 4
.SH EXAMPLES
.TP
.B sed 10q file
Print the first 10 lines of the file.
.TP
.B sed '/^$/d'
Delete empty lines from standard input.
.TP
.B sed 's/UNIX/& system/g'
Replace every instance of 
.L UNIX
by
.LR "UNIX system" .
.PP
.EX
sed 's/ *$//	\fRdrop trailing blanks\fP
/^$/d			\fRdrop empty lines\fP
s/  */\e		\fRreplace blanks by newlines\fP
/g
/^$/d' chapter*
.EE
.ns
.IP
Print the files 
.BR chapter1 ,
.BR chapter2 ,
etc. one word to a line.
.PP
.EX
nroff -ms manuscript | sed '
${
	/^$/p		\fRif last line of file is empty, print it\fP
}
//N			\fRif current line is empty, append next line\fP
/^\en$/D'		\fRif two lines are empty, delete the first\fP
.EE
.ns
.IP
Delete all but one of each group of empty lines from a
formatted manuscript.
.PP
.EX
ls /usr/* | sed '
/^$/d			\fRdelete empty lines\fP
/^[/].*:$/{		\fRlook for lines like \fP/usr/lem:
	s/:$/\e//	\fRreplace \fP:\fR by \fP/
	h		\fRhold directory name\fP
	d		\fRdon't print; get next line\fP
}
G			\fRappend held directory name\fP
s/\e(.*\e)\en\e(.*\e)/\e2\e1/'	\fRexchange file and directory\fP
.EE
.ns
.IP
List all files in user directories, as
.B ls -d /usr/*/* 
would do if it didn't cause argument list overflow.
.SH SEE ALSO
.IR ed (1), 
.IR gre (1), 
.IR awk (1), 
.IR lex (1), 
.IR cut (1), 
.IR split (1), 
.IR sam (9.1)
.SH BUGS
If input is from a pipe, buffering may consume
characters beyond a line on which a 
.L q
command is executed.