NAME
sed - stream editor
SYNOPSIS
sed [-n] script[file...]
sed [-n][-e script]...[-f script_file]...[file...]
DESCRIPTION
The sed utility is a stream editor that shall read one or more text
files, make editing changes according to a script of editing commands,
and write the results to standard output. The script shall be obtained
from either the script operand string or a combination of the option-
arguments from the -e script and -f script_file options.
OPTIONS
The sed utility shall conform to the Base Definitions volume of
IEEE Std 1003.1-2001, Section 12.2, Utility Syntax Guidelines, except
that the order of presentation of the -e and -f options is significant.
The following options shall be supported:
-e script
Add the editing commands specified by the script option-argument
to the end of the script of editing commands. The script option-
argument shall have the same properties as the script operand,
described in the OPERANDS section.
-f script_file
Add the editing commands in the file script_file to the end of
the script.
-n Suppress the default output (in which each line, after it is
examined for editing, is written to standard output). Only lines
explicitly selected for output are written.
Multiple -e and -f options may be specified. All commands shall be
added to the script in the order specified, regardless of their origin.
OPERANDS
The following operands shall be supported:
file A pathname of a file whose contents are read and edited. If
multiple file operands are specified, the named files shall be
read in the order specified and the concatenation shall be
edited. If no file operands are specified, the standard input
shall be used.
script A string to be used as the script of editing commands. The
application shall not present a script that violates the
restrictions of a text file except that the final character need
not be a <newline>.
STDIN
The standard input shall be used only if no file operands are
specified. See the INPUT FILES section.
INPUT FILES
The input files shall be text files. The script_files named by the -f
option shall consist of editing commands.
ENVIRONMENT VARIABLES
The following environment variables shall affect the execution of sed:
LANG Provide a default value for the internationalization variables
that are unset or null. (See the Base Definitions volume of
IEEE Std 1003.1-2001, Section 8.2, Internationalization
Variables for the precedence of internationalization variables
used to determine the values of locale categories.)
LC_ALL If set to a non-empty string value, override the values of all
the other internationalization variables.
LC_COLLATE
Determine the locale for the behavior of ranges, equivalence
classes, and multi-character collating elements within regular
expressions.
LC_CTYPE
Determine the locale for the interpretation of sequences of
bytes of text data as characters (for example, single-byte as
opposed to multi-byte characters in arguments and input files),
and the behavior of character classes within regular
expressions.
LC_MESSAGES
Determine the locale that should be used to affect the format
and contents of diagnostic messages written to standard error.
NLSPATH
Determine the location of message catalogs for the processing of
LC_MESSAGES .
ASYNCHRONOUS EVENTS
Default.
STDOUT
The input files shall be written to standard output, with the editing
commands specified in the script applied. If the -n option is
specified, only those input lines selected by the script shall be
written to standard output.
STDERR
The standard error shall be used only for diagnostic messages.
OUTPUT FILES
The output files shall be text files whose formats are dependent on the
editing commands given.
EXTENDED DESCRIPTION
The script shall consist of editing commands of the following form:
[address[,address]]function
where function represents a single-character command verb from the list
in Editing Commands in sed , followed by any applicable arguments.
The command can be preceded by <blank>s and/or semicolons. The function
can be preceded by <blank>s. These optional characters shall have no
effect.
In default operation, sed cyclically shall append a line of input, less
its terminating <newline>, into the pattern space. Normally the pattern
space will be empty, unless a D command terminated the last cycle. The
sed utility shall then apply in sequence all commands whose addresses
select that pattern space, and at the end of the script copy the
pattern space to standard output (except when -n is specified) and
delete the pattern space. Whenever the pattern space is written to
standard output or a named file, sed shall immediately follow it with a
<newline>.
Some of the editing commands use a hold space to save all or part of
the pattern space for subsequent retrieval. The pattern and hold spaces
shall each be able to hold at least 8192 bytes.
Addresses in sed
An address is either a decimal number that counts input lines
cumulatively across files, a ’$’ character that addresses the last line
of input, or a context address (which consists of a BRE, as described
in Regular Expressions in sed , preceded and followed by a delimiter,
usually a slash).
An editing command with no addresses shall select every pattern space.
An editing command with one address shall select each pattern space
that matches the address.
An editing command with two addresses shall select the inclusive range
from the first pattern space that matches the first address through the
next pattern space that matches the second. (If the second address is a
number less than or equal to the line number first selected, only one
line shall be selected.) Starting at the first line following the
selected range, sed shall look again for the first address. Thereafter,
the process shall be repeated. Omitting either or both of the address
components in the following form produces undefined results:
[address[,address]]
Regular Expressions in sed
The sed utility shall support the BREs described in the Base
Definitions volume of IEEE Std 1003.1-2001, Section 9.3, Basic Regular
Expressions, with the following additions:
* In a context address, the construction "\cBREc" , where c is any
character other than backslash or <newline>, shall be identical to
"/BRE/" . If the character designated by c appears following a
backslash, then it shall be considered to be that literal character,
which shall not terminate the BRE. For example, in the context
address "\xabc\xdefx" , the second x stands for itself, so that the
BRE is "abcxdef" .
* The escape sequence ’\n’ shall match a <newline> embedded in the
pattern space. A literal <newline> shall not be used in the BRE of a
context address or in the substitute function.
* If an RE is empty (that is, no pattern is specified) sed shall
behave as if the last RE used in the last command applied (either as
an address or as part of a substitute command) was specified.
Editing Commands in sed
In the following list of editing commands, the maximum number of
permissible addresses for each function is indicated by [ 0addr], [
1addr], or [ 2addr], representing zero, one, or two addresses.
The argument text shall consist of one or more lines. Each embedded
<newline> in the text shall be preceded by a backslash. Other
backslashes in text shall be removed, and the following character shall
be treated literally.
The r and w command verbs, and the w flag to the s command, take an
optional rfile (or wfile) parameter, separated from the command verb
letter or flag by one or more <blank>s; implementations may allow zero
separation as an extension.
The argument rfile or the argument wfile shall terminate the editing
command. Each wfile shall be created before processing begins.
Implementations shall support at least ten wfile arguments in the
script; the actual number (greater than or equal to 10) that is
supported by the implementation is unspecified. The use of the wfile
parameter shall cause that file to be initially created, if it does not
exist, or shall replace the contents of an existing file.
The b, r, s, t, w, y, and : command verbs shall accept additional
arguments. The following synopses indicate which arguments shall be
separated from the command verbs by a single <space>.
The a and r commands schedule text for later output. The text specified
for the a command, and the contents of the file specified for the r
command, shall be written to standard output just before the next
attempt to fetch a line of input when executing the N or n commands, or
when reaching the end of the script. If written when reaching the end
of the script, and the -n option was not specified, the text shall be
written after copying the pattern space to standard output. The
contents of the file specified for the r command shall be as of the
time the output is written, not the time the r command is applied. The
text shall be output in the order in which the a and r commands were
applied to the input.
Command verbs other than {, a, b, c, i, r, t, w, :, and # can be
followed by a semicolon, optional <blank>s, and another command verb.
However, when the s command verb is used with the w flag, following it
with another command in this manner produces undefined results.
A function can be preceded by one or more ’!’ characters, in which case
the function shall be applied if the addresses do not select the
pattern space. Zero or more <blank>s shall be accepted before the first
’!’ character. It is unspecified whether <blank>s can follow a ’!’
character, and conforming applications shall not follow a ’!’
character with <blank>s.
[2addr] {function
function
...
} Execute a list of sed functions only when the pattern space is
selected. The list of sed functions shall be surrounded by
braces and separated by <newline>s, and conform to the following
rules. The braces can be preceded or followed by <blank>s. The
functions can be preceded by <blank>s, but shall not be followed
by <blank>s. The <right-brace> shall be preceded by a <newline>
and can be preceded or followed by <blank>s.
[1addr]a\
text Write text to standard output as described previously.
[2addr]b [label]
Branch to the : function bearing the label. If label is not
specified, branch to the end of the script. The implementation
shall support labels recognized as unique up to at least 8
characters; the actual length (greater than or equal to 8) that
shall be supported by the implementation is unspecified. It is
unspecified whether exceeding a label length causes an error or
a silent truncation.
[2addr]c\
text Delete the pattern space. With a 0 or 1 address or at the end of
a 2-address range, place text on the output and start the next
cycle.
[2addr]d
Delete the pattern space and start the next cycle.
[2addr]D
Delete the initial segment of the pattern space through the
first <newline> and start the next cycle.
[2addr]g
Replace the contents of the pattern space by the contents of the
hold space.
[2addr]G
Append to the pattern space a <newline> followed by the contents
of the hold space.
[2addr]h
Replace the contents of the hold space with the contents of the
pattern space.
[2addr]H
Append to the hold space a <newline> followed by the contents of
the pattern space.
[1addr]i\
text Write text to standard output.
[2addr]l
(The letter ell.) Write the pattern space to standard output in
a visually unambiguous form. The characters listed in the Base
Definitions volume of IEEE Std 1003.1-2001, Table 5-1, Escape
Sequences and Associated Actions ( ’\\’ , ’\a’ , ’\b’ , ’\f’ ,
’\r’ , ’\t’ , ’\v’ ) shall be written as the corresponding
escape sequence; the ’\n’ in that table is not applicable. Non-
printable characters not in that table shall be written as one
three-digit octal number (with a preceding backslash) for each
byte in the character (most significant byte first). If the size
of a byte on the system is greater than 9 bits, the format used
for non-printable characters is implementation-defined.
Long lines shall be folded, with the point of folding indicated by
writing a backslash followed by a <newline>; the length at which
folding occurs is unspecified, but should be appropriate for the output
device. The end of each line shall be marked with a ’$’ .
[2addr]n
Write the pattern space to standard output if the default output
has not been suppressed, and replace the pattern space with the
next line of input, less its terminating <newline>.
If no next line of input is available, the n command verb shall branch
to the end of the script and quit without starting a new cycle.
[2addr]N
Append the next line of input, less its terminating <newline>,
to the pattern space, using an embedded <newline> to separate
the appended material from the original material. Note that the
current line number changes.
If no next line of input is available, the N command verb shall branch
to the end of the script and quit without starting a new cycle or
copying the pattern space to standard output.
[2addr]p
Write the pattern space to standard output.
[2addr]P
Write the pattern space, up to the first <newline>, to standard
output.
[1addr]q
Branch to the end of the script and quit without starting a new
cycle.
[1addr]r rfile
Copy the contents of rfile to standard output as described
previously. If rfile does not exist or cannot be read, it shall
be treated as if it were an empty file, causing no error
condition.
[2addr]s/BRE/replacement/flags
Substitute the replacement string for instances of the BRE in
the pattern space. Any character other than backslash or
<newline> can be used instead of a slash to delimit the BRE and
the replacement. Within the BRE and the replacement, the BRE
delimiter itself can be used as a literal character if it is
preceded by a backslash.
The replacement string shall be scanned from beginning to end. An
ampersand ( ’&’ ) appearing in the replacement shall be replaced by the
string matching the BRE. The special meaning of ’&’ in this context can
be suppressed by preceding it by a backslash. The characters "\n",
where n is a digit, shall be replaced by the text matched by the
corresponding backreference expression. The special meaning of "\n"
where n is a digit in this context, can be suppressed by preceding it
by a backslash. For each other backslash ( ’\’ ) encountered, the
following character shall lose its special meaning (if any). The
meaning of a ’\’ immediately followed by any character other than ’&’ ,
’\’ , a digit, or the delimiter character used for this command, is
unspecified.
A line can be split by substituting a <newline> into it. The
application shall escape the <newline> in the replacement by preceding
it by a backslash. A substitution shall be considered to have been
performed even if the replacement string is identical to the string
that it replaces. Any backslash used to alter the default meaning of a
subsequent character shall be discarded from the BRE or the replacement
before evaluating the BRE or using the replacement.
The value of flags shall be zero or more of:
n
Substitute for the nth occurrence only of the BRE found within
the pattern space.
g
Globally substitute for all non-overlapping instances of the BRE
rather than just the first one. If both g and n are specified,
the results are unspecified.
p
Write the pattern space to standard output if a replacement was
made.
w wfile
Write. Append the pattern space to wfile if a replacement was
made. A conforming application shall precede the wfile argument
with one or more <blank>s. If the w flag is not the last flag
value given in a concatenation of multiple flag values, the
results are undefined.
[2addr]t [label]
Test. Branch to the : command verb bearing the label if any
substitutions have been made since the most recent reading of an
input line or execution of a t. If label is not specified,
branch to the end of the script.
[2addr]w wfile
Append (write) the pattern space to wfile.
[2addr]x
Exchange the contents of the pattern and hold spaces.
[2addr]y/string1/string2/
Replace all occurrences of characters in string1 with the
corresponding characters in string2. If a backslash followed by
an ’n’ appear in string1 or string2, the two characters shall be
handled as a single <newline>. If the number of characters in
string1 and string2 are not equal, or if any of the characters
in string1 appear more than once, the results are undefined. Any
character other than backslash or <newline> can be used instead
of slash to delimit the strings. If the delimiter is not n,
within string1 and string2, the delimiter itself can be used as
a literal character if it is preceded by a backslash. If a
backslash character is immediately followed by a backslash
character in string1 or string2, the two backslash characters
shall be counted as a single literal backslash character. The
meaning of a backslash followed by any character that is not ’n’
, a backslash, or the delimiter character is undefined.
[0addr]:label
Do nothing. This command bears a label to which the b and t
commands branch.
[1addr]=
Write the following to standard output:
"%d\n", <current line number>
[0addr]
Ignore this empty command.
[0addr]#
Ignore the ’#’ and the remainder of the line (treat them as a
comment), with the single exception that if the first two
characters in the script are "#n" , the default output shall be
suppressed; this shall be the equivalent of specifying -n on the
command line.
EXIT STATUS
The following exit values shall be returned:
0 Successful completion.
>0 An error occurred.
CONSEQUENCES OF ERRORS
Default.
The following sections are informative.
APPLICATION USAGE
Regular expressions match entire strings, not just individual lines,
but a <newline> is matched by ’\n’ in a sed RE; a <newline> is not
allowed by the general definition of regular expression in
IEEE Std 1003.1-2001. Also note that ’\n’ cannot be used to match a
<newline> at the end of an arbitrary input line; <newline>s appear in
the pattern space as a result of the N editing command.
EXAMPLES
This sed script simulates the BSD cat -s command, squeezing excess
blank lines from standard input.
sed -n ’
# Write non-empty lines.
/./ {
p
d
}
# Write a single empty line, then look for more empty lines.
/^$/ p
# Get next line, discard the held <newline> (empty line),
# and look for more empty lines.
:Empty
/^$/ {
N
s/.//
b Empty
}
# Write the non-empty line before going back to search
# for the first in a set of empty lines.
p
RATIONALE
This volume of IEEE Std 1003.1-2001 requires implementations to support
at least ten distinct wfiles, matching historical practice on many
implementations. Implementations are encouraged to support more, but
conforming applications should not exceed this limit.
The exit status codes specified here are different from those in System
V. System V returns 2 for garbled sed commands, but returns zero with
its usage message or if the input file could not be opened. The
standard developers considered this to be a bug.
The manner in which the l command writes non-printable characters was
changed to avoid the historical backspace-overstrike method, and other
requirements to achieve unambiguous output were added. See the
RATIONALE for ed for details of the format chosen, which is the same as
that chosen for sed.
This volume of IEEE Std 1003.1-2001 requires implementations to provide
pattern and hold spaces of at least 8192 bytes, larger than the 4000
bytes spaces used by some historical implementations, but less than the
20480 bytes limit used in an early proposal. Implementations are
encouraged to allocate dynamically larger pattern and hold spaces as
needed.
The requirements for acceptance of <blank>s and <space>s in command
lines has been made more explicit than in early proposals to describe
clearly the historical practice and to remove confusion about the
phrase "protect initial blanks [sic] and tabs from the stripping that
is done on every script line" that appears in much of the historical
documentation of the sed utility description of text. (Not all
implementations are known to have stripped <blank>s from text lines,
although they all have allowed leading <blank>s preceding the address
on a command line.)
The treatment of ’#’ comments differs from the SVID which only allows a
comment as the first line of the script, but matches BSD-derived
implementations. The comment character is treated as a command, and it
has the same properties in terms of being accepted with leading
<blank>s; the BSD implementation has historically supported this.
Early proposals required that a script_file have at least one non-
comment line. Some historical implementations have behaved in
unexpected ways if this were not the case. The standard developers
considered that this was incorrect behavior and that application
developers should not have to avoid this feature. A correct
implementation of this volume of IEEE Std 1003.1-2001 shall permit
script_files that consist only of comment lines.
Early proposals indicated that if -e and -f options were intermixed,
all -e options were processed before any -f options. This has been
changed to process them in the order presented because it matches
historical practice and is more intuitive.
The treatment of the p flag to the s command differs between System V
and BSD-based systems when the default output is suppressed. In the two
examples:
echo a | sed ’s/a/A/p’
echo a | sed -n ’s/a/A/p’
this volume of IEEE Std 1003.1-2001, BSD, System V documentation, and
the SVID indicate that the first example should write two lines with A,
whereas the second should write one. Some System V systems write the A
only once in both examples because the p flag is ignored if the -n
option is not specified.
This is a case of a diametrical difference between systems that could
not be reconciled through the compromise of declaring the behavior to
be unspecified. The SVID/BSD/System V documentation behavior was
adopted for this volume of IEEE Std 1003.1-2001 because:
* No known documentation for any historic system describes the
interaction between the p flag and the -n option.
* The selected behavior is more correct as there is no technical
justification for any interaction between the p flag and the -n
option. A relationship between -n and the p flag might imply that
they are only used together, but this ignores valid scripts that
interrupt the cyclical nature of the processing through the use of
the D, d, q, or branching commands. Such scripts rely on the p
suffix to write the pattern space because they do not make use of
the default output at the "bottom" of the script.
* Because the -n option makes the p flag unnecessary, any interaction
would only be useful if sed scripts were written to run both with
and without the -n option. This is believed to be unlikely. It is
even more unlikely that programmers have coded the p flag expecting
it to be unnecessary. Because the interaction was not documented,
the likelihood of a programmer discovering the interaction and
depending on it is further decreased.
* Finally, scripts that break under the specified behavior produce too
much output instead of too little, which is easier to diagnose and
correct.
The form of the substitute command that uses the n suffix was limited
to the first 512 matches in an early proposal. This limit has been
removed because there is no reason an editor processing lines of
{LINE_MAX} length should have this restriction. The command s/a/A/2047
should be able to substitute the 2047th occurrence of a on a line.
The b, t, and : commands are documented to ignore leading white space,
but no mention is made of trailing white space. Historical
implementations of sed assigned different locations to the labels ’x’
and "x " . This is not useful, and leads to subtle programming errors,
but it is historical practice, and changing it could theoretically
break working scripts. Implementors are encouraged to provide warning
messages about labels that are never used or jumps to labels that do
not exist.
Historically, the sed ! and } editing commands did not permit multiple
commands on a single line using a semicolon as a command delimiter.
Implementations are permitted, but not required, to support this
extension.
FUTURE DIRECTIONS
None.
SEE ALSO
awk , ed , grep
COPYRIGHT
Portions of this text are reprinted and reproduced in electronic form
from IEEE Std 1003.1, 2003 Edition, Standard for Information Technology
-- Portable Operating System Interface (POSIX), The Open Group Base
Specifications Issue 6, Copyright (C) 2001-2003 by the Institute of
Electrical and Electronics Engineers, Inc and The Open Group. In the
event of any discrepancy between this version and the original IEEE and
The Open Group Standard, the original IEEE and The Open Group Standard
is the referee document. The original Standard can be obtained online
at http://www.opengroup.org/unix/online.html .