A powerful stream-oriented text editor.
sed is a stream editor and a crucial tool in text processing, working perfectly with regular expressions. During processing, it stores the current line being handled in a temporary buffer called the "pattern space," then processes the buffer's content with sed commands and sends the result to the screen. It then moves to the next line, repeating this until the end of the file. The file content itself does not change unless you use redirection to save the output. sed is primarily used for automatically editing one or more files, simplifying repetitive operations on files, and writing conversion programs.
Command Format
sed [options] 'command' file(s)
sed [options] -f scriptfile file(s)
-e<script> or --expression=<script>: Process the input text file using the specified script.
-f<scriptfile> or --file=<scriptfile>: Process the input text file using the script file.
-h or --help: Display help.
-n, --quiet, --silent: Only display the result of script processing.
-V or --version: Display version information.
File: Specifies the list of text files to be processed.
a\ # Insert text below the current line.
i\ # Insert text above the current line.
c\ # Change the selected lines to new text.
d # Delete the selected lines.
D # Delete the first line of the pattern space.
s # Substitute specified characters.
h # Copy the pattern space content to the hold buffer.
H # Append the pattern space content to the hold buffer.
g # Get the hold buffer content and replace the current pattern space text.
G # Get the hold buffer content and append it after the current pattern space text.
l # List non-printable characters.
n # Read the next input line and process it with the next command instead of the first.
N # Append the next input line to the pattern space with an embedded newline, changing the current line number.
p # Print the lines in the pattern space.
P # (Uppercase) Print the first line of the pattern space.
q # Quit sed.
b label # Branch to a labeled spot in the script; if the label doesn't exist, branch to the end of the script.
r file # Read lines from 'file'.
t label # Conditional branch; if the last 's' command succeeded, branch to the label.
T label # Error branch; if the last 's' command failed, branch to the label.
w file # Write and append the pattern space to 'file'.
W file # Write and append the first line of the pattern space to 'file'.
! # Apply the following command to lines not selected.
= # Print the current line number.
# # Comment extending to the next newline.
g # Global substitution within a line.
p # Print the line.
w # Write the line to a file.
x # Swap the contents of the pattern space and the hold buffer.
y # Translate one character to another (not used with regular expressions).
\1 # Substring match flag.
& # Matched string flag.
^ # Match the beginning of a line. E.g., /^sed/ matches all lines starting with 'sed'.
$ # Match the end of a line. E.g., /sed$/ matches all lines ending with 'sed'.
. # Match any single non-newline character. E.g., /s.d/ matches 's' followed by any character and then 'd'.
* # Match zero or more occurrences of the preceding character.
[] # Match any character within the range. E.g., /[sS]ed/ matches 'sed' or 'Sed'.
[^] # Match any character not in the range.
\(..\) # Match a substring and save it. E.g., s/\(love\)able/\1rs/ replaces 'loveable' with 'lovers'.
& # Save the searched character for replacement. E.g., s/love/**&**/ results in **love**.
\< # Match the beginning of a word.
\> # Match the end of a word.
x\{m\} # Repeat character 'x' exactly m times.
x\{m,\} # Repeat character 'x' at least m times.
x\{m,n\} # Repeat character 'x' between m and n times.
Replace a string in a file:
sed 's/book/books/' file
The -n option and p command used together print only the lines where substitution occurred:
sed -n 's/test/TEST/p' file
Use the -i option to edit the file directly, replacing all occurrences of 'book' with 'books':
sed -i 's/book/books/g' file
The /g suffix replaces all matches in each line:
sed 's/book/books/g' file
To start substitution from the Nth match, use /Ng:
echo sksksksksksk | sed 's/sk/SK/2g'
skSKSKSKSKSK
echo sksksksksksk | sed 's/sk/SK/3g'
skskSKSKSKSK
The / character is used as a delimiter in the above commands, but any character can be used:
sed 's:test:TEXT:g'
sed 's|test|TEXT|g'
If the delimiter appears within the pattern, it must be escaped:
sed 's/\/bin/\/usr\/local\/bin/g'
Delete blank lines:
sed '/^$/d' file
Delete the 2nd line:
sed '2d' file
Delete from the 2nd line to the end:
sed '2,$d' file
Delete the last line:
sed '$d' file
Delete lines starting with 'test':
sed '/^test/d' file
The regex \w\+ matches every word; replace it with [&], where & represents the matched word:
echo this is a test line | sed 's/\w\+/[&]/g'
[this] [is] [a] [test] [line]
Replace lines starting with '192.168.0.1' with itself plus 'localhost':
sed 's/^192.168.0.1/&localhost/' file
Match a part of a given pattern:
echo this is digit 7 in a number | sed 's/digit \([0-9]\)/\1/'
this is 7 in a number
digit 7 is replaced by 7. The matched substring is 7, captured by \(..\) and referenced by \1.
echo aaa BBB | sed 's/\([a-z]\+\) \([A-Z]\+\)/\2 \1/'
BBB aaa
\u: Convert the first letter to uppercase.
\U: Convert everything to uppercase.
\l: Convert the first letter to lowercase.
\L: Convert everything to lowercase.
Convert the first letter to uppercase:
sed 's/^[a-z]\+/\u&/' passwd
Convert matched characters to uppercase:
sed 's/^[a-z]\+/\U&/' passwd
sed -e 's/old/new/g' -e 's/another_old/another_new/g' file.txt
sed -e '1d' -e '/pattern/d' file.txt
Use double quotes if the expression contains variables:
test=hello
echo hello WORLD | sed "s/$test/HELLO/"
HELLO WORLD
Print all lines within the range defined by patterns 'test' and 'check':
sed -n '/test/,/check/p' file
sed -e '1,5d' -e 's/test/check/' file
Read content from 'file' and display it after the line matching 'test':
sed '/test/r file' filename
Write all lines containing 'test' in 'example' to 'file':
sed -n '/test/w file' example
Append 'this is a test line' after lines starting with 'test':
sed '/^test/a\this is a test line' file
Insert 'this is a test line' before lines starting with 'test':
sed '/^test/i\this is a test line' file
Replace lines starting with 'root' with new content:
sed '/^root/c\this is new line!' passwd
If 'test' is matched, move to the next line and replace 'aa' with 'bb':
sed '/test/{ n; s/aa/bb/; }' file
Transform characters a-e to uppercase in lines 1-10:
sed '1,10y/abcde/ABCDE/' file
Quit sed after printing the first 10 lines:
sed '10q' file
sed -e '/test/h' -e '$G' file
Lines containing 'test' are copied to the hold buffer. At the last line, the G command appends the hold buffer content to the end of the file.
Swap the contents of the pattern space and the hold buffer:
sed -e '/test/h' -e '/check/x' file
sed [options] -f scriptfile file(s)
sed -n 'p;n' test.txt # Odd lines
sed -n 'n;p' test.txt # Even lines
sed -n '/SCC/{n;p}' URFILE