Linux

SED

SED Cheat sheet

Syntaxe

$ sed [options] command [input-file]

Exemples avec pipeline

$ cat report.txt | sed 's/Nick/John/g'
$ echo '123abc' | sed 's/[0-9]+//g'

commandes multiples

$ echo "hello world" | sed -e 's/h/H/g' -e 's/w/W/g'
Hello World

Script sed

$ echo 's/h/H/g' >> hello.sed
$ echo 's/w/W/g' >> hello.sed
$ echo "hello world" | sed -f hello.sed
Hello World

Options

Option	Example	Description
-i	`sed -ibak 's/On/Off/' php.ini`	Backup and modify input file directly
-E	`sed -E 's/[0-9]+//g' input-file`	Use extended regular expressions
-n	`sed -n '3 p' config.conf`	Suppress default pattern space printing
-f	`sed -f script.sed config.conf`	Execute sed script file
-e	`sed -e 'command1' -e 'command2' input-file`	Execute multiple sed commands

Commands

Command	Example	Description
p	`sed -n '1,4 p' input.txt`	Print lines 1-4
p	`sed -n -e '1,4 p' -e '6,7 p' input.txt`	Print lines 1-4 and 6-7
d	`sed '1,4 d' input.txt`	Print lines except 1-4
w	`sed -n '1,4 w output.txt' input.txt`	Write pattern space to file
a	`sed '2 a new-line' input.txt`	Append line after
i	`sed '2 i new-line' input.txt`	Insert line before
Space commands
n		Print pattern space, empty pattern space, and read next line
x		Swap pattern space with hold space
h		Copy pattern space to hold space
H		Append pattern space to hold space
g		Copy hold space to pattern space
G		Append hold space to pattern space
Flags
$ sed 's/old/new/[flags]' [input-file]
g		Global substitution
1,2…		Substitute the nth occurrence
p		Print only the substituted line
w		Write only the substituted line to a file
I		Ignore case while searching
e		Substitute and execute in the command line
Loops commands
b label		Branch to a label (for looping)
t label		Branch to a label only on successful substitution (for looping)
:label		Label for the b and t commands (for looping)
N		Append next line to pattern space
P		Print 1st line in multi-line
D		Delete 1st line in multi-line
Misc flags
/ \| ^ @ ! #		Substitution delimiter can be any character
&		Gets the matched pattern
( ) \1 \2 \3		Group using ( and ). Use \1, \2 in replacement to refer the group

Recherche de texte

# Search for a string and only print the lines that were matched
$ sed -n '/hello/p' file.txt
 
# Case insensitive search
$ sed -n '/hello/Ip' file.txt
 
# Search for a string but only output lines that do not match
$ sed -n '/hello/!p' file.txt

Remplacement de texte

# Replace all occurrences of a string
$ sed 's/old/new/g' file.txt
 
# Replace only the nth occurrence of a string
$ sed 's/old/new/2' file.txt
 
# Replace replace a string only on the 5th line
$ sed '5 s/old/new/' file.txt
 
# Replace "world" with "universe" but only if the line begins with "hello"
$ sed '/hello/s/world/universe/' file.txt
 
# Remove "" from the end of each line
$ sed 's/\\$//' file.txt
 
# Remove all whitespace from beginning of each line
$ sed 's/^\s*//' file.txt
 
# Remove comments. Even those that are at the end of a line
$ sed 's/#.*$//' file.txt

Ajout de texte

# Append line after line 2
$ sed '2a Text after line 2' file.txt
 
# Append line at the end of the file
$ sed '$a THE END!' file.txt
 
# Append line after every 3rd line starting from line 3
$ sed '3~3a Some text' file.txt
 
# Insert text before line 5
$ sed '5i line number five' file.txt
 
# Insert "Example: " before each line that contains "hello"
$ sed '/hello/i Example: ' file.txt

Numérotation

# Number line of a file (simple left alignment)
$ sed = file.txt | sed 'N;s/\n/\t/'
 
# Number line of a file (number on left, right-aligned)
$ sed = file.txt | sed 'N; s/^/   /; s/ *\(.\{6,\}\)\n/\1  /'
 
# Number line of file, but only print numbers if line is not blank
$ sed '/./=' file.txt | sed '/./N; s/\n/ /'
 
# Count lines (emulates "wc -l")
$ sed -n '$='

Effacement de texte

# Delete line 5-7 in file
$ sed '5,7d' file.txt
 
# Delete every 2nd line starting with line 3
$ sed '3~2d' file.txt
 
# Delete the last line in file
$ sed '$d' file.txt
 
# Delete lines starting with "Hello"
$ sed '/^Hello/d' file.txt
 
# Delete all empty lines
$ sed '/^$/d' file.txt
 
# Delete lines starting with "#"
$ sed '/^#/d' file.txt

Gestion des espaces

# Double space
$ sed G
 
# Delete all blank lines and double space
$ sed '/^$/d;G'
 
# Triple space a file
$ sed 'G;G'
 
# Undo double-spacing
$ sed 'n;d'
 
# Insert a blank line above line which matches "regex"
$ sed '/regex/{x;p;x;}'
 
# Insert a blank line below line which matches "regex"
$ sed '/regex/G'
 
# Insert a blank line around line which matches "regex"
$ sed '/regex/{x;p;x;G;}'

Exemples

Liste de fichiers de moins de 30 jours, ordonnée par date dans le répertoire spécifié seulement

$ find ./* -prune -name "*<chaine>*" -mtime 30 | xargs -exec ls -ltr