Mastering sed: Essential Commands and Advanced Techniques for Text Processing

Basic sed commands

Basic syntax

sed OPTIONS… [SCRIPT] [INPUTFILE…]

Common options:

-n, --quiet: suppress automatic printing of the pattern space

-i: edit files in place (default does not modify the original file)

-e: allow multiple commands or scripts

-f /path/from/sed_script: read the script from a file

-r: use extended regular expressions

sed command options

Replacement flags g: global replacement within the line w: write the line to a file x: exchange the contents of pattern space and hold space y: transliterate characters (not a regular‑expression operation)

Single‑line commands a: append text after the current line c: replace the selected lines with new text d: delete lines (no following text) i: insert text before the current line p: print selected lines (often used with -n) s: substitute using regular expressions, e.g., 1,20s/old/new/g n: read the next input line and continue processing y: transliterate characters, e.g., y/abc/ABC/

[root@localhost ~]# vim xbz
[root@localhost ~]# cat xbz
a b c
d
c
b
a
[root@localhost ~]# sed '3abbxxxx' xbz   // add after line 3
...
[root@localhost ~]# sed '/c/abbxxxx' xbz   // add after lines matching c

c command usage

[root@localhost ~]# cat xbz
a b c
d
c
b
a
[root@localhost ~]# sed '2cxxb' xbz   // replace line 2
...
[root@localhost ~]# sed '/d/caa' xbz   // replace lines matching d

d command usage

[root@localhost ~]# cat xbz
a b c
d
c
b
a
[root@localhost ~]# sed '1d' xbz   // delete first line
...
[root@localhost ~]# sed '/c/d' xbz   // delete lines matching c

i command usage

[root@localhost ~]# cat xbz
a b c
d
c
b
a
[root@localhost ~]# sed '2i3838' xbz   // insert after line 2
...
[root@localhost ~]# sed '/c/i6868' xbz   // insert before lines matching c

p command usage

[root@localhost ~]# cat xbz
a b c
d
c
b
a
[root@localhost ~]# sed -n '/b/p' xbz   // print lines containing b

s command usage

[root@localhost ~]# cat xbz
a b c
d
c
bbb
a
[root@localhost ~]# sed 's/b/a/' xbz   // replace first b with a on each line
...
[root@localhost ~]# sed 's/b/a/g' xbz   // replace all b with a

n command usage

[root@localhost ~]# cat xbz
a b c
d
c
bbb
a
[root@localhost ~]# sed -n '/a/n;p' xbz   // print lines after a

y command usage

[root@localhost ~]# cat xbz
a b c
d
c
bbb
a
[root@localhost ~]# sed '3y/c/C/' xbz   // transliterate c to C on line 3

Advanced sed commands

Pattern space and hold space

Pattern space is the buffer that holds the current line being processed. Hold space is a secondary buffer that retains its contents until explicitly overwritten; it is not printed automatically. Commands h, H, g, and G allow interaction between these two spaces.

Multi‑line space commands

Multiple‑line space N: append the next line to the pattern space P: print up to the first newline in the pattern space D: delete up to the first newline and restart the cycle without reading a new line

[root@localhost ~]# cat test
This is the header line.
This is a data line.

This is the last line.
[root@localhost ~]# sed '/^$/{N;/header/D}' test   // delete the header line using multi‑line space

Hold space commands

Command

Abbreviation

Function

Hold

h (copy) or H (append)

Copy or append pattern space to hold space

Get

g or G

Copy or append hold space to pattern space

Exchange

x

Swap contents of hold and pattern spaces

[root@localhost ~]# cat abc
1
2
11
22
111
222
[root@localhost ~]# sed '/1/{h;d};/2/G' abc   // store line 1 in hold space, delete it, then append hold space when line 2 matches