Linux sed Command: Find and Replace Text

Find and replace

By default sed prints the result to the screen and leaves the file unchanged, so you can check it first.

sed "s/cat/dog/" pets.txt

Replaces the first cat on each line with dog.

sed "s/cat/dog/g" pets.txt

Replaces every cat on each line. The g stands for global.

sed "s/cat/dog/gI" pets.txt

Replaces every cat, ignoring case.

sed "s/\bcat\b/dog/g" pets.txt

Plain patterns also match inside other words, so category would become dogegory. \b marks the edge of a word, so this only changes cat on its own.

sed "s#/var/www#/srv/www#g" site.conf

Uses # instead of / to separate the parts, which is easier to read when the text contains slashes.

Editing a file

sed -i.bak "s/old.example.com/new.example.com/g" site.conf

Changes the file itself (-i) and keeps the original as site.conf.bak.

Showing and deleting lines

sed -n "2,4p" pets.txt

Prints only lines 2 to 4.

sed "1d" names.csv

Removes the first line, such as the heading row of a CSV file.

sed "/^#/d" config.conf

Removes comment lines that start with #.

sed -e "/^#/d" -e "/^$/d" config.conf

Removes comment lines and blank lines, leaving just the settings.

Adding lines and rearranging text

sed "1i Name,City,Amount" data.csv

Inserts a new first line.

sed "$ a # End of file" config.conf

Adds a line at the end.

sed -E "s#([0-9]{4})-([0-9]{2})-([0-9]{2})#\3/\2/\1#g" dates.txt

Changes dates like 2026-11-02 to 02/11/2026. The brackets capture each part, and \1 \2 \3 put them back in a new order.

Quick reference