Comprehensive Guide to Using sed for Searching, Deleting, and Replacing Text in Linux

sed is a powerful command‑line tool that extends the capabilities of grep by providing both search and replace functions for text streams.

The basic syntax follows the pattern sed -n 'n'p filename , where n denotes the line number to print; -n suppresses automatic output and p explicitly prints matching lines.

Common printing examples include sed -n '5'p test.txt to display line 5, sed -n '1,5'p test.txt for lines 1‑5, and sed -n '1,$'p test.txt to output the whole file.

Pattern searching uses regular expressions, e.g., sed -n '/root/'p 1.txt finds lines containing "root", sed -n '/^l/'p 1.txt matches lines starting with "l", and sed -n '/\/login$/'p 1.txt locates lines ending with "login".

Deletion commands such as sed '1'd file.txt , sed '1,3'd file.txt , and sed '/ab/'d file.txt remove specific lines or lines matching a pattern, while the -i option (e.g., sed -i 's/^/#/g file.txt ) edits the file in place.

Replacement is performed with the s command; examples include sed 's/[0-9]/a/g' 1.txt to replace all digits with "a", sed '1,10s/root/ROOT/g' 1.txt to change "root" to "ROOT" in the first ten lines, and sed -i '30,40s/^.*$/#&/g' 1.txt to comment out a range of lines.

Advanced substitutions use the -r (extended regex) flag and alternate delimiters, such as sed -r 's/([^:]+):(.*):([^:]+)/\3:\2:\1/' file.txt to reorder colon‑separated fields, or sed 's#ot#to#g' file.txt where # replaces the usual / delimiter.

Multiple commands can be chained with -e or a single -r -e expression, for example sed -e '1'p -e '/111/'p -n file.txt or sed -n -r -e '/root/'p -e '/user5/'p file.txt , allowing complex processing in one pass.