The sed command is a non-interactive stream editor that allows you to modify the content of the standard input or file. It performs an operation on each line in a pipeline. The syntax will be:
sed [OPTIONS]... {script} [input-file …]
By default, the output is displayed on stdout, but can be redirected to a file if specified.
The input-file are the files on which sed needs to be run. If no files are specified, it reads from stdin.
The script can be a command or a file with multiple commands to pass to sed, and OPTIONS to sed are described in the following table:
|
Option |
Description |
|---|---|
|
-n | |
|
-e script | |
|
-r | |
|
-l N | |
|
--posix | |
|
-u |
This loads the minimal amounts of data from input and flushes output buffers frequently |
The sed command is widely used for string substitution in a text file. Programmers frequently use this feature while renaming a variable in a huge source code. It saves a lot of programmers' time by avoiding manual renaming.
The substitution command s has the following field:
s/regex/replacement/
Here, s means perform substitution, / acts as separator, and regex is a regular expression that needs to be replaced. A simple string can also be specified here. The last field replacement is with what matched results should be replaced.
By default, sed will replace only the first occurrence of a matched pattern in a line. To replace all occurrences, use the g flag after the end of /—, that is, s/regex/replacement/g.
Some of the flags that can be used are mentioned in the following table:
|
Flag |
Description |
|---|---|
|
| |
|
| |
|
| |
|
|
We have the sed.sh file for our example. The content of this file is as follows:
$ cat sed.sh
#!/bin/bash var1="sed " var1+="command " var1+="usage" echo $var1
This is a shell script, where the variable var1 has been used at four places. Now, we want to rename the variable var1 to variable. We can do this very easily using the sed command:
$ sed -i 's/var1/variable/g' sed.sh $ cat sed.sh
#!/bin/bash variable="sed " variable+="command " variable+="usage" echo $variable
Here, the -i option is used to replace an input file.
We can also specify multiple commands to be executed for substitution using -e followed by a command.
For example, consider the sed.txt file. The content of this file is as follows:
$ cat sed.txt The sed command is widely used for string substitution in text file. Programmers frequently use this feature while renaming a variable in huge source code. It saves lot of programmers time by avoiding manual renaming.
Now, we want to replace '.' with ',' and delete the line containing a string manual:
$ sed -e 's/./,/g' -e '/manual/d' sed.txt The sed command is widely used for string substitution in text file, Programmers frequently use this feature while renaming a variable in huge source code,
In sed.txt file, the s/./,/g command first replaces '.' with ',' and /manual/d deletes further the line containing the string manual.