sed Command to Replace a String in a File

[Linux Hint]

Whenever we are working with any sort of files, it is a very common practice to make modifications to those files by finding and replacing words. Most of the text editors do provide different GUI based methods to perform these modifications. However, at times, you need a way to find and replace the words within a file via the terminal. Therefore, in this article, we will walk you through the method of using the sed command to replace a string in a text file. We will also throw light on the different scenarios in which this command can be used.

Note: We will be using Linux Mint 20 for demonstrating the usage of the sed command to replace a string in a file. You can also use any other flavor of Linux if you wish to.

Usage of sed Command:

Now we are going to show you some very interesting examples for depicting the usage of the sed command in Linux. Let us see, how this command works in different scenarios.

Creating a Text File for Demonstration:

For demonstrating the usage of the sed command, we would like to work with a text file. However, you can even create a Bash file or any other file of your choice. Secondly, we will create the file in the Home directory so that we do not have to specify any complex paths while accessing this file. For creating a text file, the following steps should be performed:

• Click on the File Manager icon located on your Linux Mint 20 taskbar:

• Now create a new text file in the Home directory by right-clicking anywhere over there and then choosing the New Document option from the cascading menu and Empty document option from the sub-cascading menu. Once the text file has been created, give a suitable name to it. For this particular example, I have named it as sed.txt.

• Double click on this file to open it and type any random text in it as shown in the image below. Save this text file by pressing Ctrl +S and then close it.

• Now launch the terminal in Linux Mint 20 as shown in the following image:

After launching the terminal in Linux Mint 20, you can go through all the examples mentioned below one by one.

Using the sed Command for replacing all Occurrences of a Given Word:

You can use the sed command for replacing all the occurrences of a given word in a text file. The syntax of achieving this goal is mentioned below:

$ sed ‘s/sed/FindWord/ReplaceWord/’ file.txt

Here, you need to replace FindWord with the word that you wish to replace and ReplaceWord with the word to be replaced. Moreover, you also need to replace “file” with the name of the file where you want to make these replacements. In this scenario, I have replaced FindWord with sed and ReplaceWord with replace. Our file name was sed.txt. It is also shown in the following image:

The successful execution of this command will show you the changes that have occurred because of running this command on your terminal:

Using the sed command for replacing the nth Occurrence of a Given Word in every Line:

The above scenario was the simplest replacement scenario, however at times, you do not want to replace all the occurrences of a word rather you want to replace only the first, second, or nth occurrence of that word in all the lines. For doing this, you can execute the following command:

$ sed ‘s/FindWord/ReplaceWord/num’ file.txt

Here, you need to replace FindWord with the word that you wish to replace and ReplaceWord with the word to be replaced. Moreover, you also need to replace “file” with the name of the file where you want to make these replacements. In this scenario, I have replaced FindWord replace and ReplaceWord with sed. Our file name was sed.txt. Also, num refers to the occurrence or the position of the word to be replaced. In the demonstrated example, I want to replace the first occurrence of replace with sed in every line as shown in the image below:

The successful execution of this command will show you the changes that have occurred because of running this command on your terminal:

Using the sed command for replacing a Given Word in a Specific Line:

At times, you only want to replace a given word in a specific line and not in the whole document. For doing that, you explicitly need to specify that particular line number in the following command:

$ sed ‘LineNum s/FindWord/ReplaceWord/’ file.txt

Here, you need to replace FindWord with the word that you wish to replace and ReplaceWord with the word to be replaced. Moreover, you also need to replace “file” with the name of the file where you want to make these replacements. In this scenario, I have replaced FindWord sed and ReplaceWord with replace. Our file name was sed.txt. Also, you need to replace LineNum with the line number of the particular line in which you want to make the replacement. In this example, we wanted to replace sed with replace in line number 2 of our file as shown in the following image:

The successful execution of this command will show you the changes that have occurred because of running this command on your terminal. You can easily verify that the said changes have only occurred on line number 2 of our file and not in the whole text.

Conclusion:

Whenever you feel the need of replacing any word in a file, the very first thing you need to do is to identify your particular requirement i.e. whether you want to replace that specific word in the whole file, or you want to replace a specific occurrence of that particular word, or you want to replace that word in a specific line. Once you have identified your particular scenario, then you can use that specific method from the examples discussed in this article.

View the full article