Linux Diff Command – Options and Examples

diff [option] file1 file2

• Normal format- This is the default output that is displayed when you compare two files without specifying any other option.
• Context format- This output format displays lines of context about the text that is different in the compared files.
• Unified format- This output format is an improved form of the context format output and it displays a smaller output.

To better explain how the Linux diff command works, we are creating two sample files that we will use in the examples in this tutorial:

1. Create the first file capitals.txt using the nano editor:

nano capitals.txt

2. Add the following lines to the first file:

Beijing
Tokyo
Mumbai
Moscow

3. Save the changes and exit.

4. Create the second file cities.txt using the nano editor:

nano cities.txt

5. Add the following lines to the second file:

New York
Los Angeles
Chicago
Huston
Phoenix

6. Save the changes and exit.

1. Compare in Normal Format

To compare the files in the normal format, specify the file names after the diff command:

diff capitals.txt cities.txt

The normal diff command output provides instructions on modifying the first file to make it identical to the second file. Let’s review the output and translate the instructions in human-readable format:

• 1d0 – You must delete the first line from the first file, otherwise, it will seem in line number 0 in the second file.
• < Beijing – The text that needs to be deleted, is specified as 1d0.
• 3c2 – In line number 3 of the first file, you must change line 2 from the second file.
• < Tokyo – The text you need to change.
• > Los Angeles – The text you need to change it to.
• 5c4,5 – In line 5 of the first file, you must change line numbers 4 and 5 from the second file.
• < Moscow – The text you need to change.
• > Huston – The text you need to change it to.
• > Tokyo - The text you need to change it to.

You can use the patch command after generating the output instructions to save the diff command output and apply the changes.

2. Compare in Context Mode

The default output of the diff command is quite complex to understand. You can modify this output to the context mode to make it easy to understand. Use the -c option with the diff command to display lines of context around the content that is different:

diff -c capitals.txt cities.txt

In this output, the lines starting with *** display information about the first file, and the lines starting with --- display information about the second file. The initial two lines of the output of the diff command display file names and modification dates. Then we have the ********** separator.

After that, we have the line range (1,5) and differences for the first file. The initial symbol indicates the change that we need to make to the first file and the rest of the line is the content. You can see the following three symbols:

• - or minus – You need to delete this content.
• ! or exclamation mark – You need to replace this content with the corresponding line in the second file.
• + or plus – You need to add this text to the first file.

If a line does not start with a symbol, then it does not require any modification.

3. Compare in Unified Format

Use the -u option to display the output in a unified format where the redundant context lines are ignored:

diff -u capitals.txt cities.txt

In this output, the lines starting with --- display information about the first file and the lines starting with +++ display information about the second file. Below that, @@ -1,5 +1,5 @@ represents the line numbers range of both the files.

The next lines display the list of changes that you need to make to make both the files identical. You can see the following two symbols:

• - or minus – You need to delete this content.
• + or plus – You need to add this text to the first file.

If a line does not start with a symbol, then it does not require any modification.

4. Ignore Case Using -i Option

The diff command in Linux, like all other Linux commands, is case-sensitive by default. This means that if the only difference between the specified file is the casing of the text, then also you will see the difference in the files. However, you can use the -i option, to ignore the case difference in the content in the two files.

For example, let's create a new file named metros.txt:

nano metros.txt

Then add the following city names to this file with two entries starting with a lowercase letter:

Beijing
New York
tokyo
Mumbai
moscow

Without any additional option, the output of the diff command shows that these files are different and lists the changes we need to make them identical:

Use the -i option to ignore casing and display results that are different in content, not just the case:

Now we don’t see any differences because the content is not different.

5. Using Patch with Diff

In a production environment, the files are updated frequently, and it is extremely difficult to keep track of all these changes. Using the diff command in combination with the patch provides you an option to directly apply the changes to the target files.

The first step is to save the differences between two files into a new file:

diff -u capitals.txt cities.txt > correction.txt

Then we check the content of the correction.txt file:

cat correction.txt

Once we have the list of changes, you can apply the patch to the second file as follows:

patch cities.txt < correction.txt

Check the contents of the second file to make sure it is identical to the first file:

cat cities.txt

Conclusion

In this tutorial, we learned how to use the diff command in Linux to compare two files and list the differences. We have included multiple examples in this article that will help you better understand how the diff command works. You can also use the patch in combination with the diff command to automate the changes to make the compared files identical.

If this resource helped you, let us know your care by a Thanks Tweet. Tweet a thanks

Did you find this article helpful?

We are glad you liked the article. Share with your friends.

Sorry about that.

How can we improve it?

About The Author

Subhash Chandra

Subhash Chandra is a professional writer and an Oracle-certified Database Administrator with over 15 years of writing experience. He has a passion for technology and loves to write how-to articles for Linux, Windows, Mac OS, Networking, Telecom, and Database. In his spare time, he enjoys swimming and playing table tennis.

LinkedIn