diff command can be used to analyze two files and output the lines that are different. It also can output a patch file which can be used to make one file be identical to the other.

Basic Usage

fool@bar:~$ diff file1 file2

Here are some very useful options.

• -b: ignore changes in the amount of white space
• -B: ignore changes whose lines are all blank
• -w: ignore all spaces

Output Format

The basic usage only outputs the different lines. The context of the different lines can also be output by using -c or -u.

• -c: context mode, output 3 lines context for both files
• -u: unified mode, output 3 lines context only once

The result can be output side by side by -y option. There are two additional options in side by side mode.

• --left-column option prints only the left column of two common lines.
• --suppress-common-lines option suppresses common lines entirely.

Create Editing Scripts

The output of diff can be used by the editing program patch to modify the contents of file1, so that it matches the contents of file2.

fool@bar:~$ diff file1 file2 > patchfile.patch
fool@bar:~$ patch file1 -i patchfile.patch -o updatedfile

The patch file can also be used to reverse the change.

fool@bar:~$ patch updatedfile -R -i patchfile.patch -o newName

For two versions of a large project, the directories can be compared.

fool@bar:~$ diff -r dir1 dir2 > patchfile.patch
fool@bar:~$ patch -p0 -i patchfile.patch
fool@bar:~$ patch -p0 -R -i patchfile.patch

Reference

computerhope linuxacademy