As you’ve seen over the course of the last two “Power Tools” columns, one reason to use the command-line GNU diffutils is that, compared to windowed difference programs, diffutils can give you much more control over how variations are found and displayed. Another compelling argument is that some diffutils can create new files from merged versions of other files - either interactively, or by using the output of one utility as input to another. You can also use diff --from-file to compare one file to two or more other files. patch can also merge files automatically.
The diff3 utility compares three files, and can also merge the changes from two other files into a common ancestor file. Let’s dig in and see how it works.
What’s different about diff3?
Most diffutils compare two files, but diff3 compares three. It’s often used to compare two differing files that are both based on a common ancestor. For instance, if two programmers copy and then modify the same program file, diff3 could show both sets of changes.
diff3 can also merge changes from two files into a third file - typically in a case like the one we just mentioned. The options -m and --merge do this.
The -e and --ed options create an ed script that lets you merge changes into a separate file. ed scripts were common before patch came on the scene, and remain effective. However, because ed scripts generally use absolute line numbers, such scripts can cause…
Please log in to view this content.
Not Yet a Member?
Register with LinuxMagazine.com and get free access to the entire archive, including:
