Linux的sort命令如何将后续传递应用于数据？

Question

There's an example in A Practical Guide to Linux Commands, Editors, and Shell Programming that looks like this.

You have a file, "fruit", that contains the following:

Pear
Pear
apple
pear
Apple

Executing this command:

sort -u -k 1f -k 1 fruit

Results in this output:

Apple
apple
Pear
pear

Why does this work? If I execute only the first pass:

sort -u -k 1f fruit

This results in:

apple
Pear

The first pass appears to eliminate lines that somehow reappear on the second pass of the former example. Also, since the first pass results in unique lines, I would expect the second pass not to happen at all.

Answer 1

By using secondary sorting key, you're telling sort "if the fields are the same, use this to compare them". So, -k1f sees Apple and apple as equal, so calls -k1 to compare them. The result isn't "equal", so -u doesn't remove anything.