Archive: January 17, 2006

Tuesday, January 17, 2006

Unix command line scripting question

I'm wondering if it's possible to do the following just using unix command line utilities. Given a text file with duplicate lines, give a printout of all unique lines in the file and a count of how many copies there are of each.

Example file:

foo
bar
foo
baz
bar


Example output:

foo    2
bar    2
baz    1


It'd be trivial for me to write a script to accomplish this. I'm just wondering if there's a simple way to do it with sort, uniq, and awk or whatever.

Update: Solutions in the comments (thanks!).

Also, that should probably be "bar 2, foo 2, baz 1" for that last example, since "bar" comes before "foo" alphabetically. Though, ironically the order I gave is what you get with the straightforward sort -rn.

Permalink | 9 Comments | Tags: Programming | id 7859

14:39 (utc-6)
edited: 2006-01-18 07:17 (utc-6)

 

1. Adam Langley on Factor:

   Factor - a programming language; very Forth like. I'm playing with it when I get a chance. I'm still not sure about stack based languages. They have nice advantages: factoring out functions is stupidly easy and when the data flow works, it's very elegant. But one cannot understand only part of a function (which is why each function must be small) and, when dealing with > 3 variables, the stack fuzz is crazy. None the less, factor has a good environment (based on jEdit, by the same author) and Erlang like concurrency.

   I've had a lot of interest in Factor lately and have been following its development. I don't have any experience in stack languages, so I want to learn its way of thinking.

     ¶ (3) Tags: [Programming]