In my original response, I showed the test files I created with the data from your OP. You didn't say what your filenames were; I had to make up names for my files. The pm indicates it's a PerlMonks file; the 1206312 is the node ID of your OP; the in is for input. Those are fairly standard naming conventions that I use; I very much doubt you use these same conventions.

My intention was to help you learn; not to do your school/job/whatever work for free. Spend some time understanding the techniques I've used, instead of blindly copying my code and expecting it to work as is. I probably have a different directory structure to you; names I've given to test files (as seen here) won't be the same as filenames on your system; I may have used a CPAN module which you'll first need to install; there could be differences between software versions which require you to write your code slightly differently; you may even have local coding standards that you need to follow.

If you're genuinely interested in learning, then you'll need to put in some effort yourself and do some troubleshooting. Investigate how %uniq changes as the script runs: from initial (my) declaration to final (printf) output. Do the same with other variables: look at their values and see how those change over the life of the program. If, on the other hand, you just want your work done for free, you're in the wrong place: see "How (Not) To Ask A Question" and, in particular, its "Do Your Own Work" section.

All of the code that I provided is very straightforward and documented. You should be able to find information on everything I've used in http://perldoc.perl.org/perl.html: I have this link bookmarked; I recommend you do the same.