CIS 199

Final Exam

Due: Mar 21, 4:25 P.M.

This final exam for this class is a take-home final consisting of two programming projects. There are several pieces to each project, and you can earn partial credit by completing some parts of one or both projects.

Important Notes

• You must work on these projects by yourself. You cannot get design ideas, code, or debugging help from anyone else, whether they are a student in the class or someone you know or someone you communicate with via the internet.
• You are allowed to use any published resources, including the textbook(s) or information from web sites.
• To turn in the exam send e-mail to conery@cs.uoregon.edu. Send your Ruby programs as attachments so I can save them and run them.
• You can send e-mail more than once, e.g. you can send in a partially working version and then send an updated version later. I will grade the latest version of each program you send.
• You must send your e-mail by the due date (Friday Mar 21 at 4:25 PM) in order for me to use it to calculate your course grade. Grades are due Monday Mar 24 so I must have everything from you by the deadline.

Project 1: March Madness

Download games.txt, a text file that has scores from the NCAA Men's Basketball Tournament for the last four years.

Each line contains the result of a single game. There are five items on each line in this file:

• the year the game was played
• the name of one team
• the number of points scored by the first team
• the name of the second team
• the number of points scored by the second team

The file is in "tab separated values" (TSV) format, which means there is a single tab character ("\t") separating the items on each line. Here are two example lines from the file:

2006   Seton Hall      66      Wichita State   86
2006   Wisconsin-Milwaukee     82      Oklahoma        74

Note that the winning team could be either the first or second team. Note also that team names can have embedded spaces or special characters so don't assume a name consists only of letters.

Your job is to write a program that identifes the team with the best overall winning percentage. For example, if a team won 4 games and lost 2 games, they played 6 games in all and their winning percentage is 4 / 6 = 0.667.

A suggested outline is as follows (but make sure you read the important notes below first before you jump in and starting coding):

• create two hashes, one named wins and one named losses, where the keys will be team names and the values will be integers
• create a third hash named teams, where the key will be a team name (the value can be anything, e.g. the object true)
• when you determine from an input line that team x won the game and team y lost the game, add 1 to wins[x] and add 1 to losses[y]; also add both x and y to the teams hash by assigning teams[x] = true and teams[y] = true
• after reading the last line, iterate over the keys in teams , and for each team x set teams[x] to the winning percentage of that team (using the values of wins[x] and losses[x])
• keep variables named best_team and best_pct to keep track of the best winning percentage and the name of the team that has that percentage (you can use the same type of loop that you used for the original MPG program from project 2)
• after processing all teams print the name of the team with the best percentage, their won-loss record, and the percentage.

Important Notes:

• Be careful with the data types -- it's easy to make the same mistakes here that a lot of people made on BMI and other projects. Make sure you convert wins and losses to floats (because 4 / 6 = 0, but 4.0 / 6.0 = 0.667).
• Also make sure you convert scores to integers. If you leave them as strings you can easily be fooled. For example if the score was 110-88 and you don't convert scores to integers, Ruby will tell you the wrong team won! That's beacause comparing two strings compares them alphabetically, and "110" < "88" (the digit "1" comes before the digit "8" in the ASCII code).
• For extra credit you can define a class named Team and save the won-lost record for a team as attributes of a Team object. You can either do this after you get the basic version working, or, if you think the program will be easier if you use objects you can submit just a single program that uses Team objects in the main loop.
• For more extra credit sort the teams by winning percentage and print a list of the top ten teams.

Project 2: iTunes Library

When you load a CD into your iTunes library you get a new directory for the CD, and inside that directory is one file for each song. For example, when I loaded Another Day on Earth by Brian Eno I got:

% ls
01 This.m4a
02 And Then So Clear.m4a
03 A Long Way Down.m4a
...

If you then use iTunes to convert the songs to a new format you get a second copy of the file. On the Mac, the second file has the same name as the first, but with " 1" attached. For example, if I originally load the CD and have it encoded with the ALE format, and then ask iTunes to convert it to MP3, this is what I get:

% ls | cat
01 This 1.m4a
01 This.m4a
02 And Then So Clear 1.m4a
02 And Then So Clear.m4a
03 A Long Way Down 1.m4a
03 A Long Way Down.m4a
...

Since I don't want to have two copies of each song in my iTunes libary, I want to move the copies to a separate library. If I want to use this new directory as my library I can switch (e.g. if I want to load the MPGs onto my iPod). But when I do this I don't want to see " 1" in the name of the song so I should rename all the files as I move them.

Your job is to write a script named movemusic.rb that moves the extra copy of each song to a new directory, renaming the copy as it does the move. To test your program, download itunes.tgz, which is a fake iTunes library. It has file names based on files in a real library, but the files are all empty (as if you really wanted a copy of my files -- I have strange tastes).

The basic program should have two command line arguments, the name of the source directory and the name of the destination directory. For example, when you unpack the tar file, you will have directory named itunes with three subdirectories, one for each of three different CDs. To separate out the MPG files to a new directory named mp3s, type

% movemusic.rb itunes mp3s

For each directory x in the source, look to see if there are any files that end with " 1". If so, make a new directory named x in the destination, and copy all the extra files to the new directory. Rename the files in the new location so they do not have the " 1" at the end.

Some extra things you can add to your program:

• print an error message and exit if the source directory or destination directory do not exist
• print a warning message and skip a subdirectory if it already exists in the destination
• have your program see if "-f" is on the command line, and if so force the move even if the subdirectory exists -- delete the current destination subdirectory and make a new one to replace it

If you're really adventurous (and have a Mac), figure out how to invoke Applescript from within your Ruby program. Use Applescript to tell iTunes to reformat a specified CD, and then move all the duplicates.