Description
In this assignment you will parse, compute, and study statistics from an obscure video-based endeavor called
“League of Legends”. As it is unlikely that you know anyone who has played this sport, we will summarize
all of the rules that you need to know to complete this homework. We certainly do not recommend you
engage in such activities before you have completed your homework!
A head-to-head match in League of Legends consists of two teams of 5 players. At the start of the match,
each player selects a champion from a list of (currently) 134 champions, each with their own abilities. Players
may coordinate to select champions with complementary strengths to improve the performance of their team
in the match. During the match, each player controls his or her champion character and battles with players
on the opposite team and also computer-controlled minion characters. When combat results in the death
of a champion, that player must wait a small period of time before their champion re-spawns and they can
re-join the match. Credit for the kill of a champion character is given to the appropriate player on the other
team, or in some cases to a general game minion. One or more of the teammates may also be given credit
for assisting. The game ends when one team destroys the other team’s Nexus.
More details on the game are available here: https://en.wikipedia.org/wiki/League_of_Legends. Data
for this homework comes from actual gameplay, made publicly available by the game author, Riot Games,
Inc.: https://developer.riotgames.com/docs/getting-started. We have heavily reduced these very
large and detailed datasets to a minimum set of information.
Here is an example of the input format for this homework (some event lines are omitted for space):
MATCH ID 1778973240
LOSING TEAM
iTcnRKaneki playing champion Syndra
T1Zed playing champion Kalista
IamZevahc playing champion Thresh
SushiNChopstix playing champion Riven
BattlecastLulu playing champion Udyr
WINNING TEAM
Dopy playing champion Graves
komimatt playing champion Brand
SRVaRiaX playing champion Lissandra
Diddydan playing champion Zac
Jrazo playing champion Blitzcrank
EVENTS
@ 277230 BattlecastLulu [ SushiNChopstix ] killed SRVaRiaX
@ 432589 Dopy [ Jrazo ] killed T1Zed
@ 444272 SushiNChopstix [ T1Zed IamZevahc ] killed Dopy
@ 448804 minion killed SushiNChopstix
@ 519016 T1Zed [ IamZevahc ] killed Dopy
. . .
@ 2410278 Dopy [ komimatt SRVaRiaX Jrazo ] killed BattlecastLulu
END
First we see the MATCH ID, then the players and champions for each team are listed. The teams are labeled
“WINNING” and “LOSING”, indicating the final result of the match. The highlights of all champion kill events
are listed chronologically. Each of these events begins with the ’@’ character, then a timestamp (in milliseconds
from the start of the match), the name of the player given primary credit for the kill, the assistants (if any)
listed in square brackets, and finally the name of the victim. At the end of the events list is the keyword
“END”. An input file will contain one or more matches.
File I/O and Command Line Arguments
Your program will expect 3 required command line arguments. The first and second are the names of the
input and output files, respectively. The third argument is the name of the output table that should be
printed, one of “players”, “champions”, or “custom”. For example, here are some valid command lines to
your program:
./lol_stats.out input_simple.txt output_simple_players.txt players
./lol_stats.out input_small.txt output_small_champions.txt champions
./lol_stats.out input_medium.txt output_medium_custom.txt custom
Statistics Collected and Output
The output will be one of three different tables, as specified on the command line.
First, we study the “players” table, containing all players who participated in one or more matches in the
input file. Each row contains the number of kills credited to this player, the number of times they died and
re-spawned during the match, the kill-to-death ratio (KDR), and the names of the champion(s) used by that
player. NOTE: To avoid divide by zero, if a player’s deaths is zero, then the KDR is equal to the kills. If a
player played more than one champion in different matches, the names of the champions should be sorted
alphabetically, separated by commas (without duplicates). Overall, the rows of this table are sorted by KDR.
If two players are tied in KDR, they are sorted by kills (highest first), then by deaths, and if tied in all of
the above, they should be alphabetically by player name. Note: We will simply use the less than < operator
on STL strings, which places numbers before capital letters, and capital letters before lowercase letters.
Here is an example of this output:
PLAYER NAME KILLS DEATHS KDR PLAYED WITH CHAMPION(S)
squirT0T 6 1 6.00 Amumu
komimatt 7 2 3.50 Brand
KelloggsBRNFLKZ 10 3 3.33 Sejuani
Diddydan 3 1 3.00 Zac
AntiToro 13 6 2.17 Tristana
Dopy 10 6 1.67 Graves
Jrazo 4 3 1.33 Blitzcrank
alittlemido 5 4 1.25 Thresh
OMGLeiChen 9 10 0.90 Graves
Jovone 7 8 0.88 Nasus
SRVaRiaX 6 7 0.86 Lissandra
iTcnRKaneki 6 7 0.86 Syndra
Stylether 10 12 0.83 LeBlanc
SushiNChopstix 6 8 0.75 Riven
T1Zed 6 8 0.75 Kalista
Indovaasion 5 10 0.50 Fizz
Zraotic 4 9 0.44 Warwick
BattlecastLulu 1 3 0.33 Udyr
IamZevahc 0 5 0.00 Thresh
Zyyke 0 7 0.00 Taric
Next, we study the “champions” table, which contains all champions that were used in one or more matches
in the input file. This table contains the number of wins & losses, the corresponding win percentage, and the
number of times that champion was killed not by another champion, but by a minion (rather embarrassing if
you think about it…). Overall this table is sorted by win percentage, highest first. If tied in win percentage,
champions should be sorted by wins, then by losses, and finally by name if all of the above are equal.
And here is an example of this type of table:
2
CHAMPION NAME WINS LOSSES WIN% MINION DEATHS
Amumu 1 0 1.00 0
Blitzcrank 1 0 1.00 0
Brand 1 0 1.00 0
LeBlanc 1 0 1.00 1
Lissandra 1 0 1.00 0
Sejuani 1 0 1.00 0
Tristana 1 0 1.00 0
Zac 1 0 1.00 0
Graves 1 1 0.50 0
Thresh 1 1 0.50 0
Fizz 0 1 0.00 0
Kalista 0 1 0.00 0
Nasus 0 1 0.00 0
Riven 0 1 0.00 1
Syndra 0 1 0.00 0
Taric 0 1 0.00 0
Udyr 0 1 0.00 0
Warwick 0 1 0.00 0
Finally, the “custom” table is a chance for you to be creative. You will collect, organize, and output some
other statistic from the provided match data. Notice that the earlier tables didn’t use the event timestamp,
or the champion kill assists. You may be interested in exploring that data. Can you identify champions who
are good supporting team members (many assists, but not many kills)? Can you identify the best carry
players, who are strong late in the match and help carry the team to victory? Or perhaps you can study
pairs of champions who are more successful when they play on a team together. Or find champions who are
particularly susceptible to a team with a specific enemy champion.
The most important task for this part of the assignment is to write a concise description (∼ 100 words) of
your new statistic. What is your hypothesis for the patterns that might emerge from your statistic? Put this
description in your README.txt along with any other notes for the grader. Be sure to tell the grader which
dataset best demonstrates your new statistic. You may also create your own dataset and include it and
your program’s output for that test case with your submission. Extra credit will be awarded to particularly
interesting statistics that require clever programming.
Useful Code
To control the formatting of your tables, you’ll want to read up on the various iomanipulators: std::setw(int),
std::setprecision(int), std::fixed, std::left, and std::right. And don’t forget about the sort
function that can be used to order the contents of a vector.
We encourage you to use the << input stream function for strings to implement all of the parsing necessary
for this assignment. Be sure to study the more complex parsing example on this website: Misc. C++
Programming Information. Note: You do not need to use getline or eof. In fact, your code should not rely
on the position of newlines or whitespace within the input file.
Program Requirements & Submission Details
Your program should involve the definition of at least one well-designed C++ class that has its own .h and
.cpp files, named appropriately.
Be sure to read the “Homework Grading Criteria” as you put the finishing touches on your solution. Use the
provided template README.txt file for notes you want the grader to read. You must do this assignment on
your own, as described in the “Collaboration Policy & Academic Integrity” handout. If you did discuss this
assignment, problem solving techniques, or error messages, etc. with anyone, please list their names in your
README.txt file.
3
