Skip to content

report_1.md

Latest commit

 

History

History
83 lines (74 loc) · 6.05 KB

report_1.md

File metadata and controls

83 lines (74 loc) · 6.05 KB

Report Part 1

By Henk Bierlee (4148053) and Kilian Grashoff (4171373)

3.1 End-to-End Testing

Exercise 1

The overall coverage varies between 76.9% and 77.4% (LauncherSmokeTest is not consistent between multiple runs).

Examples of classes that are not well-tested are CollisionInteractionMap and DefaultPlayerInteractionMap. These classes are data structures, which are only used by other code, not directly by the user. Therefore, the smoke test does not cover these classes.

Exercise 2

See package nl.tudelft.jpacman.endToEnd.

Exercise 3

Scenario 2.3 is the hardest to test, since a ghost needs to be next to the player, which can only be achieved by sleeping the thread, hard coding ghost paths or editing the board. We chose the first option.

Scenario 2.5 is also difficult, since programming a test to complete the game should be very difficult without altering the board. However, there is a way to cheat, since there is a publicly accessible method to remove Units from Squares, so all the pellets except one next to the Player can be removed, after which the game can be won in one move.

Exercise 4

Story 3 is hard to test because many methods relating to Ghost are private, since ghosts are moved by the game in normal gameplay, not by the player. Therefore many of the functionality is contained in private methods. We tested scenario 3.1 as an extra exercise.

3.2 Boundary Testing

Exercise 5

For the boundary test, a Board instance with a 10x10 grid was constructed, populated with only BasicSquare objects. Because of this, the boundaries in the actual game (getWidth() and getHeight()) have been replaced with the number 10.

Boundary conditions for x >= 0 && x < 10 && y >= 0 && y < 10

Variable Condition Type t1 t2 t3 t4 t5 t6 t7 t8
x >= 0 on 0
off -1
< 10 on 10
off 9
typical in 8 1 5 6
y >= 0 on 0
off -1
< 10 on 10
off 9
typical in 2 6 4 5
Result TRUE FALSE FALSE TRUE TRUE FALSE FALSE TRUE

Exercise 6

See nl.tudelft.jpacman.board.ParameterizedBoardTest. Checkstyle warnings are explained below.

3.3 Submit Part 1

Code submission readiness

Our tests test all user stories exactly as described, contains proper commenting and javadoc, and contains no Checkstyle or Findbugs violations except for the ones explained in the next paragraph.

Checkstyle and Findbugs violations

Checkstyle reports the following violations:

Severity Category Rule Message Line
Warning coding MagicNumber '6' is a magic number. 69
Warning coding MagicNumber '10' is a magic number. 70
Warning coding MagicNumber '4' is a magic number. 70
Warning coding MagicNumber '9' is a magic number. 70
Warning coding MagicNumber '5' is a magic number. 70
Warning coding MagicNumber '8' is a magic number. 70
Warning coding MagicNumber '5' is a magic number. 71
Warning coding MagicNumber '10' is a magic number. 71
Warning coding MagicNumber '6' is a magic number. 71
Warning coding MagicNumber '9' is a magic number. 71
These violations are all in the ParameterizedBoardTest.data() method and cannot be avoided in this test, since the magic numbers are our test parameters.

Log of test results

Running nl.tudelft.jpacman.endToEnd.MovePlayerTest
Tests run: 5, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 5.32 sec
Running nl.tudelft.jpacman.endToEnd.StartGameTest
Tests run: 1, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.039 sec
Running nl.tudelft.jpacman.endToEnd.SuspendGameTest
Tests run: 2, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.147 sec
Running nl.tudelft.jpacman.endToEnd.MoveGhostTest
Tests run: 1, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.038 sec
Running nl.tudelft.jpacman.board.BoardTest
Tests run: 1, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0 sec
Running nl.tudelft.jpacman.board.ParameterizedBoardTest
Tests run: 8, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.012 sec
Running nl.tudelft.jpacman.board.DirectionTest
Tests run: 4, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.003 sec

Results :

Tests run: 22, Failures: 0, Errors: 0, Skipped: 0

Additional adequacy in jpacman-framework

Since we have not altered jpacman-framework all additional adequacy stems from extra guarantees about the code provided by our tests.

  • StartGameTest guarantees that the game starts correctly.
  • MovePlayerTest guarantees that the player is able to move, and that moving results in gameplay events (such as consuming pellets or dying).
  • MoveGhostTest guarantees that ghosts are able to move to empty cells. The rest of these scenarios has not been implemented, since they are optional.
  • SuspendGameTest guarantees that the game can be suspended correctly, and restarted afterwards.
  • ParameterizedBoardTest is a unit test, which guarantees that the Board.withinBorders() method functions as intended, according to the one-by-one domain testing strategy.