This branch of the project uses no external libraries and, as such, uses nested
lists as an array-like object to store the state of the game. The use-numpy
branch uses the arrays from the numpy library instead. It also uses the
matplotlib library to make an animated demonstration of a Gosper Glider Gun.
My task was to implement the rules of Conway's Game of Life as explained in the document statement.
The document which accompanied this task contained 6 scenarios describing the rules of Conway's game of life. The first and fifth of these scenarios were implemented as a unit test case in the project skeleton. This inspired me to add test cases for the other 4 scenarios and apply a test-driven approach to solving the problem.
The expected behaviour for cells at the edges or corners of a Game of Life world was not specified in the document so I added a boolean option to the World class which would allow the user to switch between two different types of border behaviour. Namely whether a cell would be considered to have no neighbours outside of the world or whether the borders would wrap around to create a closed world.
In order to count the number of neighbours a given cell has a 3x3 sub-section of the game state centred on that cell is generated. For cells not on the edges of the world this is just a small section of the game state. For cells on the edges or in the corners the method depends on the boundary wrapping setting:
- For non-wrapping borders any 'neighbour' lying outside of the game state is not counted towards the number of neighbours. This is whenever a cell with either a -1 or out of range index is looked for.
- For wrapping borders a -1 index is used to access the last element of a list which causes a wrap around. An index which is past the end of the list is set to 0, this causes the wrapping in the other direction.
- It is possible to develop a way of visualising the Game of Life using only standard libraries. A renderer should be relatively simple to make with tkinter.
- The rules of Conway's Game of Life are just 1 out of many possible rules for this kind of cellular automaton. It might be interesting to allow rules to be specified.