Copyright 2016 Dean Attali. Licensed under the MIT license.
Lights Out is a puzzle game consisting of a grid of lights that are either on or off. Pressing any light will toggle it and its adjacent lights. The goal of the game is to switch all the lights off. This package provides an interface to play the game on different board sizes, both through the command line or with a visual application (Shiny app). Puzzles can also be solved using the automatic solver included. Play the game either by downloading the package or online at https://daattali.com/shiny/lightsout/.
Need Shiny help? I’m
available for consulting.
If you find
{shinyjs} useful, please consider supporting my work to
unlock rewards! ❤
lightsout
provides a Shiny app that lets you play with a
graphical user interface. To run the game locally, install the package
with install.packages("lightsout")
and run the
lightsout::launch()
command. This will launch the provided
app in a web browser.
Alternatively, you can see the app online at https://daattali.com/shiny/lightsout/.
The game looks like this:
In this image, there is a 5x5 Lights Out board. The darker green lights represent lights that are off, and the brighter green lights are lights that are on. Clicking on the Show solution button will highlight all the lights that need to be pressed in order to solve the current board.
Other than the Shiny app that lets you play the game visually, you
can also interact with lightsout
using the command
line.
You can use the random_board()
function to initialize a
new Lights Out game with a random configuration of lights. The generated
board is guaranteed to be solvable. The first argument to the function
is the board dimensions (number of rows and columns), which is
restricted to be either 3, 5, 7, or 9.
#> Lights Out 3x3 board
#> Game mode: classic
#>
#> 0 0 1
#> 1 1 1
#> 1 1 0
#>
#> Lights Out 5x5 board
#> Game mode: classic
#>
#> 0 0 0 0 1
#> 1 0 1 1 0
#> 1 1 1 1 1
#> 0 0 0 0 0
#> 0 0 1 1 0
#>
You can also use the new_board()
function to create a
new board if you want to provide your own defined set of lights. The
first argument is either a matrix or a vector of the lights, with all
values being either 0 (light off) or 1 (light on). If a vector is
provided, then it is read row-by-row rather than column-by-column.
#> Lights Out 3x3 board
#> Game mode: classic
#>
#> 0 0 0
#> 0 1 0
#> 1 1 1
#>
Any board generated by random_board()
is guaranteed to
be solvable, but since we defined this board ourselves, it’s a good idea
to make sure it’s solvable before attempting to play it.
#> [1] TRUE
Pressing a light is done with the play()
function.
Pressing a light will cause that light and all its adjacent lights to
toggle (in classic mode, which is the default). All the coordinates in
lightsout
are treated as (row,column). Let’s press the
light at (2,3) - row 2, column 3
#> Lights Out 3x3 board
#> Game mode: classic
#>
#> 0 0 1
#> 0 0 1
#> 1 1 0
#>
Notice how the light at (2,3) and its 4 neighbours are flipped.
If you want to play the non-classic game mode, where pressing a light
toggles the entire row and column, then use the
classic = FALSE
argument of either new_board()
or random_board()
. To demonstrate, here is the same board
initialized as non-classic mode, with the same light being pressed.
#> Lights Out 3x3 board
#> Game mode: entire row/column
#>
#> 0 0 1
#> 1 0 1
#> 1 1 0
#>
Note that this specific board does not actually have a solution in non-classic mode, which means that no combination of lights being pressed will result in all lights being off:
#> [1] FALSE
Notice also that lightsout
works well with the
%>%
operator that allows you to easily chain calls.
By chaining calls, you can press multiple lights easily one after the other:
#> Lights Out 3x3 board
#> Game mode: classic
#>
#> 1 1 0
#> 0 1 1
#> 1 1 0
#>
By the way, when pressing multiple lights, the order does not matter.
Pressing y
after x
will always have the same
result as pressing x
after y
.
If you want to press multiple lights in one call, you can pass a vector of row numbers and a vector of column numbers instead of a single row and column. The previous chain of presses can also be achieved with:
#> Lights Out 3x3 board
#> Game mode: classic
#>
#> 1 1 0
#> 0 1 1
#> 1 1 0
#>
Another way to press many lights with one call is to use a matrix
instead of rows and columns. The matrix must have the same dimensions as
the board, and every position in the matrix with a value of
1
means that the light in that position will be
pressed.
# Press light at (2,3) and (1,2)
board1 %>% play(matrix = matrix(nrow = 3, byrow = TRUE,
c(0, 1, 0,
0, 0, 1,
0, 0, 0)))
#> Lights Out 3x3 board
#> Game mode: classic
#>
#> 1 1 0
#> 0 1 1
#> 1 1 0
#>
Now let’s try to actually solve the board. The board we have is actually super easy - it should be obvious that a single press at (3,2) will turn off all the lights.
#> Good job, you won!
If you have a more complex board, it will generally require many more presses and will not be as trivial to solve. Let’s create a random board of size 5.
#> Lights Out 5x5 board
#> Game mode: classic
#>
#> 1 0 0 0 1
#> 1 1 0 0 1
#> 1 1 0 1 1
#> 0 0 0 0 1
#> 1 0 0 1 0
#>
This one isn’t so simple, is it? You can try playing with it for a bit.
If you want to see the solution to a board, use the
solve_board()
function:
#>
#> 0 1 1 1 0
#> 0 0 1 0 0
#> 1 1 0 0 1
#> 1 1 0 0 0
#> 1 1 1 0 0
#>
The solution is a matrix with the same dimensions as the board, with
the values of 1
representing the lights that need to be
pressed. We can verify that the solution does indeed solve the current
board:
#> Good job, you won!
There are a few algorithms for solving Lights Out puzzles. This
package (specifically the solve_board()
function)
implements the Gaussian Elimination technique, which uses linear algebra
to solve a matrix equation Ax=b
in order to derive the
solution. The matrix operations are done in modulus 2, with
b
being the current board configuration, x
being the solution vector, and A
is a special matrix that
depends on the board size and game mode. The reason that only boards
with an odd number of rows/columns are supported is simply because I
couldn’t figure out how to derive the special matrix A
for
even-sized boards.
The Gaussian Elimination strategy does not guarantee the minimum number of steps, and therefore some steps in the suggested solution may be redundant. If you are interested in more details about how this puzzle is solved mathematically, you can look at the source code or look at resources online for the exact details of the algorithm and the mathematical theory behind it.
I hope you enjoy this little game fully implemented in R, or get inspired to create Shiny games!