README
functions manipulating 2d matrices I use to generate random maps
function names might sometimes seem wonky because I don't want to create confusions with legitimate mathematical terms
this library is in a very early stage and the API will change without warning until version 1.0 is released
the code should hopefully conform to ES6 specs; if it does not then it's a bug
EXAMPLES
look under t/ for examples about how to use it
BUILD
npm run build npm run test (also runs build)
FUNCTIONS
pgm_to_matrix
in
- image_data: content of a ascii PGM file
out
- new 2D array
normalize_matrix
by default it will adjust the values to a scale between 0 and 16, but you can provide other normal_min and normal_max values, and you can provide a function; the default uses normalize_int below
in
- 2d array
- normal_min // default 0
- normal_max // default 16
- normalize_function // this should be a callback
out
- new 2d array
get_min_and_max_values
returns the min and the max values in the matrix as an object, such as { min: 0, max: 255 }
in
- 2d array
out
- object, ex. { min: 1, max: 33 }
normalize_array
elements are parseInt-ed before being changed; the result contains only integers
in
- 1d array
- normal_min
- normal_max
out
- 1d array
normalize_int
see formula http://mathforum.org/library/drmath/view/60433.html ; Math.floor is applied to the result of that formula
in (all integers)
- item
- actual_min_value
- actual_max_value
- normal_min
- normal_max
out
- integer
slice2d
in
- original
- start_x
- end_x
- start_y
- end_y
out
- new 2d array
random_matrix
returns a 2d array of the specified size with random numbers between 0 and 1 as values; if a callback is provided that value will be passed though it and the result used in the cell
in
- x : integer
- y : integer
- callback = (x) => x)
out
- new 2d array
resize
creates a new 2d array with extra values (as returned by the fill() callback ) and the original data starting with the specified offset;
allow_cropping permits allows the new size to be smaller than the original or the offset values to be negative so only part of the original 2d array values are preserved in the result
fill(x,y) is function expecting the coordinates of the element and returning a value to be set to each element not existing already in the original
in
- original : 2d array
- offset : object, { x: someValue, y: someOtherValue}
- new_size : object, { x: 10, y:10 }
- fill : fill(x, y)
- allow_cropping: boolean
out
- new 2d array
stretch
similar with scaling using bilinear interpolation, but simpler; values on the edges are kept on the edges; not using "scale" in the name so it will not create the wrong expectations
in
- orig : 2d array of integers
- new_width : int
- new_height : int
out
- new 2d array
add_matrices
in
- first: 2d array
- second: 2d array same size as first
out
- 2d array, each element sum of the elements with the same indices in the arguments
moving_average
not weighted, only doing plain average of the cells within specified distance
in
- 2d array
- distance
out
- new 2d array
sum_elements
adds all elements in the array
in
- 2d array
out
- number
replace_elements
replace elements in the matrix based on a filter function which returns a boolean
in
- matrix,
- filter = (i) => i < 0,
- transform = (i) => Math.abs(i)
out
- new 2d array with the matching elements replaced with the result of the transform function
are_equal
will throw Error if any of the parameters is not a 2d array
in
- first
- second
out
- boolean
is_2d_array
in
- 2d array
out
- boolean
squeeze TODO
downscale 2d arrays
enlarge TODO
similar with scaling using nearest neighbor interpolation but hopefully simpler
in
- orig: 2d array of integers
- new_width
- new_height
out
- new 2d array
bundle TODO
creates a new 2D array from a bunch of smaller 2d arrays by placing them one besides the other with offsets based on a single origin point and custom spacing between them
stack TODO
creates a new 2D array of specified size from a bunch of smaller 2d arrays by placing them one on top of another at specified offsets, adding the values of the overlapping elements then using that value and the coordinates in the new array to compute new values for the final result
DEPENDENCIES
Production
none
Development
- eslint
- tap
- webpack
- webpack-cli
- pngjs