maia-util/es6/unique_rows.js

Utility math and music functions supporting various applications by Music Artificial Intelligence Algorithms, Inc.

import array_equals from './array_equals'
import array_approx_equals from './array_approx_equals'
import sort_rows from './sort_rows'

export default function unique_rows(point_set, approx = false){
  // Tom Collins 16/12/2014.
  // In
  // point_set Array mandatory
  // approx Boolean optional
  // Out Array
  // The first argument to this function is an array consisting of numeric
  // arrays of uniform dimension (what I call a point set). The second
  // (optional) argument approx determines whether arrays are compared for
  // equality up to some level of tolerance (approx = true) or not (approx =
  // false). This function returns the unique elements in lexicographic order as
  // first argument. As second argument are the indices of each element from the
  // input array.

  // Catch empty arrays.
  var n = point_set.length;
  if (n == 0){
    return [[], []];
  }

  // Sort the rows of a copy of the dataset.
  var E = sort_rows(point_set);
  var F = E[0];
  var g = E[1];
  // Create a new variable that will contain the unique rows of the dataset.
  var k = point_set[0].length;
  var U = new Array(n);
  // Create a new variable that will contain the indices of each unique row in
  // the original dataset.
  var v = new Array(n);
  U[0] = F[0];
  v[0] = [g[0]];
  var i = 1; // Increment over F and g.
  var j = 1; // Increment over U and v.
  if (approx){
    while (i < n){
      if (array_approx_equals(F[i],F[i - 1])){
        v[j - 1].push(g[i]);
      }
      else{
        U[j] = F[i];
        v[j] = [g[i]];
        j++;
      }
      i++;
    }
  }
  else {
    while (i < n){
      if (array_equals(F[i],F[i - 1])){
        v[j - 1].push(g[i]);
      }
      else{
        U[j] = F[i];
        v[j] = [g[i]];
        j++;
      }
      i++;
    }
  }
  return [U.slice(0, j), v.slice(0, j)];
}