I need to solve a cubic equation (ax^3 + bx^2 + c*x + d = 0) analytically and in real numbers, preferably in pure javascript (no libs). As there could be 1 to 3 roots, I think an array of numbers is a reasonable result type.
P.S. Provided my own solution below, hope it'll be useful.
Here you go. Includes handling degenerate cases. Main algorithm is mostly from wikipedia article.
function cuberoot(x) {
var y = Math.pow(Math.abs(x), 1/3);
return x < 0 ? -y : y;
}
function solveCubic(a, b, c, d) {
if (Math.abs(a) < 1e-8) { // Quadratic case, ax^2+bx+c=0
a = b; b = c; c = d;
if (Math.abs(a) < 1e-8) { // Linear case, ax+b=0
a = b; b = c;
if (Math.abs(a) < 1e-8) // Degenerate case
return [];
return [-b/a];
}
var D = b*b - 4*a*c;
if (Math.abs(D) < 1e-8)
return [-b/(2*a)];
else if (D > 0)
return [(-b+Math.sqrt(D))/(2*a), (-b-Math.sqrt(D))/(2*a)];
return [];
}
// Convert to depressed cubic t^3+pt+q = 0 (subst x = t - b/3a)
var p = (3*a*c - b*b)/(3*a*a);
var q = (2*b*b*b - 9*a*b*c + 27*a*a*d)/(27*a*a*a);
var roots;
if (Math.abs(p) < 1e-8) { // p = 0 -> t^3 = -q -> t = -q^1/3
roots = [cuberoot(-q)];
} else if (Math.abs(q) < 1e-8) { // q = 0 -> t^3 + pt = 0 -> t(t^2+p)=0
roots = [0].concat(p < 0 ? [Math.sqrt(-p), -Math.sqrt(-p)] : []);
} else {
var D = q*q/4 + p*p*p/27;
if (Math.abs(D) < 1e-8) { // D = 0 -> two roots
roots = [-1.5*q/p, 3*q/p];
} else if (D > 0) { // Only one real root
var u = cuberoot(-q/2 - Math.sqrt(D));
roots = [u - p/(3*u)];
} else { // D < 0, three roots, but needs to use complex numbers/trigonometric solution
var u = 2*Math.sqrt(-p/3);
var t = Math.acos(3*q/p/u)/3; // D < 0 implies p < 0 and acos argument in [-1..1]
var k = 2*Math.PI/3;
roots = [u*Math.cos(t), u*Math.cos(t-k), u*Math.cos(t-2*k)];
}
}
// Convert back from depressed cubic
for (var i = 0; i < roots.length; i++)
roots[i] -= b/(3*a);
return roots;
}
Related
Lets say I have the following array of strings, var = array_of_strings["abc","abcd"]
My goal is to run a function and have this return roughly 75% (0.75). Implying that the results are roughly 75% in common. Roughly being defined as within a certain error range, let us say 5% or some settable number.
I'm currently using the the Levenshtein algorithm to compute differences in the strings, however, this is extremely slow and taxing on the CPU in my situation as the strings I'm using are thousands and thousands of lines long.
Levenshtein gives me what the differences are; and while useful in certain situations, my particular use case is simply looking to see what percentage the strings are roughly different from each other and not what each difference actually is necessarily.
The current levenshtein algorithm I'm using is below (which I borrowed from another answer here on stackoverflow). It will return how many differences it found which I can then use to calculate a percentage difference, but it's very slow! Sometimes taking a couple of seconds to run and freezes up the computer as well.
async function levenshtein(s, t) {
return new Promise((resolve, reject) => {
console.log("levenshtein active");
if (s === t) {
return 0;
}
var n = s.length, m = t.length;
if (n === 0 || m === 0) {
return n + m;
}
var x = 0, y, a, b, c, d, g, h, k;
var p = new Array(n);
for (y = 0; y < n;) {
p[y] = ++y;
}
for (; (x + 3) < m; x += 4) {
var e1 = t.charCodeAt(x);
var e2 = t.charCodeAt(x + 1);
var e3 = t.charCodeAt(x + 2);
var e4 = t.charCodeAt(x + 3);
c = x;
b = x + 1;
d = x + 2;
g = x + 3;
h = x + 4;
for (y = 0; y < n; y++) {
k = s.charCodeAt(y);
a = p[y];
if (a < c || b < c) {
c = (a > b ? b + 1 : a + 1);
}
else {
if (e1 !== k) {
c++;
}
}
if (c < b || d < b) {
b = (c > d ? d + 1 : c + 1);
}
else {
if (e2 !== k) {
b++;
}
}
if (b < d || g < d) {
d = (b > g ? g + 1 : b + 1);
}
else {
if (e3 !== k) {
d++;
}
}
if (d < g || h < g) {
g = (d > h ? h + 1 : d + 1);
}
else {
if (e4 !== k) {
g++;
}
}
p[y] = h = g;
g = d;
d = b;
b = c;
c = a;
}
}
for (; x < m;) {
var e = t.charCodeAt(x);
c = x;
d = ++x;
for (y = 0; y < n; y++) {
a = p[y];
if (a < c || d < c) {
d = (a > d ? d + 1 : a + 1);
}
else {
if (e !== s.charCodeAt(y)) {
d = c + 1;
}
else {
d = c;
}
}
p[y] = d;
c = a;
}
h = d;
}
resolve(h);
})
}
My question is, is there a way to calculate the difference faster when large string sets are used? In my case accuracy is not too important just as long as a rough difference is known of a certain percentage.
For example, if a research paper was published and I have the original paper and the students paper I want to know if roughly 10% of the students paper is plagiarized.
Maybe if I cut a random parts out of the strings this can help to save on time but this feels very dirty/inefficient.
I'm working on a game where I only need to check if there's a distance of 0 or 1 between two words and return true if that's the case. I found a general purpose levenshtein distance algorithm:
function levenshtein(s, t) {
if (s === t) { return 0; }
var n = s.length, m = t.length;
if (n === 0 || m === 0) { return n + m; }
var x = 0, y, a, b, c, d, g, h, k;
var p = new Array(n);
for (y = 0; y < n;) { p[y] = ++y; }
for (;
(x + 3) < m; x += 4) {
var e1 = t.charCodeAt(x);
var e2 = t.charCodeAt(x + 1);
var e3 = t.charCodeAt(x + 2);
var e4 = t.charCodeAt(x + 3);
c = x; b = x + 1; d = x + 2; g = x + 3; h = x + 4;
for (y = 0; y < n; y++) {
k = s.charCodeAt(y);
a = p[y];
if (a < c || b < c) { c = (a > b ? b + 1 : a + 1); }
else { if (e1 !== k) { c++; } }
if (c < b || d < b) { b = (c > d ? d + 1 : c + 1); }
else { if (e2 !== k) { b++; } }
if (b < d || g < d) { d = (b > g ? g + 1 : b + 1); }
else { if (e3 !== k) { d++; } }
if (d < g || h < g) { g = (d > h ? h + 1 : d + 1); }
else { if (e4 !== k) { g++; } }
p[y] = h = g; g = d; d = b; b = c; c = a;
}
}
for (; x < m;) {
var e = t.charCodeAt(x);
c = x;
d = ++x;
for (y = 0; y < n; y++) {
a = p[y];
if (a < c || d < c) { d = (a > d ? d + 1 : a + 1); }
else {
if (e !== s.charCodeAt(y)) { d = c + 1; }
else { d = c; }
}
p[y] = d;
c = a;
}
h = d;
}
return h;
}
Which works, but this spot is going to be a hotspot and be run potentially hundreds of thousands of times a second and I want to optimize it because I don't need a general purpose algorithm, just one that checks if there's a distance of 0 or 1.
I tried writing it and came up with this:
function closeGuess(guess, word) {
if (Math.abs(word.length - guess.length) > 1) { return false; }
var errors = 0, guessIndex = 0, wordIndex = 0;
while (guessIndex < guess.length || wordIndex < word.length) {
if (errors > 1) { return false; }
if (guess[guessIndex] !== word[wordIndex]) {
if (guess.length < word.length) { wordIndex++; }
else { guessIndex++; }
errors++;
} else {
wordIndex++;
guessIndex++;
}
}
return true;
}
But after profiling it I found that my code was twice as slow, which surprised me because I think the general purpose algorithm is O(n*m) and I think mine is O(n).
I've been testing the performance difference on this fiddle: https://jsfiddle.net/aubtze2L/3/
Are there any better algorithms I can use or any way I can optimize my code to be faster?
I don't see a more elegant way which is at the same time faster than the good old for-loop:
function lev01(a, b) {
let la = a.length;
let lb = b.length;
let d = 0;
switch (la - lb) {
case 0: // mutation
for (let i = 0; i < la; ++i) {
if (a.charAt(i) != b.charAt(i) && ++d > 1) {
return false;
}
}
return true;
case -1: // insertion
for (let i = 0; i < la + d; ++i) {
if (a.charAt(i - d) != b.charAt(i) && ++d > 1) {
return false;
}
}
return true;
case +1: // deletion
for (let i = 0; i < lb + d; ++i) {
if (a.charAt(i) != b.charAt(i - d) && ++d > 1) {
return false;
}
}
return true;
}
return false;
}
console.log(lev01("abc", "abc"));
console.log(lev01("abc", "abd"));
console.log(lev01("abc", "ab"));
console.log(lev01("abc", "abcd"));
console.log(lev01("abc", "cba"));
Performance comparison (Chrome):
80.33ms - lev01 (this answer)
234.84ms - lev
708.12ms - close
Consider the following cases:
If the difference in lengths of the terms is greater than 1, then
the Levenshtein distance between them will be greater than 1.
If the difference in lengths is exactly 1, then the shortest string must be equal to the longest string, with a single deletion (or insertion).
If the strings are the same length then you should
consider a modified version of Hamming distance which returns false
if two, different characters are found:
Here is a sample implementation:
var areSimilar;
areSimilar = function(guess, word) {
var charIndex, foundDiff, guessLength, lengthDiff, substring, wordLength, shortest, longest, shortestLength, offset;
guessLength = guess.length;
wordLength = word.length;
lengthDiff = guessLength - wordLength;
if (lengthDiff < -1 || lengthDiff > 1) {
return false;
}
if (lengthDiff !== 0) {
if (guessLength < wordLength) {
shortest = guess;
longest = word;
shortestLength = guessLength;
} else {
shortest = word;
longest = guess;
shortestLength = wordLength;
}
offset = 0;
for (charIndex = 0; charIndex < shortestLength; charIndex += 1) {
if (shortest[charIndex] !== longest[offset + charIndex]) {
if (offset > 0) {
return false; // second error
}
offset = 1;
if (shortest[charIndex] !== longest[offset + charIndex]) {
return false; // second error
}
}
}
return true; // only one error
}
foundDiff = false;
for (charIndex = 0; charIndex < guessLength; charIndex += 1) {
if (guess[charIndex] !== word[charIndex]) {
if (foundDiff) {
return false;
}
foundDiff = true;
}
}
return true;
};
I've updated your fiddle to include this method. Here are the results on my machine:
close: 154.61
lev: 176.72500000000002
sim: 32.48000000000013
Fiddle: https://jsfiddle.net/dylon/aubtze2L/11/
If you know that you are looking for distance 0 and 1, then the general purpose DP algorithm does not make sense (and by the way the algorithm you showed looks convoluted, take a look at a better explanation here).
To check that the distance is 0, all you need is to check whether 2 strings are the same. Now if the distance is one, it means that either insertion, deletion or substitution should have happened. So generate all possible deletion from the original string and check whether it is equal to second string. So you will get something like this:
for (var i = 0; i < s_1.length; i++) {
if s_2 == s_1.slice(0, i) + s_1.slice(i + 1) {
return true
}
}
For insertions and substitution you will need to know the alphabet of all characters. You can define it as a big string var alphabet = "abcde....". Now you do a similar thing, but when you introduce substitution or insertion, you also iterate over all elements in your alphabet. I am not planning to write the whole code here.
A couple of additional things. You can make a lot of micro-optimizations here. For example if the length of two strings are different by more than 1, they clearly can't have a distance 1. Another one relates to the frequencies of underlying characters in the string.
What I'm doing: I'm developing a mobile dictionary app for a number of languages
How I'm doing it: Using ionic framework with combination of some angular and some pure js (imported from a working online dictionary site of the same languages)
The problem: Our search function is an approximate search that uses a Levenstein distance calculator to rank all entries in the dictionary with respect to the query form. When the dictionary has up to 1,500 words, this isn't a problem at all on phones, but when the dictionary has around 10,000 words, there is about a 5-8 second delay before results are shown, despite it being instantaneous on a web browser using "ionic serve". When I run firebug, the javascript that takes the longest time to process are the distance calculations, so my working assumption is that this is where I should start, but I'm open to any suggestions at all.
Here's the distance calculator:
/**
* editDistance.js
*
* A simple Levenshtein distance calculator, except weighted such
* that insertions at the beginning and deletions at the end cost less.
*
* AUTHOR: Pat Littell
* LAST UPDATED: 2015-05-16
*/
var distanceCalculator = {
insertionCost : 1.0,
deletionCost : 1.0,
insertionAtBeginningCost : 0.11,
deletionAtEndCost : 0.1,
substitutionCost : 1.0,
getEditDistance : function(a, b) {
if(a.length === 0) return b.length;
if(b.length === 0) return a.length;
var matrix = [];
// var currentInsertionCost, currentDeletionCost, currentSubstitutionCost = 0;
// increment along the first column of each row
var i;
for(i = 0; i <= b.length; i++){
matrix[i] = [i * this.insertionAtBeginningCost];
}
// increment each column in the first row
var j;
for(j = 0; j <= a.length; j++){
matrix[0][j] = j;
}
// Fill in the rest of the matrix
for(i = 1; i <= b.length; i++){
for(j = 1; j <= a.length; j++){
currentInsertionCost = matrix[i][j-1] + this.insertionCost;
currentSubstitutionCost = matrix[i-1][j-1] + (b.charAt(i-1) != a.charAt(j-1) ? this.substitutionCost : 0);
currentDeletionCost = matrix[i-1][j] + (j==a.length ? this.deletionAtEndCost : this.deletionCost);
matrix[i][j] = Math.min(currentSubstitutionCost, Math.min(currentInsertionCost, currentDeletionCost));
}
}
return matrix[b.length][a.length];
},
// Given a query <a> and a series of targets <bs>, return the least distance to any target
getLeastEditDistance : function(a, bs) {
var that = this;
return Math.min.apply(null, bs.map(function(b) {
return that.getEditDistance(a,b);
}));
}
}
First of all, if you have a known dictionary you will get the fastest solution with something like a Levenshtein Automata, which will solve this in linear time to get all candidates. You can't beat this with a general purpose implementation.
With that said, this implementation of levenshtein distance is a few times faster than yours.
function distance(s, t) {
if (s === t) {
return 0;
}
var n = s.length, m = t.length;
if (n === 0 || m === 0) {
return n + m;
}
var x = 0, y, py, a, b, c, d, e, f, k;
var p = new Array(n);
for (y = 0; y < n;) {
p[y] = ++y;
}
for (; (x + 3) < m; x += 4) {
var tx0 = t.charCodeAt(x);
var tx1 = t.charCodeAt(x + 1);
var tx2 = t.charCodeAt(x + 2);
var tx3 = t.charCodeAt(x + 3);
a = x;
b = x + 1;
c = x + 2;
d = x + 3;
e = x + 4;
for (y = 0; y < n; y++) {
k = s.charCodeAt(y);
py = p[y];
if (py < a || b < a) {
a = (py > b ? b + 1 : py + 1);
}
else {
if (tx0 !== k) {
a++;
}
}
if (a < b || c < b) {
b = (a > c ? c + 1 : a + 1);
}
else {
if (tx1 !== k) {
b++;
}
}
if (b < c || d < c) {
c = (b > d ? d + 1 : b + 1);
}
else {
if (tx2 !== k) {
c++;
}
}
if (c < d || e < d) {
d = (c > e ? e + 1 : c + 1);
}
else {
if (tx3 !== k) {
d++;
}
}
p[y] = e = d;
d = c;
c = b;
b = a;
a = py;
}
}
for (; x < m;) {
tx0 = t.charCodeAt(x);
a = x;
b = ++x;
for (y = 0; y < n; y++) {
py = p[y];
if (py < a || b < a) {
b = (py > b ? b + 1 : py + 1);
}
else {
if (tx0 !== s.charCodeAt(y)) {
b = a + 1;
}
else {
b = a;
}
}
p[y] = b;
a = py;
}
f = b;
}
return f;
}
I would also not use map in getLeastEditDistance, it is very slow. Just use a normal loop. Also Math.min with many arguments is not very performant.
I am working with Levenstein distances by my self and I have not found a good way to improve performance and will not recommend using it in a non-batch application.
I suggest you use another approach by using a search tree. A binary or ternary search tree can also find near match.
A good place to start is those articles:
http://www.codeproject.com/Articles/5819/Ternary-Search-Tree-Dictionary-in-C-Faster-String
or
http://www.codeproject.com/Articles/68500/Balanced-Binary-Search-Tree-BST-Search-Delete-InOr
The code is relatively simple sp you should not use much time to port it to JavaScript.
I have an array being built like this:
var entries = ['L','L','L','L','L','L','L','L','L','L','R','R','R','R','R','R','R','R','R','R','M','M','M','M','M']
This means the array is always filled with 10 times L, 10 times R and 5 times M
The output I want to achieve is a randomly generated array, so I came up with the simple solution to just shuffle it with
function shuffle(o){
for(var j, x, i = o.length; i; j = Math.floor(Math.random() * i), x = o[--i], o[i] = o[j], o[j] = x);
return o;
}
The problem I have now is that there is a rule for the outcome, never have one of these letters more than 2 times in a row. So I thought I just use a do/while loop to shuffle until that criteria is met. But in my test runs this totally fails with long loops.
So my question is - what is the best way to build this array without depending on luck. My full program that fails is something like this
function shuffle(o){
for(var j, x, i = o.length; i; j = Math.floor(Math.random() * i), x = o[--i], o[i] = o[j], o[j] = x);
return o;
}
function createProgram(numShooters){
var programs = [];
for(var s = 0; s < numShooters; s++){
//Build array with L/R/M
for( d=0; d < 10; d++ ){
program.push('L');
program.push('R');
if(d < 5){
program.push('M');
}
}
// This will run way too long and is not reliable
//do{
// program = shuffle(program);
//}while(!checkProgram(program))
if(!checkProgram(program)){
console.log('invalid program at ' + s);
}
programs[s] = program;
}
return programs;
}
function checkProgram(program){
var len = program.length;
var last = null;
var dups = 0;
for(var i=0; i<len; len++){
if(program[i] == last){
dups++;
}else{
dups = 0;
}
if(dups == 2){
return false;
}
last = program[i];
}
return true;
}
createProgram(5);
Instead of just shuffle arrays and hope for one without duplicates, you can create them by picking characters by random and specifically avoid to pick the character that was picked previously.
If you keep track of how many there are left to pick of each character, you can control the odds for the character to pick so that the distribution is correct. If for example the first two characters are L, then there are 10 R and 5 M left to pick from (and 8 L, but they are excluded for the next pick), so there should be a 2 in 3 chance to pick an R and a 1 in 3 chance to pick and M.
This approach can run into a dead end, where the array can't be completed, so it has to start over. Running it a few hundred times I have seen something like a 10% overhead, so if you create five arrays you should by average see a retry every other time.
function createProgram(numShooters){
var programs = [];
for(var s = 0; s < numShooters; s++){
var chars = [ 'L', 'R', 'M' ];
var program;
do {
program = [];
var cnt = [ 10, 10, 5, 0 ]; // picks left
var prev = 3; // previous pick
var tot = 25; // total picks left
while (program.length < 25) {
// check for duplicates
var x = program.length >= 2 && program[program.length - 2] == program[program.length - 1] ? prev : 3;
// check if more picks are possible
if (tot - cnt[x] <= 0) {
console.log('invalid program ' + program);
break;
}
// pick from the possible
var r = Math.floor(Math.random() * (tot - cnt[x]));
// determine what character was picked
var c = 0;
while (c == x || r >= cnt[c]) {
if (c != x) r -= cnt[c];
c++;
}
program.push(chars[c]);
cnt[c]--;
tot--;
prev = c;
}
} while (program.length < 25);
programs[s] = program;
}
return programs;
}
console.log(createProgram(1).toString());
So this is the final solution I came up with, as commented Guffas solution also works nice and smooth. But after doing some tests mine is about 30% faster and more readable, but way longer so I'll accept Guffas - thanks to everybody for their input!
function createProgram(numShooters){
var programs = [];
for(var s = 0; s < numShooters; s++){
var program = buildProgram();
programs[s] = program;
}
return programs;
}
function buildProgram(){
var program = [];
var ls = fillArray('L',10);
var rs = fillArray('R',10);
var ms = fillArray('M',5);
//Use either L or R to mix into M - adds variation
var side = Math.random() > 0.5 ? ls : rs;
var otherSide = side == ls ? rs : ls;
var initProg = side.concat(ms);
initProg = shuffle(initProg);
var program = [];
//Correcting invalid positions as suggested
for(var p1 = 0; p1 < initProg.length; p1++){
if(p1 > 1 && initProg[p1-1] == initProg[p1-2] && initProg[p1-1] == initProg[p1]){
if(otherSide.length > 0){
program.push(otherSide.pop());
}else{
return buildProgram(); //impossible state, redo...
}
}
program.push(initProg[p1]);
}
//Fill into remaining other pos
for(var p2 = 0; p2 < otherSide.length; p2++){
program = addAtRandomPos(program,otherSide[p2]);
}
return program;
}
function addAtRandomPos(arr,chr){
var pos = getRandomInt( 0, arr.length - 1 );
var charCur = arr[pos];
var pprev = pos > 1 ? arr[pos-2] : null;
var prev = pos > 0 ? arr[pos-1] : null;
var next = pos < arr.length - 1 ? arr[pos] : null;
var nnext = pos < arr.length - 2 ? arr[pos+1] : null;
var str = pprev + prev + chr + next + nnext;
if(str.indexOf('MMM') !== -1 || str.indexOf('RRR') !== -1 || str.indexOf('LLL') !== -1){
return addAtRandomPos(arr,chr);
}else{
arr.splice(pos,0,chr);
}
return arr;
}
function getRandomInt (min, max) {
return Math.floor(Math.random() * (max - min + 1)) + min;
}
function fillArray(chr,num){
var arr = [];
for(i = 0; i < num; i++){
arr.push(chr);
}
return arr;
}
So I have a random javascript array of names...
[#larry,#nicholas,#notch] etc.
They all start with the # symbol. I'd like to sort them by the Levenshtein Distance so that the the ones at the top of the list are closest to the search term. At the moment, I have some javascript that uses jQuery's .grep() on it using javascript .match() method around the entered search term on key press:
(code edited since first publish)
limitArr = $.grep(imTheCallback, function(n){
return n.match(searchy.toLowerCase())
});
modArr = limitArr.sort(levenshtein(searchy.toLowerCase(), 50))
if (modArr[0].substr(0, 1) == '#') {
if (atRes.childred('div').length < 6) {
modArr.forEach(function(i){
atRes.append('<div class="oneResult">' + i + '</div>');
});
}
} else if (modArr[0].substr(0, 1) == '#') {
if (tagRes.children('div').length < 6) {
modArr.forEach(function(i){
tagRes.append('<div class="oneResult">' + i + '</div>');
});
}
}
$('.oneResult:first-child').addClass('active');
$('.oneResult').click(function(){
window.location.href = 'http://hashtag.ly/' + $(this).html();
});
It also has some if statements detecting if the array contains hashtags (#) or mentions (#). Ignore that. The imTheCallback is the array of names, either hashtags or mentions, then modArr is the array sorted. Then the .atResults and .tagResults elements are the elements that it appends each time in the array to, this forms a list of names based on the entered search terms.
I also have the Levenshtein Distance algorithm:
var levenshtein = function(min, split) {
// Levenshtein Algorithm Revisited - WebReflection
try {
split = !("0")[0]
} catch(i) {
split = true
};
return function(a, b) {
if (a == b)
return 0;
if (!a.length || !b.length)
return b.length || a.length;
if (split) {
a = a.split("");
b = b.split("")
};
var len1 = a.length + 1,
len2 = b.length + 1,
I = 0,
i = 0,
d = [[0]],
c, j, J;
while (++i < len2)
d[0][i] = i;
i = 0;
while (++i < len1) {
J = j = 0;
c = a[I];
d[i] = [i];
while(++j < len2) {
d[i][j] = min(d[I][j] + 1, d[i][J] + 1, d[I][J] + (c != b[J]));
++J;
};
++I;
};
return d[len1 - 1][len2 - 1];
}
}(Math.min, false);
How can I work with algorithm (or a similar one) into my current code to sort it without bad performance?
UPDATE:
So I'm now using James Westgate's Lev Dist function. Works WAYYYY fast. So performance is solved, the issue now is using it with source...
modArr = limitArr.sort(function(a, b){
levDist(a, searchy)
levDist(b, searchy)
});
My problem now is general understanding on using the .sort() method. Help is appreciated, thanks.
Thanks!
I wrote an inline spell checker a few years ago and implemented a Levenshtein algorithm - since it was inline and for IE8 I did quite a lot of performance optimisation.
var levDist = function(s, t) {
var d = []; //2d matrix
// Step 1
var n = s.length;
var m = t.length;
if (n == 0) return m;
if (m == 0) return n;
//Create an array of arrays in javascript (a descending loop is quicker)
for (var i = n; i >= 0; i--) d[i] = [];
// Step 2
for (var i = n; i >= 0; i--) d[i][0] = i;
for (var j = m; j >= 0; j--) d[0][j] = j;
// Step 3
for (var i = 1; i <= n; i++) {
var s_i = s.charAt(i - 1);
// Step 4
for (var j = 1; j <= m; j++) {
//Check the jagged ld total so far
if (i == j && d[i][j] > 4) return n;
var t_j = t.charAt(j - 1);
var cost = (s_i == t_j) ? 0 : 1; // Step 5
//Calculate the minimum
var mi = d[i - 1][j] + 1;
var b = d[i][j - 1] + 1;
var c = d[i - 1][j - 1] + cost;
if (b < mi) mi = b;
if (c < mi) mi = c;
d[i][j] = mi; // Step 6
//Damerau transposition
if (i > 1 && j > 1 && s_i == t.charAt(j - 2) && s.charAt(i - 2) == t_j) {
d[i][j] = Math.min(d[i][j], d[i - 2][j - 2] + cost);
}
}
}
// Step 7
return d[n][m];
}
I came to this solution:
var levenshtein = (function() {
var row2 = [];
return function(s1, s2) {
if (s1 === s2) {
return 0;
} else {
var s1_len = s1.length, s2_len = s2.length;
if (s1_len && s2_len) {
var i1 = 0, i2 = 0, a, b, c, c2, row = row2;
while (i1 < s1_len)
row[i1] = ++i1;
while (i2 < s2_len) {
c2 = s2.charCodeAt(i2);
a = i2;
++i2;
b = i2;
for (i1 = 0; i1 < s1_len; ++i1) {
c = a + (s1.charCodeAt(i1) === c2 ? 0 : 1);
a = row[i1];
b = b < a ? (b < c ? b + 1 : c) : (a < c ? a + 1 : c);
row[i1] = b;
}
}
return b;
} else {
return s1_len + s2_len;
}
}
};
})();
See also http://jsperf.com/levenshtein-distance/12
Most speed was gained by eliminating some array usages.
Updated: http://jsperf.com/levenshtein-distance/5
The new Revision annihilates all other benchmarks. I was specifically chasing Chromium/Firefox performance as I don't have an IE8/9/10 test environment, but the optimisations made should apply in general to most browsers.
Levenshtein Distance
The matrix to perform Levenshtein Distance can be reused again and again. This was an obvious target for optimisation (but be careful, this now imposes a limit on string length (unless you were to resize the matrix dynamically)).
The only option for optimisation not pursued in jsPerf Revision 5 is memoisation. Depending on your use of Levenshtein Distance, this could help drastically but was omitted due to its implementation specific nature.
// Cache the matrix. Note this implementation is limited to
// strings of 64 char or less. This could be altered to update
// dynamically, or a larger value could be used.
var matrix = [];
for (var i = 0; i < 64; i++) {
matrix[i] = [i];
matrix[i].length = 64;
}
for (var i = 0; i < 64; i++) {
matrix[0][i] = i;
}
// Functional implementation of Levenshtein Distance.
String.levenshteinDistance = function(__this, that, limit) {
var thisLength = __this.length, thatLength = that.length;
if (Math.abs(thisLength - thatLength) > (limit || 32)) return limit || 32;
if (thisLength === 0) return thatLength;
if (thatLength === 0) return thisLength;
// Calculate matrix.
var this_i, that_j, cost, min, t;
for (i = 1; i <= thisLength; ++i) {
this_i = __this[i-1];
for (j = 1; j <= thatLength; ++j) {
// Check the jagged ld total so far
if (i === j && matrix[i][j] > 4) return thisLength;
that_j = that[j-1];
cost = (this_i === that_j) ? 0 : 1; // Chars already match, no ++op to count.
// Calculate the minimum (much faster than Math.min(...)).
min = matrix[i - 1][j ] + 1; // Deletion.
if ((t = matrix[i ][j - 1] + 1 ) < min) min = t; // Insertion.
if ((t = matrix[i - 1][j - 1] + cost) < min) min = t; // Substitution.
matrix[i][j] = min; // Update matrix.
}
}
return matrix[thisLength][thatLength];
};
Damerau-Levenshtein Distance
jsperf.com/damerau-levenshtein-distance
Damerau-Levenshtein Distance is a small modification to Levenshtein Distance to include transpositions. There is very little to optimise.
// Damerau transposition.
if (i > 1 && j > 1 && this_i === that[j-2] && this[i-2] === that_j
&& (t = matrix[i-2][j-2]+cost) < matrix[i][j]) matrix[i][j] = t;
Sorting Algorithm
The second part of this answer is to choose an appropriate sort function. I will upload optimised sort functions to http://jsperf.com/sort soon.
I implemented a very performant implementation of levenshtein distance calculation if you still need this.
function levenshtein(s, t) {
if (s === t) {
return 0;
}
var n = s.length, m = t.length;
if (n === 0 || m === 0) {
return n + m;
}
var x = 0, y, a, b, c, d, g, h, k;
var p = new Array(n);
for (y = 0; y < n;) {
p[y] = ++y;
}
for (; (x + 3) < m; x += 4) {
var e1 = t.charCodeAt(x);
var e2 = t.charCodeAt(x + 1);
var e3 = t.charCodeAt(x + 2);
var e4 = t.charCodeAt(x + 3);
c = x;
b = x + 1;
d = x + 2;
g = x + 3;
h = x + 4;
for (y = 0; y < n; y++) {
k = s.charCodeAt(y);
a = p[y];
if (a < c || b < c) {
c = (a > b ? b + 1 : a + 1);
}
else {
if (e1 !== k) {
c++;
}
}
if (c < b || d < b) {
b = (c > d ? d + 1 : c + 1);
}
else {
if (e2 !== k) {
b++;
}
}
if (b < d || g < d) {
d = (b > g ? g + 1 : b + 1);
}
else {
if (e3 !== k) {
d++;
}
}
if (d < g || h < g) {
g = (d > h ? h + 1 : d + 1);
}
else {
if (e4 !== k) {
g++;
}
}
p[y] = h = g;
g = d;
d = b;
b = c;
c = a;
}
}
for (; x < m;) {
var e = t.charCodeAt(x);
c = x;
d = ++x;
for (y = 0; y < n; y++) {
a = p[y];
if (a < c || d < c) {
d = (a > d ? d + 1 : a + 1);
}
else {
if (e !== s.charCodeAt(y)) {
d = c + 1;
}
else {
d = c;
}
}
p[y] = d;
c = a;
}
h = d;
}
return h;
}
It was my answer to a similar SO question
Fastest general purpose Levenshtein Javascript implementation
Update
A improved version of the above is now on github/npm see
https://github.com/gustf/js-levenshtein
The obvious way of doing this is to map each string to a (distance, string) pair, then sort this list, then drop the distances again. This way you ensure the levenstein distance only has to be computed once. Maybe merge duplicates first, too.
I would definitely suggest using a better Levenshtein method like the one in #James Westgate's answer.
That said, DOM manipulations are often a great expense. You can certainly improve your jQuery usage.
Your loops are rather small in the example above, but concatenating the generated html for each oneResult into a single string and doing one append at the end of the loop will be much more efficient.
Your selectors are slow. $('.oneResult') will search all elements in the DOM and test their className in older IE browsers. You may want to consider something like atRes.find('.oneResult') to scope the search.
In the case of adding the click handlers, we may want to do one better avoid setting handlers on every keyup. You could leverage event delegation by setting a single handler on atRest for all results in the same block you are setting the keyup handler:
atRest.on('click', '.oneResult', function(){
window.location.href = 'http://hashtag.ly/' + $(this).html();
});
See http://api.jquery.com/on/ for more info.
I just wrote an new revision: http://jsperf.com/levenshtein-algorithms/16
function levenshtein(a, b) {
if (a === b) return 0;
var aLen = a.length;
var bLen = b.length;
if (0 === aLen) return bLen;
if (0 === bLen) return aLen;
var len = aLen + 1;
var v0 = new Array(len);
var v1 = new Array(len);
var i = 0;
var j = 0;
var c2, min, tmp;
while (i < len) v0[i] = i++;
while (j < bLen) {
c2 = b.charAt(j++);
v1[0] = j;
i = 0;
while (i < aLen) {
min = v0[i] - (a.charAt(i) === c2 ? 1 : 0);
if (v1[i] < min) min = v1[i];
if (v0[++i] < min) min = v0[i];
v1[i] = min + 1;
}
tmp = v0;
v0 = v1;
v1 = tmp;
}
return v0[aLen];
}
This revision is faster than the other ones. Works even on IE =)