Develop Reference

using Damerau-Levenshtein distance to compare sets of text in code.org

Not very knowledgeable with coding, I usually use block coding and not typing.
I've used many different Levenshtein distance codes I've found online and most of them didn't work for one reason or another
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 (i = n; i >= 0; i--) d[i][0] = i;
for (var j = m; j >= 0; j--) d[0][j] = j;
// Step 3
for (i = 1; i <= n; i++) {
var s_i = s.charAt(i - 1);
// Step 4
for (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];
};
This is all the code I’ve written (including the most recent attempt of getting the levenshtein distance)
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 (i = n; i >= 0; i--) d[i][0] = i;
for (var j = m; j >= 0; j--) d[0][j] = j;
// Step 3
for (i = 1; i <= n; i++) {
var s_i = s.charAt(i - 1);
// Step 4
for (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];
};
var S = "Hello World";
var grossWPM;
var Transparency = 1;
var Timer = 60;
var InitialTime = Timer;
var Texts = getColumn("Texts", "Texts");
var TextLength = getColumn("Texts", "Number of Characters");
var Title = getColumn("Texts", "Titles");
var Author = getColumn("Texts", "Authors");
var TextSelector = randomNumber(0, 19);
console.log("Article #" + (TextSelector + 1));
console.log(TextLength[TextSelector] + " Characters in total");
console.log(Title[TextSelector]);
console.log("By: " + Author[TextSelector]);
var Countdown;
var Countdown = 6;
//Texts are obtained from
//https://data.typeracer.com/pit/texts
onEvent("button1", "click", function( ) {
timedLoop(1000, function() {
Countdown = Countdown - 1;
setText("button1", Countdown - 0);
timedLoop(100, function() {
setText("text_area2", "");
});
if (Countdown <= 1) {
stopTimedLoop();
setTimeout(function() {
setText("button1", "GO!");
setText("text_area1", Texts[TextSelector]);
if (getText("button1") == "GO!") {
var TransparentLoop = timedLoop(100, function() {
Transparency = Transparency - 0.1;
setProperty("Warning", "text-color", rgb(77,87,95, Transparency));
if (Transparency <= 0) {
deleteElement("Warning");
showElement("label2");
stopTimedLoop(TransparentLoop);
}
});
var TimerLoop = timedLoop(1000, function() {
Timer = Timer - 1;
setText("label2", Timer);
if (Timer <= 0) {
grossWPM = (TextLength[TextSelector] / 5) / ((InitialTime - Timer) / 60);
console.log(grossWPM);
setScreen("screen2");
if (Timer == 1) {
S = " second";
} else {
S = " seconds";
}
setText("label1", "Your typing speed was approximately " + (Math.round(grossWPM) + (" WPM* with " + (Timer + (S + " left")))));
stopTimedLoop(TimerLoop);
}
});
console.log("Timer Started");
timedLoop(10, function() {
var str = getText("text_area2");
if (str.length == TextLength[TextSelector]) {
stopTimedLoop(TimerLoop);
grossWPM = (TextLength[TextSelector] / 5) / ((InitialTime - Timer) / 60);
setScreen("screen2");
levDist(str, Texts[TextSelector]);
if (Timer == 1) {
S = " second";
} else {
S = " seconds";
}
setText("label1", "Your typing speed was approximately " + (Math.round(grossWPM) + (" WPM* with " + (Timer + (S + " left")))));
if (grossWPM == 69) {
setText("label4", "Nice");
}
stopTimedLoop();
}
});
}
}, 1000);
}
});
});
Obviously not that good at this so can anyone help?
I want to compare two sets of text
Something the user types in.
Paragraph that the user was supposed to type.
This is for a WPM test and I want a way to get a measurement for WPM that includes errors the user makes while typing.
If there is a way to check this besides the Levenshtein distance please tell me, I just looked up a way to do that and Levenshtein distance seemed like the way to do so
The error given by code.org says:
ERROR: Line: 50: TypeError: d[n] is undefined

I fixed the issue, I used this code
function levenshtein(s1, s2) {
if (s1 == s2) {
return 0;
}
var s1_len = s1.length;
var s2_len = s2.length;
if (s1_len === 0) {
return s2_len;
}
if (s2_len === 0) {
return s1_len;
}
// BEGIN STATIC
var split = false;
try {
split = !('0')[0];
} catch (e) {
// Earlier IE may not support access by string index
split = true;
}
// END STATIC
if (split) {
s1 = s1.split('');
s2 = s2.split('');
}
var v0 = new Array(s1_len + 1);
var v1 = new Array(s1_len + 1);
var s1_idx = 0,
s2_idx = 0,
cost = 0;
for (s1_idx = 0; s1_idx < s1_len + 1; s1_idx++) {
v0[s1_idx] = s1_idx;
}
var char_s1 = '',
char_s2 = '';
for (s2_idx = 1; s2_idx <= s2_len; s2_idx++) {
v1[0] = s2_idx;
char_s2 = s2[s2_idx - 1];
for (s1_idx = 0; s1_idx < s1_len; s1_idx++) {
char_s1 = s1[s1_idx];
cost = (char_s1 == char_s2) ? 0 : 1;
var m_min = v0[s1_idx + 1] + 1;
var b = v1[s1_idx] + 1;
var c = v0[s1_idx] + cost;
if (b < m_min) {
m_min = b;
}
if (c < m_min) {
m_min = c;
}
v1[s1_idx + 1] = m_min;
}
var v_tmp = v0;
v0 = v1;
v1 = v_tmp;
}
return v0[s1_len];
}
and I got that code from this question
This is levenshtein distance NOT damerau-levenshtein distance

Javascript while loop (Card deck simulation)

I am having an issue with the following code that simulates a card deck.
The deck is created properly (1 array containing 4 arrays (suits) containing 13 elements each (face values)) and when I use the G.test(); function it is correctly pulling 13 random cards but then returns 39x "Empty" (A total of 52).
I hate to ask for help, but I have left the problem overnight and then some and I still cannot find the reason that this is happening. I appreciate any and all insight that can be offered.
var G = {};
G.cards = [[], [], [], []];
G.newCard = function(v) { //currently a useless function, tried a few things
return v;
};
G.deck = {
n: function() { //new deck
var x; var list = [];
list.push(G.newCard("A"));
for (x = 2; x <= 10; x += 1) {
list.push(G.newCard(x.toString()));
}
list.push(G.newCard("J"), G.newCard("Q"), G.newCard("K"));
for (x = 0; x < G.cards.length; x += 1) {
G.cards[x] = list;
}
},
d: function() { //random card - returns suit & value
var s; var c; var v; var drawn = false; var n;
s = random(0, G.cards.length);
c = random(0, G.cards[s].length);
n = 0;
while (!drawn) {
if (G.cards[s].length > 0) {
if (G.cards[s][c]) {
v = G.cards[s].splice(c, 1);
drawn = true;
} else {
c = random(0, G.cards[s].length);
}
} else {
s = (s + 1 >= G.cards.length) ? 0 : s + 1;
n += 1;
console.log(s);
if (n >= G.cards.length) {
console.log(n);
return "Empty";
}
}
}
return {s: s, v: v[0]};
},
}; //G.deck
G.test = function() {
var x; var v;
G.deck.n();
for (x = 0; x < 52; x += 1) {
v = G.deck.d();
console.log(v);
}
};

Replace
for (x = 0; x < G.cards.length; x += 1) {
G.cards[x] = list;
}
with
for (x = 0; x < G.cards.length; x += 1) {
G.cards[x] = list.slice();
}
as this prevents all elements of G.cards[x] binding to the same (single) array instance.
If all elements bind to the same instance, mutating one element equals mutating all elements. list.slice() creates a new copy of list and thus a new array instance to prevent the aforementioned issue.

I won't go through your code, but I built a code that will do what you wanted. I only built this for one deck and not multiple deck play. There are two functions, one will generate the deck, and the other will drawn cards from the deck, bases on how many hands you need and how many cards you wanted for each hand. One a card is drawn, it will not be re-drawn. I might publish a short article for how a card dealing program work or similar in the short future at http://kevinhng86.iblog.website.
function random(min, max){
return Math.floor(Math.random() * (max - min)) + min;
}
function deckGenerate(){
var output = [];
var face = {1: "A", 11: "J", 12: "Q", 13: "K"};
// Heart Space Diamond & Club;
var suit = ["H", "S", "D", "C"];
// Delimiter between card identification and suit identification.
var d = "-";
for(i = 0; i < 4; i++){
output[i] = [];
for(ind = 0; ind < 13; ind++ ){
card = (ind + 1);
output[i][ind] = (card > 10) || (card === 1)? face[card] + d + suit[i] : card.toString() + d + suit[i];
}
}
return output;
}
function randomCard(deck, hand, card){
var output = [];
var randS = 0;
var randC = 0;
if( hand * card > 52 ) throw("Too many card, I built this for one deck only");
for(i = 0; i < hand; i++){
output[i] = [];
for(ind = 0; ind < card; ind++){
randS = random(0, deck.length);
randC = random(0, deck[randS].length);
output[i][ind] = deck[randS][randC];
deck[randS].splice(randC,1);
if(deck[randS].length === 0) deck.splice(randS,1);
}
}
document.write( JSON.stringify(deck, null, 2) );
return output;
}
var deck = deckGenerate()
document.write( JSON.stringify(deck, null, 2) );
document.write("<br><br>");
var randomhands = randomCard(deck, 5, 8);
document.write("<br><br>");
document.write("<br><br>");
document.write( JSON.stringify(randomhands, null, 2) );

Trying to convert a C++ code to Javascript code

Well, I'm trying to convert this code in C++ (a custom LZSS decompressor):
// Modified LZSS routine
// - starting position at 0 rather than 0xFEE
// - optionally, additional byte for repetition count
// - dictionary writing in two passes
static bstr custom_lzss_decompress(
const bstr &input, size_t output_size, const size_t dict_capacity)
{
std::vector<u8> dict(dict_capacity);
size_t dict_size = 0;
size_t dict_pos = 0;
bstr output(output_size);
auto output_ptr = output.get<u8>();
auto output_end = output.end<const u8>();
auto input_ptr = input.get<const u8>();
auto input_end = input.end<const u8>();
u16 control = 0;
while (output_ptr < output_end)
{
control >>= 1;
if (!(control & 0x100))
control = *input_ptr++ | 0xFF00;
if (control & 1)
{
dict[dict_pos++] = *output_ptr++ = *input_ptr++;
dict_pos %= dict_capacity;
if (dict_size < dict_capacity)
dict_size++;
}
else
{
auto tmp = *reinterpret_cast<const u16*>(input_ptr);
input_ptr += 2;
auto look_behind_pos = tmp >> 4;
auto repetitions = tmp & 0xF;
if (repetitions == 0xF)
repetitions += *input_ptr++;
repetitions += 3;
auto i = repetitions;
while (i-- && output_ptr < output_end)
{
*output_ptr++ = dict[look_behind_pos++];
look_behind_pos %= dict_size;
}
auto source = &output_ptr[-repetitions];
while (source < output_ptr)
{
dict[dict_pos++] = *source++;
dict_pos %= dict_capacity;
if (dict_size < dict_capacity)
dict_size++;
}
}
}
return output;
}
Into a javascript version, and i'm actually on this:
function custom_lzss_decompress(input,osize,dictcap){
var dict = new Uint8Array(dictcap);
var output = new Uint8Array(osize);
var data = str2ab(input);
var dict_size = 0;
var dict_pos = 0;
var control = 0;
var iptr = 0;
var optr = 0;
while (iptr < osize){
control >>= 1;
if(!(control & 0x100)){
control = data[iptr] | 0xFF00;
iptr++;
}
if(control & 1){
dict[dict_pos] = output[optr] = data[iptr];
dict_pos++;
iptr++;
optr++;
dict_pos %= dictcap;
if(dict_size < dictcap)
dict_size++;
}
else{
var tmp = new Uint16Array(1);
tmp[0] = (data[iptr] << 8) + data[iptr+1];
iptr += 2;
var loop_behind_pos = new Uint16Array(1);
loop_behind_pos[0] = tmp >> 4;
var repetitions = new Uint16Array(1);
repetitions[0] = tmp & 0xF;
if(repetitions[0] == 0xF){
repetitions[0] += data[iptr];
}
iptr++;
repetitions[0] += 3;
var ai = new Uint16Array(1);
ai[0] = repetitions[0];
while(ai[0] && (optr < osize)){
ai[0]--;
output[optr] = dict[loop_behind_pos];
optr++;
loop_behind_pos++;
loop_behind_pos %= dict_size;
}
var source = new Uint16Array(1);
source[0] = output[-repetitions[0]];
while(source[0] < optr){
dict[dict_pos] = output[source];
dict_pos++;
source++;
dict_pos %= dictcap;
if(dict_size < dictcap)
dict_size++;
}
}
}
return output;
}
It runs... but doesn't works... And i can't figure out what is going on... Someone can see some bizarre error that I'm not seeing?

I see one problem, and one potential problem. The problem is in the 'repetitions[0] == 0xFhandler. You're incrementingiptr` all the time, but the original code only does it in the body of the if.
The potential problem is in the 16 bit word read, tmp[0] = (data[iptr] << 8) + data[iptr+1];. This works if the word is big-endian, but will not give the right value if the source is little-endian.

Speeding up Levenshtein distance calculation in Ionic app

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.

LogLog and HyperLogLog algorithms for counting of large cardinalities

Where can I find a valid implementation of LogLog algorithm? Have tried to implement it by myself but my draft implementation yields strange results.
Here it is:
function LogLog(max_error, max_count)
{
function log2(x)
{
return Math.log(x) / Math.LN2;
}
var m = 1.30 / max_error;
var k = Math.ceil(log2(m * m));
m = Math.pow(2, k);
var k_comp = 32 - k;
var l = log2(log2(max_count / m));
if (isNaN(l)) l = 1; else l = Math.ceil(l);
var l_mask = ((1 << l) - 1) >>> 0;
var M = [];
for (var i = 0; i < m; ++i) M[i] = 0;
function count(hash)
{
if (hash !== undefined)
{
var j = hash >>> k_comp;
var rank = 0;
for (var i = 0; i < k_comp; ++i)
{
if ((hash >>> i) & 1)
{
rank = i + 1;
break;
}
}
M[j] = Math.max(M[j], rank & l_mask);
}
else
{
var c = 0;
for (var i = 0; i < m; ++i) c += M[i];
return 0.79402 * m * Math.pow(2, c / m);
}
}
return {count: count};
}
function fnv1a(text)
{
var hash = 2166136261;
for (var i = 0; i < text.length; ++i)
{
hash ^= text.charCodeAt(i);
hash += (hash << 1) + (hash << 4) + (hash << 7) +
(hash << 8) + (hash << 24);
}
return hash >>> 0;
}
var words = ['aardvark', 'abyssinian', ... ,'zoology']; // about 2 300 words
var log_log = LogLog(0.01, 100000);
for (var i = 0; i < words.length; ++i) log_log.count(fnv1a(words[i]));
alert(log_log.count());
For unknown reason implementation is very sensitive to max_error parameter, it is the main factor that determines the magnitude of the result. I'm sure, there is some stupid mistake :)
UPDATE: This problem is solved in the newer version of algorithm. I will post its implementation later.

Here it is the updated version of the algorithm based on the newer paper:
var pow_2_32 = 0xFFFFFFFF + 1;
function HyperLogLog(std_error)
{
function log2(x)
{
return Math.log(x) / Math.LN2;
}
function rank(hash, max)
{
var r = 1;
while ((hash & 1) == 0 && r <= max) { ++r; hash >>>= 1; }
return r;
}
var m = 1.04 / std_error;
var k = Math.ceil(log2(m * m)), k_comp = 32 - k;
m = Math.pow(2, k);
var alpha_m = m == 16 ? 0.673
: m == 32 ? 0.697
: m == 64 ? 0.709
: 0.7213 / (1 + 1.079 / m);
var M = []; for (var i = 0; i < m; ++i) M[i] = 0;
function count(hash)
{
if (hash !== undefined)
{
var j = hash >>> k_comp;
M[j] = Math.max(M[j], rank(hash, k_comp));
}
else
{
var c = 0.0;
for (var i = 0; i < m; ++i) c += 1 / Math.pow(2, M[i]);
var E = alpha_m * m * m / c;
// -- make corrections
if (E <= 5/2 * m)
{
var V = 0;
for (var i = 0; i < m; ++i) if (M[i] == 0) ++V;
if (V > 0) E = m * Math.log(m / V);
}
else if (E > 1/30 * pow_2_32)
E = -pow_2_32 * Math.log(1 - E / pow_2_32);
// --
return E;
}
}
return {count: count};
}
function fnv1a(text)
{
var hash = 2166136261;
for (var i = 0; i < text.length; ++i)
{
hash ^= text.charCodeAt(i);
hash += (hash << 1) + (hash << 4) + (hash << 7) +
(hash << 8) + (hash << 24);
}
return hash >>> 0;
}
var words = ['aardvark', 'abyssinian', ..., 'zoology']; // 2336 words
var seed = Math.floor(Math.random() * pow_2_32); // make more fun
var log_log = HyperLogLog(0.065);
for (var i = 0; i < words.length; ++i) log_log.count(fnv1a(words[i]) ^ seed);
var count = log_log.count();
alert(count + ', error ' +
(count - words.length) / (words.length / 100.0) + '%');

Here is a slightly modified version which adds the merge operation.
Merge allows you to take the counters from several instances of HyperLogLog,
and determine the unique counters overall.
For example, if you have unique visitors collected on Monday, Tuesday and Wednesday,
then you can merge the buckets together and count the number of unique visitors
over the three day span:
var pow_2_32 = 0xFFFFFFFF + 1;
function HyperLogLog(std_error)
{
function log2(x)
{
return Math.log(x) / Math.LN2;
}
function rank(hash, max)
{
var r = 1;
while ((hash & 1) == 0 && r <= max) { ++r; hash >>>= 1; }
return r;
}
var m = 1.04 / std_error;
var k = Math.ceil(log2(m * m)), k_comp = 32 - k;
m = Math.pow(2, k);
var alpha_m = m == 16 ? 0.673
: m == 32 ? 0.697
: m == 64 ? 0.709
: 0.7213 / (1 + 1.079 / m);
var M = []; for (var i = 0; i < m; ++i) M[i] = 0;
function merge(other)
{
for (var i = 0; i < m; i++)
M[i] = Math.max(M[i], other.buckets[i]);
}
function count(hash)
{
if (hash !== undefined)
{
var j = hash >>> k_comp;
M[j] = Math.max(M[j], rank(hash, k_comp));
}
else
{
var c = 0.0;
for (var i = 0; i < m; ++i) c += 1 / Math.pow(2, M[i]);
var E = alpha_m * m * m / c;
// -- make corrections
if (E <= 5/2 * m)
{
var V = 0;
for (var i = 0; i < m; ++i) if (M[i] == 0) ++V;
if (V > 0) E = m * Math.log(m / V);
}
else if (E > 1/30 * pow_2_32)
E = -pow_2_32 * Math.log(1 - E / pow_2_32);
// --
return E;
}
}
return {count: count, merge: merge, buckets: M};
}
function fnv1a(text)
{
var hash = 2166136261;
for (var i = 0; i < text.length; ++i)
{
hash ^= text.charCodeAt(i);
hash += (hash << 1) + (hash << 4) + (hash << 7) +
(hash << 8) + (hash << 24);
}
return hash >>> 0;
}
Then you can do something like this:
// initialize one counter per day
var ll_monday = HyperLogLog(0.01);
var ll_tuesday = HyperLogLog(0.01);
var ll_wednesday = HyperLogLog(0.01);
// add 5000 unique values in each day
for(var i=0; i<5000; i++) ll_monday.count(fnv1a('' + Math.random()));
for(var i=0; i<5000; i++) ll_tuesday.count(fnv1a('' + Math.random()));
for(var i=0; i<5000; i++) ll_wednesday.count(fnv1a('' + Math.random()));
// add 5000 values which appear every day
for(var i=0; i<5000; i++) {ll_monday.count(fnv1a(''+i)); ll_tuesday.count(fnv1a('' + i)); ll_wednesday.count(fnv1a('' + i));}
// merge three days together
together = HyperLogLog(0.01);
together.merge(ll_monday);
together.merge(ll_tuesday);
together.merge(ll_wednesday);
// report
console.log('unique per day: ' + Math.round(ll_monday.count()) + ' ' + Math.round(ll_tuesday.count()) + ' ' + Math.round(ll_wednesday.count()));
console.log('unique numbers overall: ' + Math.round(together.count()));

We've open sourced a project called Stream-Lib that has a LogLog implementation. The work was based on this paper.

Using the js version #actual provided, I tried to implement the same in C#, which seems close enough. Just changed fnv1a function a little bit and renamed it to getHashCode. (Credit goes to Jenkins hash function, http://en.wikipedia.org/wiki/Jenkins_hash_function)
public class HyperLogLog
{
private double mapSize, alpha_m, k;
private int kComplement;
private Dictionary<int, int> Lookup = new Dictionary<int, int>();
private const double pow_2_32 = 4294967297;
public HyperLogLog(double stdError)
{
mapSize = (double)1.04 / stdError;
k = (long)Math.Ceiling(log2(mapSize * mapSize));
kComplement = 32 - (int)k;
mapSize = (long)Math.Pow(2, k);
alpha_m = mapSize == 16 ? (double)0.673
: mapSize == 32 ? (double)0.697
: mapSize == 64 ? (double)0.709
: (double)0.7213 / (double)(1 + 1.079 / mapSize);
for (int i = 0; i < mapSize; i++)
Lookup[i] = 0;
}
private static double log2(double x)
{
return Math.Log(x) / 0.69314718055994530941723212145818;//Ln2
}
private static int getRank(uint hash, int max)
{
int r = 1;
uint one = 1;
while ((hash & one) == 0 && r <= max)
{
++r;
hash >>= 1;
}
return r;
}
public static uint getHashCode(string text)
{
uint hash = 0;
for (int i = 0, l = text.Length; i < l; i++)
{
hash += (uint)text[i];
hash += hash << 10;
hash ^= hash >> 6;
}
hash += hash << 3;
hash ^= hash >> 6;
hash += hash << 16;
return hash;
}
public int Count()
{
double c = 0, E;
for (var i = 0; i < mapSize; i++)
c += 1d / Math.Pow(2, (double)Lookup[i]);
E = alpha_m * mapSize * mapSize / c;
// Make corrections & smoothen things.
if (E <= (5 / 2) * mapSize)
{
double V = 0;
for (var i = 0; i < mapSize; i++)
if (Lookup[i] == 0) V++;
if (V > 0)
E = mapSize * Math.Log(mapSize / V);
}
else
if (E > (1 / 30) * pow_2_32)
E = -pow_2_32 * Math.Log(1 - E / pow_2_32);
// Made corrections & smoothen things, or not.
return (int)E;
}
public void Add(object val)
{
uint hashCode = getHashCode(val.ToString());
int j = (int)(hashCode >> kComplement);
Lookup[j] = Math.Max(Lookup[j], getRank(hashCode, kComplement));
}
}

I know this is an old post but the #buryat implementation has moved, and is in any case incomplete, and a bit on the slow side (sorry o_o ).
I've taken the implementation used by the new Redis release which can be found here and ported it to PHP. The repo is here https://github.com/joegreen0991/HyperLogLog
<?php
class HyperLogLog {
private $HLL_P_MASK;
private $HLL_REGISTERS;
private $ALPHA;
private $registers;
public function __construct($HLL_P = 14)
{
$this->HLL_REGISTERS = (1 << $HLL_P); /* With P=14, 16384 registers. */
$this->HLL_P_MASK = ($this->HLL_REGISTERS - 1); /* Mask to index register. */
$this->ALPHA = 0.7213 / (1 + 1.079 / $this->HLL_REGISTERS);
$this->registers = new SplFixedArray($this->HLL_REGISTERS);
for ($i = 0; $i < $this->HLL_REGISTERS; $i++) {
$this->registers[$i] = 0;
}
}
public function add($v)
{
$h = crc32(md5($v));
$h |= 1 << 63; /* Make sure the loop terminates. */
$bit = $this->HLL_REGISTERS; /* First bit not used to address the register. */
$count = 1; /* Initialized to 1 since we count the "00000...1" pattern. */
while(($h & $bit) == 0) {
$count++;
$bit <<= 1;
}
/* Update the register if this element produced a longer run of zeroes. */
$index = $h & $this->HLL_P_MASK; /* Index a register inside registers. */
if ($this->registers[$index] < $count) {
$this->registers[$index] = $count;
}
}
public function export()
{
$str = '';
for ($i = 0; $i < $this->HLL_REGISTERS; $i++) {
$str .= chr($this->registers[$i]);
}
return $str;
}
public function import($str)
{
for ($i = 0; $i < $this->HLL_REGISTERS; $i++) {
$this->registers[$i] = isset($str[$i]) ? ord($str[$i]) : 0;
}
}
public function merge($str)
{
for ($i = 0; $i < $this->HLL_REGISTERS; $i++) {
if(isset($str[$i]))
{
$ord = ord($str[$i]);
if ($this->registers[$i] < $ord) {
$this->registers[$i] = $ord;
}
}
}
}
/**
* #static
* #param $arr
* #return int Number of unique items in $arr
*/
public function count() {
$E = 0;
$ez = 0;
for ($i = 0; $i < $this->HLL_REGISTERS; $i++) {
if ($this->registers[$i] !== 0) {
$E += (1.0 / pow(2, $this->registers[$i]));
} else {
$ez++;
$E += 1.0;
}
}
$E = (1 / $E) * $this->ALPHA * $this->HLL_REGISTERS * $this->HLL_REGISTERS;
/* Use the LINEARCOUNTING algorithm for small cardinalities.
* For larger values but up to 72000 HyperLogLog raw approximation is
* used since linear counting error starts to increase. However HyperLogLog
* shows a strong bias in the range 2.5*16384 - 72000, so we try to
* compensate for it. */
if ($E < $this->HLL_REGISTERS * 2.5 && $ez != 0) {
$E = $this->HLL_REGISTERS * log($this->HLL_REGISTERS / $ez);
}
else if ($this->HLL_REGISTERS == 16384 && $E < 72000) {
// We did polynomial regression of the bias for this range, this
// way we can compute the bias for a given cardinality and correct
// according to it. Only apply the correction for P=14 that's what
// we use and the value the correction was verified with.
$bias = 5.9119 * 1.0e-18 * ($E*$E*$E*$E)
-1.4253 * 1.0e-12 * ($E*$E*$E)+
1.2940 * 1.0e-7 * ($E*$E)
-5.2921 * 1.0e-3 * $E+
83.3216;
$E -= $E * ($bias/100);
}
return floor($E);
}
}

I implemented loglog and hyperloglog in JS and PHP and well-commented code https://github.com/buryat/loglog