How do you convert decimal values to their hexadecimal equivalent in JavaScript?
Convert a number to a hexadecimal string with:
hexString = yourNumber.toString(16);
And reverse the process with:
yourNumber = parseInt(hexString, 16);
If you need to handle things like bit fields or 32-bit colors, then you need to deal with signed numbers. The JavaScript function toString(16) will return a negative hexadecimal number which is usually not what you want. This function does some crazy addition to make it a positive number.
function decimalToHexString(number)
{
if (number < 0)
{
number = 0xFFFFFFFF + number + 1;
}
return number.toString(16).toUpperCase();
}
console.log(decimalToHexString(27));
console.log(decimalToHexString(48.6));
The code below will convert the decimal value d to hexadecimal. It also allows you to add padding to the hexadecimal result. So 0 will become 00 by default.
function decimalToHex(d, padding) {
var hex = Number(d).toString(16);
padding = typeof (padding) === "undefined" || padding === null ? padding = 2 : padding;
while (hex.length < padding) {
hex = "0" + hex;
}
return hex;
}
function toHex(d) {
return ("0"+(Number(d).toString(16))).slice(-2).toUpperCase()
}
For completeness, if you want the two's-complement hexadecimal representation of a negative number, you can use the zero-fill-right shift >>> operator. For instance:
> (-1).toString(16)
"-1"
> ((-2)>>>0).toString(16)
"fffffffe"
There is however one limitation: JavaScript bitwise operators treat their operands as a sequence of 32 bits, that is, you get the 32-bits two's complement.
With padding:
function dec2hex(i) {
return (i+0x10000).toString(16).substr(-4).toUpperCase();
}
The accepted answer did not take into account single digit returned hexadecimal codes. This is easily adjusted by:
function numHex(s)
{
var a = s.toString(16);
if ((a.length % 2) > 0) {
a = "0" + a;
}
return a;
}
and
function strHex(s)
{
var a = "";
for (var i=0; i<s.length; i++) {
a = a + numHex(s.charCodeAt(i));
}
return a;
}
I believe the above answers have been posted numerous times by others in one form or another. I wrap these in a toHex() function like so:
function toHex(s)
{
var re = new RegExp(/^\s*(\+|-)?((\d+(\.\d+)?)|(\.\d+))\s*$/);
if (re.test(s)) {
return '#' + strHex( s.toString());
}
else {
return 'A' + strHex(s);
}
}
Note that the numeric regular expression came from 10+ Useful JavaScript Regular Expression Functions to improve your web applications efficiency.
Update: After testing this thing several times I found an error (double quotes in the RegExp), so I fixed that. HOWEVER! After quite a bit of testing and having read the post by almaz - I realized I could not get negative numbers to work.
Further - I did some reading up on this and since all JavaScript numbers are stored as 64 bit words no matter what - I tried modifying the numHex code to get the 64 bit word. But it turns out you can not do that. If you put "3.14159265" AS A NUMBER into a variable - all you will be able to get is the "3", because the fractional portion is only accessible by multiplying the number by ten(IE:10.0) repeatedly. Or to put that another way - the hexadecimal value of 0xF causes the floating point value to be translated into an integer before it is ANDed which removes everything behind the period. Rather than taking the value as a whole (i.e.: 3.14159265) and ANDing the floating point value against the 0xF value.
So the best thing to do, in this case, is to convert the 3.14159265 into a string and then just convert the string. Because of the above, it also makes it easy to convert negative numbers because the minus sign just becomes 0x26 on the front of the value.
So what I did was on determining that the variable contains a number - just convert it to a string and convert the string. This means to everyone that on the server side you will need to unhex the incoming string and then to determine the incoming information is numeric. You can do that easily by just adding a "#" to the front of numbers and "A" to the front of a character string coming back. See the toHex() function.
Have fun!
After another year and a lot of thinking, I decided that the "toHex" function (and I also have a "fromHex" function) really needed to be revamped. The whole question was "How can I do this more efficiently?" I decided that a to/from hexadecimal function should not care if something is a fractional part but at the same time it should ensure that fractional parts are included in the string.
So then the question became, "How do you know you are working with a hexadecimal string?". The answer is simple. Use the standard pre-string information that is already recognized around the world.
In other words - use "0x". So now my toHex function looks to see if that is already there and if it is - it just returns the string that was sent to it. Otherwise, it converts the string, number, whatever. Here is the revised toHex function:
/////////////////////////////////////////////////////////////////////////////
// toHex(). Convert an ASCII string to hexadecimal.
/////////////////////////////////////////////////////////////////////////////
toHex(s)
{
if (s.substr(0,2).toLowerCase() == "0x") {
return s;
}
var l = "0123456789ABCDEF";
var o = "";
if (typeof s != "string") {
s = s.toString();
}
for (var i=0; i<s.length; i++) {
var c = s.charCodeAt(i);
o = o + l.substr((c>>4),1) + l.substr((c & 0x0f),1);
}
return "0x" + o;
}
This is a very fast function that takes into account single digits, floating point numbers, and even checks to see if the person is sending a hex value over to be hexed again. It only uses four function calls and only two of those are in the loop. To un-hex the values you use:
/////////////////////////////////////////////////////////////////////////////
// fromHex(). Convert a hex string to ASCII text.
/////////////////////////////////////////////////////////////////////////////
fromHex(s)
{
var start = 0;
var o = "";
if (s.substr(0,2).toLowerCase() == "0x") {
start = 2;
}
if (typeof s != "string") {
s = s.toString();
}
for (var i=start; i<s.length; i+=2) {
var c = s.substr(i, 2);
o = o + String.fromCharCode(parseInt(c, 16));
}
return o;
}
Like the toHex() function, the fromHex() function first looks for the "0x" and then it translates the incoming information into a string if it isn't already a string. I don't know how it wouldn't be a string - but just in case - I check. The function then goes through, grabbing two characters and translating those in to ASCII characters. If you want it to translate Unicode, you will need to change the loop to going by four(4) characters at a time. But then you also need to ensure that the string is NOT divisible by four. If it is - then it is a standard hexadecimal string. (Remember the string has "0x" on the front of it.)
A simple test script to show that -3.14159265, when converted to a string, is still -3.14159265.
<?php
echo <<<EOD
<html>
<head><title>Test</title>
<script>
var a = -3.14159265;
alert( "A = " + a );
var b = a.toString();
alert( "B = " + b );
</script>
</head>
<body>
</body>
</html>
EOD;
?>
Because of how JavaScript works in respect to the toString() function, all of those problems can be eliminated which before were causing problems. Now all strings and numbers can be converted easily. Further, such things as objects will cause an error to be generated by JavaScript itself. I believe this is about as good as it gets. The only improvement left is for W3C to just include a toHex() and fromHex() function in JavaScript.
Without the loop:
function decimalToHex(d) {
var hex = Number(d).toString(16);
hex = "000000".substr(0, 6 - hex.length) + hex;
return hex;
}
// Or "#000000".substr(0, 7 - hex.length) + hex;
// Or whatever
// *Thanks to MSDN
Also isn't it better not to use loop tests that have to be evaluated?
For example, instead of:
for (var i = 0; i < hex.length; i++){}
have
for (var i = 0, var j = hex.length; i < j; i++){}
Combining some of these good ideas for an RGB-value-to-hexadecimal function (add the # elsewhere for HTML/CSS):
function rgb2hex(r,g,b) {
if (g !== undefined)
return Number(0x1000000 + r*0x10000 + g*0x100 + b).toString(16).substring(1);
else
return Number(0x1000000 + r[0]*0x10000 + r[1]*0x100 + r[2]).toString(16).substring(1);
}
Constrained/padded to a set number of characters:
function decimalToHex(decimal, chars) {
return (decimal + Math.pow(16, chars)).toString(16).slice(-chars).toUpperCase();
}
For anyone interested, here's a JSFiddle comparing most of the answers given to this question.
And here's the method I ended up going with:
function decToHex(dec) {
return (dec + Math.pow(16, 6)).toString(16).substr(-6)
}
Also, bear in mind that if you're looking to convert from decimal to hex for use in CSS as a color data type, you might instead prefer to extract the RGB values from the decimal and use rgb().
For example (JSFiddle):
let c = 4210330 // your color in decimal format
let rgb = [(c & 0xff0000) >> 16, (c & 0x00ff00) >> 8, (c & 0x0000ff)]
// Vanilla JS:
document.getElementById('some-element').style.color = 'rgb(' + rgb + ')'
// jQuery:
$('#some-element').css('color', 'rgb(' + rgb + ')')
This sets #some-element's CSS color property to rgb(64, 62, 154).
var number = 3200;
var hexString = number.toString(16);
The 16 is the radix and there are 16 values in a hexadecimal number :-)
function dec2hex(i)
{
var result = "0000";
if (i >= 0 && i <= 15) { result = "000" + i.toString(16); }
else if (i >= 16 && i <= 255) { result = "00" + i.toString(16); }
else if (i >= 256 && i <= 4095) { result = "0" + i.toString(16); }
else if (i >= 4096 && i <= 65535) { result = i.toString(16); }
return result
}
If you want to convert a number to a hexadecimal representation of an RGBA color value, I've found this to be the most useful combination of several tips from here:
function toHexString(n) {
if(n < 0) {
n = 0xFFFFFFFF + n + 1;
}
return "0x" + ("00000000" + n.toString(16).toUpperCase()).substr(-8);
}
AFAIK comment 57807 is wrong and should be something like:
var hex = Number(d).toString(16);
instead of
var hex = parseInt(d, 16);
function decimalToHex(d, padding) {
var hex = Number(d).toString(16);
padding = typeof (padding) === "undefined" || padding === null ? padding = 2 : padding;
while (hex.length < padding) {
hex = "0" + hex;
}
return hex;
}
And if the number is negative?
Here is my version.
function hexdec (hex_string) {
hex_string=((hex_string.charAt(1)!='X' && hex_string.charAt(1)!='x')?hex_string='0X'+hex_string : hex_string);
hex_string=(hex_string.charAt(2)<8 ? hex_string =hex_string-0x00000000 : hex_string=hex_string-0xFFFFFFFF-1);
return parseInt(hex_string, 10);
}
As the accepted answer states, the easiest way to convert from decimal to hexadecimal is var hex = dec.toString(16). However, you may prefer to add a string conversion, as it ensures that string representations like "12".toString(16) work correctly.
// Avoids a hard-to-track-down bug by returning `c` instead of `12`
(+"12").toString(16);
To reverse the process you may also use the solution below, as it is even shorter.
var dec = +("0x" + hex);
It seems to be slower in Google Chrome and Firefox, but is significantly faster in Opera. See http://jsperf.com/hex-to-dec.
I'm doing conversion to hex string in a pretty large loop, so I tried several techniques in order to find the fastest one. My requirements were to have a fixed-length string as a result, and encode negative values properly (-1 => ff..f).
Simple .toString(16) didn't work for me since I needed negative values to be properly encoded. The following code is the quickest I've tested so far on 1-2 byte values (note that symbols defines the number of output symbols you want to get, that is for 4-byte integer it should be equal to 8):
var hex = ['0', '1', '2', '3', '4', '5', '6', '7', '8', '9', 'a', 'b', 'c', 'd', 'e', 'f'];
function getHexRepresentation(num, symbols) {
var result = '';
while (symbols--) {
result = hex[num & 0xF] + result;
num >>= 4;
}
return result;
}
It performs faster than .toString(16) on 1-2 byte numbers and slower on larger numbers (when symbols >= 6), but still should outperform methods that encode negative values properly.
Converting hex color numbers to hex color strings:
A simple solution with toString and ES6 padStart for converting hex color numbers to hex color strings.
const string = `#${color.toString(16).padStart(6, '0')}`;
For example:
0x000000 will become #000000
0xFFFFFF will become #FFFFFF
Check this example in a fiddle here
How to convert decimal to hexadecimal in JavaScript
I wasn't able to find a brutally clean/simple decimal to hexadecimal conversion that didn't involve a mess of functions and arrays ... so I had to make this for myself.
function DecToHex(decimal) { // Data (decimal)
length = -1; // Base string length
string = ''; // Source 'string'
characters = [ '0', '1', '2', '3', '4', '5', '6', '7', '8', '9', 'A', 'B', 'C', 'D', 'E', 'F' ]; // character array
do { // Grab each nibble in reverse order because JavaScript has no unsigned left shift
string += characters[decimal & 0xF]; // Mask byte, get that character
++length; // Increment to length of string
} while (decimal >>>= 4); // For next character shift right 4 bits, or break on 0
decimal += 'x'; // Convert that 0 into a hex prefix string -> '0x'
do
decimal += string[length];
while (length--); // Flip string forwards, with the prefixed '0x'
return (decimal); // return (hexadecimal);
}
/* Original: */
D = 3678; // Data (decimal)
C = 0xF; // Check
A = D; // Accumulate
B = -1; // Base string length
S = ''; // Source 'string'
H = '0x'; // Destination 'string'
do {
++B;
A& = C;
switch(A) {
case 0xA: A='A'
break;
case 0xB: A='B'
break;
case 0xC: A='C'
break;
case 0xD: A='D'
break;
case 0xE: A='E'
break;
case 0xF: A='F'
break;
A = (A);
}
S += A;
D >>>= 0x04;
A = D;
} while(D)
do
H += S[B];
while (B--)
S = B = A = C = D; // Zero out variables
alert(H); // H: holds hexadecimal equivalent
You can do something like this in ECMAScript 6:
const toHex = num => (num).toString(16).toUpperCase();
If you are looking for converting Large integers i.e. Numbers greater than Number.MAX_SAFE_INTEGER -- 9007199254740991, then you can use the following code
const hugeNumber = "9007199254740991873839" // Make sure its in String
const hexOfHugeNumber = BigInt(hugeNumber).toString(16);
console.log(hexOfHugeNumber)
To sum it all up;
function toHex(i, pad) {
if (typeof(pad) === 'undefined' || pad === null) {
pad = 2;
}
var strToParse = i.toString(16);
while (strToParse.length < pad) {
strToParse = "0" + strToParse;
}
var finalVal = parseInt(strToParse, 16);
if ( finalVal < 0 ) {
finalVal = 0xFFFFFFFF + finalVal + 1;
}
return finalVal;
}
However, if you don't need to convert it back to an integer at the end (i.e. for colors), then just making sure the values aren't negative should suffice.
I haven't found a clear answer, without checks if it is negative or positive, that uses two's complement (negative numbers included). For that, I show my solution to one byte:
((0xFF + number +1) & 0x0FF).toString(16);
You can use this instruction to any number bytes, only you add FF in respective places. For example, to two bytes:
((0xFFFF + number +1) & 0x0FFFF).toString(16);
If you want cast an array integer to string hexadecimal:
s = "";
for(var i = 0; i < arrayNumber.length; ++i) {
s += ((0xFF + arrayNumber[i] +1) & 0x0FF).toString(16);
}
In case you're looking to convert to a 'full' JavaScript or CSS representation, you can use something like:
numToHex = function(num) {
var r=((0xff0000&num)>>16).toString(16),
g=((0x00ff00&num)>>8).toString(16),
b=(0x0000ff&num).toString(16);
if (r.length==1) { r = '0'+r; }
if (g.length==1) { g = '0'+g; }
if (b.length==1) { b = '0'+b; }
return '0x'+r+g+b; // ('#' instead of'0x' for CSS)
};
var dec = 5974678;
console.log( numToHex(dec) ); // 0x5b2a96
This is based on Prestaul and Tod's solutions. However, this is a generalisation that accounts for varying size of a variable (e.g. Parsing signed value from a microcontroller serial log).
function decimalToPaddedHexString(number, bitsize)
{
let byteCount = Math.ceil(bitsize/8);
let maxBinValue = Math.pow(2, bitsize)-1;
/* In node.js this function fails for bitsize above 32bits */
if (bitsize > 32)
throw "number above maximum value";
/* Conversion to unsigned form based on */
if (number < 0)
number = maxBinValue + number + 1;
return "0x"+(number >>> 0).toString(16).toUpperCase().padStart(byteCount*2, '0');
}
Test script:
for (let n = 0 ; n < 64 ; n++ ) {
let s=decimalToPaddedHexString(-1, n);
console.log(`decimalToPaddedHexString(-1,${(n+"").padStart(2)}) = ${s.padStart(10)} = ${("0b"+parseInt(s).toString(2)).padStart(34)}`);
}
Test results:
decimalToPaddedHexString(-1, 0) = 0x0 = 0b0
decimalToPaddedHexString(-1, 1) = 0x01 = 0b1
decimalToPaddedHexString(-1, 2) = 0x03 = 0b11
decimalToPaddedHexString(-1, 3) = 0x07 = 0b111
decimalToPaddedHexString(-1, 4) = 0x0F = 0b1111
decimalToPaddedHexString(-1, 5) = 0x1F = 0b11111
decimalToPaddedHexString(-1, 6) = 0x3F = 0b111111
decimalToPaddedHexString(-1, 7) = 0x7F = 0b1111111
decimalToPaddedHexString(-1, 8) = 0xFF = 0b11111111
decimalToPaddedHexString(-1, 9) = 0x01FF = 0b111111111
decimalToPaddedHexString(-1,10) = 0x03FF = 0b1111111111
decimalToPaddedHexString(-1,11) = 0x07FF = 0b11111111111
decimalToPaddedHexString(-1,12) = 0x0FFF = 0b111111111111
decimalToPaddedHexString(-1,13) = 0x1FFF = 0b1111111111111
decimalToPaddedHexString(-1,14) = 0x3FFF = 0b11111111111111
decimalToPaddedHexString(-1,15) = 0x7FFF = 0b111111111111111
decimalToPaddedHexString(-1,16) = 0xFFFF = 0b1111111111111111
decimalToPaddedHexString(-1,17) = 0x01FFFF = 0b11111111111111111
decimalToPaddedHexString(-1,18) = 0x03FFFF = 0b111111111111111111
decimalToPaddedHexString(-1,19) = 0x07FFFF = 0b1111111111111111111
decimalToPaddedHexString(-1,20) = 0x0FFFFF = 0b11111111111111111111
decimalToPaddedHexString(-1,21) = 0x1FFFFF = 0b111111111111111111111
decimalToPaddedHexString(-1,22) = 0x3FFFFF = 0b1111111111111111111111
decimalToPaddedHexString(-1,23) = 0x7FFFFF = 0b11111111111111111111111
decimalToPaddedHexString(-1,24) = 0xFFFFFF = 0b111111111111111111111111
decimalToPaddedHexString(-1,25) = 0x01FFFFFF = 0b1111111111111111111111111
decimalToPaddedHexString(-1,26) = 0x03FFFFFF = 0b11111111111111111111111111
decimalToPaddedHexString(-1,27) = 0x07FFFFFF = 0b111111111111111111111111111
decimalToPaddedHexString(-1,28) = 0x0FFFFFFF = 0b1111111111111111111111111111
decimalToPaddedHexString(-1,29) = 0x1FFFFFFF = 0b11111111111111111111111111111
decimalToPaddedHexString(-1,30) = 0x3FFFFFFF = 0b111111111111111111111111111111
decimalToPaddedHexString(-1,31) = 0x7FFFFFFF = 0b1111111111111111111111111111111
decimalToPaddedHexString(-1,32) = 0xFFFFFFFF = 0b11111111111111111111111111111111
Thrown: 'number above maximum value'
Note: Not too sure why it fails above 32 bitsize
rgb(255, 255, 255) // returns FFFFFF
rgb(255, 255, 300) // returns FFFFFF
rgb(0,0,0) // returns 000000
rgb(148, 0, 211) // returns 9400D3
function rgb(...values){
return values.reduce((acc, cur) => {
let val = cur >= 255 ? 'ff' : cur <= 0 ? '00' : Number(cur).toString(16);
return acc + (val.length === 1 ? '0'+val : val);
}, '').toUpperCase();
}
Arbitrary precision
This solution take on input decimal string, and return hex string. A decimal fractions are supported. Algorithm
split number to sign (s), integer part (i) and fractional part (f) e.g for -123.75 we have s=true, i=123, f=75
integer part to hex:
if i='0' stop
get modulo: m=i%16 (in arbitrary precision)
convert m to hex digit and put to result string
for next step calc integer part i=i/16 (in arbitrary precision)
fractional part
count fractional digits n
multiply k=f*16 (in arbitrary precision)
split k to right part with n digits and put them to f, and left part with rest of digits and put them to d
convert d to hex and add to result.
finish when number of result fractional digits is enough
// #param decStr - string with non-negative integer
// #param divisor - positive integer
function dec2HexArbitrary(decStr, fracDigits=0) {
// Helper: divide arbitrary precision number by js number
// #param decStr - string with non-negative integer
// #param divisor - positive integer
function arbDivision(decStr, divisor)
{
// algorithm https://www.geeksforgeeks.org/divide-large-number-represented-string/
let ans='';
let idx = 0;
let temp = +decStr[idx];
while (temp < divisor) temp = temp * 10 + +decStr[++idx];
while (decStr.length > idx) {
ans += (temp / divisor)|0 ;
temp = (temp % divisor) * 10 + +decStr[++idx];
}
if (ans.length == 0) return "0";
return ans;
}
// Helper: calc module of arbitrary precision number
// #param decStr - string with non-negative integer
// #param mod - positive integer
function arbMod(decStr, mod) {
// algorithm https://www.geeksforgeeks.org/how-to-compute-mod-of-a-big-number/
let res = 0;
for (let i = 0; i < decStr.length; i++)
res = (res * 10 + +decStr[i]) % mod;
return res;
}
// Helper: multiply arbitrary precision integer by js number
// #param decStr - string with non-negative integer
// #param mult - positive integer
function arbMultiply(decStr, mult) {
let r='';
let m=0;
for (let i = decStr.length-1; i >=0 ; i--) {
let n = m+mult*(+decStr[i]);
r= (i ? n%10 : n) + r
m= n/10|0;
}
return r;
}
// dec2hex algorithm starts here
let h= '0123456789abcdef'; // hex 'alphabet'
let m= decStr.match(/-?(.*?)\.(.*)?/) || decStr.match(/-?(.*)/); // separate sign,integer,ractional
let i= m[1].replace(/^0+/,'').replace(/^$/,'0'); // integer part (without sign and leading zeros)
let f= (m[2]||'0').replace(/0+$/,'').replace(/^$/,'0'); // fractional part (without last zeros)
let s= decStr[0]=='-'; // sign
let r=''; // result
if(i=='0') r='0';
while(i!='0') { // integer part
r=h[arbMod(i,16)]+r;
i=arbDivision(i,16);
}
if(fracDigits) r+=".";
let n = f.length;
for(let j=0; j<fracDigits; j++) { // frac part
let k= arbMultiply(f,16);
f = k.slice(-n);
let d= k.slice(0,k.length-n);
r+= d.length ? h[+d] : '0';
}
return (s?'-':'')+r;
}
// -----------
// TESTS
// -----------
let tests = [
["0",2],
["000",2],
["123",0],
["-123",0],
["00.000",2],
["255.75",5],
["-255.75",5],
["127.999",32],
];
console.log('Input Standard Abitrary');
tests.forEach(t=> {
let nonArb = (+t[0]).toString(16).padEnd(17,' ');
let arb = dec2HexArbitrary(t[0],t[1]);
console.log(t[0].padEnd(10,' '), nonArb, arb);
});
// Long Example (40 digits after dot)
let example = "123456789012345678901234567890.09876543210987654321"
console.log(`\nLong Example:`);
console.log('dec:',example);
console.log('hex: ',dec2HexArbitrary(example,40));
The problem basically how many padding zeros to expect.
If you expect string 01 and 11 from Number 1 and 17. it's better to use Buffer as a bridge, with which number is turn into bytes, and then the hex is just an output format of it. And the bytes organization is well controlled by Buffer functions, like writeUInt32BE, writeInt16LE, etc.
import { Buffer } from 'buffer';
function toHex(n) { // 4byte
const buff = Buffer.alloc(4);
buff.writeInt32BE(n);
return buff.toString('hex');
}
> toHex(1)
'00000001'
> toHex(17)
'00000011'
> toHex(-1)
'ffffffff'
> toHex(-1212)
'fffffb44'
> toHex(1212)
'000004bc'
Here's my solution:
hex = function(number) {
return '0x' + Math.abs(number).toString(16);
}
The question says: "How to convert decimal to hexadecimal in JavaScript". While, the question does not specify that the hexadecimal string should begin with a 0x prefix, anybody who writes code should know that 0x is added to hexadecimal codes to distinguish hexadecimal codes from programmatic identifiers and other numbers (1234 could be hexadecimal, decimal, or even octal).
Therefore, to correctly answer this question, for the purpose of script-writing, you must add the 0x prefix.
The Math.abs(N) function converts negatives to positives, and as a bonus, it doesn't look like somebody ran it through a wood-chipper.
The answer I wanted, would have had a field-width specifier, so we could for example show 8/16/32/64-bit values the way you would see them listed in a hexadecimal editing application. That, is the actual, correct answer.
Related
I'd like to encode an Integer into a String using 4 different Bits ( A, B, C, D):
Therefore I've wrote a simple Int to customBase conversion you can find here:
function messageToCustomBase(message, charset) {
var base = charset.length,
integer = message,
result = ""
do {
var index = integer % base
result = charset[index] + result
integer = parseInt(integer/base)
} while (integer > 0)
return result
}
The code is working quite fine and encoding looks like this:
0 --> "A"
1 --> "B"
2 --> "C"
...
100 --> "BCBA"
...
10000 --> "CBDABAA"
But due to a special program, I am practically forced to use a special encode algorithm which converts the integer to a string which ...
1. should have as small a length as possible
2. uses maximum four different letters (4 different bits -> A, B, C, D)
3. Prevents never having the same letters next to each other
"ABADADA" -> legit
"ABCDAD" -> legit
"BABCA" -> legit
"CDDABC" -> not legit (because of "D" "D")
"BBBACAB" -> not legit (because of "B" "B" "B")
Question: > How can I avoid multiple characters following each other by editing my messageToBase function?
Example: Encoding Integer 42 will give the result "CCC".
Because in my special case "CCC" is not valid it should be encoded different (maybe to e.g. "CBCDCA", ...)
Note: The string must be able to be decoded later on so just adding random fill-characters between repeating characters after the decoding will not be a working solution.
Note: Here you can find a working fiddle for the base en/decoding
Edit: Theoretically we have to add in cases where it comes to repeating letters an additional bit ("E"). Otherwise it will just lead to complications with other Integers!
So I really got no clue how to fix this problem and any help would be very appreciated. :)
Here's a (rather brute-force) solution:
This simply iterates over your existing encoder with higher "bit depth" until it finds a solution without repeated letters. The encoding depth is prepended to the encoded value (otherwise it would be impossible to reverse the calculation, because you wouldn't have any way of knowing whether "BA" means 2 (encoded to depth B) or 3 (encoded to depth C).
This doesn't guarantee the lowest possible "bit depth", but it does keep the strings as short as possible given your encoding method.
Here's a demonstration of encoding and decoding numbers from 1 through 50:
var encode = function(number) {
if (number == 1) {
return "A-A"; // hacky workaround for endless loop in messageToCustomBase
}
var ret = "";
var bits = Math.floor(Math.log2(number) + 1); // minimum depth required for this number
var chars = "ABCDEFGHIJKLMNOPQRSTUVWXYZ";
for (var i = bits; i < 26; i++) {
var bitIdentifier = chars.charAt(i - 1);
var encoded = messageToCustomBase(number, chars.substr(0, i));
if (encoded.match(/(.)\1/)) {
// duplicate letters found, keep looking
} else {
return bitIdentifier + "-" + encoded
}
}
}
var decode = function(m) {
if (m === 'A-A') {
return 1
} // hack again
var chars = "ABCDEFGHIJKLMNOPQRSTUVWXYZ";
var parts = m.split('-');
var charset = chars.substr(0, parts[0].charCodeAt(0) - 64);
return customBaseToMessage(parts[1], charset)
}
// exactly as your original code:
function messageToCustomBase(message, charset) {
var base = charset.length,
integer = message,
result = ""
do {
var index = integer % base
result = charset[index] + result
integer = parseInt(integer / base)
} while (integer > 0)
return result
}
function customBaseToMessage(message, charset) {
var base = charset.length,
result = 0
for (var i = 0; i < message.length; i++) {
var index = charset.indexOf(message[i])
result = result * base + index
}
return result
}
// encode numbers for testing:
for (var i = 1; i < 51; i++) {
console.log(i, encode(i), decode(encode(i)));
}
You can do this by sticking to base 3 after the first loop, where the characters available are the three you didn't use for the previous character.
As a modification of your encoding code, this looks like
function messageToCustomBase(message, charset) {
var base = charset.length,
integer = message,
result = "",
previous_index = null
var index = integer % base
result = charset[index] + result
integer = parseInt(integer/base)
previous_index = index
while (integer > 0) {
var index = integer % (base - 1)
if (index >= previous_index) {
index++
}
result = charset[index] + result
integer = parseInt(integer/(base - 1))
previous_index = index
}
return result
}
This is pretty close to optimal, but not quite. The reason is that the index for the left-most character will never be zero. Basically, it's doing a "correct" conversion to base (3..4), and is refusing to have representations that start with 0. If you're willing to forgo that, you can get a bit shorter on average.
Is it possible to convert a hex number to a decimal number with a loop?
Example: input "FE" output "254"
I looked at those questions :
How to convert decimal to hex in JavaScript?
Writing a function to convert hex to decimal
Writing a function to convert hex to decimal
Writing a function to convert hex to decimal
How to convert hex to decimal in R
How to convert hex to decimal in c#.net?
And a few more that were not related to JS or loops. I searched for a solution in other languages too in case that I find a way to do it,but I didn't. The first one was the most useful one. Maybe I can devide by 16,compare the result to preset values and print the result, but I want to try with loops. How can I do it?
Maybe you are looking for something like this, knowing that it can be done with a oneliner (with parseInt)?
function hexToDec(hex) {
var result = 0, digitValue;
hex = hex.toLowerCase();
for (var i = 0; i < hex.length; i++) {
digitValue = '0123456789abcdef'.indexOf(hex[i]);
result = result * 16 + digitValue;
}
return result;
}
console.log(hexToDec('FE'));
Alternative
Maybe you want to have a go at using reduce, and ES6 arrow functions:
function hexToDec(hex) {
return hex.toLowerCase().split('').reduce( (result, ch) =>
result * 16 + '0123456789abcdefgh'.indexOf(ch), 0);
}
console.log(hexToDec('FE'));
Just another way to do it...
// The purpose of the function is to convert Hex to Decimal.
// This is done by adding each of the converted values.
function hextoDec(val) {
// Reversed the order because the added values need to 16^i for each value since 'F' is position 1 and 'E' is position 0
var hex = val.split('').reverse().join('');
// Set the Decimal variable as a integer
var dec = 0;
// Loop through the length of the hex to iterate through each character
for (i = 0; i < hex.length; i++) {
// Obtain the numeric value of the character A=10 B=11 and so on..
// you could also change this to var conv = parseInt(hex[i], 16) instead
var conv = '0123456789ABCDEF'.indexOf(hex[i]);
// Calculation performed is the converted value * (16^i) based on the position of the character
// This is then added to the original dec variable. 'FE' for example
// in Reverse order [E] = (14 * (16 ^ 0)) + [F] = (15 * (16 ^ 1))
dec += conv * Math.pow(16, i);
}
// Returns the added decimal value
return dec;
}
console.log(hextoDec('FE'));
Sorry that was backwards, and I can't find where to edit answer, so here is corrected answer:
function doit(hex) {
var num = 0;
for(var x=0;x<hex.length;x++) {
var hexdigit = parseInt(hex[x],16);
num = (num << 4) | hexdigit;
}
return num;
}
If you want to loop over every hex digit, then just loop from end to beginning, shifting each digit 4 bits to the left as you add them (each hex digit is four bits long):
function doit(hex) {
var num = 0;
for(var x=0;x<hex.length;x++) {
var hexdigit = parseInt(hex[x],16);
num = (num << 4) | hexdigit;
}
return num;
}
JavaScript can natively count in hex. I'm finding out the hard way that, in a loop, it converts hex to decimal, so for your purposes, this is great.
prepend your hex with 0x , and you can directly write a for loop.
For example, I wanted get an array of hex values for these unicode characters, but I am by default getting an array of decimal values.
Here's sample code that is converting unicode hex to dec
var arrayOfEmojis = [];
// my range here is in hex format
for (var i=0x1F600; i < 0x1F64F; i++) {
arrayOfEmojis.push('\\u{' + i + '}');
}
console.log(arrayOfEmojis.toString()); // this outputs an array of decimals
When changing to TypeScript I'm not allowed to use escape(string) anymore because it's deprecated. The reason I still use it is that the alternatives encodeURI and encodeURIComponent give a different results.
var s = "Å"
console.log(escape(s));
console.log(encodeURI(s));
console.log(encodeURIComponent(s));
I don't use this for URLs, but for a CSV export.
What are other alternatives that will give me the same result as escape(string)?
In EcmaScript spec there is algorithm:
Call ToString(string).
Compute the number of characters in Result(1).
Let R be the empty string.
Let k be 0.
If k equals Result(2), return R.
Get the character at position k within Result(1).
If Result(6) is one of the 69 nonblank ASCII characters ABCDEFGHIJKLMNOPQRSTUVWXYZ
abcdefghijklmnopqrstuvwxyz 0123456789 #*_+-./, go to step 14.
Compute the 16-bit unsigned integer that is the Unicode character encoding of Result(6).
If Result(8), is less than 256, go to step 12.
Let S be a string containing six characters “%uwxyz” where wxyz are four hexadecimal digits encoding the
value of Result(8).
Go to step 15.
Let S be a string containing three characters “%xy” where xy are two hexadecimal digits encoding the
value of Result(8).
Go to step 15.
Let S be a string containing the single character Result(6).
Let R be a new string value computed by concatenating the previous value of R and S.
Increase k by 1.
Go to step 5.
which can be coded like this:
(function(global) {
var allowed = 'ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789#*_+-./,';
global.escapeString = function(str) {
str = str.toString();
var len = str.length, R = '', k = 0, S, chr, ord;
while(k < len) {
chr = str[k];
if (allowed.indexOf(chr) != -1) {
S = chr;
} else {
ord = str.charCodeAt(k);
if (ord < 256) {
S = '%' + ("00" + ord.toString(16)).toUpperCase().slice(-2);
} else {
S = '%u' + ("0000" + ord.toString(16)).toUpperCase().slice(-4);
}
}
R += S;
k++;
}
return R;
};
})(typeof window == 'undefined' ? global : window);
For an ObjectId in MongoDB, I work with a 24 digit hexadecimal number. Because I need to keep track of a second collection, I need to add 1 to this hexadecimal number.
In my case, here's my value
var value = "55a98f19b27585d81922ba0b"
What I'm looking for is
var newValue = "55a98f19b25785d81922ba0c"
I tried to create a function for this
function hexPlusOne(hex) {
var num = (("0x" + hex) / 1) + 1;
return num.toString(16);
}
This works with smaller hex numbers
hexPlusOne("eeefab")
=> "eeefac"
but it fails miserably for my hash
hexPlusOne(value)
=> "55a98f19b275840000000000"
Is there a better way to solve this?
This version will return a string as long as the input string, so the overflow is ignored in case the input is something like "ffffffff".
function hexIncrement(str) {
var hex = str.match(/[0-9a-f]/gi);
var digit = hex.length;
var carry = 1;
while (digit-- && carry) {
var dec = parseInt(hex[digit], 16) + carry;
carry = Math.floor(dec / 16);
dec %= 16;
hex[digit] = dec.toString(16);
}
return(hex.join(""));
}
document.write(hexIncrement("55a98f19b27585d81922ba0b") + "<BR>");
document.write(hexIncrement("ffffffffffffffffffffffff"));
This version may return a string which is 1 character longer than the input string, because input like "ffffffff" carries over to become "100000000".
function hexIncrement(str) {
var hex = str.match(/[0-9a-f]/gi);
var digit = hex.length;
var carry = 1;
while (digit-- && carry) {
var dec = parseInt(hex[digit], 16) + carry;
carry = Math.floor(dec / 16);
dec %= 16;
hex[digit] = dec.toString(16);
}
if (carry) hex.unshift("1");
return(hex.join(""));
}
document.write(hexIncrement("55a98f19b27585d81922ba0b") + "<BR>");
document.write(hexIncrement("ffffffffffffffffffffffff"));
I was curious te see whether user2864740's suggestion of working with 12-digit chunks would offer any advantage. To my surprise, even though the code looks more complicated, it's actually around twice as fast. But the first version runs 500,000 times per second too, so it's not like you're going to notice in the real world.
function hexIncrement(str) {
var result = "";
var carry = 1;
while (str.length && carry) {
var hex = str.slice(-12);
if (/^f*$/i.test(hex)) {
result = hex.replace(/f/gi, "0") + result;
carry = 1;
} else {
result = ("00000000000" + (parseInt(hex, 16) + carry).toString(16)).slice(-hex.length) + result;
carry = 0;
}
str = str.slice(0,-12);
}
return(str.toLowerCase() + (carry ? "1" : "") + result);
}
document.write(hexIncrement("55a98f19b27585d81922ba0b") + "<BR>");
document.write(hexIncrement("000000000000ffffffffffff") + "<BR>");
document.write(hexIncrement("0123456789abcdef000000000000ffffffffffff"));
The error comes from attempting to covert the entire 24-digit hex value to a number first because it won't fit in the range of integers JavaScript can represent distinctly2. In doing such a conversion to a JavaScript number some accuracy is lost.
However, it can be processed as multiple (eg. two) parts: do the math on the right part and then the left part, if needed due to overflow1. (It could also be processed one digit at a time with the entire addition done manually.)
Each chunk can be 12 hex digits in size, which makes it an easy split-in-half.
1 That is, if the final num for the right part is larger than 0xffffffffffff, simply carry over (adding) one to the left part. If there is no overflow then the left part remains untouched.
2 See What is JavaScript's highest integer value that a Number can go to without losing precision?
The range is 2^53, but the incoming value is 16^24 ~ (2^4)^24 ~ 2^(4*24) ~ 2^96; still a valid number, but outside the range of integers that can be distinctly represented.
Also, use parseInt(str, 16) instead of using "0x" + str in a numeric context to force the conversion, as it makes the intent arguably more clear.
var number = 1310;
should be left alone.
var number = 120;
should be changed to "0120";
var number = 10;
should be changed to "0010";
var number = 7;
should be changed to "0007";
In all modern browsers you can use
numberStr.padStart(4, "0");
https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/String/padStart
function zeroPad(num) {
return num.toString().padStart(4, "0");
}
var numbers = [1310, 120, 10, 7];
numbers.forEach(
function(num) {
var paddedNum = zeroPad(num);
console.log(paddedNum);
}
);
function pad_with_zeroes(number, length) {
var my_string = '' + number;
while (my_string.length < length) {
my_string = '0' + my_string;
}
return my_string;
}
try these:
('0000' + number).slice(-4);
or
(number+'').padStart(4,'0');
Here's another way. Comes from something I did that needs to be done thousands of times on a page load. It's pretty CPU efficient to hard code a string of zeroes one time, and chop as many as you need for the pad as many times as needed. I do really like the power of 10 method -- that's pretty flexible.
Anyway, this is as efficient as I could come up with:
For the original question, CHOOSE ONE of the cases...
var number = 1310;
var number = 120;
var number = 10;
var number = 7;
then
// only needs to happen once
var zeroString = "00000";
// one assignment gets the padded number
var paddedNum = zeroString.substring((number + "").length, 4) + bareNum;
//output
alert("The padded number string is: " + paddedNum);
Of course you still need to validate the input. Because this ONLY works reliably under the following conditions:
Number of zeroes in the zeroString is desired_length + 1
Number of digits in your starting number is less than or equal to your desired length
Backstory:
I have a case that needs a fixed length (14 digit) zero-padded number. I wanted to see how basic I could make this. It's run tens of thousands of times on a page load, so efficiency matters. It's not quite re-usable as-is, and it's a bit inelegant. Except that it is very very simple.
For desired n digits padded string, this method requires a string of (at least) n+1 zeroes. Index 0 is the first character in the string, which won't ever be used, so really, it could be anything.
Note also that string.substring() is different from string.substr()!
var bareNum = 42 + '';
var zeroString = "000000000000000";
var paddedNum = zeroString.substring(bareNumber.length, 14) + bareNum
This pulls zeroes from zeroString starting at the position matching the length of the string, and continues to get zeroes to the necessary length of 14. As long as that "14" in the third line is a lower integer than the number of characters in zeroString, it will work.
function pad(n, len) {
return (new Array(len + 1).join('0') + n).slice(-len);
}
might not work in old IE versions.
//to: 0 - to left, 1 - to right
String.prototype.pad = function(_char, len, to) {
if (!this || !_char || this.length >= len) {
return this;
}
to = to || 0;
var ret = this;
var max = (len - this.length)/_char.length + 1;
while (--max) {
ret = (to) ? ret + _char : _char + ret;
}
return ret;
};
Usage:
someString.pad(neededChars, neededLength)
Example:
'332'.pad('0', 6); //'000332'
'332'.pad('0', 6, 1); //'332000'
An approach I like is to add 10^N to the number, where N is the number of zeros you want. Treat the resultant number as a string and slice off the zeroth digit. Of course, you'll want to be careful if your input number might be larger than your pad length, but it's still much faster than the loop method:
// You want to pad four places:
>>> var N = Math.pow(10, 4)
>>> var number = 1310
>>> number < N ? ("" + (N + number)).slice(1) : "" + number
"1310"
>>> var number = 120
>>> number < N ? ("" + (N + number)).slice(1) : "" + number
"0120"
>>> var number = 10
>>> number < N ? ("" + (N + number)).slice(1) : "" + number
"0010"
…
etc. You can make this into a function easily enough:
/**
* Pad a number with leading zeros to "pad" places:
*
* #param number: The number to pad
* #param pad: The maximum number of leading zeros
*/
function padNumber(number, pad) {
var N = Math.pow(10, pad);
return number < N ? ("" + (N + number)).slice(1) : "" + number
}
I wrote a general function for this. It takes an input control and pad length as input.
function padLeft(input, padLength) {
var num = $("#" + input).val();
$("#" + input).val(('0'.repeat(padLength) + num).slice(-padLength));
}
With RegExp/JavaScript:
var number = 7;
number = ('0000'+number).match(/\d{4}$/);
console.log(number);
With Function/RegExp/JavaScript:
var number = 7;
function padFix(n) {
return ('0000'+n).match(/\d{4}$/);
}
console.log(padFix(number));
No loop, no functions
let n = "" + 100;
let x = ("0000000000" + n).substring(n.length);//add your amount of zeros
alert(x + "-" + x.length);
Nate as the best way I found, it's just way too long to read. So I provide you with 3 simples solutions.
1. So here's my simplification of Nate's answer.
//number = 42
"0000".substring(number.toString().length, 4) + number;
2. Here's a solution that make it more reusable by using a function that takes the number and the desired length in parameters.
function pad_with_zeroes(number, len) {
var zeroes = "0".repeat(len);
return zeroes.substring(number.toString().length, len) + number;
}
// Usage: pad_with_zeroes(42,4);
// Returns "0042"
3. Here's a third solution, extending the Number prototype.
Number.prototype.toStringMinLen = function(len) {
var zeroes = "0".repeat(len);
return zeroes.substring(self.toString().length, len) + self;
}
//Usage: tmp=42; tmp.toStringMinLen(4)
Use String.JS librairy function padLeft:
S('123').padLeft(5, '0').s --> 00123