Neolithic Programming Machine

    flakeid

    1.0.0 • Public • Published

    FlakeId

    A tiny module to generate time based 64-bit unique id, inspired by Twitter id (snowflake).

    FlakeId takes 42 bit of timestamp, 10 bit of machine id (or any random number you provide), 12 bit of sequence number . As javascript is limited to 53 bit integer precision, FlakeId generates id in string format like "285124269753503744", which can be easily type casted into 64 bit bigint in database.

    Installation

    For Node

    npm install flakeid

    For Browser Include js file

    <script src="https://unpkg.com/flakeid/dist/flakeid.min.js"></script>

    Usage

    Initializtion

    const FlakeId = require('flakeid'); /* on node js only */
     
    //initiate flake
    const flake = new FlakeId({
        mid : 42, //optional, define machine id
        timeOffset : (2013-1970)*31536000*1000 //optional, define a offset time
    });

    Create a instance of flake as shown above which will be used to generate flake ids afterward.

    Id generation

    const id1 = flake.gen(); \\returns something like 285124269753503744
    const id2 = flake.gen(); \\returns something like 285124417543999488

    Options

    mid : (Default to 1) A machine id or any random id. If you are generating id in distributed system, its highly advised to provide a proper mid which is unique to different machines.

    timeOffset : (Defaults to 0) Time offset will be subtracted from current time to get the first 42 bit of id. This help in generating smaller ids.

    Method

    gen : Method to generate id from FlakeId instance.


    As js have 53bit integer precision, Flake Id uses a smart solution by Dan Vanderkam (http://www.danvk.org/hex2dec.html) to convert hex to decimal without loosing precision. Source code for converting hex to decimal is taken from http://www.danvk.org/hex2dec.html which have APACHE LICENCE

    Install

    npm i flakeid

    DownloadsWeekly Downloads

    1,692

    Version

    1.0.0

    License

    MIT

    Unpacked Size

    116 kB

    Total Files

    12

    Last publish

    Collaborators

    • sudhanshu