npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2024 – Pkg Stats / Ryan Hefner

webdb

v0.5.0

Published

Client-side database that can be synched with a remote server.

Downloads

8

Readme

WebDB v0.5.0

If the client can't get to the database, we bring the database to the client. project website

WebDB

Traditional web apps get everything they show from the server and send any user input to the server. As the user browses through the site, the same data is fetched over and over. And if the user goes offline, the app stops functioning.

What if we could download the server database to the client and perform queries against it there?

Smart subset

I know what you are thinking. There is way too much data to download the entire database right?

But think about this some more. Facebook has hundreds of milions of active users, who collectively post billions of status updates, photos, videos and chat messages each day. But most of that data is irrelevant to any single user. For most users, only those people they are friends with and the posts made by those friends, are relevant.

What if we could define a subset of our total server data, based on characteristics of the user (such as who their friends are) and actively keep a local copy of that subset synched right there on the user's machine? We would be able to offer offline functionality. We would potentially get extremely fast response times due to no network latency. Life would be better.

This is what WebDB tries to accomplish. It allows you to define a database schema on the client side (corresponding to the subset of data that is relevant to this client) and keep it synched with the server automagically.

It consists of a client-side component (that you are looking at right now) which can be used stand-alone or together with a server side component that connects it to the full-blown server database and handles synch messages.

Get WebDB

WebDB can be used directly from CDN, through a regular download, or installed with NPM.

CDN

This is the easiest way to use WebDB:

<script src="https://cdn.rawgit.com/download/webdb/0.5.0/dist/webdb.min.js"></script>

Download

If you'd rather host the file on your own server, or use the debug version:

NPM

npm install webdb

Include WebDB on your page

WebDB supports the Universal Module Definition and can be used directly from a script tag, through an AMD script loader such as RequireJS, or through a CommonJS loader.

Script tag

Plain and simple:

<script src="//cdn.rawgit.com/download/webdb/0.5.0/dist/webdb.min.js"></script>

AMD loader

define(['webdb'], function(WebDB){
	// WebDB is available here
});

If you want to load the script from CDN, configure like so:

require.config({
	paths: {
		'webdb': '//cdn.rawgit.com/download/webdb/0.5.0/dist/webdb.min',
	}
});

CJS loader

var WebDB = require('webdb');
// WebDB is available here

Use WebDB

Once loaded, you can start using WebDB. In general, you will:

  • Create a new WebDB
  • Create a new schema
  • Synch the DB manually
  • Run queries against the DB
  • Create/Read/Update/Delete items in the DB
  • Enjoy your data being auto-synched to the server as needed.

Create a new WebDB

You can use WebDB completely stand-alone or synched to a server, which makes it that much more powerful.

To create a new WebDB, we invoke the WebDB constructor function, which has this signature:

function WebDB(name, options)

The name is optional and is used to isolate different WebDBs from each other. Multiple invocations of this function using the same name will all return the same object. If not supplied, it defaults to 'webdb'. The options object is optional and can be used to enable and control synching with a remote server.

Creating a stand-alone WebDB

var db = new WebDB('lucky-luke'); // use a unique ID to isolate from other scripts

This creates a new, stand-alone WebDB named lucky-luke, using all the default options. This is a shorthand for:

var db = new WebDB('lucky-luke', {}); // note the empty second object

Which has the same effect, but explicitly passes an empty options objects (so that all will use their defaults).

Synched to a server

var db = new WebDB('lucky-luke', {synch:true, synchUrl:'/api/webdb/synch'}); 

As can be seen in the above example, to get our database synched to a server we need to set synch to true and provide a synchUrl. By default, auto-synch will be set to enabled and will start after the first call to synch(). This allows you to set up your schema and then call synch once after that is done to start the background synching process.

Database options

These are all the options that are available and the default values they get when not overridden.

synch: false                  // set this to true to allow the db to be synched to a remote counterpart
synchUrl: '/api/webdb/synch'  // set this to the correct url on your server 
synchAuto: true               // After first manual synch it will auto-synch, unless you set this to false
synchThrottleMs: 60000        // minimum time between auto-synchs, in ms (default 1 minute)
synchPollMs: 3600000          // minimum time between polling synchs, in ms (default 1 hour)
synchTimeout: 5000            // max time the request may take to complete before it is cancelled, in ms.
synchRetryCount: 2            // how many times to retry failed requests
synchRetryWait: 5000          // time to wait before retrying, in ms.

Create a new schema

To create our schema we use the method createTable, which has this signature:

function createTable(entityType, name, def)

entityType is an optional constructor function used to instantiate entities from this table. name is required and is the name of the table and def contains the table definition, which basically is a collection of column definitions.

var weapons = db.createTable('weapons', {
	'id': {type: Number, pk:true}, 
	'version': {type:Number, version:true},
	'name': {type:String, length:32, unique:true}
}); 

In this example we create a table 'weapons' with three columns. Each table needs at least one column, the primary key, marked with pk:true in the column definition. Each column needs at least a type attribute, set to the constructor function corresponding to the type. We use String and Number to represent the primitive types in the column definition, but in the actual data we use the actual primitive types. In addition to the primary key, tables in synched databases (option synch is true) also require a version column, marked with version:true. This version column is used for detecting stale objects in optimistic locking scenario's.

createTable creates a convenient alias for us on the database objects so we don't actually have to keep a variable around ourselves. We can just use the new table through the new alias:

db.weapons.set({id:1, version:null, name:'Sword'});

In this example, the records themselves will be returned as simple objects:

var sword = weapons.get(1);
sword.prototype.constructor === Object; // true

By supplying a constructor function as the entityType, we can have WebDB return records as instances of that constructor function:

function Weapon(obj) {
	this.id = obj && obj.id !== undefined ? obj.id : null;
	this.version = obj && obj.version !== undefined ? obj.version : null;
	this.name = obj && obj.name !== undefined ? obj.name : null;
}

Constructor functions should accept an initializer object used to initialize the new instance's state.

Then, we can pass the constructor function when we create the table:

var weapons = db.createTable(Weapon, 'weapons', {
	'id': {type: Number, pk:true}, 
	'version': {type:Number, version:true},
	'name': {type:String, length:32, unique:true}
}); 

Now, when we get records from this table, they will be returned as instances of the function we passed:

var sword = weapons.get(1);
sword.prototype.constructor === Object; // false
sword.prototype.constructor === Weapon; // true
sword instanceof Weapon; // true

We can, but don't have to, pass instances of Weapon to set to insert them:

weapons.set(new Weapon({id:2, version:null, name:'Axe'}));
weapons.set({id:3, version:null, name:'Spear'}); // WebDB calls new Weapon behind the scenes
var spear = weapons.get(3);
spear instanceof Weapon; // true

set will invoke the entity type constructor behind the scenes when we pass plain objects to it.

When we create columns that only have a type, we can use a shorthand notation. Instead of mycolumn: {type: Number} we can write just mycolumn: Number. We can create foreign keys to other tables by marking our column as fk:foreignTableName. Foreign key referential integrity is not checked (yet), but foreign keys automatically get a unique index. If a column is not a foreign key but is unique, we can mark it with unique:true. If we suspect we will be doing a lot of searching on a non-unique column, we can manually give it an index by marking it with index:true. The example below illustrates some of these scenarios.

var characters = db.createTable('characters', {
	'id': {type: Number, pk:true}, 
	'version': {type:Number, version:true},
	'firstName': {type:String, length:32, index:true},
	'lastName': {type:String, length:32, index:true},
	'description': String,  // shortcut for {type:String}
	'weaponOfChoice': {type:Number, fk:'weapons'}
});

Synch the database

Synching is an asynchronous process that was designed to happen automatically in the background. It of-course requires a server that implements the WebDB synch protocol. Given such a server and assuming the database was created with the right options, we can start the synching process like so:

db.synch();

As you can see, synching happens through the method synch, which has this signature:

function synch(force)

After the first call to synch, auto-synching kicks in and it's not needed to call synch again, unless you disabled auto-synching or want to force a synch to happen immediately (which can be done by passing true as the force parameter).

synch returns a Promise, which will resolve once the server response has been processed, or be rejected if any errors occurred. Again, ususally you will not need to wait for a synch (you won't even know one is happening), but every once in a while you do, such as on first load, when the database is still completely empty. Here is how you would go about that:

db.synch().then(function ok(){
	// db is succesfully synched
	// do stuff with freshly loaded data here
}).catch(function fail(error){
	// Oh no! Something bad happened!
	console.error('Synch error.', error);
})

Querying and mutating data

WebDB offers a minimalistic interface for querying and mutating data which is still very powerful. It consists of just three functions and a length property:

db.mytable.set(item)
db.mytable.get(criteria)
db.mytable.del(item)
db.mytable.length

Inserting and updating

Inserting and updating are done with the same function: set. Bulk-inserting data is easy, because set accepts multiple arguments, as well as arrays:

weapons.set(
	{id:1, name:'Revolver'},
	{id:2, name:'Shotgun'},
	{id:3, name:'Mini revolver'},
	{id:4, name:'Teeth'}
);

characters.set([
	{id:1,  firstName:'Lucky',    lastName:'Luke',   weaponOfChoice:1, description:'Shoots faster than his shadow'},
	{id:2,  firstName:'Joe',      lastName:'Dalton', weaponOfChoice:1, description:'Leader of the Dalton brothers gang'},
	{id:3,  firstName:'Jack',     lastName:'Dalton', weaponOfChoice:1, description:'Member of the Dalton brothers gang'},
	{id:4,  firstName:'William',  lastName:'Dalton', weaponOfChoice:1, description:'Member of the Dalton brothers gang'},
	{id:5,  firstName:'Averell',  lastName:'Dalton', weaponOfChoice:1, description:'Member of the Dalton brothers gang'},
	{id:6,  firstName:'Billy',    lastName:'The Kid',weaponOfChoice:1, description:'Youngest outlaw of the west'},
	{id:7,  firstName:'Buffalo',  lastName:'Bill',   weaponOfChoice:2, description:''},
	{id:8,  firstName:'Calamity', lastName:'Jane',   weaponOfChoice:2, description:''},
	{id:9,  firstName:'Pat',      lastName:'Poker',  weaponOfChoice:3, description:''},
	{id:10, firstName:'Jesse',    lastName:'James',  weaponOfChoice:1, description:''},
	{id:11, firstName:'Jolly',    lastName:'Jumper', weaponOfChoice:4, description:'The smartest horse in the world'},
	{id:12, firstName:'Rantaplan',lastName:'?',      weaponOfChoice:4, description:'The dumbest dog in the universe'},
]);

Queries

Get a character by ID:

var luckyLuke = characters.get(1);

The first parameter of get is a criteria object. If it's of a type compatible with the primary key column (eg number/Number, string/String, MyCustomId etc) then it will be interpreted as being a primary key.

You can also explicitly specify the primary key as search criteria:

var luckyLuke = characters.get({id:1});

In the same way, you can specify other columns:

var daltons = characters.get({lastName:'Dalton'});

When multiple columns are specified, they are interpreted as AND clauses:

var william = characters.get({firstName:'William', lastName:'Dalton'});

Remember, for fast searches in larger datasets, you should limit your queries to columns that have an index on them (either pk, fk, unique or index is set).

When no arguments are given, get will return all results:

var all = characters.get();

This was designed to be fast, though it's still recommended to cache the result in between multiple consecutive calls.

Counting data

Get amount of records in table:

var characterCount = characters.length; // 12;

Note: Remember, WebDB only knows about records present on the client. If you need to know the total amount of records, including those only on the server, you need an Ajax call.

Get amount of records in resultset:

var daltons = characters.get({lastName:'Dalton'}); 
var daltonCount = daltons.length; // 4;

Deleting data

Delete a single record:

var rantaplan = characters.get({firstName:'Rantaplan'});
characters.del(rantaplan);

Delete multiple records:

characters.del(characters.get({lastName:'Dalton'}));

Under construction

WebDB is currently under heavy development and not ready for production just yet. Use at your own risk!

Roadmap

These features are currently planned to be implemented in WebDB:

Copyright

Copyright 2015 by Stijn de Witt. Some rights reserved.

License

Licensed under Creative Commons Attribution 4.0 International (CC-BY-4.0).