##DoQmentDB - A Promise-Based DocumentDB Client

DoQmentDB is a tiny layer that provides the simplicity of MongoDB for DocumentDB users(support schema, hooks/middleware, atomic-transactions, udf and more).

##Table of contents:

• 
• Get Started
• Examples
• Changelog
• Database
  • create
  • insert
  • getDatabase
  • find
  • findById
  • findOrCreate
  • remove
  • use
• Collection
  • create
  • insert
  • createOrUpdate
  • upsert
  • getCollection
  • find
  • findOne
  • findById
  • findAndRemove
  • findOneAndRemove
  • findAndModify
  • findOneAndModify
  • findOrCreate
  • update
  • updateOne
• Queries
• Operations
• Schema
• Atomic Transactions
• Middleware
  • pre
  • post

#Get Started (1) You can install DoQmentDB using 2 different methods:

• clone & build this repository
• via npm: by running $ npm install doqmentdb from your terminal

(2) Add to your project:

var DoqmentDB = require('doqmentdb');

(3) Start Playing with DoqmentDB:

var DoQmentDB  = require('doqmentdb');
// Create DocumentDB connection
var connection = new (require('documentdb').DocumentClient)(HOST, OPTIONS);
// Pass connection and database-name, if `test` is not exist it will create one.
var db = new DoQmentDB(connection, 'test');
// Create a CollectionManager instance, if `users` is not exist it will create one.
var users = db.use('users');
// Using schema
users.schema(model);
// Add hooks(see `users` full example)
users.pre('save', function(next) {
  var doc = this;
  doc.createdAt = new Date().toString();
  next();
});

// Each http function returns a `Promise` with two specific methods: success and error.
users.create({ name: '6534' })
  .then(console.log);

users.findById(1)
  .then(console.log);

users.findAndRemove({ isAdmin: false })
  .then(console.log);

#Database Create a DatabaseManager by passing connection and databaseName.

var DoQmentDB  = require('doqmentdb');
// Create DocumentDB connection
var connection = new (require('documentdb').DocumentClient)(HOST, OPTIONS);
// if `test` is not exist it will create one
var db = new DoQmentDB(connection, 'test');

##create Get name and crete new collection in the used db.
Usage: db.create(string)
Aliases: insert
Returns: Object

db.create('users')
  .then(console.log);

##getDatabase Return the used database.
Usage: db.getDatabase()

db.getDatabase()
  .then(console.log);

##find find collection by given object params.
Note: to return all documents, omit params argument or pass an empty object({}).
Usage: db.find(object[optional])
Returns: Array

db.find()
  .then(console.log); // Return all collections

db.find({ id: 'users' })
  .then(console.log); // Return collections where id equal to `users`

##findById find collection by given string id.
Usage: db.findById(string)
Returns: Object

db.findById('users')
  .then(console.log);

##findOrCreate get object properties, search for collection, if it not exist create one.
Usage: db.findOrCreate(object)
Returns: Object

db.findOrCreate({ name: 'users', id: '#1' })
  .then(console.log);

##remove get collection id as a String, if it exist - remove it and return undefined, else return false.
Usage: db.remove(string)
Returns: undefined or Boolean

db.remove('test')
  .then(console.log);

##use get collection name and return CollectionManager instance.
Note: if the given collection is not exist it will create one.
Usage: var coll = db.use(string);
Returns: object instanceof CollectionManager

var users = db.use('users'); // This operation is not async

#Collection Create a CollectionManager by passing to .use function a collection name.

var users = db.use('users');
console.log(users.constructor.name); // Collection

##create get object properties, and create new document under the used collection.
Usage: users.create(object)
Aliases: insert
Returns: Object

users.create({ name: 'Ariel', admin: true })
  .then(console.log); // { name: 'Ariel', admin: true, id: '8...31', _self: ... }

##createOrUpdate create a new document under the current collection, or update an existing document with the same id.
Usage: users.createOrUpdate(object)
Aliases: upsert
Returns: Object

users.upsert({ id: 'my_user_id', admin: true })
  .then(console.log); // { id: 'my_user_id', admin: true, _self: ... }

##getCollection return the used collection.
Usage: users.getCollection()

users.getCollection()
  .then(console.log);

##find get object properties and return array of results.
Usage: users.find(object)
Note: to return all collections, omit params argument or pass an empty object({}).
Returns: Array

users.find({ active: true })
  .then(console.log);

##findOne get object properties and return the first matching result.
Usage: users.findOne(object)
Returns: Object

users.findOne({ active: true, name: 'Bar' })
  .then(console.log);

##findById find document by giving a string id.
Usage: users.findById(string)
Returns: Object

users.findById('53...3')
  .then(console.log);

##findAndRemove get object properties to search, find the equivalents and remove them.
Usage: users.findAndRemove(object)
Returns: Array
Note: if you want support atomic-transactions(i.e: do things concurrently, e.g: distributed system), you need use this method prefix with $ sign.

users.findAndRemove({ name: 'Ariel' })
  .then(console.log);

// Using stored procedure
users.$findAndRemove({ name: 'Ariel' })
  .then(console.log);

// Remove all users
users.findAndRemove({})
  .then(console.log);

##findOneAndRemove get object properties, and remove the first matching result.
Usage: users.findOneAndRemove(object)
Returns: undefined or Boolean
Note: if you want support atomic-transactions(i.e: do things concurrently, e.g: distributed system), you need use this method prefix with $ sign.

users.findOneAndRemove({ name: 'Ariel', admin: true })
  .then(console.log);

// Using stored procedure
users.$findOneAndRemove({ name: 'Ariel', admin: true })
  .then(console.log);

##findAndModify get object properties to search, find the equivalents and modify them(extend operation).
Usage: users.findAndModify(object, extend)
Aliases: update
Returns: Array
Note: if you want support atomic-transactions(i.e: do things concurrently, e.g: distributed system), you need use this method prefix with $ sign.

users.update({ name: 'Ariel', admin: true }, { admin: false })
  .then(console.log);

// Push 'a' and 'b' to `list` field(do it concurrently)
['a', 'b'].forEach(function(ch) {
  users.$update({}, { list: { $push: ch } });
});

##findOneAndModify get object properties and modify(extend operation) the first matching.
Usage: users.findOneAndModify(object, extend)
Aliases: updateOne
Returns: Object
Note: if you want support atomic-transactions(i.e: do things concurrently, e.g: distributed system), you need use this method prefix with $ sign.

users.findOneAndModify({ admin: false }, { admin: true })
  .then(console.log);

// Using stored procedure
users.$findOneAndModify({ admin: false }, { admin: true })
  .then(console.log);

##findOrCreate get object properties, search for document, if it not exist create one.
Usage: users.findOrCreate(object)
Returns: Object
Note: if you want support atomic-transactions(i.e: do things concurrently, e.g: distributed system), you need use this method prefix with $ sign.

users.findOrCreate({ admin: false, name: 'Ariel' })
  .then(console.log);

// Using stored procedure
users.$findOrCreate({ admin: false, name: 'Ariel' })
  .then(console.log);

#Queries ###Operators

• Logical & Conjunctive:
  • $or OR
  • $and AND
  • $not NOT
  • $nor NOT(... OR ...)
• Comparison:
  • $gt >
  • $gte >=
  • $lt <
  • $lte <=
  • $ne <> or !=
• UDF:
  • $in like Array.prototype.some(...)
  • $all like Array.prototype.every(...)
  • $type typeof value
  • $regex new RegExp(...).test(value)
  • $size test array.length

###Examples

users.find({ a: 1, b: 2, c: '3' })
// ... r WHERE r.a=1 AND r.b=2 AND r.c="3"

users.find({ $or: [{ a: 2, b: 3}, { c: 3 }] })
// ... r WHERE ((r.a=2 AND r.b=3) OR r.c=3)

users.find({ $not: { a: 1, b: 2, c: 3 } })
// ... r WHERE NOT(r.a=1 AND r.b=2 AND r.c=3)

users.find({ $nor: [ { a: 1 }, { b: 3 }]})
// ... r WHERE NOT(r.a=1 OR r.b=3)

users.find({ $nor: [ { a: 1, b: 1 }, { c: 3 } ] })
// ... r WHERE NOT((r.a=1 AND r.b=1) OR r.c=3)

users.find({ $not: { name: { $gt: 3 }, age: 12 } })
// ... r WHERE NOT(r.name > 3 AND r.age=12)

users.find({ $not: { name: { $ne: 'bar' } } })
// ... r WHERE NOT(r.name <> "bar")

users.find({ $or: [
        { name: { $ne: 'Ariel' } },
        { age: { $lte: 26 } },
        { $and: [
          { isAdmin: { $ne: false } },
          { isUser: { $ne: false } }
        ]}
      ]})
// ... r WHERE r.name <> "Ariel" OR r.age <= 26 OR (r.isAdmin <> false AND r.isUser <> false)

users.find({ coins: { $in: 2 } })
// ... r WHERE inUDF(r.coins, 2)

users.find({ $not: { age: { $type: 'number' } } })
// ... r WHERE NOT(typeUDF(r.age, "number"))

#Operations When using one of the update operations(e.g: .update(), .findAndModify(), etc...), you could use the build-in prototype functions(based on the type) prefixing with $ sign.
Usage: users.update({ ... }, { keyName: { $method: value } })
Note: value could be single or array of arguments.

// Find all, and push 2 to `arr` field
users.update({}, { arr: { $push: 2 } });

// Suffix all users `name` with #
users.update({}, { name: { $concat: '#' } });

// Trim the name from `foo` to `o`
users.update({  name: 'foo' }, { name: { $substr: [1,1] } });

#Schema Manage your documents with schema.
fields:

• type
  • required
  • used for type comparing, (e.g: String, Boolean, Number, etc..).
• default
  • optional
  • value fallback
• regex
  • optional
  • regex validation, (e.g: email validation - /^[a-zA-Z0-9@:%_\+.~#?&//=|/d]{10,}$/).
• error
  • optional
  • return message to fields that fail in the validation phase(regex/type). see: example
• expose
  • optional
  • expose by default is true, unless you set it to false, it's means that all the find operations returns the documents without exposing this fields. see: example

Example using schema:
schema: model.js

module.exports = {
  /**
   * @field name
   * @default no default value
   */
  name: {
    type: String,
    'default': ''
  },

  /**
   * @field email
   * @default no default value
   * @regex email, min-length = 10
   */
  email: {
    type: String,
    'default': '',
    regex: /^[a-zA-Z0-9@:%_\+.~#?&//=|/d]{10,}$/,
    error: '`email` must be type string, valid email address, and least 10 chars',
    expose: true
  },

  /**
   * @field password
   * @default no default value
   * @regex password
   */
  password: {
    type: String,
    'default': '',
    regex: /^.*(?=.{8,})(?=.*[a-zA-Z])(?=.*\d)(?=.*[!#$%&? "]).*$/,
    error: '`password` must be type string, contain 8 chars and at least one number, ' +
      'one letter and one unique character such as !#$%&? "',
    expose: false
  },

  /**
   * @field isAdmin
   * @default false
   */
  isAdmin: {
    type: Boolean,
    'default': false
  }
};

using schema(model.js)

var DoQmentDB  = require('doqmentdb');          
var model      = require('./model');            // Get model/schema
var connection = new (require('documentdb')     // Create DocumentDB connection
.DocumentClient)(CONFIG.HOST, CONFIG.OPTIONS);

var db = new DoQmentDB(connection, CONFIG.DB);  // Create DBManager 'test'
var users = db.use('users');                    // Create CollectionManager 'users'
users.schema(model);                            // Using schema

users.create({ password: 'Ar2!as_s'})
  .then(console.log)
  .catch(console.log);
  /*
   [Error:
   `email` must be type string, valid email address, and least 10 chars
   ]
   */

users.create({ name: 'Ariel', email: 'ariel.com', password: 'Ar2!as_s'})
  .then(console.log)
  .catch(console.log);
/*
 [Error:
 `email` must be type string, valid email address, and least 10 chars
 ]
 */

users.create({ name: 'Ariel', email: '[email protected]', password: 'Ar2!as_s'})
  .then(console.log)
  .catch(console.log);
/*
 {
   name: 'Ariel',
   email: '[email protected]',
   password: 'Ar2!as_s',
   id: '2eb7...c0',
    ...
 }
 */
users.find({})
  .then(console.log);
/*
 Get all documents but without exposing fields(i.e: omit `password` field)
 */

see: How to architect your models

#Middleware Middleware/Hooks are executed at the document level(create/save/insert, update, remove/delete).
There are two types of middleware, pre and post.

##pre Usage: users.pre(operation, callback)
Note: pre middleware are executed one after another, when each middleware calls next.
Example:

users.pre('save', function(next) {
  var doc = this;
  doc.createdAt = new Date().toString();
  next();
}, function(next) {
  var doc = this;
  doc.updatedAt = new Date().toString();
  next();
});

// Do something async
users.pre('save', function(next) {
  var doc = this;
  bcrypt.genSalt(10, function(err, salt) {
    bcrypt.hash(doc.password, salt, function(err, hash) {
      doc.password = hash;
      next();
    });
  });
});
// ##Note: the order is importatnt, this example order:
// `createdAt()`, `updatedAT()`, `hash/bcrypt()`, and then the `.create` operation will called

##post Usage: users.post(operation, callback)
Note: post middleware are executed in parallel.
Example:

users.post('save', function(doc) {
  logger(new Date(), doc, 'saved!')
});

#Atomic Transactions Since v0.2.6 DoQmentDB supports atomic-transactions using a built-in sporcs(i.e: stored procedures) to handle concurrently well.
Note: To perform some operation this way, you should prefix it with $.
Read More: DocumentDB - Atomic Transactions

// Lets take some example of `consuming` from two differents
// Service-Bus queues and update the same `model`/`document`
//
// Note: This also could happen in a distributed system, when two operations happens in parallel

// We have a `stores` collection that holds the `sales` and the `users`
// fields per `store`(a Document)
// We are using the `atomic` version of `update`, because we don't want to lose data
sbs.receiveQueueMessage('sales', function(msg) {
  stores.$update({ id: msg.id }, { sales: { $push: msg.sale } });
  // Polling again...
});

sbs.receiveQueueMessage('users', function(msg) {
  stores.$update({ id: msg.id }, { users: { $push: msg.user } });
  // ...
});

#Examples

• Koa DocumentDB Example - A users CRUD application using Koa and DocumentDB.
• Express DocumentDB Example - Express application using DocumentDB.

#Changelog ##0.2.9

• Schema Fix- issue #26

##0.2.8

• Add aliases: updateOne and $updateOne(the conccurent one)
• refactor the built-in stored procedure(findAndModify)

##0.2.6 Since 0.2.6 DoQmentDB support atomic transactions using DocumentDB stored procedures.
Methods that support:

• update/findAndModify
• findOneAndModify
• findOrCreate
• findAndRemove
• findOneAndRemove

If you want to use one of this methods, you should use them prefix with $ sign.

// Push 'a' and 'b' to `list` field(do it concurrently)
['a', 'b'].forEach(function(ch) {
  users.$update({}, { list: { $push: ch } });
});