RethinkDB is a new NoSQL database that was designed for realtime apps. What’s so special in it? It provides a new way of developing web applications. RethinkDB can push updated query results to the application.

This DB fits for multiplayer games, streaming analytics applications, connected devices and other cases that can benefit from realtime feeds.

You definitely should not use Rethink when you need full ACID support and when you cannot concede writing ability to data consistency.

Now RethinkDB is available for Linux and OS X and its installation is very easy (Ubuntu):

source /etc/lsb-release && echo "deb http://download.rethinkdb.com/apt $DISTRIB_CODENAME main" | sudo tee /etc/apt/sources.list.d/rethinkdb.list

wget -qO- http://download.rethinkdb.com/apt/pubkey.gpg | sudo apt-key add -

sudo apt-get update

sudo apt-get install rethinkdb

Then execute command “rethinkdb” to start the server. If everything goes OK you’ll see RethinkDB web UI by reaching http://localhost:8080

Working with RethinkDB

The RethinkDB provides ReQL-interfaces (drivers) for many programming languages. ReQL consists of sequences of language-specific commands. Let’s see how to communicate with Rethink using Python driver.

At first, install rethinkdb driver

> sudo pip install rethinkdb

> python

>>> import rethinkdb as r

>>> r.connect( "localhost", 28015).repl()

Let’s create a database:

>>> r.db_create(“library”).run()

create a table:

>>> r.db(“library”).table_create(“books”).run()

insert a record:

>>> r.db("library").table("books").insert([{'name':'The Hunger Games', 'pages':342},'name':'1984', 'pages':240}]).run()

count records:

>>> r.db("library").table("books").count().run()

list all records:

>>> for res in r.db("library").table("books").run(): 
...     print res

You will see the list of objects printed out:

{u’pages’: 342, u’id’: u’a5d62a15-4504-42bf-a3b8-e4df402413bd’, u’name’: u’1984′}

{u’pages’: 342, u’id’: u’a808f0ad-1cf0-42cc-8fcf-7732eb989be8′, u’name’: u’The Hunger Games’}

We can retrieve a single document:

>>> r.db("library").table("books").get('a808f0ad-1cf0-42cc-8fcf-7732eb989be8').run()

So here we have learned some RethinkDB basics. Let’s move to advanced features.

Changefeeds

Now let’s talk about the most interesting part of RethinkDB. If you insert “changes()” command before “run()” it will create a new changefeed thread. Changefeed will notify your app about all changes of your dataset.

A closer look. Run this code in your terminal:

>>> cursor = r.table("books").changes().run()

>>> for d in cursor:

...     print d

Then from another terminal try to modify the dataset:

>>> r.db("library").table("books").filter(r.row["name"]== "1984").delete().run()

In the first terminal changefeed notification will be displayed:

{u’old_val’: {u’pages’: 342, u’id’: u’a5d62a15-4504-42bf-a3b8-e4df402413bd’, u’name’: u’1984′}, u’new_val’: None}

Easy clustering

RethinkDB supports clustering. To create a cluster just start another server with join parameter:

> rethinkdb --port-offset 1 --directory rethinkdb_data2 --join localhost:29015