Networked Media Week 5

Remote Data Storage

Mongo and mLab

MongoDB is a NoSQL database or object store that is relatively easy to use.

mLab is a company which offers cloud based MongoDB hosting with a REST API that we can easily utilize.

Let's go through the Quick Start

Create a Single-node, Standard Line, Sandbox database with a name that makes sense for your application. Mine is called "aoactest".

Next you'll need a user to connect, click on the database and create one. Mine is "test", test; probably a bad idea.

Next, we'll want to create a "collection" and add some "documents" to start playing with the database. Mine is called "test", still probably a bad idea.

Mongojs Node Module

There are several node modules that are designed to work with mongo. I find the mongojs module to be the quickest to get started with.

To get started, as with any node module, we need to install it via npm:

npm install mongojs

Following that, we can use it in our applications and connect to our mLab hosted databases:

var mongojs = require('mongojs');
var db = mongojs("username:password@example.com:port/mydb", ["mycollection"]);

To insert a record into the database:

db.mycollection.save({"attribute_to_save":"value_to_save"}, function(err, saved) {
  if( err || !saved ) console.log("Not saved");
  else console.log("Saved");
});

To pull all records from the database:

db.mycollection.find({}, function(err, saved) {
  if (err || !saved) {
  	console.log("No results");
  }
  else {
  	saved.forEach(function(record) {
    	console.log(record);
  	});
  	
  	/* Alternatively you could loop through the records with a "for"
  	for (var i = 0; i < saved.length; i++) {
	  	console.log(saved[i]);
	}
	*/  	
  }
});

To search for records which match a query term:

db.mycollection.find({"attribute":"value_to_search_for"}, function(err, saved) {
  if( err || !saved) console.log("No results");
  else saved.forEach( function(record) {
    console.log(record);
  } );
});

Aside from a straight match as shown above, you can use a number of query operators that mongo supports. One of the most powerful is the regex or regular expression operator which allows you to use powerful regular expression string matching.
This example will do a case-insensitive match for text in an attribute:

db.mycollection.find({"attribute": { $regex: /regular_expression_pattern/, $options: 'i' }}, function(err, saved) {
  if( err || !saved) console.log("No results");
  else saved.forEach( function(record) {
    console.log(record);
  } );
});

We can use a variable such as one that comes in from a query string to define our regular expression as follows:

app.get('/search', function(req, res) {
  var query = new RegExp(req.query.q, 'i');
  db2.posts.find({"keywords": query}, 
    function(err, saved) {
      if( err || !saved) console.log("No results");
      else res.send(saved);
  });
});

Here is the search form that would work with the above route:

<form method="GET" action="/search">
<input type='text' name="q"><input type="submit" name="submit" value="Search">
</form>

More information about regular expressions:
Programming from A to Z Regular Expressions
Lynda Using Regular Expressions Tutorial

To pull a specific record, we use a result's _id attribute:

db.mycollection.findOne({
    _id: mongojs.ObjectId("58b654cf449473350d45047f")
}, function(err, record) {
    console.log(record);
});