Tutorial on MongoDB

In this tutorial we'll go over setting up mongodb local environment and some basic queries and data creation/manipulation/deletion (DCL, DML, DDL) operations

NoSQl vs SQL

Each one has its own advantages and disadvantages, NOSQL has no structure to the data, if your schema design does not contain a specific structure or relationship between entities then NOSQL is the option to consider.

But if you want to maintain a relationship between entities then SQL is the way to go

What do we mean by a relationship?

Example

Let us consider a prison database, A prison, a Prisoner, A cell are all entities. 
 A relationship between these entities can be listed as follows. 
 A prison has many prisoners, a cell has one prisoner, a prison has many cells etc..copy

"has many", "has one" are called as relationships.

If you want to the store data in a structure less format then NOSQL will come into picture, let's take an example and define it in SQL and NOSQL.

You write the prisoner entity as follows,

Should have an primary key or an identifier unique to the prisoner
name
age
DOB

Suppose this is the structure you created for the prisoner entity, if in 90% the entries to prisoner entity you did not want to enter the DOB, you still need to add a column to the table in case of SQL. But in case of NOSQL since there is no structure each entry is an independent record. One prisoner can have a DOB and name, one of them can have one name etc..

Thats the primary difference between NOSQL and SQL.

Now lets dive into NOSQL and explore how to use it.

Head over to MongoDB downloads pages, and download the server based on your platform

Mongodb Downloads

For windows

Download the msi installer, which will take a few minutes to download, once that is done, click on the msi file and follow the steps.

Install the complete tool, and follow all default steps. Lastly, there is a checkbox to install mongocompass, do not uncheck that box. We'll go over what is the use of mongocompass later in this tutorial

Wait for a couple of minutes for the installer to finish.

For Mac users

Download the zip file from mongo wesbite by selecting the platform as Macos.

Open the terminal, change your directory to the downloaded location.

Extract the zip file,

tar xzf \entire file name\copy

copy the extracted contents into the /usr/local/mongodb

sudo mv \downloaded folder\ /usr/local/mongodbcopy

change your directory to /usr/local/mongodb/

For windows:

Go into the directory where mongodb was installed, create /data/db directories.

Open command prompt in admin mode.

Change directory to mongodb location.

cd bin 
mongod --directoryperdb --dbpath C:mongodb...copy

lets start the mongo service.

net start mongodbcopy

From the bin directory, run

mongocopy

For Mac:

create two directories data and db,

mkdir -p /data/dbcopy

Change dir to db,

cd /data/dbcopy

Give the current user permissions to access the file

sudo chown 'username' /data/dbcopy

go back to the home directory, cd

Now we need to add the mongo runtime executable to the bash_profile.

Open the .bash_profile in a text editor and add the following dirs paths to the end of the file.

export MONGO_PATH=/usr/local/mongodb"
"export PATH...copy

Now we need to start the mongo server and link the client to that server, so that we can perform all the data operations.

Open two terminals, run the following on one of the terminal "mongod"

And on the other terminal run mongo

Here mongod is the daemon that receives the commands from the mongo CLI which is the client.

To check if the connection is working

Run show dbs on the client and see if you get an output

Now lets run some DCLL, DML commands.

show dbscopy

use myprofilecopy

Lets add a admin user who will be using this database

db.createUser({
                user: 'admin',
   ...copy

Collections

A collection is a set of documents.

What is a document?

A document is a record representing some data.

To create a collection, lets run:

db.createCollection("potentialContacts")copy

All the data in mongo is stored as a JSON.

Lets open a json formatter so that we can insert data into our collection.

db.potentialContacts.insert({
                    ...copy

to see if the data is inserted, lets explore our potentialContact collection

db.potentialContact.find();
            copy

if we want to add multiple items at one go, then we pass an array of data.

db.potentialContacts.insert({
                    ...copy

So suppose you haven't handled the form validation on the frontend, in the case of SQL the database will throw an error saying some fields are missing, but in the case of NOSQL you don't have to worry about that. Now lets view the data we inserted,

db.potentialContacts.find();copy

Now the data is very difficult to read, lets format the data so we can view it as a json.

db.potentialContacts.find().pretty();copy

If in case we want to update an entry, we can do it using any field, but it is not recommended to do so, because there might be multiple entries with the same value, so it is better to use the unique identifier that mongodb has assigned to that record.

db.potentialContacts.update({ name: "James" }, { e...copy

There is a problem with doing this, try to find out what the problem is. OK, the issue here is that the update the query will reset the entire entry to what you specify in the second field, check your update by finding the entries.

db.potentialContacts.find().pretty()
            copy

You can see that the previous update query on "James" record has reset the record and removed the message field. What if we want to retain the existing fields and change only certain fields in the record. We can do this using the set keyword.

db.potentialContacts.update({ name: "Danny" }, {$s...copy

Suppose we want to remove a certain field in a record

db.potentialContacts.update({ name: "Danny" }, {$u...copy

If you want to rename a column or a key

db.potentialContacts.update({$rename: { "email": "...copy

If we want to remove a record.

db.potentialContacts.remove({ name: "Danny" })
   ...copy

Now, we need some data to perform some further operations. Copy the following command or try to insert some data yourself, that way you get more familiar with using the commands.

db.potentialContacts.insert([
                {
  ...copy

Now that we have some starter data, lets look at some queries. To find entries which satify two conditions on the same record:

db.potentialContacts.find({$or: [{name: "James"}, ...copy

This is basically an OR condition, where the name should be either James or Sasha. Lets look at a more complex query. What if we want to find all the messages which are relevant or the messages which are not dummy. To find all the messages which have more than 10 characters

db.potentialContacts.find({
                    me...copy

To find all the messages which have the word opportunity in it.

db.potentialContacts.find({
                    "m...copy

Exercise

Create a collection called "subscribers"
Add 5 entries with email-id, isSubscribed field, subscriptionType
Filter all the entries where isSubscribed is true
Get the total count of subscribers who have subscribed to subscriptionType = 'newsletter'

Using mongocompass

Once you are familiar with using the command line interface, you can use the UI provided and explore your database and collections.

Open MongoCompass and follow the steps to view your database. MongoDb Connection

That brings us to the end of the mongodb tutorial.

For more info head over to MongoDb Docs Next Tutorial

Back to: React Tutorial

Next Tutorial: Node Tutorial