Notes

Databases

What is a Database?

  • without a database, if server stops for any reason, all non-hardcoded data is lost
  • a database is a collection of information/data that have an interface that allows it to be interacted with

SQL vs NoSQL

(pronounced sequel)

SQL

*relational databases

  • been around the longest --- what most people think of when they think of databases
  • tabular databases
  • must define a table ahead of time --- not flexible
  • flat 
======================
A RELATIONAL DATABASE
======================

USERS TABLE
id | name  |  age  |  city        favColor
--------------------------
1  | Tim   |  57   |  NYC         null/false/nil/etc
2  | Ira   |  24   |  Missoula    purple
3  | Sue   |  40   |  Boulder     null/false/nil/etc


COMMENTS TABLE
id |        text
-------------------------------
1  |  "lol"
2  |  "Come to Montana!"
3  |  "I love puppies!!!"
4  |  "Seriously Montana is great!"


USER COMMENTS JOIN TABLE
userId  |  commentId
-----------------------
   1    |     3
   2    |     2
   2    |     4
   3    |     1
  • every user entered must follow the 'USERS TABLE' pattern
    • we're defining users and then adding instances of users to the table
    • if wanted to add favorite color for one person, it would have to be added for everyone
  • join tables relate the data from other tables

NoSQL

  • non-relational databases
  • don't have to define patterns ahead of time --- much more flexible
  • no tables
  • things can be nested --- not flat
  • not necessarily better than SQL databases, most times they're not, but in specific situations they make sense 
==========================
A NON-RELATIONAL DATABASE
==========================

{
    name: 'Ira',
    age: 24,
    city: 'Missoula',
    commetns: [
        {text: 'Come visit Montana!'},
        {text: 'Seriously come visit Montana!!!'},
        {text: Why does no one care about Montana???'}
    ],
    favColor : 'purple'
}

{
    name: 'Tammy',
    age: 24,
    city: 'Missoula',
    commetns: [
        {text: 'Hi'},
    ],
    favFood: 'Ribeye'
}
  • looks like javascript --- called BSON (Binary Javascript Object Notation)
    • basically javascript objects
  • can can just add favorite color to Ira's object
  • can have other objects with different contents like you can see in the comparison between Ira and Tammy

MongoDB

  • a NoSQL database
  • most popular with Node and Express
    • MEAN stack is: Mongo, Express, Angular, Node
Commands
  • mongod - running MongoDB (instructions here are Mac specific)
    • Docs
    • Option 1: run manually in the background
      • to start: mongod --config /usr/local/etc/mongod.conf --fork
      • to stop:
    • Option 2: run (i.e. the mongod process) as a MacOS service using brew (recommended)
      • to start:brew services start mongodb-community@4.4
      • to stop: brew services stop mongodb-community@4.4
  • mongo - opens the MongoDB shell
    • shell is good for testing things out and debugging, but not where we will write most of our code --- most will be written in Express code
  • help - shows some basic features
  • show dbs - shows database names
    • admin and local are the two default dbs it knows about
    • in TWDB we'll make a separate DB for each project to keep things self-contained
  • use - type command and name of database --- if it exists it will use it --- if it doesn't exist it will make it and then use it
    • DBs won't show until they have something in them
    • we add things in by creating collections
  • insert - don't have to create a new collection and then add to it, we can insert immediately
    • example: > db.dogs.insert({name: 'Rusty', breed: 'Mutt'})
      • db refers to the db we're on
      • dogs is the collection
      • after insertion, will get a WriteResult
  • show collections - see all collections in DB
  • find - to show items in collections --- also called retrieving or reading
    • items will have long unique "_id"s assigned by mongo
    • > db.dogs.find() shows all the dogs in the collection
    • example: > db.dogs.find({name: 'Rusty'})
      • will find only the specific entries with dogs named 'Rusty'
  • update - edit a previously entered entry
    • takes two things: something to select by, and thing to update
    • example: > db.dogs.update({name: 'Rusty'}, {breed: 'Corgie')
      • this updates but overwrites the whole entry, so the name is gone and the breed is now the only thing in that entry
    • example: > db.dogs.update({name: 'Rusty'}, {breed: 'Corgie')
      • > db.dogs.update({name: 'Rusty'}, {$set: {name: 'Tater', isCute: true}})
      • keeps name but changes it to 'Tater' and adds isCute to the entry --- but with the $set, the breed wasn't overwritten like the name was in the previous example
  • remove - removes entries from the collection
    • will remove everything that matches
    • example: > db.dogs.remove({breed: 'Mutt'})
  • drop - deletes a whole collection
    • example: > db.dogs.drop()
    • do occasionally when making big changes and have data you don't care about
  • a common acronym is CRUD - Create Read Update Destroy
    • in Mongo the CRUD commands are: insert, find, update, remove
Mongoose
  • it is an ODM - Object Data Mapper
    • a way to write javascript in our files that will interact with our database
  • MongoDB object modelling for Node.js
  • "...provides a straight-forward, schema-based solution to model you application data. It includes built-in type casting, validation, query building, business logic hooks and more, out of the box." (from the Mongoose Homepage)
  • it is a package that makes it easier to interact with MongoDb inside our javascript files
    • can be done without it, but it makes it easier --- like jQuery makes it easier to interact with the DOM but we don't need it
  • see example in repo: the-web-developer-bootcamp/13_databases

Copyright © 2022