Skip to content

Latest commit

 

History

History
297 lines (186 loc) · 6.62 KB

README.md

File metadata and controls

297 lines (186 loc) · 6.62 KB
Raw

Document-Based Databases

Slides

Demo

Warm-Up

  1. In what situations would we want to use mocks to test our code?
  2. If I am testing a call to an API, which function should I mock?

Learning Outcomes

  1. Use MongoDB Compass Connect to create and view items in a database
  2. Get comfortable with Mongo queries
  3. Understand how to connect and use MongoDB in a Python web app

Document-Based Databases

Overview

Document-Based Databases:

  • Store data in documents similar to JSON
  • Also called NoSql or non-relational because they don't use the more traditional, relational methods of storage

MongoDB

  • Can have several databases. Use 1 database for 1 project.

  • A database can have several collections. Each collection groups objects of the same type: User, BlogPost, Item, etc.

    • User resource would be stored in users collection
    • BlogPost resource would be stored in blog_posts collection

  • Each collection supports CRUD operations

What is CRUD?

  • Create, Read, Update, Delete

  • Represents the operations we can do on any one type of object

  • Example - Users: Create a new user, show all users, update a user, delete a user

Identifiers

MongoDB gives each document a unique identification number with the key _id.

It's sort of like dropping clothes off at the dry-cleaners: you are given a ticket with a unique identification number so that you can find your item again.

Example

So, if we saved a new project like this to a MongoDB database:

# Python Dictionary Object
{ "title": "A New Project" }

Then it will save something like this:

# MongoDB Object
{
  "_id": ObjectId("507f1f77bcf86cd799439011"),
  "title": "A New Project"
}

What is an ObjectId?

An ObjectId is neither a string nor a number. It is a special class that we can import from bson.objectid.

BSON stands for "bytes JSON" and is how our MongoDB data is typically stored.

>>> from bson.objectid import ObjectId

>>> my_id = ObjectId(b'foo-bar-quux')
ObjectId('666f6f2d6261722d71757578')
>>> str(my_id)
'666f6f2d6261722d71757578'

MongoDB Compass

What is MongoDB Compass?

It is a desktop tool that allows us to easily create, read, update, and delete (CRUD) database objects in MongoDB. It will save you a lot of time when writing your Flask applications!

You can download it here.

Demo

Compass

Activity: Create a Database

  1. Download MongoDB Compass.
  2. Create a database called test_db with a collection songs.
  3. Enter data for your top 5 favorite songs, including the title, album, and artist.
  4. Practice Creating, Updating, and Deleting objects.

Break [10 mins]

PyMongo

About PyMongo

PyMongo is a Python library to do CRUD operations on our database from a Python app.

We can install it with:

$ pip3 install pymongo

A First Program

We can access data from our database as follows:

# Import pymongo and initialize client
from pymongo import MongoClient
client = MongoClient()

# Access our database and collection
db = client['test_db']
collection = db['songs']

# Find a song
my_song = songs.find_one({'title': 'Hot N Cold'})
print(my_song)

Activity: Run our First Program [5 mins]

  1. Create a new Python file called songs.py
  2. Copy the starter code from the previous slide into the file
  3. Run the code!

Collection Operations

Operations Overview

Each CRUD operation has at least one corresponding function in PyMongo. You can read a complete list of available operations here.

CRUD Operation PyMongo Operation
Create collection.insert_one()
Read collection.find_one()
Update collection.update_one()
Delete collection.delete_one()

Read Operations

We can read all songs, returned as a list-like object:

all_songs = songs_collection.find({'artist': 'Rihanna'})

# We can loop over this just like a list
for song in all_songs:
    print(song)

Or query for only one song:

song = songs_collection.find_one({'title': 'Umbrella'})
print(song)

Create Operations

Create a JSON object containing the data, then call insert_one.

new_song = {
    'name': 'Hot N Cold',
    'artist': 'Katy Perry',
    'rating': 5
}

result = songs_collection.insert_one(new_song)

Update Operations

Here, we must pass in a JSON object with the field $set.

query = {'title': 'Hot N Cold'}
new_values = {
    '$set': {
        'rating': 4
    }
}
result = songs_collection.update_one(query, new_values)

Delete Operations

query = {'title': 'Friday'}
result = songs_collection.delete_one(query)

Activity: PyMongo Operations [15 mins]

In your songs.py program, complete the following:

  1. Create another song and put it into the database
  2. Print a list of all songs
  3. Find a song with a particular title, and print it
  4. Update a song with a new artist or rating
  5. Delete a song

Vibe Check

Go to https://make.sc/bew1.1-vibe-check and fill out the form.

🎉 Shout-Outs 🎊

Have any shout-outs? Show appreciation for your peers!

Homework

  • Start on the Playlister tutorial - due Tuesday, Nov. 26
  • Quiz on Thursday covering APIs, JSON, & Testing

Resources