M220P: MongoDB for Python Developers

 

 

 

Chapter 0: Introduction and Setup

MongoDB URI

 

Given the following MongoDB URI:

 

mongodb+srv://brigitte:bardot@xyz-1234.srv.net/admin

 

which of the following statements is true ?

- The password used to authenticate is bardot

 

 

 

Chapter 0: Introduction and Setup

Setting Up Atlas

 

mongorestore --gzip --archive=sampledata.archive.gz

 

https://www.mongodb.com/docs/database-tools/mongorestore/

mongorestore — MongoDB Database Tools

 

https://www.mongodb.com/try?jmp=university#compass 

Try MongoDB Atlas Products

https://www.mongodb.com/docs/compass/master/aggregation-pipeline-builder/

Aggregation Pipeline Builder — MongoDB Compass

https://www.mongodb.com/docs/compass/master/export-pipeline-to-language/

Export Pipeline to Specific Language — MongoDB Compass

 

 

Chapter 0: Introduction and Setup

Lab (Ungraded): Create or Reuse Atlas Cluster

 

 

Chapter 0: Introduction and Setup

OVERVIEW

 

 

 

Chapter 0: Introduction and Setup

README: Setting Up mflix

 

 

 

# navigate to the mflix-python directory

cd mflix-python

 

# enable the "conda" command in Terminal

echo ". /anaconda3/etc/profile.d/conda.sh" >> ~/.bash_profile

source ~/.bash_profile

 

# create a new environment for MFlix

conda create --name mflix

 

# activate the environment

conda activate mflix

 

pip install -r requirements.txt

 

mv dotini_unix .ini  # on Unix
rename dotini_win .ini # on Windows

 

 

 

from bson.py3compat import abc, string_type, PY3, text_type

ImportError: cannot import name 'abc' from 'bson.py3compat' (C:\Users\formin\AppData\Local\Programs\Python\Python39\lib\site-packages\bson\py3compat.py)

 

-> pip uninstall bson

-> uninstall pymongo

-> pip install pymongo

 

pytest -m connection

 

 

 

Chapter 0: Introduction and Setup

Ticket: Connection

 

python run.py

 

Now proceed to the status page to run the full suite of integration tests and get your validation code.

After passing the relevant tests, what is the validation code for Connection?

 

 

 

 

 

 

-> 5a9026003a466d5ac6497a9d

 

 

 

 

Chapter 1: Driver Setup

Introduction to Chapter 1

 

Chapter 1: Driver Setup

MFlix Application Architecture

 

During this course, which of the following files in mflix will you have to edit?

-> db.py

 

 

 

Chapter 1: Driver Setup

MongoClient

 

Chapter 1: Driver Setup

Basic Reads

 

What do the methods find() and find_one() have in common?

- They are used to query documents in a collection.

- They accept a query predicate.

- They accept a field projection.

 

 

 

 

User Story

"As a user, I'd like to be able to search movies by country and see a list of movie titles. I should be able to specify a comma-separated list of countries to search multiple countries."

 

Task

Implement the get_movies_by_country method in db.py to search movies by country and use projection to return the title and _id field. The _id field will be returned by default.

You can find examples in notebooks/your_first_read.ipynb.

 

MFlix Functionality

Once you complete this ticket, the UI will allow movie searches by one or more countries.

 

Testing and Running the Application

Make sure to look at the tests in test_projection.py to understand what is expected.

You can run the unit tests for this ticket by running:

 

pytest -m projection

 

Once the unit tests are passing, run the application with:

 

python run.py

 

Now proceed to the status page to run the full suite of integration tests and get your validation code.

After passing the relevant unit tests, what is the validation code for Projection?

 

 

def get_movies_by_country(countries):

"""

Finds and returns movies by country.

Returns a list of dictionaries, each dictionary contains a title and an _id.

"""

try:

 

"""

Ticket: Projection

 

Write a query that matches movies with the countries in the "countries"

list, but only returns the title and _id of each movie.

 

Remember that in MongoDB, the $in operator can be used with a list to

match one or more values of a specific field.

"""

 

# TODO: Projection

# Find movies matching the "countries" list, but only return the title

# and _id. Do not include a limit in your own implementation, it is

# included here to avoid sending 46000 documents down the wire.

#return list(db.movies.find().limit(1))

return list(db.movies.find({"countries": {"$in": countries}},{"title":1}))

except Exception as e:

return e

 

 

 

-> 5a94762f949291c47fa6474d

 

 

 

Chapter 1: Driver Setup

Ticket: Text and Subfield Search

 

User Story

"As a user, I'd like to be able to search movies by cast members, genre, or perform a text search of the plot summary, full plot, and title."

 

Task

For this ticket, you will need to modify the method build_query_sort_project in db.py to allow the following movie search criteria:

genres: finds movies that include any of the wanted genres.

Already, the build_query_sort_project method is able to return results for two different types of movie search criteria:

text: performs a text search in the movies collection

cast: finds movies that include any of the wanted cast

You just need to construct the query that queries the movies collection by the genres field.

 

Hint

Check the implementation of similar formats of search criteria - the genres query should be similar.

 

MFlix Functionality

Once you complete this ticket, the UI will allow movie searches by members of the cast, movie genres, movie title, and plot summary.

A text index was created for you when you restored the collections with mongorestore, so these queries will be performant once they are implemented.

 

Testing and Running the Application

Make sure to look at the tests in test_text_and_subfield_search.py to understand what is expected.

You can run the unit tests for this ticket with:

 

pytest -m text_and_subfield_search

 

Once the unit tests are passing, run the application with:

 

python run.py

 

Now proceed to the status page to run the full suite of integration tests and get your validation code.

After passing the relevant tests, what is the validation code for Text and Subfield Search?

 

def build_query_sort_project(filters):

"""

Builds the `query` predicate, `sort` and `projection` attributes for a given

filters dictionary.

"""

query = {}

# The field "tomatoes.viewer.numReviews" only exists in the movies we want

# to display on the front page of MFlix, because they are famous or

# aesthetically pleasing. When we sort on it, the movies containing this

# field will be displayed at the top of the page.

sort = [("tomatoes.viewer.numReviews", DESCENDING), ("_id", ASCENDING)]

project = None

if filters:

if "text" in filters:

query = {"$text": {"$search": filters["text"]}}

meta_score = {"$meta": "textScore"}

sort = [("score", meta_score)]

project = {"score": meta_score}

elif "cast" in filters:

query = {"cast": {"$in": filters["cast"]}}

elif "genres" in filters:

 

"""

Ticket: Text and Subfield Search

 

Given a genre in the "filters" object, construct a query that

searches MongoDB for movies with that genre.

"""

 

# TODO: Text and Subfield Search

# Construct a query that will search for the chosen genre.

#query = {}

query = {"genres": {"$in": filters["genres"]}}

return query, sort, project

 

 

def get_movies(filters, page, movies_per_page):

"""

Returns a cursor to a list of movie documents.

 

Based on the page number and the number of movies per page, the result may

be skipped and limited.

 

The `filters` from the API are passed to the `build_query_sort_project`

method, which constructs a query, sort, and projection, and then that query

is executed by this method (`get_movies`).

 

Returns 2 elements in a tuple: (movies, total_num_movies)

"""

query, sort, project = build_query_sort_project(filters)

if project:

cursor = db.movies.find(query, project).sort(sort)

else:

cursor = db.movies.find(query).sort(sort)

 

total_num_movies = 0

if page == 0:

total_num_movies = db.movies.count_documents(query)

"""

Ticket: Paging

 

Before this method returns back to the API, use the "movies_per_page"

argument to decide how many movies get displayed per page. The "page"

argument will decide which page

 

Paging can be implemented by using the skip() and limit() methods against

the Pymongo cursor.

"""

 

# TODO: Paging

# Use the cursor to only return the movies that belong on the current page.

#movies = cursor.limit(movies_per_page)

if page == 0:

movies = cursor.limit(movies_per_page)

else:

movies = cursor.skip(int(page) * int(movies_per_page)).limit(movies_per_page)

 

return (list(movies), total_num_movies)

 

 

-> 5a96a6a29c453a40d04922cc