Top 100 Podcasts API

Setup

Open a terminal.
Create a virtual environment (this is optional)
Install the dependencies.

pip3 install -r requirements.txt

How to run the code?

Type in the terminal:

python3 app.py

First, the app creates all necessary tables in the database and it displays the message Populating Database... when this is finished, it will display Database has been populated!!!.

IMPORTANT: Copy the generated URI (in this case http://localhost:5000).

How to create a new user?

To use this API you must be authenticated. You need to create a new user in order to receive a token that will be used to call all the services.

Open Postman (or another REST client).
Use the endpoint http://localhost:5000/api/create-user with POST method.
Add a new header to request called 'Content-Type' with 'application/json' as value.
Add a JSON to the request's body like the following:

If all of this is OK you will see the message New user created!.

Log in

To log in into the API follow these steps:

Use the endpoint http://localhost:5000/api/login with POST method.
In Authorization select Basic Auth.
Provide your username and password like the following:

If login is valid, a token will be generated.

IMPORTANT: make sure to COPY IT!!

How to use this token?

To use all of the following endpoints, you must use the generated token.

In all services do the following:

Add a new header called 'x-access-token' with the token value.

IMPORTANT: The token has is valid for 30 minutes.

Structure code

The code is structured like the following image:

In the root folder (podcast-api) there are two files:

app.py this is the main app and contains all the functions to use on the endpoints. This is in charge of manipulating all the requests and as well as validating the generated token.
requirements.txt this contains all the dependencies that are needed for the application to work.
extensions.py use to avoid a cyclic import in the SQLAlchemy and Marshmallow objects.

In the models folders there are three files:

__init__.py to define the models folder as a Python package and thus be able to import the following to files.
models.py to define the database tables as a classes.
schema.py to define the schemas to use and to select the fields of each model to return.

In the database folder there are three files:

__init__.py to define the database folder as a Python package
config.py to define the app database configuration such as secret key, database location, etc.
database_loader.py to load the gathered records from the iTunes API and save them to the database.

Database and Schema creation

Schema

The database has four tables: user, podcast, genre and an intermediate table for many-to-many relationship called podcast_genre.

The app gets all of results to itunes API, then it gets all unique genres from those. At last, the app adds all genres and podcasts to the database.

Solution to exercises

Exercise 1

Service to provide a search lookup within the podcasts.

The function search filters all podcasts by a name using the LIKE operator because this one makes a search for a specific pattern.

Use the endpoint http://localhost:5000/api/search with the GET method.
Send a JSON containing a name key and a string as a value.

Exercise 2

Service that would allow to save the top 20 podcasts to a separate JSON file

The function save_top_20 gets all podcasts from database sorted in ascending order by id column and save this in files/top_20.json.

Use the endpoint http://localhost:5000/api/top-20 with the POST method.

Exercise 3

Service to replace the top 20 podcasts for the bottom 20 to said JSON file.

The function replace_top_20 gets all podcasts from database sorted in descending order by id column and save this in the same JSON file called files/top_20.json.

Use the endpoint http://localhost:5000/api/replace-top-20 with the POST method.

Exercise 4

Service to remove a podcast, using a given identifier.

The function delete_podcast needs an id to search the podcast to delete. You can use any id from the podcasts previously saved within the JSON file.

If the specified podcast does not exist, this service return an error message else return a success message with Podcast with id ### has been deleted.

Use the endpoint http://localhost:5000/api/<id> with the DELETE method
Replace <id> with a podcast id you want to delete.

Exercise 5

Service to return the podcasts grouped by genres, even if they appear as duplicates within some of categories

The function group_by_genre uses a query to join the three related tables sorted by genre name. After this, this records are inserted in a dictionary to return the response.

Use the endpoint http://localhost:5000/api/group-by-genre with the GET method

yradunchev / podcasts-api