So far, our RESTful API has performed CRUD operations on a simple in-memory dictionary that acted as a data repository. The dictionary is never persisted and, therefore, the data is lost whenever we restart our Flask development server.

Now, we want to create a more complex RESTful API with Flask RESTful to interact with a database model that allows us to work with notifications that are grouped into notification categories. In our previous RESTful API, we used a string attribute to specify the notification category for a notification. In this case, we want to be able to easily retrieve all the notifications that belong to a specific notification category and, therefore, we will have a relationship between a notification and a notification category.

We must be able to perform CRUD operations on different related...

The following table shows the HTTP verbs, the scope, and the semantics for the methods that our new API must support. Each method is composed by an HTTP verb and a scope, and all the methods have well-defined meanings for all the resources and collections:

Make sure you quit Flask's development server. You just need to press Ctrl + C in the Terminal or Command Prompt window in which it is running.

Now, we will install a number of additional packages. Make sure you have activated the virtual environment we created in the previous chapter and named Flask01. After you activate the virtual environment, it is time to run numerous commands, which will be the same for either macOS, Linux, or Windows.

Now, we will edit the existing requirements.txt file to specify the additional set of packages that our application requires to be installed in any supported platform. This way, it will be extremely easy to repeat the installation of the specified packages with their versions in any new virtual environment.

Use your favorite editor to edit the existing text...

Now, we will create the PostgreSQL 10.5 database that we will use as a repository for our API. You will have to download and install a PostgreSQL database server if you aren't already running it in your computer or on a development server. You can download and install this database management system from its web page (http://www.postgresql.org). If you are working with macOS, Postgres.app provides a really easy way to install and use PostgreSQL on this operating system. You can refer to it from http://postgresapp.com. If you are working with Windows, EnterpriseDB and BigSQL provide graphics installers that simplify the configuration process on modern Windows server or desktop versions (visit https://www.postgresql.org/download/windows for more information).

Notice that the examples have been tested with PostgreSQL 10.5 on macOS, Linux, and Windows.

If you are using the same virtual environment we have created for the previous example, or you downloaded the code sample, the service folder already exists. If you created a new virtual environment, create a folder named service within the root folder for the  virtual environment created.

Create a new config.py file within the service folder. The following lines show the code that declares variables that determine the configuration for Flask and SQLAlchemy. The SQL_ALCHEMY_DATABASE_URI variable generates an SQLAlchemy URI for the PostgreSQL database. Make sure you specify the desired database name in the value for DB_NAME and that you configure the user, password, host, and port based on your PostgreSQL configuration. If you followed the previous steps, use the settings specified in these steps. The code file for the sample is included in the restful_python_2_02_01...

Now, we will create the models that we will use to represent and persist the notification categories, notifications, and their relationships in the PostgreSQL database.

Open the service/models.py file and replace its contents with the following code. The lines that declare fields related to other models are highlighted in the code listing. If you created a new virtual environment, create a new models.py file within the service folder. The code file for the sample is included in the restful_python_2_02_01 folder, in the Flask01/service/models.py file:

from marshmallow import Schema, fields, pre_load 
from marshmallow import validate 
from flask_sqlalchemy import SQLAlchemy 
from flask_marshmallow import Marshmallow 
 
 
orm = SQLAlchemy() 
ma = Marshmallow() 
 
 
class ResourceAddUpdateDelete():    
    def add(self, resource): 
       ...

Now, we will create the Flask-Marshmallow schemas that we will use to validate, serialize, and deserialize the previously declared NotificationCategory and Notification models, and their relationships.

Open the models.py file within the service folder and add the following code after the last line. The lines that declare fields related to other schemas are highlighted in the code listing. The code file for the sample is included in the restful_python_2_02_01 folder, in the Flask01/service/models.py file:

class NotificationCategorySchema(ma.Schema): 
    id = fields.Integer(dump_only=True) 
    # Minimum length = 3 characters 
    name = fields.String(required=True,  
        validate=validate.Length(3)) 
    url = ma.URLFor('service.notificationcategoryresource',  
        id='<id>',  
 ...

Now, we will create the resources that compose our main building blocks for the RESTful API. First, we will create a few instances that we will use in the different resources. Create a new views.py file within the services folder and add the following lines. Notice that the code imports the HttpStatus enum declared in the http_status.py module that we created in the previous chapter. The code file for the sample is included in the restful_python_2_02_01 folder, in the Flask01/service/views.py file:

from flask import Blueprint, request, jsonify, make_response 
from flask_restful import Api, Resource 
from http_status import HttpStatus 
from models import orm, NotificationCategory, NotificationCategorySchema, Notification, NotificationSchema 
from sqlalchemy.exc import SQLAlchemyError 
 
 
service_blueprint = Blueprint('service...

The following table shows the method of our previously created classes that we want to be executed for each combination of HTTP verb and scope: