Hi, I'm Naresh Surisetty, an enthusiastic Python developer based in India. I'm especially interested in developing scalable apps in the cloud - a main reason I'm drawn to experimenting with serverless architecture. In this post I'll share a Serverless project using Python EVE and MongoDB.

What is Python EVE? Python EVE is an open source Python REST API framework. Python EVE allows for HATEOAS (Hypermedia As The Engine Of Application State), and includes pagination support, support for multiple content types (JSON, XML, plain text, ...), if-Match/ETag and MongoDB GridFS.

Get Started Progamming

1. Installation

We'll use pip to install EVE.

pip install eve

We'll create our MongoDB database using mLab. If you don't have an existing mLab account go ahead and create one as we'll be using mLab hosted database throughout the post. (There's an option for a free Sandbox account.) Make sure you create a database, as well.

What is mLab? mLab is a fully managed cloud database service that hosts MongoDB databases. mLab runs on cloud providers Amazon, Google, and Microsoft Azure, and has partnered with PaaS (Platform-as-a-Service) providers. Find more help getting started in the mLab Docs & Quick Start Guide.

2. Creating a Basic API

Create a project named eve-api-project. Navigate to eve-api-project folder and create a file called run.py. Copy the following code:

run.py

import os
from eve import Eve
  
# AWS lambda, sensible DB connection settings are stored in environment variables.
MONGO_HOST = os.environ.get('MONGO_HOST')
MONGO_PORT = os.environ.get('MONGO_PORT')
MONGO_USERNAME = os.environ.get('MONGO_USERNAME')
MONGO_PASSWORD = os.environ.get('MONGO_PASSWORD')
MONGO_DBNAME = os.environ.get('MONGO_DBNAME')
  
# AWS 
api_settings = {
      'MONGO_HOST': MONGO_HOST,
      'MONGO_PORT': MONGO_PORT,
      'MONGO_USERNAME' : MONGO_USERNAME,
      'MONGO_PASSWORD' : MONGO_PASSWORD,
      'MONGO_DBNAME': MONGO_DBNAME,
      'RESOURCE_METHODS' : ['GET', 'POST', 'DELETE'],
      'ITEM_METHODS' : ['GET', 'PATCH', 'DELETE'],
      'EXTENDED_MEDIA_INFO' : ['content_type', 'name', 'length'],
      'RETURN_MEDIA_AS_BASE64_STRING' : False,
      'RETURN_MEDIA_AS_URL': True,
      'CACHE_CONTROL' : 'max-age=20',
      'CACHE_EXPIRES' : 20,
      'DOMAIN' : {'people': {
                  'item_title': 'person',
                  'additional_lookup':
                      {
                          'url': 'regex("[\w]+")',
                          'field': 'lastname'
                      },
                  'schema':
                      {
                          'firstname': {
                              'type': 'string',
                              'minlength': 1,
                              'maxlength': 10,
                          },
                          'lastname': {
                              'type': 'string',
                              'minlength': 1,
                              'maxlength': 15,
                              'required': True,
                              'unique': True,
                          },
                      }
                  }
              }
}
app = Eve(settings=api_settings)  

Lets break down the run.py file.

• Initially we've imported the EVE module

  import os
  from eve import Eve

• MONGO_HOST, MONGO_PORT, MONGO_USERNAME, MONGO_PASSWORD and MONGO_DBNAME defines the variable for storing our MongoDB host, port, username, password and database name that we'll configure in our serverless.yml file as part of environment section.
• api_settings dictionary defines all the configs and schemas for our API. You can use any name for defining your dictionary. For ease I've defined it as api_settings.
• DOMAIN section under api_settings defines the endpoints and their respective schemas for API.

  {'people': {
                'item_title': 'person',
                'additional_lookup':
                    {
                        'url': 'regex("[\w]+")',
                        'field': 'lastname'
                    },
                'schema':
                    {
                        'firstname': {
                            'type': 'string',
                            'minlength': 1,
                            'maxlength': 10,
                        },
                        'lastname': {
                            'type': 'string',
                            'minlength': 1,
                            'maxlength': 15,
                            'required': True,
                            'unique': True,
                        },
                    }
             }
   }

  The above mentioned people schema defines EVE API to create an endpoint with name people and add the following structure for the API defined. We have two fields in the API named firstname and lastname. Different validations can be performed upon the schema such as minlength, maxlenth, type and many more that can be referred at EVE Allowed Validations.
• Pass the api_settings as a parameter for settings in Eve.

  app = Eve(settings=api_settings)

  This registers all the configs and schemas defined under api_settings.

3. Defining the Serverless File

Create a file named serverless.yml under the same project directory.

serverless.yml

service: eve-api

frameworkVersion: ">=1.1.0 <2.0.0"

provider:
  name: aws
  runtime: python2.7
  environment:
    MONGO_HOST: "api.db.com"
    MONGO_PORT: 27104
    MONGO_USERNAME : "api-user"
    MONGO_PASSWORD : "api-pass"
    MONGO_DBNAME: "api-db"
  iamRoleStatements:
    -  Effect: "Allow"
       Action:
         - "lambda:InvokeFunction"
       Resource: "*"

plugins:
  - serverless-wsgi

functions:
  api:
    handler: wsgi.handler
    events:
      - http: ANY {proxy+}

custom:
  wsgi:
    app: run.app

Lets break down the serverless.yml file.

• If we get into the environment section under the provider section you can find MONGO_HOST, MONGO_USERNAME, MONGO_PASSWORD, MONGO_DBNAME and update them as per your MongoDB credentials. These values will be defined as Environment values in AWS Lambda.
• Under plugins section we added serverless-wsgi as a plugin which is used to build and deploy Python WSGI applications.

  npm install --save severless-wsgi
  

  What is Serverless WSGI? The Serverless WSGI plugin is used to deploy WSGI applications (Flask/Django/Pyramid etc.) and bundle Python packages.

4. Test the Service

Fire up your console and run serverless wsgi serve in the project directory to see your live API at http://localhost:5000 .

$ curl -i http://localhost:5000/people
HTTP/1.1 200 OK

The response payload will look something like this:

{
    "_items": [
        {
            "firstname": "Naresh",
            "lastname": "Surisetty",
            "_id": "50bf198338345b1c604faf31",
            "_updated": "Wed, 05 Dec 2012 09:53:07 GMT",
            "_created": "Wed, 05 Dec 2012 09:53:07 GMT",
            "_etag": "ec5e8200b8fa0596afe9ca71a87f23e71ca30e2d",
            "_links": {
                "self": {"href": "people/50bf198338345b1c604faf31", "title": "person"},
            },
        },
        ...
    ],
    "_meta": {
        "max_results": 25,
        "total": 70,
        "page": 1
    },
    "_links": {
        "self": {"href": "people", "title": "people"},
        "parent": {"href": "/", "title": "home"}
    }
}

The _items list contains the requested data. Along with its own fields, other fields such as _created, _updated, _etag, _id, are automatically handled by the API (clients don’t need to provide them when adding/editing resources).

The _meta field provides pagination data and will only be there if Pagination has been enabled (it is by default) and there is at least one document being returned. The _links list provides HATEOAS directives.

5. Deploy the Service

Deploy the service with serverless deploy. If you need to setup Serverless, please see the Serverless Quick Start Guide for instructions.

Conclusion

Python EVE and MongoDB enhanced with Serverless is an awesome bundle for REST API development. You can further enhance the REST API with more additions such as Adding Authentication , OAuth2, Swaggger documentation and even more. Check out more resources on GitHub.