This example application implements a very basic "pet shop" REST service using the Connexion Python library.

Connexion is a framework on top of Flask to automagically handle your REST API requests based on Swagger 2.0 Specification files in YAML.

👉 Please also have a look at the Connexion Example with Redis database and Kubernetes deployment manifests.

This example application shows various features supported by the Connexion library:

• mapping of REST operations to Python functions (using the operationId in swagger.yaml)
  • maps path, query and body parameters to keyword arguments
• bundled Swagger UI (served on /ui/ path)
• automatic JSON serialization for application/json content type
• schema validation for the HTTP request body and query parameters:
  • required object properties
  • primitive JSON types (string, integers, etc)
  • date/time values
  • string lengths
  • minimum/maximum values
  • regular expression patterns
• gevent WSGI server
• OAuth2 protection

The example application only needs very few files:

• swagger.yaml: the pet shop REST API Swagger definition
• app.py: implementation of the pet shop operations with in-memory storage
• Pipfile: list of required Python libraries (used by Pipenv)
• Dockerfile: to build the example as a runnable Docker image
• test.sh: shell script to execute example HTTP requests against the pet shop API

You can run the Python application directly on your local operating system (this requires Python 3 and Pipenv):

$ pipenv install --dev && pipenv shell
$ ./app.py # start the HTTP server
$ xdg-open http://localhost:8080/ui/
$ ./test.sh # do some test HTTP requests

You can build the example application as a Docker image and run it:

$ docker build -t connexion-example .
$ docker run -d -p 8080:8080 connexion-example
$ ./test.sh # do some test HTTP requests

To enable OAuth2 security (token verification), you need to pass the URL to the "tokeninfo" endpoint:

$ docker run -d -p 8080:8080 -e HTTP_TOKENINFO_URL=https://auth.example.org/tokeninfo connexion-example

You can use the Flask WSGI app with any WSGI container, e.g. using Flask with uWSGI:

$ sudo pip3 install uwsgi
$ uwsgi --http :8080 -w app

You can run uwsgi with a large number of worker processes to get high concurrency. This obviously makes no sense for the in-memory pet store example (every worker would have its own pet store dictionary):

$ uwsgi --http :8080 -w app -p 16  # use 16 worker processes

See the uWSGI documentation for more information.