/messages-demo-python

Primary LanguagePython

Messages App

Description

-- Messages app exposes CRUD operation for message resource of type string (min_length=1 and max_length=255)
The following is the list of apis:

API Desscription
[GET] /api/v1/messages Get a list of messages
[GET] /api/v1/messages/{id} Get a message by id
[POST] /api/v1/messages Create a message
[PUT] /api/v1/messages/{id} Update a message
[DELETE] /api/v1/messages/{id} Delete a message

You can also access a more detailed list of available APIs by visiting /api/v1/docs/ url after the application is deployed.

Project Structure:

messages webservice app consists of two layers: controller layer and repository layer.

  • controller layer contains the logic related to CRUD operations on message resource and error handling. This layer is represented as api folder.
  • repository layer contains of message data model and its properties. This layer is represented as models.py file.

The following provides more details about the directory structure of the messages app:

├── resources
│   └── images
├── Dockerfile
├── README.md
├── skaffold.yaml
├── wsgi.py
├── app
│   ├── config.py
│   ├── __init__.py
│   ├── models.py
│   ├── main
│   ├── api
│   └── exceptions.py
├── tests
│   ├── ...
├── entrypoint.sh
├── migrations
│   ├── ...
├── requirements.txt
└── helm
    ├── ...

Files:

  • wsgi.py: is the entry to run the application. We are using unicorn http server which is a wsgi compliant http server
  • skaffold.yaml: includes configuration for skafold to deploy the app in k8s local cluster
  • requirements.txt: includes projct dependencies

Directories:

  • resources: includes images used in README.md file
  • app: contains api related logic for message resource
  • tests: contains unit tests
  • migrations: contains migration scripts generated by flask_migration lib
  • helm: contains helm delpyment templatess

Architecture overview

Build and Deployment

The easiest way to build and deploy the messages app is to build and run it as a standalone docker container. In this mode, for simplicity, I am using an sqlite database for persistence.

To use Mysql database for persistence, you can deploy the messages app using helm. For the purpose of this demo, I am using minikube as local kubernetes.

Please refer to Perequisites section to prepare your local enviroment before deployment of the app.

There are three options to build and deploy messages app locally

  1. Standalone
    To build and deploy the app as a standalone docker container:

    • docker build -t messages-webservice:latest .
      Note: . means you are ruuning this command at the root directory of the project
    • docker run --rm --env ENV_NAME=DEV -p 80:80 -it messages-webservice:latest

  2. k8s with helm and skaffold
    To build and deploy the app in minikub cluster using skaffold to run helm charts:

    • skaffold run

  3. k8s with helm
    To build and deploy the app in minikub cluster using helm:

    • docker build -t messages-webservice:latest .
    • helm repo add bitnami https://charts.bitnami.com/bitnami
    • helm install db bitnami/mysql --set auth.rootPassword=passwordroot --version 8.5.0 -f <path_to_project_root>/messages/helm/db/values.yaml --kube-context minikube
    • helm install mw <path_to_project_root>/messages/helm/messages -f <path_to_project_root>/messages/helm/messages/values.yaml --kube-context minikube

Perequisites

The first perequisite, no matter which of above-mentioned options you chose for your depmloyment is that you run Docker on your local enviroment. Please follow this link to setup Docker if you don't have it already https://www.docker.com/get-started

If you are building and deploying the messages app as a standalone docker container, no further setup is required. In this mode, for simplicity, sqlite databse is used to store the message.

If you are building and deploying the messages app in local k8s with Mysql database, follow the following steps:

  1. Install minikube https://minikube.sigs.k8s.io/docs/start/
    If you are osx user, run brew install minikube Follow these steps after minikube is installed :

    • Run minikube start
    • Run minikube addons enable ingress
    • Append the following line (messages app hostname) to your hosts file /etc/hosts
      <minikube_ip> messages-webservice
      replace <minikube_ip> with minikube VM IP and to get minikube IP run: minikube ip

  2. Install helm https://helm.sh/docs/intro/quickstart/
    If you are osx user, run brew install helm

  3. If you want to use skaffold to run helm charts, install skaffold https://skaffold.dev/docs/install/
    If you are osx user, run brew install skaffold

k8s deployment overview

Unit Test

To run unit tests, follow the these steps:

  1. Build docker images by running the following command at the root path of the messages app:
    • docker build -t messages-webservice:latest .
  2. Run the tests:
    • docker run messages-webservice:latest python -m pytest -v tests -x

Note

These assumtions are made to determine if a message is palindrome:

  • message is a string
  • message should contain only alphanumeric character
  • leading and trailing whitespaces are consided as part of message
  • the operation is not case sensitive