Ash822 / goweb

HTTP Server

• HTTP Server
  • Getting started
    • Running using Makefile
    • Running using Docker
  • About
    • Endpoints
      • Assumption
    • Architecture
    • Testing
    • CI/CD
    • API documentation

Getting started

Prerequisites:

• git
• go 1.16.3
• docker
• docker hub account

Running using Makefile

Commands required for testing, building the project and for serving swagger client as added to the Makefile.

Commands	Tasks
make test	To execute all the test in the project.
make run	To run the server. The server will be listening to port 8000.
swagger-gen	To build swagger spec yaml and serve it.

Running using Docker

Alternatively, the application can also be run using docker.

Use the following command to pull the docker image.

docker image pull ash822/goweb

Run the application using the following command.

docker run -p 8000:8000 ash822/goweb:latest

About

This repository showcases a simple HTTP server serving REST endpoints, built using the Golang net/http package and gorilla/mux as http router.

Endpoints

It supports the following endpoints.

The POST endpoints take a json with the text that needs to be checked as palindrome.

Request body:
{
  "Text": "kayak"
}

When the resource gets created, the service performs check whether the given string is a palindrome and adds a boolean flag indicating it. Also, a random UUID is generated as id of the document in order to fetch it.

Success Response: 

{
  "id": "df6711c5-d061-4e7b-b27f-548fac86fa4f",
  "text": "kayak",
  "palindrome": true
}

The id can be used to retrieve the message back.

For more information on API documentation refer the API documentation subsection.

Security

A simple middleware is added to authenticate the requests based on API keys. Currently, the API key is hardcoded to topgun.

Add the API key header X-API-Key to authenticate the requests.

Assumption

1. The palindrome check is case-insensitive. Both Kayak and kaYAK will be a valid palindrome.
2. The message resources are stored in-memory in a map data structure, there is no persistence connection to store the data. Hence, if the service is restarted, all the data gets lost.
3. To simplify the development, the get all messages endpoint /messages does not support filter, sorting and pagination.
4. The deployment has been configured to push the docker images to my personal docker hub account. However, I made the repository public and anyone can pull images from the repository if they have a valid docker hub account.

Architecture

The data flow is designed to better manage the dependencies as follows.

The router links the path to the handler functions defined in the controller. The service controls the business logic and persist the data using the repository.

Testing

Gomega is used as the matcher library for assertions. In order to test the service, GoMock is used for mocking the repository.

The repository interface is decorated with mockgen annotations. The mock repository code can be generated using make mock-gen command.

The REST endpoints are tested using net/http/httptest package.

To execute the tests, run make test command.

CI CD

The project is setup with Continuous Integration and Continuous Deployment (to Docker hub) using Github Actions.

On every pull request to master branch and on every push to master branch, the workflow gets executed. The workflow checks out the latest, build and run tests.

As the last step, the application builds and tags a docker image, and pushes the image to Docker hub registry.

Visit Actions tab to check the jobs.

API documentation

To generate the swagger file and to serve the swagger client, run make swagger-run command.

Refer: swagger.yaml