HTTP server for managing a list of users

Descrption

HTTP server using Gin-Gonic framework and implementing CRUD operations. It operates on User entity. The whole project is dockerized. It also contains implemetation of seek pagination with sorting and filtering

User entity

Attributes:

• name
• surname
• gender
• age
• address
• created_at

name and surname are the user's unique identifier. There can be only one user with given name and surname

Endpoints

• GET /v1/users/:user_id - return User entity in JSON format
• DELETE /v1/users/:user_id - delete User entity
• PUT /v1/users/:user_id - update User entity. Request in JSON format
• POST /v1/users - create User entity. Request in JSON format
• GET /v1/users - search Users using sorting, filtering and seek pagination

Exmples

• GET /v1/users - return up to 30 users sort by id ascending
• GET /v1/users?limit=100&sort=name:desc - return up to 100 users sort by name descending
• GET /v1/users?gender=male&limit=100&sort=age:asc - return up to 100 users sort by age ascending with gender male
• GET /v1/users?name=sonny&gender=male&limit=100&sort=age:desc - return up to 100 users sort by age descending with gender male and name like %sonny% case insensitive
• GET /v1/users?name=sonny&gender=male&limit=100&sort=age:desc&min_age=30 - return up to 100 users sort by age descending with gender male and name like %sonny% case insensitive with age >= 30
• GET /v1/users?name=sonny&gender=male&limit=100&sort=age:asc&max_age=30 - return up to 100 users sort by age ascending with gender male and name like %sonny% case insensitive with age <= 30
• GET /v1/users?name=sonny&gender=male&limit=100&sort=age:asc&min_age=30&max_age=45 - return up to 100 users sort by age ascending with gender male and name like %sonny% case insensitive with age >= 30 and age <= 45
• GET /v1/users?limit=100&sort=created_at:desc - return up to 100 users sort by created_at descending

There is also Pagination object in response to know how to query next or previous page

Testing

Unit tests

To start unit tests please call command:

make go_test_unit

The unit tests do not have any dependencies

Integration tests

To start integration tests please call command:

make docker_build_image && make application_test

The command:

• builds application docker image
• starts postgres DB in docker container
• starts application itself in docker container
• executes goose migration to preapre DB schema
• inserts test data into postgres
• executes integration tests in docker container

Run Service Locally

To run service locally please call command:

application_run

Start you browser and enter url:

http://localhost:8080/v1/users

Errors reporting

In case of errors there will be returned custom error object in JSON format with custom error code and message.

For example in case of validation error there is returned HTTP code 400 with detailed error code and message Examples:

• {"code":2140004,"message":"limit can not be negative"}
• {"code":2040001,"message":"name can't be empty"}
• {"code":2040001,"message":"afterID can not be negative"}

Project structure

• app - aplication code
  • api - definition of response and request objects
  • config - application configuration object
  • controller - controller layer
  • dao - repository layer
  • db - db helpers
  • httperrors - definition of custom http error object and predefined application errors
  • middleware - gin-gonic middleware
  • model - database models
  • service -service layer
• build - docker and docker-compose files to build, run and test application
• test - integration tests