mvillarrealb / poi-api

Points of Interest api with Go

Points of Interest api with GO

A restful api for creating Points of interest using uber H3 :) and Go

Exposed Resources

The following Resources & Operations are exposed in our Api:

Path	Verb	Description
/categories	GET	List all categories
/categories	POST	Create a category
/categories/{id}	GET	Get a category by id
/categories/{id}	PUT	Update a category
/categories/{id}	DELETE	Delete a Category
/geo_code	GET	Get the area of a given latitude, longitude
/poi	GET	List all Points of interest
/poi	POST	Create a point of interest
/poi/{id}	GET	Get a single point of interest
/poi/{id}	DELETE	Delete a point of interest

Module And dependencies

The following dependencies were used for the api:

• github.com/go-playground/validator
• github.com/gorilla/mux
• gorm.io/driver/postgres
• github.com/gorilla/handlers
• gorm.io/gorm
• github.com/uber/h3-go

Directory Structure

This is the directory structure to support our api

|
|-|Dockerfile
|-|go.mod #Go module file
|-|main.go #Application entry point
|-|api # Api related directory
|---|common #Common functions
|---|models # GORM models
|---|routes # Mux routes
|---|api.go # Main api

Api entrypoint

Our api entrypoint will create a new Api reference passing as parameters the following environment variables

Environment	Description
DATABASE_PORT	Database instance port
DATABASE_HOST	Database hostname/ip address
DATABASE_USERNAME	Database user
DATABASE_PASSWORD	Database password

//main.go
package main

import (
	"os"
	"poi-api/api"
	"strconv"
)

func main() {
	port, err := strconv.Atoi(os.Getenv("DATABASE_PORT"))
	if err != nil {
		panic("Invalid Database port")
	} else {
		app := api.New(
			os.Getenv("DATABASE_HOST"),
			port,
			os.Getenv("DATABASE_USERNAME"),
			os.Getenv("DATABASE_PASSWORD"))
		app.Run(":8080")
	}
}

Run the go application locally

# It's required a a postgresql database instance

DATABASE_PORT=5432 DATABASE_HOST=127.0.0.1 DATABASE_USERNAME=postgres DATABASE_PASSWORD=casa1234 go run main.go 

Load Some data

In order to use the geocode endpoint and create a referenced point of interest load the data file, this includes all divisions on the city of Lima. I've indexed every area of the city with h3 hexagons using 11 as resolution for the index.

Build & Run Docker Image locally

docker build -t poi-api:0.0.1 .

docker run --name poi-api --network=host -e DATABASE_PASSWORD=casa1234 poi-api:0.0.1

---

Run the Project on GCP Cloud Run

To run the project on Cloud Run follow the tutorial

Use the postman collection

Use the poi-e2e.postman_collection.json collection to run some tests and play with the api