miguel91it/backend-challenge

Golioth Backend Code Challenge

Golioth Challange

Project developed by João Miguel to solve the Golioth Challenge.

Following it will be described how the project was structured and hot to run it.

Project Structure

The project is a monorepo containing a `src/ folder wich has:

* main.go file               -> responsible to start the gateway application
* Dockerfile file            -> responsible to build the gateway container image
* gateway/ folder            -> responsible for grouping all files related to the gateway application
* goliothMongo/ folder       -> responsible for implementing the mongodb client for this project
* devices-simulation/ folder -> responsible for grouping all files related to the devices simulation application
* docker-compose.yaml file   -> responsible for launching all the 3 containers (gateway, devices and mongodb)
* .env file                  -> responsible for providing environment variables for the gateway and devices containers

Always you want to run the gateway application and the devices simulation, you'll use the docker-compose tool. To see how to run the solution, go to the end of this file.

Gateway Module

Gateway Module is composed by 3 files:

* db.go       -> responsible to describe the Repository Interface necessary for the gateway
* gateway.go  -> responsible for launching the gateway server
* handlers.go -> responsible to group the API endpoints handlers
* weather.go  -> responsible for modeling the weather entity

The Weather Telemetry API has the following entity (model):

* Weather

Weather Entity

This entity groups data related to the environment weather sent by the devices.

* Device_id           (string)  -> MAC Addres as device id
* Timestamp           (int)     -> Timestamp of the measure
* SoilMoisture        (float64) -> Rate of soil moisture (must be between 0 and 100%)
* ExternalTemperature (float64) -> External temperature in celsius degree
* ExternalHumidity    (float64) -> Rate of external humidity (must be between 0 and 100%)

Gateway Endpoints

Gateway has 2 visible endpoints to make a Request:

[POST] api/v1/WeatherTelemetry

    Endpoint to send the weather data from a device

[POST] api/v1/GetWeather

    Endpoint to get weather data from a datetime filter

GoliothMongo Module

GoliothMongo Module is composed by one file:

* mongo.go  -> lib responsible to create a mongodb client such way that implements the Interface Repository

Devices Module

Devices Module is composed by 3 main files:

* devices.go   -> contains a function to provides a list of MAC Address to the simualtion
* telemetry.go -> contains the functions responsible for sending telemetry
* main.go      -> responsible for coordinate the creation of one goroutine for each device and for the interval of sending management during the time of simulation desired.

How to run the project

The project is containerized using a Golang image and for run it you must follow the next steps:

• First of all, clone the repository into your machine

git clone https://github.com/miguel91it/backend-challenge.git

• Next you must enter in the src folder:

cd backend-challenge/src

• Once inside at the root folder of the project, you must build and run all the containers (mongodb, gateway and devices) using docker compose:

sudo docker-compose up --build gateway devices

The command above will build and run all the 3 containers but only gateway and devices containers will be attached and logged into standard output. This way you'll be able to see both logs during the gateway and devices containers executions. Mongodb service doesn't need to be explicit in the command because it's a dependency and therefore it will be started automatically with gateway container.

• Finally, to stop and remove all containers

sudo docker-compose rm -sfv

• If you want to change the devices simulation parameters like number of devices, duration or interval, goes to the file backend-challenge/src/.env and change de varibales located in the DEVICES SIMULATION ENVS

How to test the Gateway Server

It's possible to test manualy the telemetry endpoint of the API using command line tools like Curl or GUI tools like Postman.

• Following we have some example of how to send a telemetry data to the gateway using Curl:

POST: localhost:28000/api/v1/WeatherTelemetry

    Example of Request body:

        {
            "id":              "1",
            "timestamp":       12345678,
            "soil_moisture":   10,
            "ext_temperature": 10.1,
            "ext_humidity":    1.3
	    }

Full Curl command:

curl --request POST 'localhost:28000/api/v1/WeatherTelemetry' \
--header 'Content-Type: application/json' \
--data-raw '{
"id": "02:42:ac:17:00:03",
"timestamp": 1629855252,
"soil_moisture": 10,
"ext_temperature": 10.1,
"ext_humidity": 1.3
}'

Experiment change soil_moisture, ext_temperature and ext_humidity values.

• Following we have some example of how to send a telemetry data to the gateway using Curl:

POST: localhost:28000/api/v1/GetWeather

    Example of Request body:

        {
            "start_date": "2021-08-26T00:27:00",
            "end_date": "2021-08-26T12:24:47",
        }

Full Curl command:

curl --location --request POST 'localhost:28000/api/v1/GetWeather' \
--header 'Content-Type: application/json' \
--data-raw '{
"start_date": "2021-08-26T00:27:00",
"end_date": "2021-08-26T12:24:47"
}'

OBS: for now it's MANDATORY pass start_date and end_date, it will be fixed in the future.

OBS: pay attention to the hour OFFSET when writing the filter because the containerized application persists the data in UTC timezone.

Experiment change soil_moisture, ext_temperature and ext_humidity values.