Parsyl Backend Code Challenge

Pre-requisites

An IDE or text editor of your choice
Docker

Your Task

Parsyl ingests a lot of data from hardware in order to present that information to users. Your task will be to create an API that ingests log messages, transforms them for a database and then aggregates that data for display via other endpoints. Review the "user stories" below for instructions.

Our stack is written in Go and we’d love to see a solution in that but you are welcome to build your solution in whatever language and libraries you are most comfortable with. Google and the internet are at your disposal.

You are free to spend as much time as you’d like to solve this challenge, however, we don’t expect more than 2-3 hours.

Considerations

When developing your solution, please consider the following:

Structure of your endpoints
Quality of your code
Testability
Can your solution be easily configured and deployed (deployment not actually required)?

It’s okay if you don’t finish the entire challenge. We favor high quality code over completing all the stories! One well built story is better than three poorly built ones. It's not uncommon for the first story to take significantly longer than the second or third.

It’s okay if you get stuck. Please reach out and we will answer questions or try to help you get moving again. There is an FAQ which may contain an answer but don't hesitate to reach out.

Getting Started

Make sure Docker is running
Build and start the database Docker image: make bootstrap
Write and structure your project and code how you see fit
- Reference sample_log_messages.txt for a bunch of sample log messages
- sample_requests contains a couple cURL requests that can be used to test your work
Stop the docker containers when you’re finished: make stop

User Stories To Complete

You can complete the stories any way you see fit but we recommend they be completed in order with distinct commits (or a merge commit) for each.

Story #1: Ingest log messages

A log message is stored appropriately in the database
Logs are ingested through the /ingest endpoint
Temperature values should be converted to farenheit before saving to the database
- TempF = (2*TempC) + 30

The request format should have the contract (see sample_requests below):

 {
 	"data": [
 		// Log event string
 		//"YYARKx|2022-03-22T21:42:02.362Z|-8"
 	]
 }

Implementation Note: See the detail about the database and log structure below.

Story #2: Display aggregated data by device id and date

Endpoint exists at /device/{id} to retrieve log messages aggregated by device id (see sample_request below)
Datetimes can be in UTC
Log events should be sorted ASC by log date

Response should have the contract:

 {
 	"deviceId": String,
 	"averageTemperature": Number,
 	"mostRecentLogDate": Datetime,
 	"logs": [
 		{
 			"logDate": Datetime,
 			"temperature": Number,
 			"humidity": Number
 		}
 	]
 }

Story #3: Flag records according to business rules

Log messages that have a temperature greater than 32 degrees Fahrenheit should be flagged

Response should have the contract:

{
	"deviceId": String,
	"averageTemperature": Number,
	"mostRecentLogDate": Datetime,
	"totalAlerts": Number, // <-- New field
	"logs": [
		{
			"logDate": Datetime,
			"temperature": Number,
			"alert": true // <-- New field
		}
	]
}

Data Structures

Log Messages

The log messages are pipe delimited strings that have the following structure:

device_id|datetime|temperature_celsius

Datetime is in UTC
Temperature is in celsius

Note, this structure cannot be modified!

Database

The database stores a row per log message. The Logs table has the following schema:

Column	Type	Constraints
event_id	INT	Primary Key, auto increment
event_date	DATETIME	Not Null
device_id	VARCHAR(6)	Not Null
temp_farenheit	INT

Note, this schema cannot be changed!

Once Postgres is spun up in Docker, you connect to it with any postgres library you choose. The schema will be in place but there will be no data. The connection information should be:

Host: http://localhost
Port: 5432
User: pguser
Password: pgpassword
Database: code_challenge

Frequently Asked Questions

Can I run queries against the database with a client?

You can run queries directly. If you use a client (e.g. Postico), you can use the connection information provided in the README. If you prefer to use psql you can connect to the running docker container:

Connect to the container: docker exec -it db bash
Connect to the database: psql --user pguser --db=code_challenge
Submit queries directly (e.g. SELECT * FROM logs)

I get the error: Error response from daemon: Conflict. The container name "/db" is already in use...

This is because you stopped the container without removing it. If you want to start where you left off previously, you can simply run docker start db. Alternately, if you want to start from a clean install of the database then you can run docker rm db and then run make bootstrap again.

This issue and solution will work for the generator container as well.

Sample Requests

Some sample cURL requests are provided to help you interact with the API you build. They default to http://localhost:8000. You can also copy these commands into a REST client (e.g. Postman) if you prefer to use a GUI instead of the terminal.

GET /ingest

curl --location --request POST 'http://localhost:8000/ingest' \
--header 'Content-Type: application/json' \
--data-raw '{
    "data": [
        "YYARKx|2022-03-22T21:42:02.362Z|-8",
        "YYARKx|2022-03-22T21:42:04.372Z|-1",
        "YYARKx|2022-03-22T21:42:50.572Z|7"
    ]
}'

POST /device/{id}

curl --location --request GET 'http://localhost:8000/device/YYARKx'

eddwinpaz / device-svc