In this challenge you will process a CSV-like data stream through TCP and append it in realtime to a file as JSON Lines.
- Linux: Ubuntu 18+ recommended (can be a virtual machine).
- Docker + compose: Install Docker Engine on Ubuntu, Install Docker Compose
To start the datasource with random data:
docker-compose up
Wait until the code compiles and the service shows Listening at tcp://0.0.0.0:9999
Then you can test it with:
nc localhost 9999
Example:
leomar@intelie:dev-challenge$ nc localhost 9999
2020-07-13T15:06:58.872-03:00,0455,0.996195,"(a=0110110011010110,b=8d567ba0-002d0bd4::8d567ba0fd2f2bf6;c=8d476ab1)"
2020-07-13T15:06:59.374-03:00,0456,0.994522,"(a=0001111110110000,b=c26f1757-00d5a6c4::c26f175774f7a6e6;c=c27e0646)"
...
The csv-like stream generates 2 rows per second with the following columns:
- timestamp - ISO 8601 timestamp to be converted to milliseconds of unix time (example);
- index - An integer that increases at each event;
- signalwave - A floating point with a sine wave from a sensor;
- metadata - key/value list that needs to be converted to json map objects (hint: you can use regular expressions).
If you receive these rows:
2020-07-13T15:06:58.872-03:00,0455,0.996195,"(a=0110110011010110,b=8d567ba0-002d0bd4::8d567ba0fd2f2bf6;c=8d476ab1)"
2020-07-13T15:06:59.374-03:00,0456,0.994522,"(a=0001111110110000,b=c26f1757-00d5a6c4::c26f175774f7a6e6;c=c27e0646)"
2020-07-13T18:24:19.921Z,1676,-0.173648,"(a=1100010101110110,b=1d6c2fab-00d1e736::1d6c2fab2bf3e736;c=1d7d3eba)"
Your code should append the following lines to a JSON file:
{"timestamp": 1594663618872, "index": 455, "signalwave": 0.996195, "metadata": { "a": "0110110011010110", "b": ["8d567ba0", "002d0bd4", "8d567ba0fd2f2bf6"], "c": "8d476ab1"}}
{"timestamp": 1594663619374, "index": 456, "signalwave": 0.994522, "metadata": { "a": "0001111110110000", "b": ["c26f1757", "00d5a6c4", "c26f175774f7a6e6"], "c": "c27e0646"}}
{"timestamp": 1594664659921, "index": 1676, "signalwave": -0.173648, "metadata": { "a": "1100010101110110", "b": ["1d6c2fab", "00d1e736", "1d6c2fab2bf3e736"], "c": "1d7d3eba"}}
You can use the language you want: C/C++, Go, Java, JavaScript (Node.js/Deno), Python 3.6+, Rust, etc, but you need to send us the code and instructions so we are able to deploy in a clean Ubuntu VM.
- Create a container to compile/run your code and add it to the docker-compose.yml.
- After appending the object to the json file, send it to a mongodb container so anyone can see it with mongo-express web interface.
Also remember to add the instructions how to use it.
To solve this challenge, you may fork this repository, then send us a link with your implementation. Alternatively, if you do not want to have this repo on your profile (we totally get it), send us a git patch file with your changes.
If you are already in the hiring process, you may send it to whoever is your contact at Intelie. If you wish to apply for a job at Intelie, please send your solution to [email protected].