Tech Test

Before running this project, ensure you have the following installed:

• Node.js
• npm or Yarn

Installation

To install the project dependencies, run the following command:

npm install

or

yarn install

Build

To build the project, run the following command:

yarn build

Usage

To run the project after it has been built, use the following command:

yarn start

Add your sample file in the sample folder.

By default, the project uses the sample file located at SAMPLE_PATH. To provide a different sample file, modify the SAMPLE_PATH environment variable accordingly.

Testing

Running the script will test the project with TEST.MD values (and some extra)

./test.sh

Thought process

What & Why

I started with a naïve approach where I processed the file using a Map (for dates) combined with a Set (for distinct queries) to get the results.

• The approx. time on my machine is ~5s to do all the calculation.
• The look-up is instant (31ms for 1M ops).

This could lead to scalability issues with bigger datasets but in my opinion acceptable for datasets of few days.

Alternative designs

I tried using a prefix-trie approach, but after testing I found out it's less efficient:

• More processing time for calculation (~8s)
• Look-up take more time (~210ms for 1M ops)
• Bigger size in memory (a bit more than Map/set approach)

I also thought doing on-demand calculation for very huge datasets using an approximation approach

We know that a logfile is chronologically ordered hence we could try to find the beginning of the calculation based on the precision asked with a bisection method.