corrux Data Innovation Team Hiring Challenge

First, thanks for your interest in corrux! We put together this coding challenge to give you a chance to show us your skills by developing an algorithm that works with geotemporal data. You may use any programming language and any libraries to solve this task. We recommend that you make use of a geospatial library for the programming language you choose. We have collected a list of a few library choices to get you started:

• Shapely (Python)
• NetTopologySuite (C#)
• JTS Topolgy Suite (Java)
• geo (Rust)
• GEOS (C/C++)

Problem description

In this problem we make use of two concepts: Locations and Sites. A Location is the combination of a point in time and a position in space, a Site is the combination of an interval in time and an area in space.

Locations

We represent a Location as a JSON object with two attributes: datetime and position. datetime follows ISO 8601 for its time representation, while position adheres to GeoJSON to represent a point in space. Here is an example:

{
    "datetime": "2020-08-27T13:53:59Z",
    "position": {
        "type": "Point",
        "coordinates": [11.5756, 48.1371]
    }
}

Sites

We represent a Site as a JSON object with three attributes: start_datetime, end_datetime, geofence. start_datetime and end_datetime again follow ISO 8601 and represent the start and end of the time interval respectively, where start_datetime is included in the interval and end_datetime is not. geofence also follows GeoJSON, but it represents a Polygon in space that describes the area that belongs to the site. Here is an example:

{
    "start_datetime": "2020-08-01T00:00:00Z",
    "end_datetime": "2020-09-01T00:00:00Z",
    "geofence": {
        "type": "Polygon",
        "coordinates": [
            [
                [11.5748, 48.1372],
                [11.5758, 48.1370],
                [11.5764, 48.1372],
                [11.5749, 48.1375],
                [11.5748, 48.1372]
            ]
        ]
    }
}

Relationship Of Locations And Sites

For a Location l and a Site S, we say S contains l if and only if the time of l is in the time interval of S and the position of l is in the geofence of S.

Your Task

Find all locations that are affected by a modification of the definition of a site. That is, which locations are no longer contained by it and which locations have been added because of the modification?

Included files

• locations.json: The dataset of locations to evaluate your algorithm on
• site_old.json: The old definition of the site
• site_new.json: The new definition of the site

Bonus Round

• Create a visualization of your algorithm! Think about how to show the following:
  • The way it decides which locations are part of the solution
  • Visualizing time and space
• Can you find a streaming solution? Assuming disk space is infinite, a streaming solution must be able to read locations from an infinitely large input file and write the solution to an infinitely large output file, but it can only use a finite amount of RAM.