README.md

Backend Description

The backend API is built on Python Flask microservice, and is served by Nginx and uWSGI.

Backend endpoints

The backend API has the following endpoints and methods:

Endpoint	Method	Options
/available-service-areas	GET
/areas?	GET	params: zipcounties=[{"county": "sanFrancisco", "zip": "94102"},{"county": "sanFrancisco", "zip": "94103"}]
/areas	POST	body: zipcounties=[{"county": "sanFrancisco", "zip": "94102"},{"county": "sanFrancisco", "zip": "94103"}]
/areas	POST	body: zipcounty_file=<zipcounty_csv_file>

The return object of /available-service-areas endpoint is the following:

{
  "result": [
    {
        "Alameda": {
            "zips": ["94501", "94502", ...]
        },
        "Marin": {
            "zips": ["91501", "97332", ...]
        },
        ...
    }
  ]
}

For /areas endpoint, input (for GET or POST) must have at least one of the following columns (regardless of whitespaces or case):

• zip or zipcode
• county or countyname

And the return object is in the following format:

{
    "result": [
        {
            "areaInfo": {
                "countyName": "San Francisco",
                "zipCode": "94105"
            },
            "availabilityStatus": {
                "isServiceAreaAvailable": true,
                "message": "Service area available."
            },
            "boundary": {
                "geometry": {
                    "coordinates": [
                        [
                            [
                                -117.23250138140729,
                                33.19171791814623
                            ],
                            [
                                -117.23260861869265,
                                33.19171791814623
                            ],
                            [
                                -117.23250138140729,
                                33.19180808215501
                            ],
                            [
                                -117.23250138140729,
                                33.19171791814623
                            ]
                        ]
                    ],
                    "type": "Polygon"
                }
            },
            "points": [point_1, ..., point_200]
        },
        {
            "areaInfo": {
                "countyName": "San Diego",
                "zipCode": "94130"
            },
            "availabilityStatus": {
                "isServiceAreaAvailable": false,
                "message": "Service area unavailable."
            },
            "boundary": null,
            "points": []
        },
        ....
    ]
}

Exceptions

If the input file is not parsable by the backend, it throws a HTTP 400 error with the following message:

Invalid CSV file. Was not able to parse.

If the payload key is invalid (e.g. zipcountyfile instead of zipcounty_file), it throws a HTTP 400 error with the following message:

Payload key is invalid.

Each point is a geojson object of type Point:

{
    "geometry": {
        "coordinates": [
            -122.39987717199995,
            37.78892673000007
        ],
        "type": "Point"
    },
    "properties": {
        "county": "sanFrancisco",
        "residents": 1870,
        "zip": "94105"
    },
    "type": "Feature"
}