add county data source for (US only)

[ralyodio]

https://sfc-project-files.s3.amazonaws.com/project-feeds/covid19_us_cases_us.json

[ExpDev07]

This looks like something which is a bit far off right now.

[ralyodio]

https://lionbridge.ai/datasets/coronavirus-datasets-from-every-country/

[JKSenthil]

I have created an api for county data: https://github.com/JKSenthil/coronavirus-county-api

Interestingly, both my and @ExpDev07 api use Python/Flask. Perhaps I can merge a county endpoint to this project? Shall I make a pull request?

[ExpDev07]

@JKSenthil There should be some discussions surrounding this and how it should be implemented. There is a basis for adding multiple datasources (see https://github.com/ExpDev07/coronavirus-tracker-api/tree/master/app/services/location).

I'm thinking we'll add a query param to the locations endpoints that allows you to pick your data-source of choice. E.g: /v2/locations?source=jhu would mean the jhu service would be used to fetch the locations, and /v2/locations?source=1point3acres would mean for example yours.

This can easily be achieved just by attaching a middleware and dynamically retrieving the service-instance from app.services.

[JKSenthil]

That sounds good. I think that /v2/locations can default to the jhu source (for compatibility), and /v2/locations?source=1point3acres can be used to access the US county data.