search

backtrackbaba / covid-api

API's to interact with COVID19 dataset by John Hopkins University
https://covidapi.info/
MIT License
77 stars 19 forks source link

alternative data source? #2

Open pdobsan opened 4 years ago

pdobsan commented 4 years ago

Thanks for your API/service. Among JSON API-s, I like it most so far.

Many people is disappointed by JSU's dropping the recovered category. As a consequence your source (pomper's) is probably going to drop it too. In the ongoing discussion I have come across an alternative data source which provides the recovered category and also looks like a clean CSV times series data-set: https://github.com/ulklc/covid19-timeseries

You might want to consider it.

Having said that, I am still hoping that JSU reconsiders since I need the state/province resolution.

backtrackbaba commented 4 years ago

Hi @pdobsan ,

Thank you for the suggestion. I'll check out the repo and its own sources. Maybe we could roll out v2 of the API with the new data source.

I agree that JHU, shouldn't have dropped the recovered category. I'm currently tackling an issue where newer countries are being added as null which is hindering my code. Fixing it right now.

Wanted to know, are you using our endpoints? Is there any feedback or suggestions that you would llike to pass on?

pdobsan commented 4 years ago

Wanted to know, are you using our endpoints?

Not yet, just have discovered it.

Is there any feedback or suggestions that you would llike to pass on?

Actually there is one, I have just discovered. When you provide historical data, I would much prefer an array of {date: , confiremed: ...} objects to the current format of object keyed with an ever increasing list of dates. Like it is in the original json source provided by @pomber. I don't see any reason to change that.