Gets the latest data from the Johns Hopkins COVID-19 tracker, via GitHub, via Nick

corona.py

# This script will get the latest data from the Jons Hopkins GitHub repo https://github.com/CSSEGISandData
# Loads into data frame (doesn't save) and parses dates, - in long form by default.
# Modified from Nick Evershed's version https://gist.github.com/nickjevershed
import requests
import pandas as pd
import datetime
from io import BytesIO

def getData(_long=True):

    #dict of observation type and paths
    files = {"confirmed" : "time_series_19-covid-Confirmed.csv",
             "deaths" : "time_series_19-covid-Deaths.csv",
             "recovered": "time_series_19-covid-Recovered.csv"}

    headers = {'Accept': 'application/vnd.github.v3.raw'}

    df_list = []

    for obs, path in files.items():
        url = "https://api.github.com/repos/CSSEGISandData/COVID-19/contents/csse_covid_19_data/csse_covid_19_time_series/{path}".format(path=path)
        r = requests.get(url, headers=headers)
        _io = BytesIO(r.content)
        df = pd.read_csv(_io)
        date_range = parse_dates(df)
        df['Observation'] = obs
        df_list.append(df)
    dx = pd.concat(df_list)

    if _long:
        variables = list(df.columns[:4]) + ["Observation"]
        return dx.melt(value_vars=date_range, id_vars=variables, var_name="Date")
    else:
        return dx

def parse_dates(df):
   names = {date: datetime.datetime.strptime('1/22/20', "%m/%d/%y").date() for date in df.columns[4:]}
   df.rename(columns=names, inplace=True)
   return names.values()