Script that gets Brazil's cases from elhenrico's repo

utils/scripts/brazil_data.py

@@ -0,0 +1,109 @@
+import pandas as pd
+import sys
+
+if 1 < len(sys.argv) < 3:
+    init_date = sys.argv[1]
+    try:
+        init_date = pd.to_datetime(init_date, format="%Y/%m/%d")
+    except ValueError:
+        print("date format incorrect")
+        sys.exit(0)
+else:
+    print("Usage brazil_data.py <initial_date>\n date format: %Y/%m/%d example: 2020/03/25")
+    sys.exit(0)
+
+
+confirmed_url = "https://raw.githubusercontent.com/elhenrico/covid19-Brazil-timeseries/master/confirmed-cases.csv"
+deaths_url = "https://raw.githubusercontent.com/elhenrico/covid19-Brazil-timeseries/master/deaths-new.csv"
+
+confirmed = pd.read_csv(confirmed_url)
+deaths = pd.read_csv(deaths_url)
+
+###
+compare = pd.read_csv("https://raw.githubusercontent.com/DataScienceResearchPeru/covid-19_latinoamerica/master/latam_covid_19_data/latam_covid_19_daily_reports/2020-03-08.csv")
+brazil_compare = compare[compare.Country=="Brazil"]
+###
+
+
+string_a = "áéíóúäëïöüâêîôûã"  # character to be replace
+string_b = "aeiouaeiouaeioua"  # character to replace with
+
+def remove_tildes(string):
+    for index, tilde in enumerate(string_a):
+        string = string.replace(tilde, string_b[index])
+    return string
+
+
+confirmed.iloc[:,0] = confirmed.iloc[:,0].apply(remove_tildes)
+deaths.iloc[:,0] = deaths.iloc[:,0].apply(remove_tildes)
+
+confirmed = confirmed.rename(columns={"Unnamed: 0": "Subdivision", "Unnamed: 1": "Code"})
+deaths = deaths.rename(columns={"Unnamed: 0": "Subdivision", "Unnamed: 1": "Code"})
+#print(confirmed)
+
+sub_brazil = sorted(brazil_compare.Subdivision.unique())
+sub_repo = sorted(confirmed.iloc[:,0].unique())
+
+#print(sub_brazil)
+
+#print("Subdivision brazil:", len(sub_brazil))
+#print("Subdivisions:", len(sub_repo))
+
+other_subdivisions = list(set(sub_repo) - (set(sub_brazil)))  # Subdivisiones other than those listed in the main repo
+#print(other_subdivisions)
+
+confirmed = confirmed[~confirmed.Subdivision.isin(other_subdivisions)].sort_values("Subdivision")
+deaths = deaths[~deaths.Subdivision.isin(other_subdivisions)].sort_values("Subdivision")
+
+del confirmed["Code"]
+del deaths["Code"]
+
+confirmed_columns = confirmed.columns[1:]
+deaths_columns = deaths.columns[1:]
+
+#print(" ".join(confirmed_columns))
+#print(" ".join(deaths_columns))
+
+columns_order = ["ISO 3166-2 Code", "Country", "Subdivision", "Last Update", "Confirmed", "Deaths", "Recovered"]
+
+for column in confirmed_columns:
+    day, month = map(int, column.split("/"))
+    date = f"{2020}-{month:02}-{day:02}"
+
+    actual_date = pd.to_datetime(date, format="%Y/%m/%d")
+
+    if actual_date < init_date:
+        print("Skipping:", actual_date)
+        continue
+
+    daily_report_path = f"../../latam_covid_19_data/latam_covid_19_daily_reports/{date}.csv"
+    daily_report = pd.read_csv(daily_report_path)
+
+    brazil = daily_report[daily_report.Country=="Brazil"]
+    brazil_index = daily_report[daily_report.Country=="Brazil"].index
+
+    del brazil["Deaths"]
+    del brazil["Confirmed"]
+
+    confirmed_subset = confirmed.loc[:, ["Subdivision", column]]
+    deaths_subset = deaths.loc[:, ["Subdivision", column]]
+    confirmed_subset = confirmed_subset.rename(columns={column: "Confirmed"})
+    deaths_subset = deaths_subset.rename(columns={column: "Deaths"})
+    dc = pd.merge(confirmed_subset, deaths_subset, how="left", on="Subdivision")  # Deaths confirmed
+    final = pd.merge(brazil, dc, how="left", on="Subdivision").fillna(0)
+    final = final[columns_order].set_index(brazil_index)
+
+    final["Last Update"] = date
+
+    daily_report.Deaths = daily_report.Deaths.fillna(0)
+    daily_report.Confirmed = daily_report.Confirmed.fillna(0)
+    daily_report.Recovered = daily_report.Recovered.fillna(0)
+
+    daily_report.Deaths = daily_report.Deaths.astype("int64")
+    daily_report.Confirmed = daily_report.Confirmed.astype("int64")
+    daily_report.Recovered = daily_report.Recovered.astype("int64")
+
+
+    daily_report.update(final)
+    print(daily_report[daily_report.Country=="Brazil"])
+    daily_report.to_csv(daily_report_path, index=False)