benedikt/rki-impfparser
1
0
Fork 0
Code Releases Activity

fix: Refactored parts of the script into functions

Browse source
This commit is contained in:
Benedikt Bastin 2021-04-29 13:50:48 +02:00
parent e13e8f1df6
commit 0e0ad8bc0f
1 changed files with 118 additions and 109 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

63
plot.py
View file

@ -52,28 +52,33 @@ plt.rcParams["figure.figsize"] = [11.69, 8.27]
# Download
data_filename = '{}/{}_Impfquotenmonitoring.xlsx'.format(data_folder, filename_now)
def download_rki(filename_prefix):
data_filename = '{}/{}_Impfquotenmonitoring.xlsx'.format(data_folder, filename_prefix)
r = req.get('https://www.rki.de/DE/Content/InfAZ/N/Neuartiges_Coronavirus/Daten/Impfquotenmonitoring.xlsx?__blob=publicationFile')
r = req.get('https://www.rki.de/DE/Content/InfAZ/N/Neuartiges_Coronavirus/Daten/Impfquotenmonitoring.xlsx?__blob=publicationFile')
with open(data_filename, 'wb') as outfile:
with open(data_filename, 'wb') as outfile:
outfile.write(r.content)
#data_filename = 'data/20210118151908_Impfquotenmonitoring.xlsx'
return data_filename
rki_file = pd.read_excel(data_filename, sheet_name=None, engine='openpyxl')
data_filename = download_rki(filename_now)
raw_data = rki_file['Impfungen_proTag']
def parse_rki(filename):
impfungen = raw_data[:-1].dropna(subset=['Datum']).fillna(0)
rki_file = pd.read_excel(filename, sheet_name=None, engine='openpyxl')
impfungen.drop(impfungen.tail(1).index,inplace=True) # remove Gesamt row
raw_data = rki_file['Impfungen_proTag']
dates = impfungen['Datum']
impfungen = raw_data[:-1].dropna(subset=['Datum']).fillna(0)
start_of_reporting_date = dates.iloc[0].date()
impfungen.drop(impfungen.tail(1).index,inplace=True) # remove Gesamt row
def calculate_vaccination_data(data):
dates = impfungen['Datum']
start_of_reporting_date = dates.iloc[0].date()
def calculate_vaccination_data(data):
total = int(np.sum(data))
total_percentage = float(total) / einwohner_deutschland * 100
@ -169,32 +174,36 @@ def calculate_vaccination_data(data):
'vaccinations_last_week_vaccination_percentage': vaccinations_by_week[Week.thisweek() - 1] / total * 100
}
if 'Erstimpfung' in impfungen:
if 'Erstimpfung' in impfungen:
raw_first_vaccinations = impfungen['Erstimpfung']
elif 'Einmal geimpft' in impfungen:
elif 'Einmal geimpft' in impfungen:
raw_first_vaccinations = impfungen['Einmal geimpft']
elif 'Begonnene Impfserie' in impfungen:
elif 'Begonnene Impfserie' in impfungen:
raw_first_vaccinations = impfungen['Begonnene Impfserie']
if 'Zweitimpfung' in impfungen:
if 'Zweitimpfung' in impfungen:
raw_second_vaccinations = impfungen['Zweitimpfung']
elif 'Vollständig geimpft' in impfungen:
elif 'Vollständig geimpft' in impfungen:
raw_second_vaccinations = impfungen['Vollständig geimpft']
data_first_vaccination = calculate_vaccination_data(raw_first_vaccinations)
data_second_vaccination = calculate_vaccination_data(raw_second_vaccinations)
data_first_vaccination = calculate_vaccination_data(raw_first_vaccinations)
data_second_vaccination = calculate_vaccination_data(raw_second_vaccinations)
# Stand aus Daten auslesen
#stand = dates.iloc[-1]
#print_stand = stand.isoformat()
# Stand aus Daten auslesen
#stand = dates.iloc[-1]
#print_stand = stand.isoformat()
# Stand aus offiziellen Angaben auslesen
stand = rki_file['Erläuterung'].iloc[1][0]
# Stand aus offiziellen Angaben auslesen
stand = rki_file['Erläuterung'].iloc[1][0]
stand_regex = re.compile('^Datenstand: (\d\d.\d\d.\d\d\d\d, \d?\d:\d\d) Uhr$')
m = stand_regex.match(stand)
stand_date = datetime.datetime.strptime(m.groups()[0], '%d.%m.%Y, %H:%M')
print_stand = stand_date.isoformat()
stand_regex = re.compile('^Datenstand: (\d\d.\d\d.\d\d\d\d, \d?\d:\d\d) Uhr$')
m = stand_regex.match(stand)
stand_date = datetime.datetime.strptime(m.groups()[0], '%d.%m.%Y, %H:%M')
print_stand = stand_date.isoformat()
return dates, start_of_reporting_date, data_first_vaccination, data_second_vaccination, stand_date, print_stand
dates, start_of_reporting_date, data_first_vaccination, data_second_vaccination, stand_date, print_stand = parse_rki(filename=data_filename)
filename_stand = stand_date.strftime("%Y%m%d%H%M%S")