def retrieve_hosp_data():
countries = parse_countries(2)
hospData = {}
# For now, always get the data from yesterday. We could make fancier check if today's data is already available
file_name, headers = urlretrieve(URL_Hosp)
workbook = xlrd.open_workbook(file_name)
worksheet = workbook.sheet_by_name('Data (table)')
i = 0
Ix = {}
for c in worksheet.row_values(0):
Ix[c] = i
i += 1
for row_index in range(1, worksheet.nrows):
row = worksheet.row_values(row_index)
country = row[Ix['COUNTRY_REGION']]
if country in countries:
country = countries[country]
else:
#print(f'could not find country {country}')
continue
hospData[country] = {'hospitalBeds': stoi(row[Ix['VALUE']]), 'srcHospitalBeds': 'WHO Hospital data from https://gateway.euro.who.int/en/indicators/hfa_479-5061-number-of-acute-care-hospital-beds/'}
# second data set, for Asia
with open('hospital-data/ICU_asia.tsv', 'r') as fd:
rdr = csv.reader(fd, delimiter='\t')
hdr = next(rdr)
for row in rdr:
country = row[0].strip()
if stoi(row[1]):
hospData[country] = {'hospitalBeds': stoi(row[1]), 'srcHospitalBeds': URL_ICU_ASIA}
# Eurostat, newer than from WHO (2017 vs 2015)
with open('hospital-data/hospital_eurostat.tsv', 'r') as fd:
rdr = csv.reader(fd, delimiter='\t')
hdr = next(rdr)
for row in rdr:
country = row[0].strip()
if not country in hospData:
hospData[country] = {'hospitalBeds': stoi(row[1]), 'srcHospitalBeds': URL_Hosp_Eurostat}
else:
#print(f'found new data for {country}: old: {hospData[country]["hospitalBeds"]}, new: {stoi(row[1])}')
hospData[country] = {'hospitalBeds': stoi(row[1]), 'srcHospitalBeds': URL_Hosp_Eurostat}
return hospData
def retrieve_pop_data():
countries = parse_countries(2)
countries1 = parse_countries(1)
countries['XKX'] = countries['RKS']
countries1['BLM'] = countries1['BL']
popData = {}
file_name, headers = urlretrieve(URL_Pop)
workbook = xlrd.open_workbook(file_name)
worksheet = workbook.sheet_by_index(0)
i = 0
Ix = {}
for c in worksheet.row_values(0):
Ix[c] = i
i += 1
for row_index in range(1, worksheet.nrows):
row = worksheet.row_values(row_index)
popData2018 = stoi(row[Ix['popData2018']])
# replace country name if we have the "official" one in country_codes.csv
country = row[Ix['countryterritoryCode']]
if country == [] or not country:
country = countries1[row[Ix['geoId']]]
elif country in countries:
country = countries[country]
else:
#print(f'could not find country {country}')
continue
popData[country] = {'populationServed': popData2018, 'srcPopulation': 'ECDC'}
return popData
def update_cia_facts(hospData, popData):
r = get(URL_CIA)
if not r.ok:
print(f"Failed to fetch {URL}", file=sys.stderr)
exit(1)
db = json.loads(r.text)
countries = parse_countries(2)
for country in db["countries"]:
country2 = None
matches = [r for s,r in zip([x.lower() for x in countries.values()],[x for x in countries.values()]) if country.replace("_"," ") in s]
if len(matches) == 0:
# manual map to compensate
if country in ciaMap:
country2 = ciaMap[country]
else:
#print(f'no match found for {country}')
continue
elif len(matches) == 1:
country2 = matches[0]
else:
# only few cases, handle manually
for n in matches:
if country.replace("_"," ") == n.lower():
country2 = n
if country == 'united_states':
country2 = 'United States of America'
elif country == 'virgin_islands':
country2 = 'Virgin Islands (U.S.)'
if not country2:
print(f'{country} matches {matches}')
continue
if not 'people' in db['countries'][country]['data']:
continue
population = stoi(db["countries"][country]["data"]["people"]["population"]["total"])
source = db["countries"][country]["metadata"]["source"]
if not country2 in hospData and 'hospital_bed_density' in db['countries'][country]['data']['people']:
hospData[country2] = {}
hospData[country2]['hospitalBeds'] = int(population/1000*float(db['countries'][country]['data']['people']['hospital_bed_density']['beds_per_1000_population']))
if 'date' in db['countries'][country]['data']['people']['hospital_bed_density']:
hospData[country2]['srcHospitalBeds'] = f"Computed from {db['countries'][country]['data']['people']['hospital_bed_density']['date']} data in CIA fact book {source}"
else:
hospData[country2]['srcHospitalBeds'] = f"Computed from data in CIA fact book {source}"
if not country2 in popData and 'population' in db['countries'][country]['data']['people']:
popData[country2] = {}
popData[country2]['populationServed'] = population
if 'date' in db['countries'][country]['data']['people']['population']:
popData[country2]['srcPopulation'] = f"{db['countries'][country]['data']['people']['population']['date']} data in CIA fact book {source}"
else:
popData[country2]['srcPopulation'] = f"Data in CIA fact book {source}"
return hospData, popData
def retrieve_icu_data():
icuData = {}
with open('hospital-data/ICU_capacity.tsv', 'r') as fd:
rdr = csv.reader(fd, delimiter='\t')
hdr = next(rdr)
for row in rdr:
country = row[0].strip()
icuData[country] = {'ICUBeds': stoi(row[5]), 'srcICUBeds': URL_ICU}
# second data set, for Asia
with open('hospital-data/ICU_asia.tsv', 'r') as fd:
rdr = csv.reader(fd, delimiter='\t')
hdr = next(rdr)
for row in rdr:
country = row[0].strip()
icuData[country] = {'ICUBeds': stoi(row[5]), 'srcICUBeds': URL_ICU_ASIA}
return icuData
def parse_tsv(tsv, location):
rdr = csv.reader(tsv, delimiter='\t')
hdr = next(rdr)
idx = {}
for col in cols:
try:
idx[col] = hdr.index(col)
except:
print(col, cols, hdr)
return None, False
data = defaultdict(list)
for row in rdr:
data[location].append({c:stoi(row[idx[c]]) if i > 0 else row[idx[c]] for i, c in enumerate(cols)})
return data, True
def get_old_data():
known = {}
with open('populationData.tsv', 'r') as fd:
rdr = csv.reader(fd, delimiter='\t')
hdr = next(rdr)
i = 0
Ix = {}
for c in hdr:
Ix[c] = i
i += 1
for row in rdr:
name = row[0]
if not name in known:
known[name] = {}
for n in Ix:
if n == 'name':
continue
if 'name' in n or 'ageDistribution' in n or 'hemisphere' in n or 'src' in n:
known[name][n] = row[Ix[n]]
else:
known[name][n] = stoi(row[Ix[n]])
return known