#!//usr/bin/env python3
+import sys
+if len(sys.argv) != 2:
+ print('Expecting infections table file path as only argument.')
+ exit(1)
+infections_table = sys.argv[1]
+
# District population numbers as per Wikipedia.
district_pops = {
'CW': 342332,
'sum': 3754418,
}
-f = open('daily_infections_table.txt', 'r')
+f = open(infections_table, 'r')
lines = f.readlines()
f.close()
-# Parse first table file line for the names and order of districts.
+# Basic input validation.
+import datetime
+header_elements = lines[0].split()
+if set(header_elements) != district_pops.keys() or \
+ len(header_elements) != len(district_pops.keys()):
+ raise Exception('infections table: invalid header')
+line_count = 0
+for line in lines[1:]:
+ line_count += 1
+ fields = line.split()
+ if len(header_elements) != len(fields) - 1:
+ raise Exception('infections table: too many elements on line %s',
+ line_count)
+ try:
+ datetime.date.fromisoformat(fields[0])
+ except ValueError:
+ raise Exception('infections table: bad ISO date on line %s',
+ line_count)
+ for field in fields[1:]:
+ try:
+ int(field)
+ except ValueError:
+ raise Exception('infections table: bad value on line %s',
+ line_count)
+
+# Parse first table file line for the names and order of districts.
db = {}
sorted_districts = []
for header in lines[0].split():
db[district][date] = {'new_infections': int(district_data)}
sorted_dates.sort()
+# In LaGeSo's data, the last "district" is actually the sum of all districts /
+# the whole of Berlin.
+#
# Fail on any day where the "sum" district's new infections are not the proper
# sum of the individual districts new infections. Yes, sometimes Lageso sends
# data that is troubled in this way. It will then have to be fixed manually in
# the table file, since we should have a human look at what mistake was
-# probably made.
+# probably made.
for date in sorted_dates:
sum_district = sorted_districts[-1]
day_sum = 0
if i < 6:
continue
date = sorted_dates[i]
- week_dates = []
+ week_dates = []
for j in range(7):
- week_dates += [sorted_dates[i - j]]
+ week_dates += [sorted_dates[i - j]]
for district in sorted_districts:
district_pop = district_pops[district]
week_sum = 0
db[district][date]['week_incidence'] = (week_sum / district_pop) * 100000
# Explain what this is.
-intro = """
-Table of Berlin's Corona infection number development by districts, daily
-updated around 9pm.
+intro = """Table of Berlin's Corona infection number development by districts.
+Updated daily around 9pm.
Abbrevations/explanations:
+
CW: Charlottenburg-Wilmersdorf
FK: Friedrichshain-Kreuzberg
Li: Lichtenberg
if weekday_count == 0:
print(' '*11, ' '.join(sorted_districts[:-1]),
sorted_districts[-1], 'wsum', ' wavg', 'winc')
- week_start_date = date
+ week_start_date = date
- # Day data line.
+ # Day data line.
new_infections = []
for district in sorted_districts:
new_infections += [db[district][date]['new_infections']]
sum_district = sorted_districts[-1]
sum_district_data = db[sum_district][date]
if 'week_sum' in sum_district_data:
- week_sum = '%4s' % sum_district_data['week_sum']
+ week_sum = '%4s' % sum_district_data['week_sum']
if 'week_average' in sum_district_data:
- week_avg = '%5.1f' % sum_district_data['week_average']
+ week_avg = '%5.1f' % sum_district_data['week_average']
if 'week_incidence' in sum_district_data:
- week_inc = '%4.1f' % sum_district_data['week_incidence']
+ week_inc = '%4.1f' % sum_district_data['week_incidence']
print(date, ' '.join(['%3s' % infections for infections in new_infections]),
- week_sum, week_avg, week_inc)
+ week_sum, week_avg, week_inc)
# Maintain 7-day cycle.
weekday_count += 1