1 #!//usr/bin/env python3
3 # District population numbers as per Wikipedia.
20 # Map abbreviations to full names.
22 'CW': 'Charlottenburg-Wilmersdorf',
23 'FK': 'Friedrichshain-Kreuzberg',
25 'MH': 'Marzahn-Hellersdorf',
29 'Re': 'Reinickendorf',
31 'SZ': 'Steglitz-Zehlendorf',
32 'TS': 'Tempelhof-Schöneberg',
33 'TK': 'Treptow-Köpenick',
34 'sum': 'all of Berlin',
35 'wsum': 'sum for last 7 days',
36 'wavg': 'per-day average of new infections for last 7 days',
37 'winc': 'incidence (x per 100k inhabitants) of new infections for last 7 days',
40 # Read infections table path and output type.
42 if len(sys.argv) != 3:
43 print('Expecting infections table file path and output type as only arguments.')
45 infections_table = sys.argv[1]
46 output_type = sys.argv[2]
48 # Read infections table file lines.
49 f = open(infections_table, 'r')
53 # Basic input validation.
55 header_elements = lines[0].split()
56 if set(header_elements) != district_pops.keys() or \
57 len(header_elements) != len(district_pops.keys()):
58 raise Exception('infections table: invalid header')
60 for line in lines[1:]:
63 if len(header_elements) != len(fields) - 1:
64 raise Exception('infections table: too many elements on line %s',
67 datetime.date.fromisoformat(fields[0])
69 raise Exception('infections table: bad ISO date on line %s',
71 for field in fields[1:]:
75 raise Exception('infections table: bad value on line %s',
78 # Parse first table file line for the names and order of districts.
81 for header in lines[0].split():
82 sorted_districts += [header]
85 # Seed DB with daily new infections data per district, per date.
87 for line in lines[1:]:
90 sorted_dates += [date]
91 for i in range(len(sorted_districts)):
92 district = sorted_districts[i]
93 district_data = fields[i + 1]
94 db[district][date] = {'new_infections': int(district_data)}
97 # In LaGeSo's data, the last "district" is actually the sum of all districts /
98 # the whole of Berlin.
100 # Fail on any day where the "sum" district's new infections are not the proper
101 # sum of the individual districts new infections. Yes, sometimes Lageso sends
102 # data that is troubled in this way. It will then have to be fixed manually in
103 # the table file, since we should have a human look at what mistake was
105 for date in sorted_dates:
106 sum_district = sorted_districts[-1]
108 for district in sorted_districts[:-1]:
109 day_sum += db[district][date]['new_infections']
110 if day_sum != db[sum_district][date]['new_infections']:
111 raise Exception('Questionable district infection sum in %s' % date)
113 # Enhance DB with data about weekly sums, averages, incidences per day. Ignore
114 # days that have less than 6 predecessors (we can only know a weekly average if
115 # we have a whole week of data).
116 for i in range(len(sorted_dates)):
119 date = sorted_dates[i]
122 week_dates += [sorted_dates[i - j]]
123 for district in sorted_districts:
124 district_pop = district_pops[district]
126 for week_date in week_dates:
127 week_sum += db[district][week_date]['new_infections']
128 db[district][date]['week_sum'] = week_sum
129 db[district][date]['week_average'] = week_sum / 7
130 db[district][date]['week_incidence'] = (week_sum / district_pop) * 100000
132 # Optimized for web browser viewing.
133 if output_type == 'html':
134 print("""<!DOCTYPE html>
138 table, tr, th, td { border: 1px solid black; text-align: center; }
139 .day_row:nth-child(7n+2) { background-color: yellow; }
140 .district_name { writing-mode: vertical-rl; transform: rotate(180deg); }
141 .bonus_data th { font-weight: normal; }
142 .new_infections { font-weight: bold; }
144 <title>Berlin's Corona infection numbers, development by districts</title>
146 <h1>Berlin's Corona infection numbers, development by districts</h1>
147 <p>Updated daily at 9pm based on data from the "Senatsverwaltung für Gesundheit, Pflege und Gleichstellung". <a href="https://plomlompom.com/repos/?p=berlin-corona-table">Source code</a>. <a href="berlin_corona.txt">Text view optimized for terminal curl</a>.</p>
151 sorted_dates.reverse()
152 sum_district = sorted_districts[-1]
153 for district in sorted_districts:
154 long_form = translate[district]
155 if sum_district == district:
156 print('<th>%s</th>' % long_form)
158 print('<th class="district_name">%s</th>' % long_form)
160 for date in sorted_dates:
161 print('<tr class="day_row">')
162 print('<td>%s</td>' % date)
163 long_wsum = translate['wsum']
164 long_wavg = translate['wavg']
165 long_winc = translate['winc']
166 for district in sorted_districts:
167 district_data = db[district][date]
168 week_sum = week_avg = week_inc = '(not enough data)'
169 new_infections = district_data['new_infections']
170 if 'week_sum' in district_data:
171 week_sum = '%s' % district_data['week_sum']
172 if 'week_average' in district_data:
173 week_avg = '%.1f' % district_data['week_average']
174 if 'week_incidence' in district_data:
175 week_inc = '%.1f' % district_data['week_incidence']
177 print('<span class="new_infections">%s</span>' % new_infections)
178 if district != sum_district:
179 print('<details><summary></summary>')
180 print('<table class="bonus_data">')
181 print('<tr><th>%s</th><td>%s</td></tr>' % (long_wsum, week_sum))
182 print('<tr><th>%s</th><td>%s</td></tr>' % (long_wavg, week_avg))
183 print('<tr><th>%s</th><td>%s</td></tr>' % (long_winc, week_inc))
185 if district != sum_district:
192 # Optimized for in-terminal curl.
193 elif output_type == 'txt':
195 # Explain what this is.
197 """Table of Berlin's Corona infection number development by districts.
198 Updated daily at 9pm based on data from the "Senatsverwaltung für Gesundheit, Pflege und Gleichstellung".
200 Abbrevations/explanations:
203 intro += "%s: %s\n" % (k, translate[k])
205 Source code: https://plomlompom.com/repos/?p=berlin-corona-table
207 HTML view: https://plomlompom.com/berlin_corona.html
211 # Output table of enhanced daily infection data, newest on top,
212 # separated into 7-day units.
213 sorted_dates.reverse()
215 sum_district = sorted_districts[-1]
216 for date in sorted_dates:
219 if weekday_count == 0:
220 print(' '*11, ' '.join(sorted_districts[:-1]),
221 sorted_districts[-1], 'wsum', ' wavg', 'winc')
222 week_start_date = date
226 for district in sorted_districts:
227 new_infections += [db[district][date]['new_infections']]
228 week_sum = week_avg = week_inc = ''
229 sum_district_data = db[sum_district][date]
230 if 'week_sum' in sum_district_data:
231 week_sum = '%4s' % sum_district_data['week_sum']
232 if 'week_average' in sum_district_data:
233 week_avg = '%5.1f' % sum_district_data['week_average']
234 if 'week_incidence' in sum_district_data:
235 week_inc = '%4.1f' % sum_district_data['week_incidence']
236 print(date, ' '.join(['%3s' % infections
237 for infections in new_infections]),
238 week_sum, week_avg, week_inc)
240 # Maintain 7-day cycle.
242 if weekday_count != 7:
246 # After each 7 days, print summary for individual districts.
250 for district in sorted_districts[:-1]:
251 weekly_sums += [db[district][week_start_date]['week_sum']]
252 weekly_avgs += [db[district][week_start_date]['week_average']]
253 weekly_incs += [db[district][week_start_date]['week_incidence']]
255 print('district stats for week from %s to %s:' % (date, week_start_date))
256 print(' '*7, ' '.join(sorted_districts[:-1]))
257 print('wsum', ' '.join(['%5.1f' % wsum for wsum in weekly_sums]))
258 print('wavg', ' '.join(['%5.1f' % wavg for wavg in weekly_avgs]))
259 print('winc', ' '.join(['%5.1f' % winc for winc in weekly_incs]))