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 .day_row:nth-child(7n+3) > td { border-top: 1px solid black; }
139 .vertical_header { writing-mode: vertical-rl; transform: rotate(180deg); font-weight: normal; }
140 .fixed_head { position: sticky; top: 0; background-color: white; }
141 .bold { font-weight: bold }
143 <title>Berlin's Corona infection numbers, development by districts</title>
145 <a href="/">home</a> · <a href="/contact.html">contact</a> · <a href="/privacy.html">privacy</a>
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>
150 <th colspan=2></th>""")
151 sorted_dates.reverse()
152 long_wsum = translate['wsum']
153 long_wavg = translate['wavg']
154 long_winc = translate['winc']
155 sum_district = sorted_districts[-1]
156 for district in sorted_districts:
157 long_form = translate[district]
158 # Wrap in div because the vertical orientation otherwise fails
160 print('<th><div class="vertical_header">%s</div></th>' % long_form)
162 print('<tr class="fixed_head">')
163 # In Chromium, the th only stay fixed if also given this class.
164 print('<th class="fixed_head">date</th>')
165 print('<th class="fixed_head"></th>')
166 for district in sorted_districts:
167 print('<th class="fixed_head">%s</th>' % district)
169 for date in sorted_dates:
170 print('<tr class="day_row">')
171 print('<td>%s</td>' % date)
173 print('<tr><th><abbr title="new">+<abbr></th></tr>')
174 print('<tr><th><abbr title="%s">Σ</abbr></th></tr>' % long_wsum)
175 print('<tr><th><abbr title="%s">Ø</abbr></th></tr>' % long_wavg)
176 print('<tr><th><abbr title="%s">i</abbr></th></tr>' % long_winc)
177 print('</table></td>')
178 for district in sorted_districts:
179 district_data = db[district][date]
180 week_sum = week_avg = week_inc = '(not enough data)'
181 new_infections = district_data['new_infections']
182 if 'week_sum' in district_data:
183 week_sum = '%s' % district_data['week_sum']
184 if 'week_average' in district_data:
185 week_avg = '%.1f' % district_data['week_average']
186 if 'week_incidence' in district_data:
187 week_inc = '%.1f' % district_data['week_incidence']
190 print('<tr><td class="bold">%s</td></tr>' % new_infections)
191 print('<tr><td>%s</td></tr>' % week_sum)
192 print('<tr><td>%s</td></tr>' % week_avg)
193 print('<tr><td>%s</td></tr>' % week_inc)
200 # Optimized for in-terminal curl.
201 elif output_type == 'txt':
203 # Explain what this is.
205 """Table of Berlin's Corona infection number development by districts.
206 Updated daily at 9pm based on data from the "Senatsverwaltung für Gesundheit, Pflege und Gleichstellung".
208 Abbrevations/explanations:
211 intro += "%s: %s\n" % (k, translate[k])
213 Source code: https://plomlompom.com/repos/?p=berlin-corona-table
215 HTML view: https://plomlompom.com/berlin_corona.html
219 # Output table of enhanced daily infection data, newest on top,
220 # separated into 7-day units.
221 sorted_dates.reverse()
223 sum_district = sorted_districts[-1]
224 for date in sorted_dates:
227 if weekday_count == 0:
228 print(' '*11, ' '.join(sorted_districts[:-1]),
229 sorted_districts[-1], 'wsum', ' wavg', 'winc')
230 week_start_date = date
234 for district in sorted_districts:
235 new_infections += [db[district][date]['new_infections']]
236 week_sum = week_avg = week_inc = ''
237 sum_district_data = db[sum_district][date]
238 if 'week_sum' in sum_district_data:
239 week_sum = '%4s' % sum_district_data['week_sum']
240 if 'week_average' in sum_district_data:
241 week_avg = '%5.1f' % sum_district_data['week_average']
242 if 'week_incidence' in sum_district_data:
243 week_inc = '%4.1f' % sum_district_data['week_incidence']
244 print(date, ' '.join(['%3s' % infections
245 for infections in new_infections]),
246 week_sum, week_avg, week_inc)
248 # Maintain 7-day cycle.
250 if weekday_count != 7:
254 # After each 7 days, print summary for individual districts.
258 for district in sorted_districts[:-1]:
259 weekly_sums += [db[district][week_start_date]['week_sum']]
260 weekly_avgs += [db[district][week_start_date]['week_average']]
261 weekly_incs += [db[district][week_start_date]['week_incidence']]
263 print('district stats for week from %s to %s:' % (date, week_start_date))
264 print(' '*7, ' '.join(sorted_districts[:-1]))
265 print('wsum', ' '.join(['%5.1f' % wsum for wsum in weekly_sums]))
266 print('wavg', ' '.join(['%5.1f' % wavg for wavg in weekly_avgs]))
267 print('winc', ' '.join(['%5.1f' % winc for winc in weekly_incs]))