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 '+': 'new infections counted that day',
36 'Σ': 'sum of new infections for last 7 days',
37 'Ø': 'per-day average of new infections for last 7 days',
38 'i': 'incidence (x per 100k inhabitants) of new infections for last 7 days',
41 # Read infections table path and output type.
43 if len(sys.argv) != 3:
44 print('Expecting infections table file path and output type as only arguments.')
46 infections_table = sys.argv[1]
47 output_type = sys.argv[2]
49 # Read infections table file lines.
50 f = open(infections_table, 'r')
54 # Basic input validation.
56 header_elements = lines[0].split()
57 if set(header_elements) != district_pops.keys() or \
58 len(header_elements) != len(district_pops.keys()):
59 raise Exception('infections table: invalid header')
61 for line in lines[1:]:
64 if len(header_elements) != len(fields) - 1:
65 raise Exception('infections table: too many elements on line %s',
68 datetime.date.fromisoformat(fields[0])
70 raise Exception('infections table: bad ISO date on line %s',
72 for field in fields[1:]:
76 raise Exception('infections table: bad value on line %s',
79 # Parse first table file line for the names and order of districts.
82 for header in lines[0].split():
83 sorted_districts += [header]
86 # Seed DB with daily new infections data per district, per date.
88 for line in lines[1:]:
91 sorted_dates += [date]
92 for i in range(len(sorted_districts)):
93 district = sorted_districts[i]
94 district_data = fields[i + 1]
95 db[district][date] = {'new_infections': int(district_data)}
98 # In LaGeSo's data, the last "district" is actually the sum of all districts /
99 # the whole of Berlin. For our district order, move it in front of the other
100 # districts, as its numbers are the most interesting, so in the table views
101 # we want to see it first.
102 sum_district = sorted_districts.pop()
103 sorted_districts.insert(0, sum_district)
105 # Fail on any day where the "sum" district's new infections are not the proper
106 # sum of the individual districts new infections. Yes, sometimes Lageso sends
107 # data that is troubled in this way. It will then have to be fixed manually in
108 # the table file, since we should have a human look at what mistake was
110 for date in sorted_dates:
112 for district in [d for d in sorted_districts if not d==sum_district]:
113 day_sum += db[district][date]['new_infections']
114 if day_sum != db[sum_district][date]['new_infections']:
115 raise Exception('Questionable district infection sum in %s' % date)
117 # Enhance DB with data about weekly sums, averages, incidences per day. Ignore
118 # days that have less than 6 predecessors (we can only know a weekly average if
119 # we have a whole week of data).
120 for i in range(len(sorted_dates)):
123 date = sorted_dates[i]
126 week_dates += [sorted_dates[i - j]]
127 for district in sorted_districts:
128 district_pop = district_pops[district]
130 for week_date in week_dates:
131 week_sum += db[district][week_date]['new_infections']
132 db[district][date]['week_sum'] = week_sum
133 db[district][date]['week_average'] = week_sum / 7
134 db[district][date]['week_incidence'] = (week_sum / district_pop) * 100000
136 # Optimized for web browser viewing.
138 if output_type == 'html':
139 print("""<!DOCTYPE html>
143 th { text-align: left; vertical-align: bottom; }
144 .vertical_header { writing-mode: vertical-rl; transform: rotate(180deg); font-weight: normal; }
145 .repeated_head th { padding-top: 0.5em; border-bottom: 1px solid black; }
146 .bold { font-weight: bold }
147 .date { vertical-align: top; padding-top: 0.5em; }
149 <title>Berlin's Corona infection numbers, development by districts</title>
151 <a href="/">home</a> · <a href="/contact.html">contact</a> · <a href="/privacy.html">privacy</a>
152 <h1>Berlin's Corona infection numbers, development by districts</h1>
153 <p>Updated daily at 7pm 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">Plain text view (optimized for terminal curl)</a>.</p>
156 <th colspan=2></th>""")
157 sorted_dates.reverse()
158 for district in sorted_districts:
159 # Wrap in div because the vertical orientation otherwise fails
161 print('<th><div class="vertical_header">%s</div></th>' %
165 for date in sorted_dates:
166 if weekday_count == 0:
167 print('<tr class="repeated_head">')
168 print('<th>date</th>')
169 print('<th><a href="#symbols">?</a></th>')
170 for district in sorted_districts:
171 print('<th><abbr title="%s">%s</abbr></th>' %
172 (translate[district], district))
174 print('<tr class="day_row">')
175 weekday = calendar.day_name[datetime.date.fromisoformat(date).weekday()]
176 print('<td class="date">%s<br />%s</td>' % (date, weekday))
178 for abbr in ['+', 'Σ', 'Ø', 'i']:
179 print('<tr><th><abbr title="%s">%s</abbr></th></tr>' %
180 (translate[abbr], abbr))
181 print('</table></td>')
182 for district in sorted_districts:
183 district_data = db[district][date]
184 week_sum = week_avg = week_inc = '?'
185 new_infections = district_data['new_infections']
186 if 'week_sum' in district_data:
187 week_sum = '%s' % district_data['week_sum']
188 if 'week_average' in district_data:
189 week_avg = '%.1f' % district_data['week_average']
190 if 'week_incidence' in district_data:
191 week_inc = '%.1f' % district_data['week_incidence']
194 print('<tr><td class="bold">%s</td></tr>' % new_infections)
195 print('<tr><td>%s</td></tr>' % week_sum)
196 print('<tr><td>%s</td></tr>' % week_avg)
197 print('<tr><td>%s</td></tr>' % week_inc)
202 if weekday_count != 7:
206 print('<h3 id="symbols">Symbols</h3>')
208 for abbr in ['+', 'Σ', 'Ø', 'i']:
209 print('<dt>%s</dt><dd>%s</dd>' % (abbr, translate[abbr]))
213 # Optimized for in-terminal curl.
214 elif output_type == 'txt':
216 # Explain what this is.
218 """Table of Berlin's Corona infection number development by districts.
219 Updated daily at 7pm based on data from the "Senatsverwaltung für Gesundheit, Pflege und Gleichstellung".
221 Abbrevations/explanations:
224 intro += "%s: %s\n" % (k, translate[k])
226 Source code: https://plomlompom.com/repos/?p=berlin-corona-table
227 HTML view: https://plomlompom.com/berlin_corona.html"""
230 # Output table of enhanced daily infection data, newest on top,
231 # separated into 7-day units.
232 sorted_dates.reverse()
234 for date in sorted_dates:
237 if weekday_count == 0:
239 print(' '*13, ' '.join(sorted_districts))
243 weekday = calendar.day_name[datetime.date.fromisoformat(date).weekday()]
244 print('%s (%s)' % (date, weekday))
246 weekly_sum_strings = []
247 weekly_avg_strings = []
248 weekly_inc_strings = []
249 for district in sorted_districts:
250 district_day_data = db[district][date]
251 new_infections += [district_day_data['new_infections']]
252 wsum_string = ' '*3 + '?'
253 wavg_string = winc_string = ' '*4 + '?'
254 if 'week_sum' in district_day_data:
255 wsum_string = '%4s' % district_day_data['week_sum']
256 weekly_sum_strings += [wsum_string]
257 if 'week_average' in district_day_data:
258 wavg_string = '%5.1f' % district_day_data['week_average']
259 weekly_avg_strings += [wavg_string]
260 if 'week_incidence' in district_day_data:
261 winc_string = '%5.1f' % district_day_data['week_incidence']
262 weekly_inc_strings += [winc_string]
263 print('+', ' '*11, ' '.join(['%3s' % i for i in new_infections]))
264 print('Σ', ' '*10, ' '.join(weekly_sum_strings))
265 print('Ø', ' '*9, ''.join(weekly_avg_strings))
266 print('i', ' '*9, ''.join(weekly_inc_strings))
268 if weekday_count != 7: