#!//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,
}
+# Map abbreviations to full names.
+translate = {
+ 'CW': 'Charlottenburg-Wilmersdorf',
+ 'FK': 'Friedrichshain-Kreuzberg',
+ 'Li': 'Lichtenberg',
+ 'MH': 'Marzahn-Hellersdorf',
+ 'Mi': 'Mitte',
+ 'Ne': 'Neukölln',
+ 'Pa': 'Pankow',
+ 'Re': 'Reinickendorf',
+ 'Sp': 'Spandau',
+ 'SZ': 'Steglitz-Zehlendorf',
+ 'TS': 'Tempelhof-Schöneberg',
+ 'TK': 'Treptow-Köpenick',
+ 'sum': 'all of Berlin',
+ '+': 'new infections counted that day',
+ 'Σ': 'sum of new infections for last 7 days',
+ 'Ø': 'per-day average of new infections for last 7 days',
+ 'i': 'incidence (x per 100k inhabitants) of new infections for last 7 days',
+}
+
+# Read infections table path and output type.
+import sys
+if len(sys.argv) != 3:
+ print('Expecting infections table file path and output type as only arguments.')
+ exit(1)
+infections_table = sys.argv[1]
+output_type = sys.argv[2]
+
+# Read infections table file lines.
f = open(infections_table, 'r')
lines = f.readlines()
f.close()
+# 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 = []
sorted_dates.sort()
# In LaGeSo's data, the last "district" is actually the sum of all districts /
-# the whole of Berlin.
-#
+# the whole of Berlin. For our district order, move it in front of the other
+# districts, as its numbers are the most interesting, so in the table views
+# we want to see it first.
+sum_district = sorted_districts.pop()
+sorted_districts.insert(0, sum_district)
+
# 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.
for date in sorted_dates:
- sum_district = sorted_districts[-1]
day_sum = 0
- for district in sorted_districts[:-1]:
+ for district in [d for d in sorted_districts if not d==sum_district]:
day_sum += db[district][date]['new_infections']
if day_sum != db[sum_district][date]['new_infections']:
raise Exception('Questionable district infection sum in %s' % date)
db[district][date]['week_average'] = week_sum / 7
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.
-Updated daily around 9pm.
+# Optimized for web browser viewing.
+import calendar
+if output_type == 'html':
+ print("""<!DOCTYPE html>
+<html>
+<head>
+<style>
+th { text-align: left; vertical-align: bottom; }
+.vertical_header { writing-mode: vertical-rl; transform: rotate(180deg); font-weight: normal; }
+.repeated_head th { padding-top: 0.5em; border-bottom: 1px solid black; }
+.bold { font-weight: bold }
+.date { vertical-align: top; padding-top: 0.5em; }
+</style>
+<title>Berlin's Corona infection numbers, development by districts</title>
+</head>
+<a href="/">home</a> · <a href="/contact.html">contact</a> · <a href="/privacy.html">privacy</a>
+<h1>Berlin's Corona infection numbers, development by districts</h1>
+<p><del>Updated daily at 7pm based on data from the "Senatsverwaltung für Gesundheit, Pflege und Gleichstellung".</del> Updates currently inactive – check out <a href="https://www.berlin.de/corona/lagebericht/desktop/corona.html#bezirke">the new dashboard</a> instead that they offer by themselves.</p>
+<p><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>
+<table>
+<tr>
+<th colspan=2></th>""")
+ sorted_dates.reverse()
+ for district in sorted_districts:
+ # Wrap in div because the vertical orientation otherwise fails
+ # in Chromium.
+ print('<th><div class="vertical_header">%s</div></th>' %
+ translate[district])
+ print('</tr>')
+ weekday_count = 0
+ for date in sorted_dates:
+ if weekday_count == 0:
+ print('<tr class="repeated_head">')
+ print('<th>date</th>')
+ print('<th><a href="#symbols">?</a></th>')
+ for district in sorted_districts:
+ print('<th><abbr title="%s">%s</abbr></th>' %
+ (translate[district], district))
+ print('</tr>')
+ print('<tr class="day_row">')
+ weekday = calendar.day_name[datetime.date.fromisoformat(date).weekday()]
+ print('<td class="date">%s<br />%s</td>' % (date, weekday))
+ print('<td><table>')
+ for abbr in ['+', 'Σ', 'Ø', 'i']:
+ print('<tr><th><abbr title="%s">%s</abbr></th></tr>' %
+ (translate[abbr], abbr))
+ print('</table></td>')
+ for district in sorted_districts:
+ district_data = db[district][date]
+ week_sum = week_avg = week_inc = '?'
+ new_infections = district_data['new_infections']
+ if 'week_sum' in district_data:
+ week_sum = '%s' % district_data['week_sum']
+ if 'week_average' in district_data:
+ week_avg = '%.1f' % district_data['week_average']
+ if 'week_incidence' in district_data:
+ week_inc = '%.1f' % district_data['week_incidence']
+ print('<td>')
+ print('<table>')
+ print('<tr><td class="bold">%s</td></tr>' % new_infections)
+ print('<tr><td>%s</td></tr>' % week_sum)
+ print('<tr><td>%s</td></tr>' % week_avg)
+ print('<tr><td>%s</td></tr>' % week_inc)
+ print('</table>')
+ print('</td>')
+ print('</tr>')
+ weekday_count += 1
+ if weekday_count != 7:
+ continue
+ weekday_count = 0
+ print('</table>')
+ print('<h3 id="symbols">Symbols</h3>')
+ print('<dl>')
+ for abbr in ['+', 'Σ', 'Ø', 'i']:
+ print('<dt>%s</dt><dd>%s</dd>' % (abbr, translate[abbr]))
+ print('</dl>')
+ print('</html>')
-Abbrevations/explanations:
+# Optimized for in-terminal curl.
+elif output_type == 'txt':
-CW: Charlottenburg-Wilmersdorf
-FK: Friedrichshain-Kreuzberg
-Li: Lichtenberg
-MH: Marzahn-Hellersdorf
-Mi: Mitte
-Ne: Neukölln
-Pa: Pankow
-Re: Reinickendorf
-Sp: Spandau
-SZ: Steglitz-Zehlendorf
-TS: Tempelhof-Schöneberg
-TK: Treptow-Köpenick
-sum: sum for all the districts
-wsum: sum for last 7 days
-wavg: per-day average of new infections for last 7 days
-winc: incidence (x per 100k inhabitants) of new infections for last 7 days
+ # Explain what this is.
+ intro = \
+"""Table of Berlin's Corona infection number development by districts.
+NO LONGER Updated daily at 7pm based on data from the "Senatsverwaltung für Gesundheit, Pflege und Gleichstellung".
+Currently inactive. Instead check out the new dashboard they offer: https://www.berlin.de/corona/lagebericht/desktop/corona.html#bezirke
-Source code: https://plomlompom.com/repos/?p=berlin-corona-table
+Abbrevations/explanations:
"""
-print(intro)
-
-# Output table of enhanced daily infection data, newest on top, separated into
-# 7-day units.
-sorted_dates.reverse()
-weekday_count = 0
-for date in sorted_dates:
-
- # Week table header.
- if weekday_count == 0:
- print(' '*11, ' '.join(sorted_districts[:-1]),
- sorted_districts[-1], 'wsum', ' wavg', 'winc')
- week_start_date = date
+ for k in translate:
+ intro += "%s: %s\n" % (k, translate[k])
+ intro += """
+Source code: https://plomlompom.com/repos/?p=berlin-corona-table
+HTML view: https://plomlompom.com/berlin_corona.html"""
+ print(intro)
- # Day data line.
- new_infections = []
- for district in sorted_districts:
- new_infections += [db[district][date]['new_infections']]
- week_sum = week_avg = week_inc = ''
- 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']
- if 'week_average' in sum_district_data:
- 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']
- print(date, ' '.join(['%3s' % infections for infections in new_infections]),
- week_sum, week_avg, week_inc)
-
- # Maintain 7-day cycle.
- weekday_count += 1
- if weekday_count != 7:
- continue
+ # Output table of enhanced daily infection data, newest on top,
+ # separated into 7-day units.
+ sorted_dates.reverse()
weekday_count = 0
+ for date in sorted_dates:
+
+ # Week table header.
+ if weekday_count == 0:
+ print()
+ print(' '*13, ' '.join(sorted_districts))
+ print('-'*77)
- # After each 7 days, print summary for individual districts.
- weekly_sums = []
- weekly_avgs = []
- weekly_incs = []
- for district in sorted_districts[:-1]:
- weekly_sums += [db[district][week_start_date]['week_sum']]
- weekly_avgs += [db[district][week_start_date]['week_average']]
- weekly_incs += [db[district][week_start_date]['week_incidence']]
- print()
- print('district stats for week from %s to %s:' % (date, week_start_date))
- print(' '*7, ' '.join(sorted_districts[:-1]))
- print('wsum', ' '.join(['%5.1f' % wsum for wsum in weekly_sums]))
- print('wavg', ' '.join(['%5.1f' % wavg for wavg in weekly_avgs]))
- print('winc', ' '.join(['%5.1f' % winc for winc in weekly_incs]))
- print()
+ # Day table.
+ weekday = calendar.day_name[datetime.date.fromisoformat(date).weekday()]
+ print('%s (%s)' % (date, weekday))
+ new_infections = []
+ weekly_sum_strings = []
+ weekly_avg_strings = []
+ weekly_inc_strings = []
+ for district in sorted_districts:
+ district_day_data = db[district][date]
+ new_infections += [district_day_data['new_infections']]
+ wsum_string = ' '*3 + '?'
+ wavg_string = winc_string = ' '*4 + '?'
+ if 'week_sum' in district_day_data:
+ wsum_string = '%4s' % district_day_data['week_sum']
+ weekly_sum_strings += [wsum_string]
+ if 'week_average' in district_day_data:
+ wavg_string = '%5.1f' % district_day_data['week_average']
+ weekly_avg_strings += [wavg_string]
+ if 'week_incidence' in district_day_data:
+ winc_string = '%5.1f' % district_day_data['week_incidence']
+ weekly_inc_strings += [winc_string]
+ print('+', ' '*11, ' '.join(['%3s' % i for i in new_infections]))
+ print('Σ', ' '*10, ' '.join(weekly_sum_strings))
+ print('Ø', ' '*9, ''.join(weekly_avg_strings))
+ print('i', ' '*9, ''.join(weekly_inc_strings))
+ weekday_count += 1
+ if weekday_count != 7:
+ continue
+ weekday_count = 0
projects / berlin-corona-table / blobdiff