X-Git-Url: https://plomlompom.com/repos/do_day?a=blobdiff_plain;f=enhance_table.py;h=5287c03a3ea30796251eb1fda25455daa25b8632;hb=f0145a1f179e5fd3e61ab123fcc8ad967586fa90;hp=2287b734ca7ccfc57e96056333f37b280d804f78;hpb=ce10254e128792ce0866820c1ed36ea14b1dc61d;p=berlin-corona-table diff --git a/enhance_table.py b/enhance_table.py index 2287b73..5287c03 100755 --- a/enhance_table.py +++ b/enhance_table.py @@ -17,11 +17,65 @@ district_pops = { 'sum': 3754418, } -f = open('daily_infections_table.txt', 'r') +# 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', + 'wsum': 'sum of new infections 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', +} + +# 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() -# 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(): @@ -40,15 +94,21 @@ for line in lines[1:]: db[district][date] = {'new_infections': int(district_data)} sorted_dates.sort() +# Define and move sum_district from end to start. +sum_district = sorted_districts.pop() +sorted_districts.insert(0, sum_district) + +# 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 - 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) @@ -60,9 +120,9 @@ for i in range(len(sorted_dates)): 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 @@ -72,80 +132,135 @@ for i in range(len(sorted_dates)): 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. -Update daily around 9pm. +# Optimized for web browser viewing. +if output_type == 'html': + print(""" + +
+ +Updated daily at 9pm based on data from the "Senatsverwaltung für Gesundheit, Pflege und Gleichstellung". Source code. Text view optimized for terminal curl.
+""") + sorted_dates.reverse() + for district in sorted_districts: + long_form = translate[district] + # Wrap in div because the vertical orientation otherwise fails + # in Chromium. + print(' | %s | ' % long_form)
+ print('|||||||
---|---|---|---|---|---|---|---|---|
date | ') + print('') + for district in sorted_districts: + print(' | %s | ' % district) + print('||||||
%s | ' % date) + print('
| ')
+ 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('')
+ print('
| ')
+ print('