telegramstats.py

import json
import operator
from datetime import datetime, timedelta
import iso8601 as iso8601
import pytz as pytz
from collections import defaultdict

if __name__ == '__main__':
    with open('result.json', encoding="utf-8") as f:
        data = json.load(f)

    name = 'HIER EIGENEN NAMEN EINSETZEN'

    results = []
    for chat in data['chats']['list']:
        if 'name' not in chat:
            continue

        delta = timedelta(days=1)

        count = defaultdict(lambda: 0)
        last_msg_time = pytz.UTC.localize(datetime.min)
        for msg in chat['messages']:
            if not msg['type'] == 'message':
                continue
            t = iso8601.parse_date(msg['date'])
            if t > last_msg_time + delta:
                count[msg['from']] += 1
            last_msg_time = t
        me = count[name]
        others = list(count.keys())
        others.remove(name)
        if len(others) == 0:
            chat_name = chat['name']
            other = 0
        elif len(others) == 1:
            chat_name = others[0]
            other = count[chat_name]
        else:
            print('ERROR: Wrong Participant count')
            print('Skipping %s' % chat['name'])
            continue

        results.append({
            'name': chat_name,
            'me': me,
            'other': other,
            'ratio': me / (me + other),
            'sum': me + other,
            'messages': len(chat['messages'])
        })
    results = [x for x in results if x['messages'] >= 100]
    results.sort(key=operator.itemgetter('ratio'), reverse=True)
    for i, r in enumerate(results):
        print('%2d. %.2f (%3d:%3d, %3d, %5d) %s' % (i, r['ratio'], r['me'], r['other'], r['sum'], r['messages'], r['name']))