Script to find your most used short messages on Telegram (can be used as a base to do more useful analysis)

get_data.py

import json
import asyncio
import logging
from collections import defaultdict

from telethon import TelegramClient, utils


logging.basicConfig(level=logging.INFO)


async def main():
    client = TelegramClient('user', 6, "eb06d4abfb49dc3eeb1aeb98ae0f581e")
    await client.start()
    my_id = utils.get_peer_id(await client.get_me())

    async with client.takeout(
        finalize=False, users=True, chats=True, megagroups=True
    ) as takeout:
        async for d in takeout.iter_dialogs():
            print(f'downloading {d.name}...')
            num_counted = 0
            num_processed = 0
            async for m in takeout.iter_messages(d, wait_time=0):
                if num_processed % 10 == 0:
                    print(f'\rprocessed {num_processed} ({num_counted} counted) messsages', end='')
                num_processed += 1
                if m.from_id != my_id:
                    continue
                if not m.message:
                    continue
                if len(m.message) > 64:  # arbitrary definition of "short messages"
                    continue
                message_counts[m.message] += 1
                num_counted += 1
            print()


message_counts = defaultdict(int)
asyncio.run(main())
with open('data.json', 'w') as f:
    json.dump(message_counts, f)

print_data.py

import json

with open('data.json') as f:
    counts = json.load(f)

sorted_msgs = sorted(counts, key=lambda m: counts[m], reverse=True)

# print the top 100 messages
top_msgs = sorted_msgs[:100]
print('\n'.join(f'{counts[m]} {m}' for m in top_msgs if counts[m] != 1)