This script fetch the News and corresponding reactions from a given telegram channel.

getting_news.py

import pandas as pd
from telethon import TelegramClient
import datetime
import re

# Replace these with your own values
api_id = 1111 # Your API ID
api_hash = 'Your API hash'
channel_username = 'Akharinkhabar'

# setting initial date to one day before 7 october
initial_date = datetime.datetime(2023 , 10 , 6)

def extract_title_remove_water_mark(text):
    text = re.sub(r"@Akharinkhabar.+$", "", text)
    title_text = ""
    # Step 2: Extract the title (first line that starts with emoji and bold text)
    title = re.search(r"^.+$", text ,re.MULTILINE)
    if title :
        title_text = title.group()
    text = text.strip()
    return title_text ,text  

def get_reaction(message ) : 
    message_reaction = {}
    if message.reactions:
        for reaction in message.reactions.results:
            count = reaction.count  # Get the count of that reaction
            emoji_code = reaction.reaction.emoticon # Final

            if emoji_code in message_reaction : 
                message_reaction[emoji_code] += count
            else : 
                message_reaction[emoji_code]= count
    return message_reaction

# Create the client and connect
client = TelegramClient('anon', api_id, api_hash)

async def fetch_messages_to_csv(channel_username, batch_size=100):
    await client.start()

    channel = await client.get_entity(channel_username)

    messages_data = []

    async for message in client.iter_messages(channel , offset_date=initial_date , reverse= True) : 
        try :
            message_reaction = get_reaction(message= message)
        except : 
            message_reaction = {}
        try:
            title , text = extract_title_remove_water_mark(message.raw_text) 
        except :
            title = ""
            text = ""
        message_info = {
            'id': f"{channel_username}-{message.id}", 
            'channel': channel_username,
            'date': message.date,
            'views': message.views if message.views is not None else 0,
            'reply_to': f"{channel_username}-{message.id}" if message.is_reply else None , 
            'title' : title,
            'text': text,
            'reactions' : message_reaction
        }
        messages_data.append(message_info)

        if len(messages_data) >= batch_size:
            # Convert to DataFrame and save to CSV
            df = pd.DataFrame(messages_data)
            df.to_csv(f'{channel_username}_messages.csv', mode='a', header=not pd.io.common.file_exists(f'{channel_username}_messages.csv'), index=False)
            messages_data = []  # Reset the list for the next batch
            exit(1)
    # Save any remaining messages that didn't fill a full batch
    if messages_data:
        df = pd.DataFrame(messages_data)
        df.to_csv(f'{channel_username}_messages.csv', mode='a', header=not pd.io.common.file_exists(f'{channel_username}_messages.csv'), index=False)

# Run the client
with client:
    client.loop.run_until_complete(fetch_messages_to_csv(channel_username, batch_size=100))