CODEWITHSUNDEEP
pythonpython-telegram-bot
Divyam Bansal
Divyam Bansal

Reputation: 153

Telegram bot html parsemode is giving string instead of parsing it

I want to send a message with the twitter name as text hyperlinked with the tweet. I have been trying to use html parsemode but instead of treating my string as HTML, it is simply returning the entire string. The code that I have written is given below.

import os
# from dotenv import load_dotenv
# load_dotenv()
import requests
import json
import tweepy
from nltk.tokenize import WordPunctTokenizer
import re
from bs4 import BeautifulSoup
import config
from textblob import TextBlob, Word, Blobber
from telegram import ParseMode


token= "5236830904:AAHjxyq08dXuJzKofXmPkJ30X5V1lNOUhIc"
consumer_key= config.api_key
consumer_secret= config.api_secret_key
access_token= config.access_token
access_token_secret= config.access_token_secret

botsUrl= "https://api.telegram.org/bot{}".format(token)
auth = tweepy.OAuthHandler(consumer_key, consumer_secret)
auth.set_access_token(access_token, access_token_secret)
api = tweepy.API(auth, wait_on_rate_limit=True)

def clean_tweets(twt):
    # twt = re.sub('#ethereum', 'ethereum', twt)
    # twt = re.sub('#Ethereum', 'Ethereum', twt)
    token = WordPunctTokenizer()  
    twt = re.sub('#[A-Za-z0-9]+ ','', twt) #removes any string with a '#' character
    twt = re.sub('\\n', '', twt)
    twt = re.sub('&;','and',twt)
    twt = re.sub('@[A-Za-z0-9]+ ','', twt)
    twt = re.sub('https?:\/\/\S+','',twt) #Removes any hyperlinks
    regex_pattern = re.compile(pattern = "["
        u"\U0001F600-\U0001F64F"  # emoticons
        u"\U0001F300-\U0001F5FF"  # symbols & pictographs
        u"\U0001F680-\U0001F6FF"  # transport & map symbols
        u"\U0001F1E0-\U0001F1FF"  # flags (iOS)
                           "]+", flags = re.UNICODE)
    twt = re.sub(regex_pattern,'',twt)
    pattern = re.compile(r'(https?://)?(www\.)?(\w+\.)?(\w+)(\.\w+)(/.+)?')
    twt = re.sub(pattern,'',twt)
    re_list = ['@[A-Za-z0-9_]+', '#']
    combined_re = re.compile( '|'.join( re_list) )
    twt = re.sub(combined_re,'',twt)
    del_amp = BeautifulSoup(twt, 'lxml')
    del_amp_text = del_amp.get_text()
    del_link_mentions = re.sub(combined_re, '', del_amp_text)
    del_emoticons = re.sub(regex_pattern, '', del_link_mentions)
    lower_case = del_emoticons.lower()
    words = token.tokenize(lower_case)
    result_words = [x for x in words if len(x) > 2]
    return (" ".join(result_words)).strip()

def subjectivity(twt):
    return TextBlob(twt).sentiment.subjectivity

#Function to get the polarity

def getPolarity(twt):
    return TextBlob(twt).sentiment.polarity

def getSentiment(score):
    if score<0:
        return 'Negative'
    elif score == 0:
        return 'Neutral'
    else:
        return 'Positive'

def giveUpdate(offset=None):
    url = botsUrl+ "/getupdates?timeout=100"
    if offset:
        url = botsUrl+ "/getupdates?offset={}&timeout=100".format(offset+1)
    resp= requests.get(url)
    return json.loads(resp.content)

def sendMessage(msg, chat_id):
    url= botsUrl+ "/sendMessage?chat_id={}&text={}".format(chat_id,msg,parse_mode = ParseMode.HTML)
    resp= requests.get(url)
    return "sent message"


def getReply(msg):
    tweets= tweepy.Cursor(api.search, q= "#{} -filter:retweets".format(msg)).items(5)
    all_tweet= []
    for tw in tweets:
        screen_name = tw.user.screen_name
        text = tw.text
        id = str(tw.id)
        hyperlink =  "<a href='https://twitter.com/twitter/statuses/"+id+"'>"+screen_name+"</a>"
        sentiment = getSentiment(getPolarity(text))
        finalTweet = hyperlink+' - '+text+' -- '+sentiment
        all_tweet.append(finalTweet)
    return all_tweet



id_=None
while True:
    update= giveUpdate(offset=id_)
    update= update['result']

    if update:
        for item in update:
            id_= item['update_id']
            msg= item['message']['text']
            chat_id= item['message']['from']['id']
            if msg:
                reply= getReply(msg)
                for tw in reply:
                    print(sendMessage(tw, chat_id))

The reply that I'm getting on the bot is as follows:

enter image description here

SOLUTION

For anyone stuck in the same situation, here's something that'd help:

So whenever we use a method to operate the telegram bot, the bot sends a url request to the server which then gets updated on the chat screen. Here, the HTML parsemode is passed as an argument but because of string formatting, it never gets passed as a parameter. As I explained earlier, we can manually send the HTML parsemode parameter in the url string which would be picked up by the bot. So the effective change would be:

hyperlink =  "<a href='https://twitter.com/twitter/statuses/"+id+"'>"+screen_name+"</a>"+"&parse_mode=HTML"

Upvotes: 0

Views: 3113

Answers (1)

CallMeStag
CallMeStag

Reputation: 7060

First things first: Revoke the bot token that you posted as part of your code snippet.

About your problem:

url= botsUrl+ "/sendMessage?chat_id={}&text={}".format(chat_id,msg,parse_mode = ParseMode.HTML)

Here you have only two slots in the string that .format can fill, so it will just ignore the parse_mode parameter.

Moreover, I noticed that you use the python-telegram-bot package, but don't really use it. You only use telegram.ParseMode.HTML, which in fact is just a shortcut for typing 'HTML'. The real value of the package is that you don't need to implement methods such as sendMessage or giveUpdate yourself. The telegram.ext package can also take care of fetching updates for you continuously. See here and here. It's ofc up to you whether or not you want to use this functionality or make the requests to Telegram manually - in the latter case I would just recommend to not use python-telegram-bot at all.

Disclaimer: I'm currently the maintainer of python-telegram-bot.

Upvotes: 1

Related Questions