Skip to content

Instantly share code, notes, and snippets.

Embed
What would you like to do?
Parse the html export of a Telegram chat
''' The parse_telegram_export function in this gist parses the html export of a Telegram chat.
You should have [BeautifulSoup](https://www.crummy.com/software/BeautifulSoup/bs4/) and
[dateutil](https://dateutil.readthedocs.io/) installed. It extracts the sender name, the
message date and time, the message text and the links in the message.
'''
from bs4 import BeautifulSoup
import dateutil
def parse_telegram_export(html_str, tz_name=None):
''' Parses a Telegram html export.
Params:
- html_str (str): The html string containing the Telegram export.
- tz_name (str|None): The name of the timezone where the export was made (eg. "Italy/Rome").
If None, no time zone will be set for the resulting datetime.
Returns (generator<(str, datetime.datetime, str, list<str>)>): A generator object
that yields a (from_name, date, text, links) tuple for each messages in the export,
where from_name is the sender name, date and text are the date and text of the message
and links is a list of the links eventually found in the message.
'''
soup = BeautifulSoup(html_str, 'html.parser')
tz = dateutil.tz.gettz(tz_name) if tz_name else None
for div in soup.select("div.message.default"):
body = div.find('div', class_='body')
from_name_ = body.find('div', class_='from_name')
if from_name_ is not None:
from_name = from_name_.string.strip()
text = body.find('div', class_='text').get_text().strip()
links = [l.get('href') for l in body.find_all('a')]
raw_date = body.find('div', class_='date')['title']
naiv_date = dateutil.parser.parse(raw_date)
date = naiv_date.astimezone(tz) if tz else naiv_date
yield (from_name, date, text, links)
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment