bellingcat/snscrape
1
0
Fork 0
mirror of https://github.com/bellingcat/snscrape.git synced 2026-06-10 19:38:29 +03:00
Code Issues Packages Projects Releases Wiki Activity

fixed edge case where members information wasnt included

Browse Source
This commit is contained in:
Tristan Lee
2022-11-30 12:59:38 -06:00
parent 73f10a4f24
commit cbdfeed812
1 changed files with 7 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files

13
snscrape/modules/telegram.py
View File

@@ -270,12 +270,13 @@ class TelegramChannelScraper(snscrape.base.Scraper):
raise snscrape.base.ScraperException(f'Got status code {r.status_code}')
soup = bs4.BeautifulSoup(r.text, 'lxml')
membersDiv = soup.find('div', class_ = 'tgme_page_extra')
if membersDiv.text.split(',')[0].endswith((' members', ' subscribers')):
membersStr = ''.join(membersDiv.text.split(',')[0].split(' ')[:-1])
if membersStr == 'no':
kwargs['members'] = 0
else:
kwargs['members'] = int(membersStr)
if membersDiv is not None:
if membersDiv.text.split(',')[0].endswith((' members', ' subscribers')):
membersStr = ''.join(membersDiv.text.split(',')[0].split(' ')[:-1])
if membersStr == 'no':
kwargs['members'] = 0
else:
kwargs['members'] = int(membersStr)
photoImg = soup.find('img', class_ = 'tgme_page_photo_image')
if photoImg is not None:
kwargs['photo'] = photoImg.attrs['src']