bellingcat/snscrape
1
0
Fork 0
mirror of https://github.com/bellingcat/snscrape.git synced 2026-06-12 20:38:29 +03:00
Code Issues Packages Projects Releases Wiki Activity

add the original tweet source

Browse Source
This commit is contained in:
sahrul
2020-10-27 13:21:21 +07:00
parent d65f0434da
commit d2dce37fa0
1 changed files with 4 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
snscrape/modules/twitter.py
View File

@@ -34,8 +34,9 @@ class Tweet(snscrape.base.Item):
quoteCount: int quoteCount: int
conversationId: int conversationId: int
lang: str lang: str
sourceUrl: str source: str
sourceLabel: str sourceUrl: typing.Optional[str] = None
sourceLabel: typing.Optional[str] = None
media: typing.Optional[typing.List['Medium']] = None media: typing.Optional[typing.List['Medium']] = None
retweetedTweet: typing.Optional['Tweet'] = None retweetedTweet: typing.Optional['Tweet'] = None
quotedTweet: typing.Optional['Tweet'] = None quotedTweet: typing.Optional['Tweet'] = None
@@ -315,6 +316,7 @@ class TwitterAPIScraper(snscrape.base.Scraper):
kwargs['quoteCount'] = tweet['quote_count'] kwargs['quoteCount'] = tweet['quote_count']
kwargs['conversationId'] = tweet['conversation_id'] if 'conversation_id' in tweet else int(tweet['conversation_id_str']) kwargs['conversationId'] = tweet['conversation_id'] if 'conversation_id' in tweet else int(tweet['conversation_id_str'])
kwargs['lang'] = tweet['lang'] kwargs['lang'] = tweet['lang']
kwargs['source'] = tweet['source']
if (match := re.search(r'href=[\'"]?([^\'" >]+)', tweet['source'])): if (match := re.search(r'href=[\'"]?([^\'" >]+)', tweet['source'])):
kwargs['sourceUrl'] = match.group(1) kwargs['sourceUrl'] = match.group(1)
if (match := re.search(r'>([^<]*)<', tweet['source'])): if (match := re.search(r'>([^<]*)<', tweet['source'])):