depau/youtube-dl
1
0
Fork 0
mirror of https://github.com/ytdl-org/youtube-dl.git synced 2024-12-22 16:57:40 +00:00
Code Issues Releases Wiki Activity

Twitter: get and describe video from status urls

Browse source
This commit is contained in:
fnord 2015-07-21 16:38:40 -05:00
parent c84683c88b
commit f57f84f606
1 changed files with 44 additions and 0 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

44
youtube_dl/extractor/twitter.py
View file

@ -70,3 +70,47 @@ class TwitterCardIE(InfoExtractor):
'duration': duration,
'formats': formats,
}
class TwitterIE(TwitterCardIE):
_VALID_URL = r'https?://(?:www|m|mobile)?\.?twitter\.com/(?P<id>[^/]+/status/\d+)'
_TESTS = [{
'url': 'https://m.twitter.com/thereaIbanksy/status/614301758345490432',
'md5': '8bbccb487bd7a31349b775915fcd412f',
'info_dict': {
'id': '614301758345490432',
'ext': 'mp4',
'title': 'thereaIbanksy - This time lapse is so pretty \U0001f60d\U0001f60d',
'thumbnail': 're:^https?://.*\.jpg',
'duration': 29.5,
'description': 'banksy on Twitter: "This time lapse is so pretty \U0001f60d\U0001f60d http://t.co/QB8DDbqiR1"',
'uploader': 'banksy',
'uploader_id': 'thereaIbanksy',
},
}]
def _real_extract(self, url):
id = self._match_id(url)
username, twid = re.match(r'([^/]+)/status/(\d+)', id).groups()
name = username
url = re.sub(r'https?://(m|mobile)\.', 'https://', url)
webpage = self._download_webpage(url, 'tweet: ' + url)
description = unescapeHTML(self._search_regex('<title>\s*(.+?)\s*</title>', webpage, 'title'))
title = description.replace('\n', ' ')
splitdesc = re.match(r'^(.+?)\s*on Twitter:\s* "(.+?)"$', title)
if splitdesc:
name, title = splitdesc.groups()
title = re.sub(r'\s*https?://[^ ]+', '', title) # strip 'https -_t.co_BJYgOjSeGA' junk from filenames
card_id = self._search_regex(r'["\']/i/cards/tfw/v1/(\d+)', webpage, '/i/card/...')
card_url = 'https://twitter.com/i/cards/tfw/v1/' + card_id
return {
'_type': 'url_transparent',
'ie_key': 'TwitterCard',
'uploader_id': username,
'uploader': name,
'url': card_url,
'webpage_url': url,
'description': description,
'title': username + ' - ' + title,
}