Merge branch 'pr-twitter' of https://github.com/atomicdryad/youtube-dl into atomicdryad-pr-twitter

This commit is contained in:
Yen Chi Hsuan 2015-10-18 16:49:56 +08:00
commit c571dea953
2 changed files with 82 additions and 11 deletions

View File

@ -690,7 +690,7 @@
TwitchBookmarksIE, TwitchBookmarksIE,
TwitchStreamIE, TwitchStreamIE,
) )
from .twitter import TwitterCardIE from .twitter import TwitterCardIE, TwitterIE
from .ubu import UbuIE from .ubu import UbuIE
from .udemy import ( from .udemy import (
UdemyIE, UdemyIE,

View File

@ -12,17 +12,30 @@
class TwitterCardIE(InfoExtractor): class TwitterCardIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?twitter\.com/i/cards/tfw/v1/(?P<id>\d+)' _VALID_URL = r'https?://(?:www\.)?twitter\.com/i/cards/tfw/v1/(?P<id>\d+)'
_TEST = { _TESTS = [
'url': 'https://twitter.com/i/cards/tfw/v1/560070183650213889', {
'md5': 'a74f50b310c83170319ba16de6955192', 'url': 'https://twitter.com/i/cards/tfw/v1/560070183650213889',
'info_dict': { 'md5': 'a74f50b310c83170319ba16de6955192',
'id': '560070183650213889', 'info_dict': {
'ext': 'mp4', 'id': '560070183650213889',
'title': 'TwitterCard', 'ext': 'mp4',
'thumbnail': 're:^https?://.*\.jpg$', 'title': 'TwitterCard',
'duration': 30.033, 'thumbnail': 're:^https?://.*\.jpg$',
'duration': 30.033,
}
}, },
} {
'url': 'https://twitter.com/i/cards/tfw/v1/623160978427936768',
'md5': '7ee2a553b63d1bccba97fbed97d9e1c8',
'info_dict': {
'id': '623160978427936768',
'ext': 'mp4',
'title': 'TwitterCard',
'thumbnail': 're:^https?://.*\.jpg',
'duration': 80.155,
},
}
]
def _real_extract(self, url): def _real_extract(self, url):
video_id = self._match_id(url) video_id = self._match_id(url)
@ -44,6 +57,20 @@ def _real_extract(self, url):
unescapeHTML(self._search_regex( unescapeHTML(self._search_regex(
r'data-player-config="([^"]+)"', webpage, 'data player config')), r'data-player-config="([^"]+)"', webpage, 'data player config')),
video_id) video_id)
if 'playlist' not in config:
if 'vmapUrl' in config:
webpage = self._download_webpage(config['vmapUrl'], video_id + ' (xml)')
video_url = self._search_regex(
r'<MediaFile>\s*<!\[CDATA\[(https?://.+?)\]\]>', webpage, 'data player config (xml)')
f = {
'url': video_url,
}
ext = re.search(r'\.([a-z0-9]{2,4})(\?.+)?$', video_url)
if ext:
f['ext'] = ext.group(1)
formats.append(f)
break # same video regardless of UA
continue
video_url = config['playlist'][0]['source'] video_url = config['playlist'][0]['source']
@ -70,3 +97,47 @@ def _real_extract(self, url):
'duration': duration, 'duration': duration,
'formats': formats, 'formats': formats,
} }
class TwitterIE(TwitterCardIE):
_VALID_URL = r'https?://(?:www|m|mobile)?\.?twitter\.com/(?P<id>[^/]+/status/\d+)'
_TESTS = [{
'url': 'https://m.twitter.com/thereaIbanksy/status/614301758345490432',
'md5': '8bbccb487bd7a31349b775915fcd412f',
'info_dict': {
'id': '614301758345490432',
'ext': 'mp4',
'title': 'thereaIbanksy - This time lapse is so pretty \U0001f60d\U0001f60d',
'thumbnail': 're:^https?://.*\.jpg',
'duration': 29.5,
'description': 'banksy on Twitter: "This time lapse is so pretty \U0001f60d\U0001f60d http://t.co/QB8DDbqiR1"',
'uploader': 'banksy',
'uploader_id': 'thereaIbanksy',
},
}]
def _real_extract(self, url):
id = self._match_id(url)
username, twid = re.match(r'([^/]+)/status/(\d+)', id).groups()
name = username
url = re.sub(r'https?://(m|mobile)\.', 'https://', url)
webpage = self._download_webpage(url, 'tweet: ' + url)
description = unescapeHTML(self._search_regex('<title>\s*(.+?)\s*</title>', webpage, 'title'))
title = description.replace('\n', ' ')
splitdesc = re.match(r'^(.+?)\s*on Twitter:\s* "(.+?)"$', 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,
}