X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;ds=sidebyside;f=youtube_dl%2Fextractor%2Ftwitter.py;h=ea673054fdc7135a203cca8db00dc128344b0829;hb=6461f2b7ec5fa0114d4bb38ca27c0f72edff8e23;hp=3627d49f1bd4da65690eb9b9448c329a1391ce83;hpb=b8972bd69da48509333a8239fc0b755dccb1a681;p=youtube-dl diff --git a/youtube_dl/extractor/twitter.py b/youtube_dl/extractor/twitter.py index 3627d49f1..ea673054f 100644 --- a/youtube_dl/extractor/twitter.py +++ b/youtube_dl/extractor/twitter.py @@ -10,7 +10,6 @@ from ..utils import ( remove_end, int_or_none, ExtractorError, - sanitized_Request, ) @@ -87,78 +86,79 @@ class TwitterCardIE(TwitterBaseIE): def _real_extract(self, url): video_id = self._match_id(url) - # Different formats served for different User-Agents - USER_AGENTS = [ - 'Mozilla/5.0 (X11; Linux x86_64; rv:10.0) Gecko/20150101 Firefox/20.0 (Chrome)', # mp4 - 'Mozilla/5.0 (Windows NT 5.2; WOW64; rv:38.0) Gecko/20100101 Firefox/38.0', # webm - ] - config = None formats = [] duration = None - for user_agent in USER_AGENTS: - request = sanitized_Request(url) - request.add_header('User-Agent', user_agent) - webpage = self._download_webpage(request, video_id) - - iframe_url = self._html_search_regex( - r']+src="((?:https?:)?//(?:www.youtube.com/embed/[^"]+|(?:www\.)?vine\.co/v/\w+/card))"', - webpage, 'video iframe', default=None) - if iframe_url: - return self.url_result(iframe_url) - - config = self._parse_json(self._html_search_regex( - r'data-(?:player-)?config="([^"]+)"', webpage, 'data player config'), - video_id) - - playlist = config.get('playlist') - if playlist: - video_url = playlist[0]['source'] - - f = { - 'url': video_url, - } - - m = re.search(r'/(?P\d+)x(?P\d+)/', video_url) - if m: - f.update({ - 'width': int(m.group('width')), - 'height': int(m.group('height')), - }) - formats.append(f) - continue - - vmap_url = config.get('vmapUrl') or config.get('vmap_url') - if vmap_url: - formats.append({ - 'url': self._get_vmap_video_url(vmap_url, video_id), + webpage = self._download_webpage(url, video_id) + + iframe_url = self._html_search_regex( + r']+src="((?:https?:)?//(?:www.youtube.com/embed/[^"]+|(?:www\.)?vine\.co/v/\w+/card))"', + webpage, 'video iframe', default=None) + if iframe_url: + return self.url_result(iframe_url) + + config = self._parse_json(self._html_search_regex( + r'data-(?:player-)?config="([^"]+)"', webpage, 'data player config'), + video_id) + + if config.get('source_type') == 'vine': + return self.url_result(config['player_url'], 'Vine') + + def _search_dimensions_in_video_url(a_format, video_url): + m = re.search(r'/(?P\d+)x(?P\d+)/', video_url) + if m: + a_format.update({ + 'width': int(m.group('width')), + 'height': int(m.group('height')), }) - break # same video regardless of UA - - media_info = config.get('status', {}).get('entities', [{}])[0].get('mediaInfo', {}) - if media_info: - for media_variant in media_info['variants']: - media_url = media_variant['url'] - if media_url.endswith('.m3u8'): - formats.extend(self._extract_m3u8_formats(media_url, video_id, ext='mp4', m3u8_id='hls')) - elif media_url.endswith('.mpd'): - formats.extend(self._extract_mpd_formats(media_url, video_id, mpd_id='dash')) - else: - vbr = int_or_none(media_variant.get('bitRate'), scale=1000) - a_format = { - 'url': media_url, - 'format_id': 'http-%d' % vbr if vbr else 'http', - 'vbr': vbr, - } - # Reported bitRate may be zero - if not a_format['vbr']: - del a_format['vbr'] - - formats.append(a_format) - - duration = float_or_none(media_info.get('duration', {}).get('nanos'), scale=1e9) - break # same video regardless of UA + + video_url = config.get('video_url') or config.get('playlist', [{}])[0].get('source') + + if video_url: + f = { + 'url': video_url, + } + + _search_dimensions_in_video_url(f, video_url) + + formats.append(f) + + vmap_url = config.get('vmapUrl') or config.get('vmap_url') + if vmap_url: + formats.append({ + 'url': self._get_vmap_video_url(vmap_url, video_id), + }) + + media_info = None + + for entity in config.get('status', {}).get('entities', []): + if 'mediaInfo' in entity: + media_info = entity['mediaInfo'] + + if media_info: + for media_variant in media_info['variants']: + media_url = media_variant['url'] + if media_url.endswith('.m3u8'): + formats.extend(self._extract_m3u8_formats(media_url, video_id, ext='mp4', m3u8_id='hls')) + elif media_url.endswith('.mpd'): + formats.extend(self._extract_mpd_formats(media_url, video_id, mpd_id='dash')) + else: + vbr = int_or_none(media_variant.get('bitRate'), scale=1000) + a_format = { + 'url': media_url, + 'format_id': 'http-%d' % vbr if vbr else 'http', + 'vbr': vbr, + } + # Reported bitRate may be zero + if not a_format['vbr']: + del a_format['vbr'] + + _search_dimensions_in_video_url(a_format, media_url) + + formats.append(a_format) + + duration = float_or_none(media_info.get('duration', {}).get('nanos'), scale=1e9) self._sort_formats(formats) @@ -187,7 +187,6 @@ class TwitterIE(InfoExtractor): 'ext': 'mp4', 'title': 'FREE THE NIPPLE - FTN supporters on Hollywood Blvd today!', 'thumbnail': 're:^https?://.*\.jpg', - 'duration': 12.922, 'description': 'FREE THE NIPPLE on Twitter: "FTN supporters on Hollywood Blvd today! http://t.co/c7jHH749xJ"', 'uploader': 'FREE THE NIPPLE', 'uploader_id': 'freethenipple', @@ -234,6 +233,44 @@ class TwitterIE(InfoExtractor): # Test case of TwitterCardIE 'skip_download': True, }, + }, { + 'url': 'https://twitter.com/jaydingeer/status/700207533655363584', + 'md5': '', + 'info_dict': { + 'id': '700207533655363584', + 'ext': 'mp4', + 'title': 'jay - BEAT PROD: @suhmeduh #Damndaniel', + 'description': 'jay on Twitter: "BEAT PROD: @suhmeduh https://t.co/HBrQ4AfpvZ #Damndaniel https://t.co/byBooq2ejZ"', + 'thumbnail': 're:^https?://.*\.jpg', + 'uploader': 'jay', + 'uploader_id': 'jaydingeer', + }, + 'params': { + 'skip_download': True, # requires ffmpeg + }, + }, { + 'url': 'https://twitter.com/Filmdrunk/status/713801302971588609', + 'md5': '89a15ed345d13b86e9a5a5e051fa308a', + 'info_dict': { + 'id': 'MIOxnrUteUd', + 'ext': 'mp4', + 'title': 'Dr.Pepperの飲み方 #japanese #バカ #ドクペ #電動ガン', + 'uploader': 'TAKUMA', + 'uploader_id': '1004126642786242560', + 'upload_date': '20140615', + }, + 'add_ie': ['Vine'], + }, { + 'url': 'https://twitter.com/captainamerica/status/719944021058060289', + # md5 constantly changes + 'info_dict': { + 'id': '719944021058060289', + 'ext': 'mp4', + 'title': 'Captain America - @King0fNerd Are you sure you made the right choice? Find out in theaters.', + 'description': 'Captain America on Twitter: "@King0fNerd Are you sure you made the right choice? Find out in theaters. https://t.co/GpgYi9xMJI"', + 'uploader_id': 'captainamerica', + 'uploader': 'Captain America', + }, }] def _real_extract(self, url): @@ -258,17 +295,6 @@ class TwitterIE(InfoExtractor): 'title': username + ' - ' + title, } - card_id = self._search_regex( - r'["\']/i/cards/tfw/v1/(\d+)', webpage, 'twitter card url', default=None) - if card_id: - card_url = 'https://twitter.com/i/cards/tfw/v1/' + card_id - info.update({ - '_type': 'url_transparent', - 'ie_key': 'TwitterCard', - 'url': card_url, - }) - return info - mobj = re.search(r'''(?x) ]+class="animated-gif"(?P[^>]+)>\s* ]+video-src="(?P[^"]+)"