diff options
Diffstat (limited to 'youtube_dl/extractor/twitter.py')
-rw-r--r-- | youtube_dl/extractor/twitter.py | 97 |
1 files changed, 70 insertions, 27 deletions
diff --git a/youtube_dl/extractor/twitter.py b/youtube_dl/extractor/twitter.py index 1f32ea2eb..ac0b221b4 100644 --- a/youtube_dl/extractor/twitter.py +++ b/youtube_dl/extractor/twitter.py @@ -4,7 +4,9 @@ from __future__ import unicode_literals import re from .common import InfoExtractor +from ..compat import compat_urlparse from ..utils import ( + determine_ext, float_or_none, xpath_text, remove_end, @@ -12,6 +14,8 @@ from ..utils import ( ExtractorError, ) +from .periscope import PeriscopeIE + class TwitterBaseIE(InfoExtractor): def _get_vmap_video_url(self, vmap_url, video_id): @@ -21,7 +25,7 @@ class TwitterBaseIE(InfoExtractor): class TwitterCardIE(TwitterBaseIE): IE_NAME = 'twitter:card' - _VALID_URL = r'https?://(?:www\.)?twitter\.com/i/(?:cards/tfw/v1|videos/tweet)/(?P<id>\d+)' + _VALID_URL = r'https?://(?:www\.)?twitter\.com/i/(?:cards/tfw/v1|videos(?:/tweet)?)/(?P<id>\d+)' _TESTS = [ { 'url': 'https://twitter.com/i/cards/tfw/v1/560070183650213889', @@ -47,12 +51,12 @@ class TwitterCardIE(TwitterBaseIE): }, { 'url': 'https://twitter.com/i/cards/tfw/v1/654001591733886977', - 'md5': 'd4724ffe6d2437886d004fa5de1043b3', + 'md5': 'b6d9683dd3f48e340ded81c0e917ad46', 'info_dict': { 'id': 'dq4Oj5quskI', 'ext': 'mp4', 'title': 'Ubuntu 11.10 Overview', - 'description': 'Take a quick peek at what\'s new and improved in Ubuntu 11.10.\n\nOnce installed take a look at 10 Things to Do After Installing: http://www.omgubuntu.co.uk/2011/10/10-things-to-do-after-installing-ubuntu-11-10/', + 'description': 'md5:a831e97fa384863d6e26ce48d1c43376', 'upload_date': '20111013', 'uploader': 'OMG! Ubuntu!', 'uploader_id': 'omgubuntu', @@ -80,6 +84,9 @@ class TwitterCardIE(TwitterBaseIE): 'title': 'Twitter web player', 'thumbnail': 're:^https?://.*\.jpg', }, + }, { + 'url': 'https://twitter.com/i/videos/752274308186120192', + 'only_matching': True, }, ] @@ -99,12 +106,17 @@ class TwitterCardIE(TwitterBaseIE): return self.url_result(iframe_url) config = self._parse_json(self._html_search_regex( - r'data-(?:player-)?config="([^"]+)"', webpage, 'data player config'), + r'data-(?:player-)?config="([^"]+)"', webpage, + 'data player config', default='{}'), video_id) if config.get('source_type') == 'vine': return self.url_result(config['player_url'], 'Vine') + periscope_url = PeriscopeIE._extract_url(webpage) + if periscope_url: + return self.url_result(periscope_url, PeriscopeIE.ie_key()) + def _search_dimensions_in_video_url(a_format, video_url): m = re.search(r'/(?P<width>\d+)x(?P<height>\d+)/', video_url) if m: @@ -116,13 +128,16 @@ class TwitterCardIE(TwitterBaseIE): video_url = config.get('video_url') or config.get('playlist', [{}])[0].get('source') if video_url: - f = { - 'url': video_url, - } + if determine_ext(video_url) == 'm3u8': + formats.extend(self._extract_m3u8_formats(video_url, video_id, ext='mp4', m3u8_id='hls')) + else: + f = { + 'url': video_url, + } - _search_dimensions_in_video_url(f, video_url) + _search_dimensions_in_video_url(f, video_url) - formats.append(f) + formats.append(f) vmap_url = config.get('vmapUrl') or config.get('vmap_url') if vmap_url: @@ -207,6 +222,7 @@ class TwitterIE(InfoExtractor): 'uploader_id': 'giphz', }, 'expected_warnings': ['height', 'width'], + 'skip': 'Account suspended', }, { 'url': 'https://twitter.com/starwars/status/665052190608723968', 'md5': '39b7199856dee6cd4432e72c74bc69d4', @@ -239,10 +255,10 @@ class TwitterIE(InfoExtractor): '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"', + 'title': 'JG - BEAT PROD: @suhmeduh #Damndaniel', + 'description': 'JG on Twitter: "BEAT PROD: @suhmeduh https://t.co/HBrQ4AfpvZ #Damndaniel https://t.co/byBooq2ejZ"', 'thumbnail': 're:^https?://.*\.jpg', - 'uploader': 'jay', + 'uploader': 'JG', 'uploader_id': 'jaydingeer', }, 'params': { @@ -260,6 +276,31 @@ class TwitterIE(InfoExtractor): 'upload_date': '20140615', }, 'add_ie': ['Vine'], + }, { + 'url': 'https://twitter.com/captainamerica/status/719944021058060289', + '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', + }, + 'params': { + 'skip_download': True, # requires ffmpeg + }, + }, { + 'url': 'https://twitter.com/OPP_HSD/status/779210622571536384', + 'info_dict': { + 'id': '1zqKVVlkqLaKB', + 'ext': 'mp4', + 'title': 'Sgt Kerry Schmidt - Ontario Provincial Police - Road rage, mischief, assault, rollover and fire in one occurrence', + 'upload_date': '20160923', + 'uploader_id': 'OPP_HSD', + 'uploader': 'Sgt Kerry Schmidt - Ontario Provincial Police', + 'timestamp': 1474613214, + }, + 'add_ie': ['Periscope'], }] def _real_extract(self, url): @@ -267,7 +308,11 @@ class TwitterIE(InfoExtractor): user_id = mobj.group('user_id') twid = mobj.group('id') - webpage = self._download_webpage(self._TEMPLATE_URL % (user_id, twid), twid) + webpage, urlh = self._download_webpage_handle( + self._TEMPLATE_URL % (user_id, twid), twid) + + if 'twitter.com/account/suspended' in urlh.geturl(): + raise ExtractorError('Account suspended by Twitter.', expected=True) username = remove_end(self._og_search_title(webpage), ' on Twitter') @@ -284,17 +329,6 @@ 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) <video[^>]+class="animated-gif"(?P<more_info>[^>]+)>\s* <source[^>]+video-src="(?P<url>[^"]+)" @@ -317,13 +351,22 @@ class TwitterIE(InfoExtractor): }) return info + twitter_card_url = None if 'class="PlayableMedia' in webpage: + twitter_card_url = '%s//twitter.com/i/videos/tweet/%s' % (self.http_scheme(), twid) + else: + twitter_card_iframe_url = self._search_regex( + r'data-full-card-iframe-url=([\'"])(?P<url>(?:(?!\1).)+)\1', + webpage, 'Twitter card iframe URL', default=None, group='url') + if twitter_card_iframe_url: + twitter_card_url = compat_urlparse.urljoin(url, twitter_card_iframe_url) + + if twitter_card_url: info.update({ '_type': 'url_transparent', 'ie_key': 'TwitterCard', - 'url': '%s//twitter.com/i/videos/tweet/%s' % (self.http_scheme(), twid), + 'url': twitter_card_url, }) - return info raise ExtractorError('There\'s no video in this tweet.') @@ -331,7 +374,7 @@ class TwitterIE(InfoExtractor): class TwitterAmplifyIE(TwitterBaseIE): IE_NAME = 'twitter:amplify' - _VALID_URL = 'https?://amp\.twimg\.com/v/(?P<id>[0-9a-f\-]{36})' + _VALID_URL = r'https?://amp\.twimg\.com/v/(?P<id>[0-9a-f\-]{36})' _TEST = { 'url': 'https://amp.twimg.com/v/0ba0c3c7-0af3-4c0a-bed5-7efd1ffa2951', |