From 2441c1aab152cd81d53f0a6fca982af9f8c8de10 Mon Sep 17 00:00:00 2001 From: =?utf8?q?Sergey=20M=E2=80=A4?= Date: Mon, 23 Apr 2018 00:16:52 +0700 Subject: [PATCH] [breakcom] Fix extraction (closes #16254) --- youtube_dl/extractor/breakcom.py | 136 ++++++++++--------------------- 1 file changed, 41 insertions(+), 95 deletions(-) diff --git a/youtube_dl/extractor/breakcom.py b/youtube_dl/extractor/breakcom.py index 5a87c2661..70d16767f 100644 --- a/youtube_dl/extractor/breakcom.py +++ b/youtube_dl/extractor/breakcom.py @@ -3,15 +3,13 @@ from __future__ import unicode_literals import re from .common import InfoExtractor +from .youtube import YoutubeIE from ..compat import compat_str -from ..utils import ( - int_or_none, - parse_age_limit, -) +from ..utils import int_or_none class BreakIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?(?Pbreak|screenjunkies)\.com/video/(?P[^/]+?)(?:-(?P\d+))?(?:[/?#&]|$)' + _VALID_URL = r'https?://(?:www\.)?break\.com/video/(?P[^/]+?)(?:-(?P\d+))?(?:[/?#&]|$)' _TESTS = [{ 'url': 'http://www.break.com/video/when-girls-act-like-guys-2468056', 'info_dict': { @@ -19,125 +17,73 @@ class BreakIE(InfoExtractor): 'ext': 'mp4', 'title': 'When Girls Act Like D-Bags', 'age_limit': 13, - } - }, { - 'url': 'http://www.screenjunkies.com/video/best-quentin-tarantino-movie-2841915', - 'md5': '5c2b686bec3d43de42bde9ec047536b0', - 'info_dict': { - 'id': '2841915', - 'display_id': 'best-quentin-tarantino-movie', - 'ext': 'mp4', - 'title': 'Best Quentin Tarantino Movie', - 'thumbnail': r're:^https?://.*\.jpg', - 'duration': 3671, - 'age_limit': 13, - 'tags': list, - }, - }, { - 'url': 'http://www.screenjunkies.com/video/honest-trailers-the-dark-knight', - 'info_dict': { - 'id': '2348808', - 'display_id': 'honest-trailers-the-dark-knight', - 'ext': 'mp4', - 'title': 'Honest Trailers - The Dark Knight', - 'thumbnail': r're:^https?://.*\.(?:jpg|png)', - 'age_limit': 10, - 'tags': list, }, }, { - # requires subscription but worked around - 'url': 'http://www.screenjunkies.com/video/knocking-dead-ep-1-the-show-so-far-3003285', + # youtube embed + 'url': 'http://www.break.com/video/someone-forgot-boat-brakes-work', 'info_dict': { - 'id': '3003285', - 'display_id': 'knocking-dead-ep-1-the-show-so-far', + 'id': 'RrrDLdeL2HQ', 'ext': 'mp4', - 'title': 'State of The Dead Recap: Knocking Dead Pilot', - 'thumbnail': r're:^https?://.*\.jpg', - 'duration': 3307, - 'age_limit': 13, - 'tags': list, + 'title': 'Whale Watching Boat Crashing Into San Diego Dock', + 'description': 'md5:afc1b2772f0a8468be51dd80eb021069', + 'upload_date': '20160331', + 'uploader': 'Steve Holden', + 'uploader_id': 'sdholden07', }, + 'params': { + 'skip_download': True, + } }, { 'url': 'http://www.break.com/video/ugc/baby-flex-2773063', 'only_matching': True, }] - _DEFAULT_BITRATES = (48, 150, 320, 496, 864, 2240, 3264) - def _real_extract(self, url): - site, display_id, video_id = re.match(self._VALID_URL, url).groups() + display_id, video_id = re.match(self._VALID_URL, url).groups() - if not video_id: - webpage = self._download_webpage(url, display_id) - video_id = self._search_regex( - (r'src=["\']/embed/(\d+)', r'data-video-content-id=["\'](\d+)'), - webpage, 'video id') + webpage = self._download_webpage(url, display_id) - webpage = self._download_webpage( - 'http://www.%s.com/embed/%s' % (site, video_id), - display_id, 'Downloading video embed page') - embed_vars = self._parse_json( + youtube_url = YoutubeIE._extract_url(webpage) + if youtube_url: + return self.url_result(youtube_url, ie=YoutubeIE.ie_key()) + + content = self._parse_json( self._search_regex( - r'(?s)embedVars\s*=\s*({.+?})\s*', webpage, 'embed vars'), + r'(?s)content["\']\s*:\s*(\[.+?\])\s*[,\n]', webpage, + 'content'), display_id) - youtube_id = embed_vars.get('youtubeId') - if youtube_id: - return self.url_result(youtube_id, 'Youtube') - - title = embed_vars['contentName'] - formats = [] - bitrates = [] - for f in embed_vars.get('media', []): - if not f.get('uri') or f.get('mediaPurpose') != 'play': + for video in content: + video_url = video.get('url') + if not video_url or not isinstance(video_url, compat_str): continue - bitrate = int_or_none(f.get('bitRate')) - if bitrate: - bitrates.append(bitrate) + bitrate = int_or_none(self._search_regex( + r'(\d+)_kbps', video_url, 'tbr', default=None)) formats.append({ - 'url': f['uri'], + 'url': video_url, 'format_id': 'http-%d' % bitrate if bitrate else 'http', - 'width': int_or_none(f.get('width')), - 'height': int_or_none(f.get('height')), 'tbr': bitrate, - 'format': 'mp4', }) + self._sort_formats(formats) - if not bitrates: - # When subscriptionLevel > 0, i.e. plus subscription is required - # media list will be empty. However, hds and hls uris are still - # available. We can grab them assuming bitrates to be default. - bitrates = self._DEFAULT_BITRATES - - auth_token = embed_vars.get('AuthToken') + title = self._search_regex( + (r'title["\']\s*:\s*(["\'])(?P(?:(?!\1).)+)\1', + r']*>(?P[^<]+)'), webpage, 'title', group='value') - def construct_manifest_url(base_url, ext): - pieces = [base_url] - pieces.extend([compat_str(b) for b in bitrates]) - pieces.append('_kbps.mp4.%s?%s' % (ext, auth_token)) - return ','.join(pieces) + def get(key, name): + return int_or_none(self._search_regex( + r'%s["\']\s*:\s*["\'](\d+)' % key, webpage, name, + default=None)) - if bitrates and auth_token: - hds_url = embed_vars.get('hdsUri') - if hds_url: - formats.extend(self._extract_f4m_formats( - construct_manifest_url(hds_url, 'f4m'), - display_id, f4m_id='hds', fatal=False)) - hls_url = embed_vars.get('hlsUri') - if hls_url: - formats.extend(self._extract_m3u8_formats( - construct_manifest_url(hls_url, 'm3u8'), - display_id, 'mp4', entry_protocol='m3u8_native', m3u8_id='hls', fatal=False)) - self._sort_formats(formats) + age_limit = get('ratings', 'age limit') + video_id = video_id or get('pid', 'video id') or display_id return { 'id': video_id, 'display_id': display_id, 'title': title, - 'thumbnail': embed_vars.get('thumbUri'), - 'duration': int_or_none(embed_vars.get('videoLengthInSeconds')) or None, - 'age_limit': parse_age_limit(embed_vars.get('audienceRating')), - 'tags': embed_vars.get('tags', '').split(','), + 'thumbnail': self._og_search_thumbnail(webpage), + 'age_limit': age_limit, 'formats': formats, } -- 2.22.2