X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/33cd347759d6d999325ebf3c69b7ed5692c343b2..1f17a37b9b95db09a420a1f52cf18723ce4eb8b5:/youtube_dl/extractor/breakcom.py diff --git a/youtube_dl/extractor/breakcom.py b/youtube_dl/extractor/breakcom.py index 53a898d..5a87c26 100644 --- a/youtube_dl/extractor/breakcom.py +++ b/youtube_dl/extractor/breakcom.py @@ -1,38 +1,143 @@ +from __future__ import unicode_literals + import re -import json from .common import InfoExtractor -from ..utils import determine_ext +from ..compat import compat_str +from ..utils import ( + int_or_none, + parse_age_limit, +) class BreakIE(InfoExtractor): - _VALID_URL = r'(?:http://)?(?:www\.)?break\.com/video/([^/]+)' - _TEST = { - u'url': u'http://www.break.com/video/when-girls-act-like-guys-2468056', - u'file': u'2468056.mp4', - u'md5': u'a3513fb1547fba4fb6cfac1bffc6c46b', - u'info_dict': { - u"title": u"When Girls Act Like D-Bags" + _VALID_URL = r'https?://(?:www\.)?(?Pbreak|screenjunkies)\.com/video/(?P[^/]+?)(?:-(?P\d+))?(?:[/?#&]|$)' + _TESTS = [{ + 'url': 'http://www.break.com/video/when-girls-act-like-guys-2468056', + 'info_dict': { + 'id': '2468056', + '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', + 'info_dict': { + 'id': '3003285', + 'display_id': 'knocking-dead-ep-1-the-show-so-far', + 'ext': 'mp4', + 'title': 'State of The Dead Recap: Knocking Dead Pilot', + 'thumbnail': r're:^https?://.*\.jpg', + 'duration': 3307, + 'age_limit': 13, + 'tags': list, + }, + }, { + '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): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group(1).split("-")[-1] - embed_url = 'http://www.break.com/embed/%s' % video_id - webpage = self._download_webpage(embed_url, video_id) - info_json = self._search_regex(r'var embedVars = ({.*?});', webpage, - u'info json', flags=re.DOTALL) - info = json.loads(info_json) - video_url = info['videoUri'] - m_youtube = re.search(r'(https?://www\.youtube\.com/watch\?v=.*)', video_url) - if m_youtube is not None: - return self.url_result(m_youtube.group(1), 'Youtube') - final_url = video_url + '?' + info['AuthToken'] - return [{ - 'id': video_id, - 'url': final_url, - 'ext': determine_ext(final_url), - 'title': info['contentName'], - 'thumbnail': info['thumbUri'], - }] + site, 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( + 'http://www.%s.com/embed/%s' % (site, video_id), + display_id, 'Downloading video embed page') + embed_vars = self._parse_json( + self._search_regex( + r'(?s)embedVars\s*=\s*({.+?})\s*', webpage, 'embed vars'), + 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': + continue + bitrate = int_or_none(f.get('bitRate')) + if bitrate: + bitrates.append(bitrate) + formats.append({ + 'url': f['uri'], + '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', + }) + + 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') + + 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) + + 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) + + 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(','), + 'formats': formats, + }