From 221ce325296657ebb50719f64914ab63a3238d4b Mon Sep 17 00:00:00 2001 From: Remita Amine Date: Fri, 6 Jan 2017 11:25:48 +0100 Subject: [PATCH] [break] merge BreakIE and ScreenJunkiesIE --- youtube_dl/extractor/breakcom.py | 125 ++++++++++++++++++----- youtube_dl/extractor/extractors.py | 1 - youtube_dl/extractor/screenjunkies.py | 138 -------------------------- 3 files changed, 102 insertions(+), 162 deletions(-) delete mode 100644 youtube_dl/extractor/screenjunkies.py diff --git a/youtube_dl/extractor/breakcom.py b/youtube_dl/extractor/breakcom.py index 725859b4d..5a87c2661 100644 --- a/youtube_dl/extractor/breakcom.py +++ b/youtube_dl/extractor/breakcom.py @@ -1,9 +1,9 @@ from __future__ import unicode_literals import re -import json from .common import InfoExtractor +from ..compat import compat_str from ..utils import ( int_or_none, parse_age_limit, @@ -11,7 +11,7 @@ from ..utils import ( class BreakIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?break\.com/video/(?:[^/]+/)*.+-(?P\d+)' + _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': { @@ -20,45 +20,124 @@ class BreakIE(InfoExtractor): '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): - video_id = self._match_id(url) + 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.break.com/embed/%s' % video_id, video_id) - info = json.loads(self._search_regex( - r'var embedVars = ({.*})\s*?', - webpage, 'info json', flags=re.DOTALL)) + '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 = info.get('youtubeId') + youtube_id = embed_vars.get('youtubeId') if youtube_id: return self.url_result(youtube_id, 'Youtube') - formats = [{ - 'url': media['uri'] + '?' + info['AuthToken'], - 'tbr': media['bitRate'], - 'width': media['width'], - 'height': media['height'], - } for media in info['media'] if media.get('mediaPurpose') == 'play'] + title = embed_vars['contentName'] - if not formats: + 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': info['videoUri'] + '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', }) - 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') - duration = int_or_none(info.get('videoLengthInSeconds')) - age_limit = parse_age_limit(info.get('audienceRating')) + 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, - 'title': info['contentName'], - 'thumbnail': info['thumbUri'], - 'duration': duration, - 'age_limit': age_limit, + '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, } diff --git a/youtube_dl/extractor/extractors.py b/youtube_dl/extractor/extractors.py index 3017bf56c..923fb6f60 100644 --- a/youtube_dl/extractor/extractors.py +++ b/youtube_dl/extractor/extractors.py @@ -813,7 +813,6 @@ from .sbs import SBSIE from .scivee import SciVeeIE from .screencast import ScreencastIE from .screencastomatic import ScreencastOMaticIE -from .screenjunkies import ScreenJunkiesIE from .seeker import SeekerIE from .senateisvp import SenateISVPIE from .sendtonews import SendtoNewsIE diff --git a/youtube_dl/extractor/screenjunkies.py b/youtube_dl/extractor/screenjunkies.py deleted file mode 100644 index 50aea65cb..000000000 --- a/youtube_dl/extractor/screenjunkies.py +++ /dev/null @@ -1,138 +0,0 @@ -from __future__ import unicode_literals - -import re - -from .common import InfoExtractor -from ..compat import compat_str -from ..utils import ( - int_or_none, - parse_age_limit, -) - - -class ScreenJunkiesIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?screenjunkies\.com/video/(?P[^/]+?)(?:-(?P\d+))?(?:[/?#&]|$)' - _TESTS = [{ - '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', - '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': 'Knocking Dead Ep 1: State of The Dead Recap', - 'thumbnail': r're:^https?://.*\.jpg', - 'duration': 3307, - 'age_limit': 13, - 'tags': list, - }, - }] - - _DEFAULT_BITRATES = (48, 150, 496, 864, 2240) - - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') - display_id = mobj.group('display_id') - - 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.screenjunkies.com/embed/%s' % 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) - - 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: - f4m_formats = self._extract_f4m_formats( - construct_manifest_url(hds_url, 'f4m'), - display_id, f4m_id='hds', fatal=False) - if len(f4m_formats) == len(bitrates): - for f, bitrate in zip(f4m_formats, bitrates): - if not f.get('tbr'): - f['format_id'] = 'hds-%d' % bitrate - f['tbr'] = bitrate - # TODO: fix f4m downloader to handle manifests without bitrates if possible - # formats.extend(f4m_formats) - - 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, - } -- 2.40.0