From 67167920db50e818c9fca20579c8a05eb2218f86 Mon Sep 17 00:00:00 2001 From: remitamine Date: Fri, 29 Apr 2016 11:14:42 +0100 Subject: [PATCH] [viewlift] replace SnagFilms extractors - add support for other sites that use the same logic - improve format extraction and sorting --- youtube_dl/extractor/extractors.py | 8 ++--- youtube_dl/extractor/generic.py | 10 +++--- .../extractor/{snagfilms.py => viewlift.py} | 35 +++++++++++++------ 3 files changed, 34 insertions(+), 19 deletions(-) rename youtube_dl/extractor/{snagfilms.py => viewlift.py} (81%) diff --git a/youtube_dl/extractor/extractors.py b/youtube_dl/extractor/extractors.py index 3adcd41c4b..b1b7f9b428 100644 --- a/youtube_dl/extractor/extractors.py +++ b/youtube_dl/extractor/extractors.py @@ -673,10 +673,6 @@ SmotriUserIE, SmotriBroadcastIE, ) -from .snagfilms import ( - SnagFilmsIE, - SnagFilmsEmbedIE, -) from .snotr import SnotrIE from .sohu import SohuIE from .soundcloud import ( @@ -879,6 +875,10 @@ ) from .vidzi import VidziIE from .vier import VierIE, VierVideosIE +from .viewlift import ( + ViewLiftIE, + ViewLiftEmbedIE, +) from .viewster import ViewsterIE from .viidea import ViideaIE from .vimeo import ( diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py index a95501d86b..0f1eb7fa64 100644 --- a/youtube_dl/extractor/generic.py +++ b/youtube_dl/extractor/generic.py @@ -51,7 +51,7 @@ from .vimeo import VimeoIE from .dailymotion import DailymotionCloudIE from .onionstudios import OnionStudiosIE -from .snagfilms import SnagFilmsEmbedIE +from .viewlift import ViewLiftEmbedIE from .screenwavemedia import ScreenwaveMediaIE from .mtv import MTVServicesEmbeddedIE from .pladform import PladformIE @@ -1924,10 +1924,10 @@ def _playlist_from_matches(matches, getter=None, ie=None): if onionstudios_url: return self.url_result(onionstudios_url) - # Look for SnagFilms embeds - snagfilms_url = SnagFilmsEmbedIE._extract_url(webpage) - if snagfilms_url: - return self.url_result(snagfilms_url) + # Look for ViewLift embeds + viewlift_url = ViewLiftEmbedIE._extract_url(webpage) + if viewlift_url: + return self.url_result(viewlift_url) # Look for JWPlatform embeds jwplatform_url = JWPlatformIE._extract_url(webpage) diff --git a/youtube_dl/extractor/snagfilms.py b/youtube_dl/extractor/viewlift.py similarity index 81% rename from youtube_dl/extractor/snagfilms.py rename to youtube_dl/extractor/viewlift.py index 6977afb278..dd4a13a4a8 100644 --- a/youtube_dl/extractor/snagfilms.py +++ b/youtube_dl/extractor/viewlift.py @@ -13,8 +13,12 @@ ) -class SnagFilmsEmbedIE(InfoExtractor): - _VALID_URL = r'https?://(?:(?:www|embed)\.)?snagfilms\.com/embed/player\?.*\bfilmId=(?P[\da-f-]{36})' +class ViewLiftBaseIE(InfoExtractor): + _DOMAINS_REGEX = '(?:snagfilms|snagxtreme|funnyforfree|kiddovid|winnersview|monumentalsportsnetwork|vayafilm)\.com|kesari\.tv' + + +class ViewLiftEmbedIE(ViewLiftBaseIE): + _VALID_URL = r'https?://(?:(?:www|embed)\.)?(?:%s)/embed/player\?.*\bfilmId=(?P[\da-f-]{36})' % ViewLiftBaseIE._DOMAINS_REGEX _TESTS = [{ 'url': 'http://embed.snagfilms.com/embed/player?filmId=74849a00-85a9-11e1-9660-123139220831&w=500', 'md5': '2924e9215c6eff7a55ed35b72276bd93', @@ -40,7 +44,7 @@ class SnagFilmsEmbedIE(InfoExtractor): @staticmethod def _extract_url(webpage): mobj = re.search( - r']+?src=(["\'])(?P(?:https?:)?//(?:embed\.)?snagfilms\.com/embed/player.+?)\1', + r']+?src=(["\'])(?P(?:https?:)?//(?:embed\.)?(?:%s)/embed/player.+?)\1' % ViewLiftBaseIE._DOMAINS_REGEX, webpage) if mobj: return mobj.group('url') @@ -55,6 +59,7 @@ def _real_extract(self, url): 'Film %s is not playable in your area.' % video_id, expected=True) formats = [] + has_bitrate = False for source in self._parse_json(js_to_json(self._search_regex( r'(?s)sources:\s*(\[.+?\]),', webpage, 'json')), video_id): file_ = source.get('file') @@ -63,22 +68,25 @@ def _real_extract(self, url): type_ = source.get('type') ext = determine_ext(file_) format_id = source.get('label') or ext - if all(v == 'm3u8' for v in (type_, ext)): + if all(v == 'm3u8' or v == 'hls' for v in (type_, ext)): formats.extend(self._extract_m3u8_formats( file_, video_id, 'mp4', m3u8_id='hls')) else: bitrate = int_or_none(self._search_regex( [r'(\d+)kbps', r'_\d{1,2}x\d{1,2}_(\d{3,})\.%s' % ext], file_, 'bitrate', default=None)) + if not has_bitrate and bitrate: + has_bitrate = True height = int_or_none(self._search_regex( r'^(\d+)[pP]$', format_id, 'height', default=None)) formats.append({ 'url': file_, - 'format_id': format_id, + 'format_id': 'http-%s%s' % (format_id, ('-%dk' % bitrate if bitrate else '')), 'tbr': bitrate, 'height': height, }) - self._sort_formats(formats) + field_preference = None if has_bitrate else ('height', 'tbr', 'format_id') + self._sort_formats(formats, field_preference) title = self._search_regex( [r"title\s*:\s*'([^']+)'", r'([^<]+)'], @@ -91,8 +99,8 @@ def _real_extract(self, url): } -class SnagFilmsIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?snagfilms\.com/(?:films/title|show)/(?P[^?#]+)' +class ViewLiftIE(ViewLiftBaseIE): + _VALID_URL = r'https?://(?:www\.)?(?P%s)/(?:films/title|show|(?:news/)?videos?)/(?P[^?#]+)' % ViewLiftBaseIE._DOMAINS_REGEX _TESTS = [{ 'url': 'http://www.snagfilms.com/films/title/lost_for_life', 'md5': '19844f897b35af219773fd63bdec2942', @@ -127,10 +135,16 @@ class SnagFilmsIE(InfoExtractor): # Film is not available. 'url': 'http://www.snagfilms.com/show/augie_alone/flirting', 'only_matching': True, + }, { + 'url': 'http://www.winnersview.com/videos/the-good-son', + 'only_matching': True, + }, { + 'url': 'http://www.kesari.tv/news/video/1461919076414', + 'only_matching': True, }] def _real_extract(self, url): - display_id = self._match_id(url) + domain, display_id = re.match(self._VALID_URL, url).groups() webpage = self._download_webpage(url, display_id) @@ -170,7 +184,7 @@ def _real_extract(self, url): return { '_type': 'url_transparent', - 'url': 'http://embed.snagfilms.com/embed/player?filmId=%s' % film_id, + 'url': 'http://%s/embed/player?filmId=%s' % (domain, film_id), 'id': film_id, 'display_id': display_id, 'title': title, @@ -178,4 +192,5 @@ def _real_extract(self, url): 'thumbnail': thumbnail, 'duration': duration, 'categories': categories, + 'ie_key': 'ViewLiftEmbed', }