X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/9dc487f48b50767cf540fa36c3de2c386fd74c04..a5a28b1eabe0533f26d23c4157c97144215b22b7:/youtube_dl/extractor/jwplatform.py?ds=sidebyside diff --git a/youtube_dl/extractor/jwplatform.py b/youtube_dl/extractor/jwplatform.py index e44e311..63d0dc9 100644 --- a/youtube_dl/extractor/jwplatform.py +++ b/youtube_dl/extractor/jwplatform.py @@ -4,101 +4,9 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..utils import ( - determine_ext, - float_or_none, - int_or_none, -) -class JWPlatformBaseIE(InfoExtractor): - @staticmethod - def _find_jwplayer_data(webpage): - # TODO: Merge this with JWPlayer-related codes in generic.py - - mobj = re.search( - 'jwplayer\((?P[\'"])[^\'" ]+(?P=quote)\)\.setup\((?P[^)]+)\)', - webpage) - if mobj: - return mobj.group('options') - - def _extract_jwplayer_data(self, webpage, video_id, *args, **kwargs): - jwplayer_data = self._parse_json( - self._find_jwplayer_data(webpage), video_id) - return self._parse_jwplayer_data( - jwplayer_data, video_id, *args, **kwargs) - - def _parse_jwplayer_data(self, jwplayer_data, video_id, require_title=True, m3u8_id=None, rtmp_params=None): - # JWPlayer backward compatibility: flattened playlists - # https://github.com/jwplayer/jwplayer/blob/v7.4.3/src/js/api/config.js#L81-L96 - if 'playlist' not in jwplayer_data: - jwplayer_data = {'playlist': [jwplayer_data]} - - video_data = jwplayer_data['playlist'][0] - - # JWPlayer backward compatibility: flattened sources - # https://github.com/jwplayer/jwplayer/blob/v7.4.3/src/js/playlist/item.js#L29-L35 - if 'sources' not in video_data: - video_data['sources'] = [video_data] - - formats = [] - for source in video_data['sources']: - source_url = self._proto_relative_url(source['file']) - source_type = source.get('type') or '' - if source_type in ('application/vnd.apple.mpegurl', 'hls') or determine_ext(source_url) == 'm3u8': - formats.extend(self._extract_m3u8_formats( - source_url, video_id, 'mp4', 'm3u8_native', m3u8_id=m3u8_id, fatal=False)) - elif source_type.startswith('audio'): - formats.append({ - 'url': source_url, - 'vcodec': 'none', - }) - else: - a_format = { - 'url': source_url, - 'width': int_or_none(source.get('width')), - 'height': int_or_none(source.get('height')), - } - if source_url.startswith('rtmp'): - a_format['ext'] = 'flv', - - # See com/longtailvideo/jwplayer/media/RTMPMediaProvider.as - # of jwplayer.flash.swf - rtmp_url_parts = re.split( - r'((?:mp4|mp3|flv):)', source_url, 1) - if len(rtmp_url_parts) == 3: - rtmp_url, prefix, play_path = rtmp_url_parts - a_format.update({ - 'url': rtmp_url, - 'play_path': prefix + play_path, - }) - if rtmp_params: - a_format.update(rtmp_params) - formats.append(a_format) - self._sort_formats(formats) - - subtitles = {} - tracks = video_data.get('tracks') - if tracks and isinstance(tracks, list): - for track in tracks: - if track.get('file') and track.get('kind') == 'captions': - subtitles.setdefault(track.get('label') or 'en', []).append({ - 'url': self._proto_relative_url(track['file']) - }) - - return { - 'id': video_id, - 'title': video_data['title'] if require_title else video_data.get('title'), - 'description': video_data.get('description'), - 'thumbnail': self._proto_relative_url(video_data.get('image')), - 'timestamp': int_or_none(video_data.get('pubdate')), - 'duration': float_or_none(jwplayer_data.get('duration')), - 'subtitles': subtitles, - 'formats': formats, - } - - -class JWPlatformIE(JWPlatformBaseIE): +class JWPlatformIE(InfoExtractor): _VALID_URL = r'(?:https?://content\.jwplatform\.com/(?:feeds|players|jw6)/|jwplatform:)(?P[a-zA-Z0-9]{8})' _TEST = { 'url': 'http://content.jwplatform.com/players/nPripu9l-ALJ3XQCI.js', @@ -115,11 +23,14 @@ class JWPlatformIE(JWPlatformBaseIE): @staticmethod def _extract_url(webpage): - mobj = re.search( - r']+?src=["\'](?P(?:https?:)?//content.jwplatform.com/players/[a-zA-Z0-9]{8})', + urls = JWPlatformIE._extract_urls(webpage) + return urls[0] if urls else None + + @staticmethod + def _extract_urls(webpage): + return re.findall( + r'<(?:script|iframe)[^>]+?src=["\']((?:https?:)?//content\.jwplatform\.com/players/[a-zA-Z0-9]{8})', webpage) - if mobj: - return mobj.group('url') def _real_extract(self, url): video_id = self._match_id(url)