2 from __future__ import unicode_literals
6 from .common import InfoExtractor
7 from ..compat import compat_urlparse
17 class JWPlatformBaseIE(InfoExtractor):
19 def _find_jwplayer_data(webpage):
20 # TODO: Merge this with JWPlayer-related codes in generic.py
23 r'jwplayer\((?P<quote>[\'"])[^\'" ]+(?P=quote)\)\.setup\s*\((?P<options>[^)]+)\)',
26 return mobj.group('options')
28 def _extract_jwplayer_data(self, webpage, video_id, *args, **kwargs):
29 jwplayer_data = self._parse_json(
30 self._find_jwplayer_data(webpage), video_id,
31 transform_source=js_to_json)
32 return self._parse_jwplayer_data(
33 jwplayer_data, video_id, *args, **kwargs)
35 def _parse_jwplayer_data(self, jwplayer_data, video_id=None, require_title=True, m3u8_id=None, rtmp_params=None, base_url=None):
36 # JWPlayer backward compatibility: flattened playlists
37 # https://github.com/jwplayer/jwplayer/blob/v7.4.3/src/js/api/config.js#L81-L96
38 if 'playlist' not in jwplayer_data:
39 jwplayer_data = {'playlist': [jwplayer_data]}
43 # JWPlayer backward compatibility: single playlist item
44 # https://github.com/jwplayer/jwplayer/blob/v7.7.0/src/js/playlist/playlist.js#L10
45 if not isinstance(jwplayer_data['playlist'], list):
46 jwplayer_data['playlist'] = [jwplayer_data['playlist']]
48 for video_data in jwplayer_data['playlist']:
49 # JWPlayer backward compatibility: flattened sources
50 # https://github.com/jwplayer/jwplayer/blob/v7.4.3/src/js/playlist/item.js#L29-L35
51 if 'sources' not in video_data:
52 video_data['sources'] = [video_data]
54 this_video_id = video_id or video_data['mediaid']
57 for source in video_data['sources']:
58 source_url = self._proto_relative_url(source['file'])
60 source_url = compat_urlparse.urljoin(base_url, source_url)
61 source_type = source.get('type') or ''
62 ext = mimetype2ext(source_type) or determine_ext(source_url)
63 if source_type == 'hls' or ext == 'm3u8':
64 formats.extend(self._extract_m3u8_formats(
65 source_url, this_video_id, 'mp4', 'm3u8_native', m3u8_id=m3u8_id, fatal=False))
66 # https://github.com/jwplayer/jwplayer/blob/master/src/js/providers/default.js#L67
67 elif source_type.startswith('audio') or ext in ('oga', 'aac', 'mp3', 'mpeg', 'vorbis'):
74 height = int_or_none(source.get('height'))
76 # Often no height is provided but there is a label in
78 height = int_or_none(self._search_regex(
79 r'^(\d{3,})[pP]$', source.get('label') or '',
80 'height', default=None))
83 'width': int_or_none(source.get('width')),
87 if source_url.startswith('rtmp'):
88 a_format['ext'] = 'flv'
90 # See com/longtailvideo/jwplayer/media/RTMPMediaProvider.as
91 # of jwplayer.flash.swf
92 rtmp_url_parts = re.split(
93 r'((?:mp4|mp3|flv):)', source_url, 1)
94 if len(rtmp_url_parts) == 3:
95 rtmp_url, prefix, play_path = rtmp_url_parts
98 'play_path': prefix + play_path,
101 a_format.update(rtmp_params)
102 formats.append(a_format)
103 self._sort_formats(formats)
106 tracks = video_data.get('tracks')
107 if tracks and isinstance(tracks, list):
109 if track.get('file') and track.get('kind') == 'captions':
110 subtitles.setdefault(track.get('label') or 'en', []).append({
111 'url': self._proto_relative_url(track['file'])
116 'title': video_data['title'] if require_title else video_data.get('title'),
117 'description': video_data.get('description'),
118 'thumbnail': self._proto_relative_url(video_data.get('image')),
119 'timestamp': int_or_none(video_data.get('pubdate')),
120 'duration': float_or_none(jwplayer_data.get('duration')),
121 'subtitles': subtitles,
124 if len(entries) == 1:
127 return self.playlist_result(entries)
130 class JWPlatformIE(JWPlatformBaseIE):
131 _VALID_URL = r'(?:https?://content\.jwplatform\.com/(?:feeds|players|jw6)/|jwplatform:)(?P<id>[a-zA-Z0-9]{8})'
133 'url': 'http://content.jwplatform.com/players/nPripu9l-ALJ3XQCI.js',
134 'md5': 'fa8899fa601eb7c83a64e9d568bdf325',
138 'title': 'Big Buck Bunny Trailer',
139 'description': 'Big Buck Bunny is a short animated film by the Blender Institute. It is made using free and open source software.',
140 'upload_date': '20081127',
141 'timestamp': 1227796140,
146 def _extract_url(webpage):
148 r'<script[^>]+?src=["\'](?P<url>(?:https?:)?//content.jwplatform.com/players/[a-zA-Z0-9]{8})',
151 return mobj.group('url')
153 def _real_extract(self, url):
154 video_id = self._match_id(url)
155 json_data = self._download_json('http://content.jwplatform.com/feeds/%s.json' % video_id, video_id)
156 return self._parse_jwplayer_data(json_data, video_id)