2 from __future__ import unicode_literals
6 from .common import InfoExtractor
7 from ..compat import compat_str
19 class TurnerBaseIE(InfoExtractor):
20 def _extract_timestamp(self, video_data):
21 return int_or_none(xpath_attr(video_data, 'dateCreated', 'uts'))
23 def _extract_cvp_info(self, data_src, video_id, path_data={}):
24 video_data = self._download_xml(data_src, video_id)
25 video_id = video_data.attrib['id']
26 title = xpath_text(video_data, 'headline', fatal=True)
27 content_id = xpath_text(video_data, 'contentId') or video_id
28 # rtmp_src = xpath_text(video_data, 'akamai/src')
30 # splited_rtmp_src = rtmp_src.split(',')
31 # if len(splited_rtmp_src) == 2:
32 # rtmp_src = splited_rtmp_src[1]
33 # aifp = xpath_text(video_data, 'akamai/aifp', default='')
39 r'(?P<width>[0-9]+)x(?P<height>[0-9]+)(?:_(?P<bitrate>[0-9]+))?')
40 # Possible formats locations: files/file, files/groupFiles/files
42 for video_file in video_data.findall('.//file'):
43 video_url = video_file.text.strip()
46 ext = determine_ext(video_url)
47 if video_url.startswith('/mp4:protected/'):
49 # TODO Correct extraction for these files
50 # protected_path_data = path_data.get('protected')
51 # if not protected_path_data or not rtmp_src:
53 # protected_path = self._search_regex(
54 # r'/mp4:(.+)\.[a-z0-9]', video_url, 'secure path')
55 # auth = self._download_webpage(
56 # protected_path_data['tokenizer_src'], query={
57 # 'path': protected_path,
58 # 'videoId': content_id,
61 # token = xpath_text(auth, 'token')
64 # video_url = rtmp_src + video_url + '?' + token
65 elif video_url.startswith('/secure/'):
66 secure_path_data = path_data.get('secure')
67 if not secure_path_data:
69 video_url = secure_path_data['media_src'] + video_url
70 secure_path = self._search_regex(r'https?://[^/]+(.+/)', video_url, 'secure path') + '*'
71 token = tokens.get(secure_path)
73 auth = self._download_xml(
74 secure_path_data['tokenizer_src'], video_id, query={
76 'videoId': content_id,
78 error_msg = xpath_text(auth, 'error/msg')
80 raise ExtractorError(error_msg, expected=True)
81 token = xpath_text(auth, 'token')
84 tokens[secure_path] = token
85 video_url = video_url + '?hdnea=' + token
86 elif not re.match('https?://', video_url):
87 base_path_data = path_data.get(ext, path_data.get('default', {}))
88 media_src = base_path_data.get('media_src')
91 video_url = media_src + video_url
94 urls.append(video_url)
95 format_id = video_file.get('bitrate')
97 formats.extend(self._extract_smil_formats(
98 video_url, video_id, fatal=False))
100 formats.extend(self._extract_m3u8_formats(
101 video_url, video_id, 'mp4',
102 m3u8_id=format_id or 'hls', fatal=False))
104 formats.extend(self._extract_f4m_formats(
105 update_url_query(video_url, {'hdcore': '3.7.0'}),
106 video_id, f4m_id=format_id or 'hds', fatal=False))
109 'format_id': format_id,
113 mobj = rex.search(format_id + video_url)
116 'width': int(mobj.group('width')),
117 'height': int(mobj.group('height')),
118 'tbr': int_or_none(mobj.group('bitrate')),
120 elif isinstance(format_id, compat_str):
121 if format_id.isdigit():
122 f['tbr'] = int(format_id)
124 mobj = re.match(r'ios_(audio|[0-9]+)$', format_id)
126 if mobj.group(1) == 'audio':
132 f['tbr'] = int(mobj.group(1))
134 self._sort_formats(formats)
137 for source in video_data.findall('closedCaptions/source'):
138 for track in source.findall('track'):
139 track_url = track.get('url')
140 if not isinstance(track_url, compat_str) or track_url.endswith('/big'):
142 lang = track.get('lang') or track.get('label') or 'en'
143 subtitles.setdefault(lang, []).append({
149 }.get(source.get('format'))
153 'id': image.get('cut'),
155 'width': int_or_none(image.get('width')),
156 'height': int_or_none(image.get('height')),
157 } for image in video_data.findall('images/image')]
163 'subtitles': subtitles,
164 'thumbnails': thumbnails,
165 'description': xpath_text(video_data, 'description'),
166 'duration': parse_duration(xpath_text(video_data, 'length') or xpath_text(video_data, 'trt')),
167 'timestamp': self._extract_timestamp(video_data),
168 'upload_date': xpath_attr(video_data, 'metas', 'version'),
169 'series': xpath_text(video_data, 'showTitle'),
170 'season_number': int_or_none(xpath_text(video_data, 'seasonNumber')),
171 'episode_number': int_or_none(xpath_text(video_data, 'episodeNumber')),