]>
Commit | Line | Data |
---|---|---|
29825140 RA |
1 | import re |
2 | ||
c38f0681 | 3 | from .adobepass import AdobePassIE |
cd10b3ea | 4 | from ..compat import compat_str |
29825140 | 5 | from ..utils import ( |
29f7c58a | 6 | fix_xml_ampersands, |
29825140 RA |
7 | xpath_text, |
8 | int_or_none, | |
9 | determine_ext, | |
e0d42dd4 | 10 | float_or_none, |
29825140 RA |
11 | parse_duration, |
12 | xpath_attr, | |
13 | update_url_query, | |
bdcc046d | 14 | ExtractorError, |
04c09f19 | 15 | strip_or_none, |
3052a30d | 16 | url_or_none, |
29825140 RA |
17 | ) |
18 | ||
19 | ||
c38f0681 | 20 | class TurnerBaseIE(AdobePassIE): |
b6f78d76 RA |
21 | _AKAMAI_SPE_TOKEN_CACHE = {} |
22 | ||
da30a20a RA |
23 | def _extract_timestamp(self, video_data): |
24 | return int_or_none(xpath_attr(video_data, 'dateCreated', 'uts')) | |
25 | ||
e0d42dd4 | 26 | def _add_akamai_spe_token(self, tokenizer_src, video_url, content_id, ap_data, custom_tokenizer_query=None): |
b6f78d76 RA |
27 | secure_path = self._search_regex(r'https?://[^/]+(.+/)', video_url, 'secure path') + '*' |
28 | token = self._AKAMAI_SPE_TOKEN_CACHE.get(secure_path) | |
29 | if not token: | |
30 | query = { | |
31 | 'path': secure_path, | |
b6f78d76 | 32 | } |
e0d42dd4 RA |
33 | if custom_tokenizer_query: |
34 | query.update(custom_tokenizer_query) | |
35 | else: | |
36 | query['videoId'] = content_id | |
b6f78d76 RA |
37 | if ap_data.get('auth_required'): |
38 | query['accessToken'] = self._extract_mvpd_auth(ap_data['url'], content_id, ap_data['site_name'], ap_data['site_name']) | |
39 | auth = self._download_xml( | |
40 | tokenizer_src, content_id, query=query) | |
41 | error_msg = xpath_text(auth, 'error/msg') | |
42 | if error_msg: | |
43 | raise ExtractorError(error_msg, expected=True) | |
44 | token = xpath_text(auth, 'token') | |
45 | if not token: | |
46 | return video_url | |
47 | self._AKAMAI_SPE_TOKEN_CACHE[secure_path] = token | |
48 | return video_url + '?hdnea=' + token | |
49 | ||
29f7c58a | 50 | def _extract_cvp_info(self, data_src, video_id, path_data={}, ap_data={}, fatal=False): |
51 | video_data = self._download_xml( | |
52 | data_src, video_id, | |
53 | transform_source=lambda s: fix_xml_ampersands(s).strip(), | |
54 | fatal=fatal) | |
55 | if not video_data: | |
56 | return {} | |
3c77a54d | 57 | video_id = video_data.attrib['id'] |
29825140 | 58 | title = xpath_text(video_data, 'headline', fatal=True) |
bdcc046d | 59 | content_id = xpath_text(video_data, 'contentId') or video_id |
29825140 RA |
60 | # rtmp_src = xpath_text(video_data, 'akamai/src') |
61 | # if rtmp_src: | |
a0566bbf | 62 | # split_rtmp_src = rtmp_src.split(',') |
63 | # if len(split_rtmp_src) == 2: | |
64 | # rtmp_src = split_rtmp_src[1] | |
29825140 RA |
65 | # aifp = xpath_text(video_data, 'akamai/aifp', default='') |
66 | ||
29825140 RA |
67 | urls = [] |
68 | formats = [] | |
29f7c58a | 69 | thumbnails = [] |
70 | subtitles = {} | |
cd10b3ea S |
71 | rex = re.compile( |
72 | r'(?P<width>[0-9]+)x(?P<height>[0-9]+)(?:_(?P<bitrate>[0-9]+))?') | |
73 | # Possible formats locations: files/file, files/groupFiles/files | |
74 | # and maybe others | |
75 | for video_file in video_data.findall('.//file'): | |
29f7c58a | 76 | video_url = url_or_none(video_file.text.strip()) |
29825140 RA |
77 | if not video_url: |
78 | continue | |
79 | ext = determine_ext(video_url) | |
80 | if video_url.startswith('/mp4:protected/'): | |
81 | continue | |
82 | # TODO Correct extraction for these files | |
83 | # protected_path_data = path_data.get('protected') | |
84 | # if not protected_path_data or not rtmp_src: | |
85 | # continue | |
86 | # protected_path = self._search_regex( | |
87 | # r'/mp4:(.+)\.[a-z0-9]', video_url, 'secure path') | |
88 | # auth = self._download_webpage( | |
89 | # protected_path_data['tokenizer_src'], query={ | |
90 | # 'path': protected_path, | |
bdcc046d | 91 | # 'videoId': content_id, |
29825140 RA |
92 | # 'aifp': aifp, |
93 | # }) | |
94 | # token = xpath_text(auth, 'token') | |
95 | # if not token: | |
96 | # continue | |
97 | # video_url = rtmp_src + video_url + '?' + token | |
98 | elif video_url.startswith('/secure/'): | |
99 | secure_path_data = path_data.get('secure') | |
100 | if not secure_path_data: | |
101 | continue | |
b6f78d76 RA |
102 | video_url = self._add_akamai_spe_token( |
103 | secure_path_data['tokenizer_src'], | |
104 | secure_path_data['media_src'] + video_url, | |
105 | content_id, ap_data) | |
29825140 RA |
106 | elif not re.match('https?://', video_url): |
107 | base_path_data = path_data.get(ext, path_data.get('default', {})) | |
108 | media_src = base_path_data.get('media_src') | |
109 | if not media_src: | |
110 | continue | |
111 | video_url = media_src + video_url | |
112 | if video_url in urls: | |
113 | continue | |
114 | urls.append(video_url) | |
cd10b3ea | 115 | format_id = video_file.get('bitrate') |
29f7c58a | 116 | if ext in ('scc', 'srt', 'vtt'): |
117 | subtitles.setdefault('en', []).append({ | |
118 | 'ext': ext, | |
119 | 'url': video_url, | |
120 | }) | |
121 | elif ext == 'png': | |
122 | thumbnails.append({ | |
123 | 'id': format_id, | |
124 | 'url': video_url, | |
125 | }) | |
126 | elif ext == 'smil': | |
cd10b3ea S |
127 | formats.extend(self._extract_smil_formats( |
128 | video_url, video_id, fatal=False)) | |
29f7c58a | 129 | elif re.match(r'https?://[^/]+\.akamaihd\.net/[iz]/', video_url): |
130 | formats.extend(self._extract_akamai_formats( | |
131 | video_url, video_id, { | |
132 | 'hds': path_data.get('f4m', {}).get('host'), | |
133 | # nba.cdn.turner.com, ht.cdn.turner.com, ht2.cdn.turner.com | |
134 | # ht3.cdn.turner.com, i.cdn.turner.com, s.cdn.turner.com | |
135 | # ssl.cdn.turner.com | |
136 | 'http': 'pmd.cdn.turner.com', | |
137 | })) | |
29825140 | 138 | elif ext == 'm3u8': |
36fce548 | 139 | m3u8_formats = self._extract_m3u8_formats( |
bdcc046d | 140 | video_url, video_id, 'mp4', |
36fce548 RA |
141 | m3u8_id=format_id or 'hls', fatal=False) |
142 | if '/secure/' in video_url and '?hdnea=' in video_url: | |
143 | for f in m3u8_formats: | |
0a5a191a | 144 | f['downloader_options'] = {'ffmpeg_args': ['-seekable', '0']} |
36fce548 | 145 | formats.extend(m3u8_formats) |
29825140 RA |
146 | elif ext == 'f4m': |
147 | formats.extend(self._extract_f4m_formats( | |
148 | update_url_query(video_url, {'hdcore': '3.7.0'}), | |
cd10b3ea | 149 | video_id, f4m_id=format_id or 'hds', fatal=False)) |
29825140 RA |
150 | else: |
151 | f = { | |
152 | 'format_id': format_id, | |
153 | 'url': video_url, | |
154 | 'ext': ext, | |
155 | } | |
29f7c58a | 156 | mobj = rex.search(video_url) |
29825140 RA |
157 | if mobj: |
158 | f.update({ | |
159 | 'width': int(mobj.group('width')), | |
160 | 'height': int(mobj.group('height')), | |
161 | 'tbr': int_or_none(mobj.group('bitrate')), | |
162 | }) | |
cd10b3ea S |
163 | elif isinstance(format_id, compat_str): |
164 | if format_id.isdigit(): | |
165 | f['tbr'] = int(format_id) | |
166 | else: | |
167 | mobj = re.match(r'ios_(audio|[0-9]+)$', format_id) | |
168 | if mobj: | |
169 | if mobj.group(1) == 'audio': | |
170 | f.update({ | |
171 | 'vcodec': 'none', | |
172 | 'ext': 'm4a', | |
173 | }) | |
174 | else: | |
175 | f['tbr'] = int(mobj.group(1)) | |
29825140 RA |
176 | formats.append(f) |
177 | self._sort_formats(formats) | |
178 | ||
29825140 RA |
179 | for source in video_data.findall('closedCaptions/source'): |
180 | for track in source.findall('track'): | |
3052a30d S |
181 | track_url = url_or_none(track.get('url')) |
182 | if not track_url or track_url.endswith('/big'): | |
29825140 | 183 | continue |
547993dc S |
184 | lang = track.get('lang') or track.get('label') or 'en' |
185 | subtitles.setdefault(lang, []).append({ | |
186 | 'url': track_url, | |
29825140 RA |
187 | 'ext': { |
188 | 'scc': 'scc', | |
189 | 'webvtt': 'vtt', | |
190 | 'smptett': 'tt', | |
191 | }.get(source.get('format')) | |
192 | }) | |
193 | ||
29f7c58a | 194 | thumbnails.extend({ |
195 | 'id': image.get('cut') or image.get('name'), | |
29825140 RA |
196 | 'url': image.text, |
197 | 'width': int_or_none(image.get('width')), | |
198 | 'height': int_or_none(image.get('height')), | |
29f7c58a | 199 | } for image in video_data.findall('images/image')) |
29825140 | 200 | |
04c09f19 RA |
201 | is_live = xpath_text(video_data, 'isLive') == 'true' |
202 | ||
29825140 RA |
203 | return { |
204 | 'id': video_id, | |
39ca3b5c | 205 | 'title': title, |
29825140 RA |
206 | 'formats': formats, |
207 | 'subtitles': subtitles, | |
208 | 'thumbnails': thumbnails, | |
04c09f19 RA |
209 | 'thumbnail': xpath_text(video_data, 'poster'), |
210 | 'description': strip_or_none(xpath_text(video_data, 'description')), | |
29825140 | 211 | 'duration': parse_duration(xpath_text(video_data, 'length') or xpath_text(video_data, 'trt')), |
da30a20a | 212 | 'timestamp': self._extract_timestamp(video_data), |
29825140 RA |
213 | 'upload_date': xpath_attr(video_data, 'metas', 'version'), |
214 | 'series': xpath_text(video_data, 'showTitle'), | |
215 | 'season_number': int_or_none(xpath_text(video_data, 'seasonNumber')), | |
216 | 'episode_number': int_or_none(xpath_text(video_data, 'episodeNumber')), | |
04c09f19 | 217 | 'is_live': is_live, |
29825140 | 218 | } |
e0d42dd4 RA |
219 | |
220 | def _extract_ngtv_info(self, media_id, tokenizer_query, ap_data=None): | |
1e79316e | 221 | is_live = ap_data.get('is_live') |
e0d42dd4 RA |
222 | streams_data = self._download_json( |
223 | 'http://medium.ngtv.io/media/%s/tv' % media_id, | |
224 | media_id)['media']['tv'] | |
225 | duration = None | |
226 | chapters = [] | |
227 | formats = [] | |
228 | for supported_type in ('unprotected', 'bulkaes'): | |
229 | stream_data = streams_data.get(supported_type, {}) | |
230 | m3u8_url = stream_data.get('secureUrl') or stream_data.get('url') | |
231 | if not m3u8_url: | |
232 | continue | |
233 | if stream_data.get('playlistProtection') == 'spe': | |
234 | m3u8_url = self._add_akamai_spe_token( | |
235 | 'http://token.ngtv.io/token/token_spe', | |
236 | m3u8_url, media_id, ap_data or {}, tokenizer_query) | |
237 | formats.extend(self._extract_m3u8_formats( | |
1e79316e | 238 | m3u8_url, media_id, 'mp4', m3u8_id='hls', live=is_live, fatal=False)) |
e0d42dd4 RA |
239 | |
240 | duration = float_or_none(stream_data.get('totalRuntime')) | |
241 | ||
1e79316e | 242 | if not chapters and not is_live: |
e0d42dd4 RA |
243 | for chapter in stream_data.get('contentSegments', []): |
244 | start_time = float_or_none(chapter.get('start')) | |
245 | chapter_duration = float_or_none(chapter.get('duration')) | |
246 | if start_time is None or chapter_duration is None: | |
247 | continue | |
248 | chapters.append({ | |
249 | 'start_time': start_time, | |
250 | 'end_time': start_time + chapter_duration, | |
251 | }) | |
252 | self._sort_formats(formats) | |
253 | ||
254 | return { | |
255 | 'formats': formats, | |
256 | 'chapters': chapters, | |
257 | 'duration': duration, | |
258 | } |