1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 import re
5
6 from .common import InfoExtractor
7 from ..compat import (
8 compat_urllib_parse_unquote,
9 compat_urllib_parse_urlparse,
10 )
11 from ..utils import (
12 ExtractorError,
13 float_or_none,
14 sanitized_Request,
15 str_or_none,
16 traverse_obj,
17 urlencode_postdata,
18 USER_AGENTS,
19 )
20
21
22 class CeskaTelevizeIE(InfoExtractor):
23 _VALID_URL = r'https?://(?:www\.)?ceskatelevize\.cz/(?:ivysilani|porady|zive)/(?:[^/?#&]+/)*(?P<id>[^/#?]+)'
24 _TESTS = [{
25 'url': 'http://www.ceskatelevize.cz/ivysilani/10441294653-hyde-park-civilizace/215411058090502/bonus/20641-bonus-01-en',
26 'info_dict': {
27 'id': '61924494877028507',
28 'ext': 'mp4',
29 'title': 'Bonus 01 - En - Hyde Park Civilizace',
30 'description': 'English Subtittles',
31 'thumbnail': r're:^https?://.*\.jpg',
32 'duration': 81.3,
33 },
34 'params': {
35 # m3u8 download
36 'skip_download': True,
37 },
38 }, {
39 # live stream
40 'url': 'http://www.ceskatelevize.cz/zive/ct1/',
41 'info_dict': {
42 'id': '102',
43 'ext': 'mp4',
44 'title': r'ČT1 - živé vysílání online',
45 'description': 'Sledujte živé vysílání kanálu ČT1 online. Vybírat si můžete i z dalších kanálů České televize na kterémkoli z vašich zařízení.',
46 'is_live': True,
47 },
48 'params': {
49 # m3u8 download
50 'skip_download': True,
51 },
52 }, {
53 # another
54 'url': 'http://www.ceskatelevize.cz/ivysilani/zive/ct4/',
55 'only_matching': True,
56 'info_dict': {
57 'id': 402,
58 'ext': 'mp4',
59 'title': r're:^ČT Sport \d{4}-\d{2}-\d{2} \d{2}:\d{2}$',
60 'is_live': True,
61 },
62 # 'skip': 'Georestricted to Czech Republic',
63 }, {
64 'url': 'http://www.ceskatelevize.cz/ivysilani/embed/iFramePlayer.php?hash=d6a3e1370d2e4fa76296b90bad4dfc19673b641e&IDEC=217 562 22150/0004&channelID=1&width=100%25',
65 'only_matching': True,
66 }, {
67 # video with 18+ caution trailer
68 'url': 'http://www.ceskatelevize.cz/porady/10520528904-queer/215562210900007-bogotart/',
69 'info_dict': {
70 'id': '215562210900007-bogotart',
71 'title': 'Bogotart - Queer',
72 'description': 'Hlavní město Kolumbie v doprovodu queer umělců. Vroucí svět plný vášně, sebevědomí, ale i násilí a bolesti',
73 },
74 'playlist': [{
75 'info_dict': {
76 'id': '61924494877311053',
77 'ext': 'mp4',
78 'title': 'Bogotart - Queer (Varování 18+)',
79 'duration': 11.9,
80 },
81 }, {
82 'info_dict': {
83 'id': '61924494877068022',
84 'ext': 'mp4',
85 'title': 'Bogotart - Queer (Queer)',
86 'thumbnail': r're:^https?://.*\.jpg',
87 'duration': 1558.3,
88 },
89 }],
90 'params': {
91 # m3u8 download
92 'skip_download': True,
93 },
94 }, {
95 # iframe embed
96 'url': 'http://www.ceskatelevize.cz/porady/10614999031-neviditelni/21251212048/',
97 'only_matching': True,
98 }]
99
100 def _search_nextjs_data(self, webpage, video_id, **kw):
101 return self._parse_json(
102 self._search_regex(
103 r'(?s)<script[^>]+id=[\'"]__NEXT_DATA__[\'"][^>]*>([^<]+)</script>',
104 webpage, 'next.js data', **kw),
105 video_id, **kw)
106
107 def _real_extract(self, url):
108 playlist_id = self._match_id(url)
109 webpage, urlh = self._download_webpage_handle(url, playlist_id)
110 parsed_url = compat_urllib_parse_urlparse(urlh.geturl())
111 site_name = self._og_search_property('site_name', webpage, fatal=False, default='Česká televize')
112 playlist_title = self._og_search_title(webpage, default=None)
113 if site_name and playlist_title:
114 playlist_title = re.split(r'\s*[—|]\s*%s' % (site_name, ), playlist_title, 1)[0]
115 playlist_description = self._og_search_description(webpage, default=None)
116 if playlist_description:
117 playlist_description = playlist_description.replace('\xa0', ' ')
118
119 type_ = 'IDEC'
120 if re.search(r'(^/porady|/zive)/', parsed_url.path):
121 next_data = self._search_nextjs_data(webpage, playlist_id)
122 if '/zive/' in parsed_url.path:
123 idec = traverse_obj(next_data, ('props', 'pageProps', 'data', 'liveBroadcast', 'current', 'idec'), get_all=False)
124 else:
125 idec = traverse_obj(next_data, ('props', 'pageProps', 'data', ('show', 'mediaMeta'), 'idec'), get_all=False)
126 if not idec:
127 idec = traverse_obj(next_data, ('props', 'pageProps', 'data', 'videobonusDetail', 'bonusId'), get_all=False)
128 if idec:
129 type_ = 'bonus'
130 if not idec:
131 raise ExtractorError('Failed to find IDEC id')
132 iframe_hash = self._download_webpage(
133 'https://www.ceskatelevize.cz/v-api/iframe-hash/',
134 playlist_id, note='Getting IFRAME hash')
135 query = {'hash': iframe_hash, 'origin': 'iVysilani', 'autoStart': 'true', type_: idec, }
136 webpage = self._download_webpage(
137 'https://www.ceskatelevize.cz/ivysilani/embed/iFramePlayer.php',
138 playlist_id, note='Downloading player', query=query)
139
140 NOT_AVAILABLE_STRING = 'This content is not available at your territory due to limited copyright.'
141 if '%s</p>' % NOT_AVAILABLE_STRING in webpage:
142 self.raise_geo_restricted(NOT_AVAILABLE_STRING)
143 if any(not_found in webpage for not_found in ('Neplatný parametr pro videopřehrávač', 'IDEC nebyl nalezen', )):
144 raise ExtractorError('no video with IDEC available', video_id=idec, expected=True)
145
146 type_ = None
147 episode_id = None
148
149 playlist = self._parse_json(
150 self._search_regex(
151 r'getPlaylistUrl\(\[({.+?})\]', webpage, 'playlist',
152 default='{}'), playlist_id)
153 if playlist:
154 type_ = playlist.get('type')
155 episode_id = playlist.get('id')
156
157 if not type_:
158 type_ = self._html_search_regex(
159 r'getPlaylistUrl\(\[\{"type":"(.+?)","id":".+?"\}\],',
160 webpage, 'type')
161 if not episode_id:
162 episode_id = self._html_search_regex(
163 r'getPlaylistUrl\(\[\{"type":".+?","id":"(.+?)"\}\],',
164 webpage, 'episode_id')
165
166 data = {
167 'playlist[0][type]': type_,
168 'playlist[0][id]': episode_id,
169 'requestUrl': parsed_url.path,
170 'requestSource': 'iVysilani',
171 }
172
173 entries = []
174
175 for user_agent in (None, USER_AGENTS['Safari']):
176 req = sanitized_Request(
177 'https://www.ceskatelevize.cz/ivysilani/ajax/get-client-playlist/',
178 data=urlencode_postdata(data))
179
180 req.add_header('Content-type', 'application/x-www-form-urlencoded')
181 req.add_header('x-addr', '127.0.0.1')
182 req.add_header('X-Requested-With', 'XMLHttpRequest')
183 if user_agent:
184 req.add_header('User-Agent', user_agent)
185 req.add_header('Referer', url)
186
187 playlistpage = self._download_json(req, playlist_id, fatal=False)
188
189 if not playlistpage:
190 continue
191
192 playlist_url = playlistpage['url']
193 if playlist_url == 'error_region':
194 raise ExtractorError(NOT_AVAILABLE_STRING, expected=True)
195
196 req = sanitized_Request(compat_urllib_parse_unquote(playlist_url))
197 req.add_header('Referer', url)
198
199 playlist = self._download_json(req, playlist_id, fatal=False)
200 if not playlist:
201 continue
202
203 playlist = playlist.get('playlist')
204 if not isinstance(playlist, list):
205 continue
206
207 playlist_len = len(playlist)
208
209 for num, item in enumerate(playlist):
210 is_live = item.get('type') == 'LIVE'
211 formats = []
212 for format_id, stream_url in item.get('streamUrls', {}).items():
213 if 'drmOnly=true' in stream_url:
214 continue
215 if 'playerType=flash' in stream_url:
216 stream_formats = self._extract_m3u8_formats(
217 stream_url, playlist_id, 'mp4', 'm3u8_native',
218 m3u8_id='hls-%s' % format_id, fatal=False)
219 else:
220 stream_formats = self._extract_mpd_formats(
221 stream_url, playlist_id,
222 mpd_id='dash-%s' % format_id, fatal=False)
223 # See https://github.com/ytdl-org/youtube-dl/issues/12119#issuecomment-280037031
224 if format_id == 'audioDescription':
225 for f in stream_formats:
226 f['source_preference'] = -10
227 formats.extend(stream_formats)
228
229 if user_agent and len(entries) == playlist_len:
230 entries[num]['formats'].extend(formats)
231 continue
232
233 item_id = str_or_none(item.get('id') or item['assetId'])
234 title = item['title']
235
236 duration = float_or_none(item.get('duration'))
237 thumbnail = item.get('previewImageUrl')
238
239 subtitles = {}
240 if item.get('type') == 'VOD':
241 subs = item.get('subtitles')
242 if subs:
243 subtitles = self.extract_subtitles(episode_id, subs)
244
245 if playlist_len == 1:
246 final_title = playlist_title or title
247 else:
248 final_title = '%s (%s)' % (playlist_title, title)
249
250 entries.append({
251 'id': item_id,
252 'title': final_title,
253 'description': playlist_description if playlist_len == 1 else None,
254 'thumbnail': thumbnail,
255 'duration': duration,
256 'formats': formats,
257 'subtitles': subtitles,
258 'is_live': is_live,
259 })
260
261 for e in entries:
262 self._sort_formats(e['formats'])
263
264 if len(entries) == 1:
265 return entries[0]
266 return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)
267
268 def _get_subtitles(self, episode_id, subs):
269 original_subtitles = self._download_webpage(
270 subs[0]['url'], episode_id, 'Downloading subtitles')
271 srt_subs = self._fix_subtitles(original_subtitles)
272 return {
273 'cs': [{
274 'ext': 'srt',
275 'data': srt_subs,
276 }]
277 }
278
279 @staticmethod
280 def _fix_subtitles(subtitles):
281 """ Convert millisecond-based subtitles to SRT """
282
283 def _msectotimecode(msec):
284 """ Helper utility to convert milliseconds to timecode """
285 components = []
286 for divider in [1000, 60, 60, 100]:
287 components.append(msec % divider)
288 msec //= divider
289 return '{3:02}:{2:02}:{1:02},{0:03}'.format(*components)
290
291 def _fix_subtitle(subtitle):
292 for line in subtitle.splitlines():
293 m = re.match(r'^\s*([0-9]+);\s*([0-9]+)\s+([0-9]+)\s*$', line)
294 if m:
295 yield m.group(1)
296 start, stop = (_msectotimecode(int(t)) for t in m.groups()[1:])
297 yield '{0} --> {1}'.format(start, stop)
298 else:
299 yield line
300
301 return '\r\n'.join(_fix_subtitle(subtitles))
|