yt_dlp/extractor/onet.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     NO_DEFAULT,
   6     ExtractorError,
   7     determine_ext,
   8     float_or_none,
   9     get_element_by_class,
  10     int_or_none,
  11     js_to_json,
  12     parse_iso8601,
  13     remove_start,
  14     strip_or_none,
  15     url_basename,
  16 )
  17
  18
  19 class OnetBaseIE(InfoExtractor):
  20     _URL_BASE_RE = r'https?://(?:(?:www\.)?onet\.tv|onet100\.vod\.pl)/[a-z]/'
  21
  22     def _search_mvp_id(self, webpage):
  23         return self._search_regex(
  24             r'id=(["\'])mvp:(?P<id>.+?)\1', webpage, 'mvp id', group='id')
  25
  26     def _extract_from_id(self, video_id, webpage=None):
  27         response = self._download_json(
  28             'http://qi.ckm.onetapi.pl/', video_id,
  29             query={
  30                 'body[id]': video_id,
  31                 'body[jsonrpc]': '2.0',
  32                 'body[method]': 'get_asset_detail',
  33                 'body[params][ID_Publikacji]': video_id,
  34                 'body[params][Service]': 'www.onet.pl',
  35                 'content-type': 'application/jsonp',
  36                 'x-onet-app': 'player.front.onetapi.pl',
  37             })
  38
  39         error = response.get('error')
  40         if error:
  41             raise ExtractorError(
  42                 '{} said: {}'.format(self.IE_NAME, error['message']), expected=True)
  43
  44         video = response['result'].get('0')
  45
  46         formats = []
  47         for format_type, formats_dict in video['formats'].items():
  48             if not isinstance(formats_dict, dict):
  49                 continue
  50             for format_id, format_list in formats_dict.items():
  51                 if not isinstance(format_list, list):
  52                     continue
  53                 for f in format_list:
  54                     video_url = f.get('url')
  55                     if not video_url:
  56                         continue
  57                     ext = determine_ext(video_url)
  58                     if format_id.startswith('ism'):
  59                         formats.extend(self._extract_ism_formats(
  60                             video_url, video_id, 'mss', fatal=False))
  61                     elif ext == 'mpd':
  62                         formats.extend(self._extract_mpd_formats(
  63                             video_url, video_id, mpd_id='dash', fatal=False))
  64                     elif format_id.startswith('hls'):
  65                         formats.extend(self._extract_m3u8_formats(
  66                             video_url, video_id, 'mp4', 'm3u8_native',
  67                             m3u8_id='hls', fatal=False))
  68                     else:
  69                         http_f = {
  70                             'url': video_url,
  71                             'format_id': format_id,
  72                             'abr': float_or_none(f.get('audio_bitrate')),
  73                         }
  74                         if format_type == 'audio':
  75                             http_f['vcodec'] = 'none'
  76                         else:
  77                             http_f.update({
  78                                 'height': int_or_none(f.get('vertical_resolution')),
  79                                 'width': int_or_none(f.get('horizontal_resolution')),
  80                                 'vbr': float_or_none(f.get('video_bitrate')),
  81                             })
  82                         formats.append(http_f)
  83
  84         meta = video.get('meta', {})
  85
  86         title = (self._og_search_title(
  87             webpage, default=None) if webpage else None) or meta['title']
  88         description = (self._og_search_description(
  89             webpage, default=None) if webpage else None) or meta.get('description')
  90         duration = meta.get('length') or meta.get('lenght')
  91         timestamp = parse_iso8601(meta.get('addDate'), ' ')
  92
  93         return {
  94             'id': video_id,
  95             'title': title,
  96             'description': description,
  97             'duration': duration,
  98             'timestamp': timestamp,
  99             'formats': formats,
 100         }
 101
 102
 103 class OnetMVPIE(OnetBaseIE):
 104     _VALID_URL = r'onetmvp:(?P<id>\d+\.\d+)'
 105
 106     _TEST = {
 107         'url': 'onetmvp:381027.1509591944',
 108         'only_matching': True,
 109     }
 110
 111     def _real_extract(self, url):
 112         return self._extract_from_id(self._match_id(url))
 113
 114
 115 class OnetIE(OnetBaseIE):
 116     _VALID_URL = OnetBaseIE._URL_BASE_RE + r'[a-z]+/(?P<display_id>[0-9a-z-]+)/(?P<id>[0-9a-z]+)'
 117     IE_NAME = 'onet.tv'
 118
 119     _TESTS = [{
 120         'url': 'http://onet.tv/k/openerfestival/open-er-festival-2016-najdziwniejsze-wymagania-gwiazd/qbpyqc',
 121         'md5': '436102770fb095c75b8bb0392d3da9ff',
 122         'info_dict': {
 123             'id': 'qbpyqc',
 124             'display_id': 'open-er-festival-2016-najdziwniejsze-wymagania-gwiazd',
 125             'ext': 'mp4',
 126             'title': 'Open\'er Festival 2016: najdziwniejsze wymagania gwiazd',
 127             'description': 'Trzy samochody, których nigdy nie użyto, prywatne spa, hotel dekorowany czarnym suknem czy nielegalne używki. Organizatorzy koncertów i festiwali muszą stawać przed nie lada wyzwaniem zapraszając gwia...',
 128             'upload_date': '20160705',
 129             'timestamp': 1467721580,
 130         },
 131     }, {
 132         'url': 'https://onet100.vod.pl/k/openerfestival/open-er-festival-2016-najdziwniejsze-wymagania-gwiazd/qbpyqc',
 133         'only_matching': True,
 134     }]
 135
 136     def _real_extract(self, url):
 137         mobj = self._match_valid_url(url)
 138         display_id, video_id = mobj.group('display_id', 'id')
 139
 140         webpage = self._download_webpage(url, display_id)
 141
 142         mvp_id = self._search_mvp_id(webpage)
 143
 144         info_dict = self._extract_from_id(mvp_id, webpage)
 145         info_dict.update({
 146             'id': video_id,
 147             'display_id': display_id,
 148         })
 149
 150         return info_dict
 151
 152
 153 class OnetChannelIE(OnetBaseIE):
 154     _VALID_URL = OnetBaseIE._URL_BASE_RE + r'(?P<id>[a-z]+)(?:[?#]|$)'
 155     IE_NAME = 'onet.tv:channel'
 156
 157     _TESTS = [{
 158         'url': 'http://onet.tv/k/openerfestival',
 159         'info_dict': {
 160             'id': 'openerfestival',
 161             'title': "Open'er Festival",
 162             'description': "Tak było na Open'er Festival 2016! Oglądaj nasze reportaże i wywiady z artystami.",
 163         },
 164         'playlist_mincount': 35,
 165     }, {
 166         'url': 'https://onet100.vod.pl/k/openerfestival',
 167         'only_matching': True,
 168     }]
 169
 170     def _real_extract(self, url):
 171         channel_id = self._match_id(url)
 172
 173         webpage = self._download_webpage(url, channel_id)
 174
 175         current_clip_info = self._parse_json(self._search_regex(
 176             r'var\s+currentClip\s*=\s*({[^}]+})', webpage, 'video info'), channel_id,
 177             transform_source=lambda s: js_to_json(re.sub(r'\'\s*\+\s*\'', '', s)))
 178         video_id = remove_start(current_clip_info['ckmId'], 'mvp:')
 179         video_name = url_basename(current_clip_info['url'])
 180
 181         if not self._yes_playlist(channel_id, video_name, playlist_label='channel'):
 182             return self._extract_from_id(video_id, webpage)
 183
 184         matches = re.findall(
 185             rf'<a[^>]+href=[\'"]({self._URL_BASE_RE}[a-z]+/[0-9a-z-]+/[0-9a-z]+)',
 186             webpage)
 187         entries = [
 188             self.url_result(video_link, OnetIE.ie_key())
 189             for video_link in matches]
 190
 191         channel_title = strip_or_none(get_element_by_class('o_channelName', webpage))
 192         channel_description = strip_or_none(get_element_by_class('o_channelDesc', webpage))
 193         return self.playlist_result(entries, channel_id, channel_title, channel_description)
 194
 195
 196 class OnetPlIE(InfoExtractor):
 197     _VALID_URL = r'https?://(?:[^/]+\.)?(?:onet|businessinsider\.com|plejada)\.pl/(?:[^/]+/)+(?P<id>[0-9a-z]+)'
 198     IE_NAME = 'onet.pl'
 199
 200     _TESTS = [{
 201         'url': 'http://eurosport.onet.pl/zimowe/skoki-narciarskie/ziobro-wygral-kwalifikacje-w-pjongczangu/9ckrly',
 202         'md5': 'b94021eb56214c3969380388b6e73cb0',
 203         'info_dict': {
 204             'id': '1561707.1685479',
 205             'ext': 'mp4',
 206             'title': 'Ziobro wygrał kwalifikacje w Pjongczangu',
 207             'description': 'md5:61fb0740084d2d702ea96512a03585b4',
 208             'upload_date': '20170214',
 209             'timestamp': 1487078046,
 210         },
 211     }, {
 212         # embedded via pulsembed
 213         'url': 'http://film.onet.pl/pensjonat-nad-rozlewiskiem-relacja-z-planu-serialu/y428n0',
 214         'info_dict': {
 215             'id': '501235.965429946',
 216             'ext': 'mp4',
 217             'title': '"Pensjonat nad rozlewiskiem": relacja z planu serialu',
 218             'upload_date': '20170622',
 219             'timestamp': 1498159955,
 220         },
 221         'params': {
 222             'skip_download': True,
 223         },
 224     }, {
 225         'url': 'http://film.onet.pl/zwiastuny/ghost-in-the-shell-drugi-zwiastun-pl/5q6yl3',
 226         'only_matching': True,
 227     }, {
 228         'url': 'http://moto.onet.pl/jak-wybierane-sa-miejsca-na-fotoradary/6rs04e',
 229         'only_matching': True,
 230     }, {
 231         'url': 'http://businessinsider.com.pl/wideo/scenariusz-na-koniec-swiata-wedlug-nasa/dwnqptk',
 232         'only_matching': True,
 233     }, {
 234         'url': 'http://plejada.pl/weronika-rosati-o-swoim-domniemanym-slubie/n2bq89',
 235         'only_matching': True,
 236     }]
 237
 238     def _search_mvp_id(self, webpage, default=NO_DEFAULT):
 239         return self._search_regex(
 240             r'data-(?:params-)?mvp=["\'](\d+\.\d+)', webpage, 'mvp id',
 241             default=default)
 242
 243     def _real_extract(self, url):
 244         video_id = self._match_id(url)
 245
 246         webpage = self._download_webpage(url, video_id)
 247
 248         mvp_id = self._search_mvp_id(webpage, default=None)
 249
 250         if not mvp_id:
 251             pulsembed_url = self._search_regex(
 252                 r'data-src=(["\'])(?P<url>(?:https?:)?//pulsembed\.eu/.+?)\1',
 253                 webpage, 'pulsembed url', group='url')
 254             webpage = self._download_webpage(
 255                 pulsembed_url, video_id, 'Downloading pulsembed webpage')
 256             mvp_id = self._search_mvp_id(webpage)
 257
 258         return self.url_result(
 259             f'onetmvp:{mvp_id}', OnetMVPIE.ie_key(), video_id=mvp_id)