Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/onet.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     determine_ext,
   9     ExtractorError,
  10     float_or_none,
  11     get_element_by_class,
  12     int_or_none,
  13     js_to_json,
  14     NO_DEFAULT,
  15     parse_iso8601,
  16     remove_start,
  17     strip_or_none,
  18     url_basename,
  19 )
  20
  21
  22 class OnetBaseIE(InfoExtractor):
  23     _URL_BASE_RE = r'https?://(?:(?:www\.)?onet\.tv|onet100\.vod\.pl)/[a-z]/'
  24
  25     def _search_mvp_id(self, webpage):
  26         return self._search_regex(
  27             r'id=(["\'])mvp:(?P<id>.+?)\1', webpage, 'mvp id', group='id')
  28
  29     def _extract_from_id(self, video_id, webpage=None):
  30         response = self._download_json(
  31             'http://qi.ckm.onetapi.pl/', video_id,
  32             query={
  33                 'body[id]': video_id,
  34                 'body[jsonrpc]': '2.0',
  35                 'body[method]': 'get_asset_detail',
  36                 'body[params][ID_Publikacji]': video_id,
  37                 'body[params][Service]': 'www.onet.pl',
  38                 'content-type': 'application/jsonp',
  39                 'x-onet-app': 'player.front.onetapi.pl',
  40             })
  41
  42         error = response.get('error')
  43         if error:
  44             raise ExtractorError(
  45                 '%s said: %s' % (self.IE_NAME, error['message']), expected=True)
  46
  47         video = response['result'].get('0')
  48
  49         formats = []
  50         for format_type, formats_dict in video['formats'].items():
  51             if not isinstance(formats_dict, dict):
  52                 continue
  53             for format_id, format_list in formats_dict.items():
  54                 if not isinstance(format_list, list):
  55                     continue
  56                 for f in format_list:
  57                     video_url = f.get('url')
  58                     if not video_url:
  59                         continue
  60                     ext = determine_ext(video_url)
  61                     if format_id.startswith('ism'):
  62                         formats.extend(self._extract_ism_formats(
  63                             video_url, video_id, 'mss', fatal=False))
  64                     elif ext == 'mpd':
  65                         formats.extend(self._extract_mpd_formats(
  66                             video_url, video_id, mpd_id='dash', fatal=False))
  67                     elif format_id.startswith('hls'):
  68                         formats.extend(self._extract_m3u8_formats(
  69                             video_url, video_id, 'mp4', 'm3u8_native',
  70                             m3u8_id='hls', fatal=False))
  71                     else:
  72                         http_f = {
  73                             'url': video_url,
  74                             'format_id': format_id,
  75                             'abr': float_or_none(f.get('audio_bitrate')),
  76                         }
  77                         if format_type == 'audio':
  78                             http_f['vcodec'] = 'none'
  79                         else:
  80                             http_f.update({
  81                                 'height': int_or_none(f.get('vertical_resolution')),
  82                                 'width': int_or_none(f.get('horizontal_resolution')),
  83                                 'vbr': float_or_none(f.get('video_bitrate')),
  84                             })
  85                         formats.append(http_f)
  86         self._sort_formats(formats)
  87
  88         meta = video.get('meta', {})
  89
  90         title = (self._og_search_title(
  91             webpage, default=None) if webpage else None) or meta['title']
  92         description = (self._og_search_description(
  93             webpage, default=None) if webpage else None) or meta.get('description')
  94         duration = meta.get('length') or meta.get('lenght')
  95         timestamp = parse_iso8601(meta.get('addDate'), ' ')
  96
  97         return {
  98             'id': video_id,
  99             'title': title,
 100             'description': description,
 101             'duration': duration,
 102             'timestamp': timestamp,
 103             'formats': formats,
 104         }
 105
 106
 107 class OnetMVPIE(OnetBaseIE):
 108     _VALID_URL = r'onetmvp:(?P<id>\d+\.\d+)'
 109
 110     _TEST = {
 111         'url': 'onetmvp:381027.1509591944',
 112         'only_matching': True,
 113     }
 114
 115     def _real_extract(self, url):
 116         return self._extract_from_id(self._match_id(url))
 117
 118
 119 class OnetIE(OnetBaseIE):
 120     _VALID_URL = OnetBaseIE._URL_BASE_RE + r'[a-z]+/(?P<display_id>[0-9a-z-]+)/(?P<id>[0-9a-z]+)'
 121     IE_NAME = 'onet.tv'
 122
 123     _TESTS = [{
 124         'url': 'http://onet.tv/k/openerfestival/open-er-festival-2016-najdziwniejsze-wymagania-gwiazd/qbpyqc',
 125         'md5': '436102770fb095c75b8bb0392d3da9ff',
 126         'info_dict': {
 127             'id': 'qbpyqc',
 128             'display_id': 'open-er-festival-2016-najdziwniejsze-wymagania-gwiazd',
 129             'ext': 'mp4',
 130             'title': 'Open\'er Festival 2016: najdziwniejsze wymagania gwiazd',
 131             '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...',
 132             'upload_date': '20160705',
 133             'timestamp': 1467721580,
 134         },
 135     }, {
 136         'url': 'https://onet100.vod.pl/k/openerfestival/open-er-festival-2016-najdziwniejsze-wymagania-gwiazd/qbpyqc',
 137         'only_matching': True,
 138     }]
 139
 140     def _real_extract(self, url):
 141         mobj = re.match(self._VALID_URL, url)
 142         display_id, video_id = mobj.group('display_id', 'id')
 143
 144         webpage = self._download_webpage(url, display_id)
 145
 146         mvp_id = self._search_mvp_id(webpage)
 147
 148         info_dict = self._extract_from_id(mvp_id, webpage)
 149         info_dict.update({
 150             'id': video_id,
 151             'display_id': display_id,
 152         })
 153
 154         return info_dict
 155
 156
 157 class OnetChannelIE(OnetBaseIE):
 158     _VALID_URL = OnetBaseIE._URL_BASE_RE + r'(?P<id>[a-z]+)(?:[?#]|$)'
 159     IE_NAME = 'onet.tv:channel'
 160
 161     _TESTS = [{
 162         'url': 'http://onet.tv/k/openerfestival',
 163         'info_dict': {
 164             'id': 'openerfestival',
 165             'title': "Open'er Festival",
 166             'description': "Tak było na Open'er Festival 2016! Oglądaj nasze reportaże i wywiady z artystami.",
 167         },
 168         'playlist_mincount': 35,
 169     }, {
 170         'url': 'https://onet100.vod.pl/k/openerfestival',
 171         'only_matching': True,
 172     }]
 173
 174     def _real_extract(self, url):
 175         channel_id = self._match_id(url)
 176
 177         webpage = self._download_webpage(url, channel_id)
 178
 179         current_clip_info = self._parse_json(self._search_regex(
 180             r'var\s+currentClip\s*=\s*({[^}]+})', webpage, 'video info'), channel_id,
 181             transform_source=lambda s: js_to_json(re.sub(r'\'\s*\+\s*\'', '', s)))
 182         video_id = remove_start(current_clip_info['ckmId'], 'mvp:')
 183         video_name = url_basename(current_clip_info['url'])
 184
 185         if self._downloader.params.get('noplaylist'):
 186             self.to_screen(
 187                 'Downloading just video %s because of --no-playlist' % video_name)
 188             return self._extract_from_id(video_id, webpage)
 189
 190         self.to_screen(
 191             'Downloading channel %s - add --no-playlist to just download video %s' % (
 192                 channel_id, video_name))
 193         matches = re.findall(
 194             r'<a[^>]+href=[\'"](%s[a-z]+/[0-9a-z-]+/[0-9a-z]+)' % self._URL_BASE_RE,
 195             webpage)
 196         entries = [
 197             self.url_result(video_link, OnetIE.ie_key())
 198             for video_link in matches]
 199
 200         channel_title = strip_or_none(get_element_by_class('o_channelName', webpage))
 201         channel_description = strip_or_none(get_element_by_class('o_channelDesc', webpage))
 202         return self.playlist_result(entries, channel_id, channel_title, channel_description)
 203
 204
 205 class OnetPlIE(InfoExtractor):
 206     _VALID_URL = r'https?://(?:[^/]+\.)?(?:onet|businessinsider\.com|plejada)\.pl/(?:[^/]+/)+(?P<id>[0-9a-z]+)'
 207     IE_NAME = 'onet.pl'
 208
 209     _TESTS = [{
 210         'url': 'http://eurosport.onet.pl/zimowe/skoki-narciarskie/ziobro-wygral-kwalifikacje-w-pjongczangu/9ckrly',
 211         'md5': 'b94021eb56214c3969380388b6e73cb0',
 212         'info_dict': {
 213             'id': '1561707.1685479',
 214             'ext': 'mp4',
 215             'title': 'Ziobro wygrał kwalifikacje w Pjongczangu',
 216             'description': 'md5:61fb0740084d2d702ea96512a03585b4',
 217             'upload_date': '20170214',
 218             'timestamp': 1487078046,
 219         },
 220     }, {
 221         # embedded via pulsembed
 222         'url': 'http://film.onet.pl/pensjonat-nad-rozlewiskiem-relacja-z-planu-serialu/y428n0',
 223         'info_dict': {
 224             'id': '501235.965429946',
 225             'ext': 'mp4',
 226             'title': '"Pensjonat nad rozlewiskiem": relacja z planu serialu',
 227             'upload_date': '20170622',
 228             'timestamp': 1498159955,
 229         },
 230         'params': {
 231             'skip_download': True,
 232         },
 233     }, {
 234         'url': 'http://film.onet.pl/zwiastuny/ghost-in-the-shell-drugi-zwiastun-pl/5q6yl3',
 235         'only_matching': True,
 236     }, {
 237         'url': 'http://moto.onet.pl/jak-wybierane-sa-miejsca-na-fotoradary/6rs04e',
 238         'only_matching': True,
 239     }, {
 240         'url': 'http://businessinsider.com.pl/wideo/scenariusz-na-koniec-swiata-wedlug-nasa/dwnqptk',
 241         'only_matching': True,
 242     }, {
 243         'url': 'http://plejada.pl/weronika-rosati-o-swoim-domniemanym-slubie/n2bq89',
 244         'only_matching': True,
 245     }]
 246
 247     def _search_mvp_id(self, webpage, default=NO_DEFAULT):
 248         return self._search_regex(
 249             r'data-(?:params-)?mvp=["\'](\d+\.\d+)', webpage, 'mvp id',
 250             default=default)
 251
 252     def _real_extract(self, url):
 253         video_id = self._match_id(url)
 254
 255         webpage = self._download_webpage(url, video_id)
 256
 257         mvp_id = self._search_mvp_id(webpage, default=None)
 258
 259         if not mvp_id:
 260             pulsembed_url = self._search_regex(
 261                 r'data-src=(["\'])(?P<url>(?:https?:)?//pulsembed\.eu/.+?)\1',
 262                 webpage, 'pulsembed url', group='url')
 263             webpage = self._download_webpage(
 264                 pulsembed_url, video_id, 'Downloading pulsembed webpage')
 265             mvp_id = self._search_mvp_id(webpage)
 266
 267         return self.url_result(
 268             'onetmvp:%s' % mvp_id, OnetMVPIE.ie_key(), video_id=mvp_id)