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     def _search_mvp_id(self, webpage):
  24         return self._search_regex(
  25             r'id=(["\'])mvp:(?P<id>.+?)\1', webpage, 'mvp id', group='id')
  26
  27     def _extract_from_id(self, video_id, webpage=None):
  28         response = self._download_json(
  29             'http://qi.ckm.onetapi.pl/', video_id,
  30             query={
  31                 'body[id]': video_id,
  32                 'body[jsonrpc]': '2.0',
  33                 'body[method]': 'get_asset_detail',
  34                 'body[params][ID_Publikacji]': video_id,
  35                 'body[params][Service]': 'www.onet.pl',
  36                 'content-type': 'application/jsonp',
  37                 'x-onet-app': 'player.front.onetapi.pl',
  38             })
  39
  40         error = response.get('error')
  41         if error:
  42             raise ExtractorError(
  43                 '%s said: %s' % (self.IE_NAME, error['message']), expected=True)
  44
  45         video = response['result'].get('0')
  46
  47         formats = []
  48         for _, formats_dict in video['formats'].items():
  49             if not isinstance(formats_dict, dict):
  50                 continue
  51             for format_id, format_list in formats_dict.items():
  52                 if not isinstance(format_list, list):
  53                     continue
  54                 for f in format_list:
  55                     video_url = f.get('url')
  56                     if not video_url:
  57                         continue
  58                     ext = determine_ext(video_url)
  59                     if format_id == 'ism':
  60                         formats.extend(self._extract_ism_formats(
  61                             video_url, video_id, 'mss', fatal=False))
  62                     elif ext == 'mpd':
  63                         formats.extend(self._extract_mpd_formats(
  64                             video_url, video_id, mpd_id='dash', fatal=False))
  65                     else:
  66                         formats.append({
  67                             'url': video_url,
  68                             'format_id': format_id,
  69                             'height': int_or_none(f.get('vertical_resolution')),
  70                             'width': int_or_none(f.get('horizontal_resolution')),
  71                             'abr': float_or_none(f.get('audio_bitrate')),
  72                             'vbr': float_or_none(f.get('video_bitrate')),
  73                         })
  74         self._sort_formats(formats)
  75
  76         meta = video.get('meta', {})
  77
  78         title = (self._og_search_title(
  79             webpage, default=None) if webpage else None) or meta['title']
  80         description = (self._og_search_description(
  81             webpage, default=None) if webpage else None) or meta.get('description')
  82         duration = meta.get('length') or meta.get('lenght')
  83         timestamp = parse_iso8601(meta.get('addDate'), ' ')
  84
  85         return {
  86             'id': video_id,
  87             'title': title,
  88             'description': description,
  89             'duration': duration,
  90             'timestamp': timestamp,
  91             'formats': formats,
  92         }
  93
  94
  95 class OnetMVPIE(OnetBaseIE):
  96     _VALID_URL = r'onetmvp:(?P<id>\d+\.\d+)'
  97
  98     _TEST = {
  99         'url': 'onetmvp:381027.1509591944',
 100         'only_matching': True,
 101     }
 102
 103     def _real_extract(self, url):
 104         return self._extract_from_id(self._match_id(url))
 105
 106
 107 class OnetIE(OnetBaseIE):
 108     _VALID_URL = r'https?://(?:www\.)?onet\.tv/[a-z]/[a-z]+/(?P<display_id>[0-9a-z-]+)/(?P<id>[0-9a-z]+)'
 109     IE_NAME = 'onet.tv'
 110
 111     _TEST = {
 112         'url': 'http://onet.tv/k/openerfestival/open-er-festival-2016-najdziwniejsze-wymagania-gwiazd/qbpyqc',
 113         'md5': 'e3ffbf47590032ac3f27249204173d50',
 114         'info_dict': {
 115             'id': 'qbpyqc',
 116             'display_id': 'open-er-festival-2016-najdziwniejsze-wymagania-gwiazd',
 117             'ext': 'mp4',
 118             'title': 'Open\'er Festival 2016: najdziwniejsze wymagania gwiazd',
 119             '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...',
 120             'upload_date': '20160705',
 121             'timestamp': 1467721580,
 122         },
 123     }
 124
 125     def _real_extract(self, url):
 126         mobj = re.match(self._VALID_URL, url)
 127         display_id, video_id = mobj.group('display_id', 'id')
 128
 129         webpage = self._download_webpage(url, display_id)
 130
 131         mvp_id = self._search_mvp_id(webpage)
 132
 133         info_dict = self._extract_from_id(mvp_id, webpage)
 134         info_dict.update({
 135             'id': video_id,
 136             'display_id': display_id,
 137         })
 138
 139         return info_dict
 140
 141
 142 class OnetChannelIE(OnetBaseIE):
 143     _VALID_URL = r'https?://(?:www\.)?onet\.tv/[a-z]/(?P<id>[a-z]+)(?:[?#]|$)'
 144     IE_NAME = 'onet.tv:channel'
 145
 146     _TEST = {
 147         'url': 'http://onet.tv/k/openerfestival',
 148         'info_dict': {
 149             'id': 'openerfestival',
 150             'title': 'Open\'er Festival Live',
 151             'description': 'Dziękujemy, że oglądaliście transmisje. Zobaczcie nasze relacje i wywiady z artystami.',
 152         },
 153         'playlist_mincount': 46,
 154     }
 155
 156     def _real_extract(self, url):
 157         channel_id = self._match_id(url)
 158
 159         webpage = self._download_webpage(url, channel_id)
 160
 161         current_clip_info = self._parse_json(self._search_regex(
 162             r'var\s+currentClip\s*=\s*({[^}]+})', webpage, 'video info'), channel_id,
 163             transform_source=lambda s: js_to_json(re.sub(r'\'\s*\+\s*\'', '', s)))
 164         video_id = remove_start(current_clip_info['ckmId'], 'mvp:')
 165         video_name = url_basename(current_clip_info['url'])
 166
 167         if self._downloader.params.get('noplaylist'):
 168             self.to_screen(
 169                 'Downloading just video %s because of --no-playlist' % video_name)
 170             return self._extract_from_id(video_id, webpage)
 171
 172         self.to_screen(
 173             'Downloading channel %s - add --no-playlist to just download video %s' % (
 174                 channel_id, video_name))
 175         matches = re.findall(
 176             r'<a[^>]+href=[\'"](https?://(?:www\.)?onet\.tv/[a-z]/[a-z]+/[0-9a-z-]+/[0-9a-z]+)',
 177             webpage)
 178         entries = [
 179             self.url_result(video_link, OnetIE.ie_key())
 180             for video_link in matches]
 181
 182         channel_title = strip_or_none(get_element_by_class('o_channelName', webpage))
 183         channel_description = strip_or_none(get_element_by_class('o_channelDesc', webpage))
 184         return self.playlist_result(entries, channel_id, channel_title, channel_description)
 185
 186
 187 class OnetPlIE(InfoExtractor):
 188     _VALID_URL = r'https?://(?:[^/]+\.)?(?:onet|businessinsider\.com|plejada)\.pl/(?:[^/]+/)+(?P<id>[0-9a-z]+)'
 189     IE_NAME = 'onet.pl'
 190
 191     _TESTS = [{
 192         'url': 'http://eurosport.onet.pl/zimowe/skoki-narciarskie/ziobro-wygral-kwalifikacje-w-pjongczangu/9ckrly',
 193         'md5': 'b94021eb56214c3969380388b6e73cb0',
 194         'info_dict': {
 195             'id': '1561707.1685479',
 196             'ext': 'mp4',
 197             'title': 'Ziobro wygrał kwalifikacje w Pjongczangu',
 198             'description': 'md5:61fb0740084d2d702ea96512a03585b4',
 199             'upload_date': '20170214',
 200             'timestamp': 1487078046,
 201         },
 202     }, {
 203         # embedded via pulsembed
 204         'url': 'http://film.onet.pl/pensjonat-nad-rozlewiskiem-relacja-z-planu-serialu/y428n0',
 205         'info_dict': {
 206             'id': '501235.965429946',
 207             'ext': 'mp4',
 208             'title': '"Pensjonat nad rozlewiskiem": relacja z planu serialu',
 209             'upload_date': '20170622',
 210             'timestamp': 1498159955,
 211         },
 212         'params': {
 213             'skip_download': True,
 214         },
 215     }, {
 216         'url': 'http://film.onet.pl/zwiastuny/ghost-in-the-shell-drugi-zwiastun-pl/5q6yl3',
 217         'only_matching': True,
 218     }, {
 219         'url': 'http://moto.onet.pl/jak-wybierane-sa-miejsca-na-fotoradary/6rs04e',
 220         'only_matching': True,
 221     }, {
 222         'url': 'http://businessinsider.com.pl/wideo/scenariusz-na-koniec-swiata-wedlug-nasa/dwnqptk',
 223         'only_matching': True,
 224     }, {
 225         'url': 'http://plejada.pl/weronika-rosati-o-swoim-domniemanym-slubie/n2bq89',
 226         'only_matching': True,
 227     }]
 228
 229     def _search_mvp_id(self, webpage, default=NO_DEFAULT):
 230         return self._search_regex(
 231             r'data-(?:params-)?mvp=["\'](\d+\.\d+)', webpage, 'mvp id',
 232             default=default)
 233
 234     def _real_extract(self, url):
 235         video_id = self._match_id(url)
 236
 237         webpage = self._download_webpage(url, video_id)
 238
 239         mvp_id = self._search_mvp_id(webpage, default=None)
 240
 241         if not mvp_id:
 242             pulsembed_url = self._search_regex(
 243                 r'data-src=(["\'])(?P<url>(?:https?:)?//pulsembed\.eu/.+?)\1',
 244                 webpage, 'pulsembed url', group='url')
 245             webpage = self._download_webpage(
 246                 pulsembed_url, video_id, 'Downloading pulsembed webpage')
 247             mvp_id = self._search_mvp_id(webpage)
 248
 249         return self.url_result(
 250             'onetmvp:%s' % mvp_id, OnetMVPIE.ie_key(), video_id=mvp_id)