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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .srgssr import SRGSSRIE
   7 from ..compat import compat_str
   8 from ..utils import (
   9     int_or_none,
  10     parse_duration,
  11     parse_iso8601,
  12     unescapeHTML,
  13     determine_ext,
  14 )
  15
  16
  17 class RTSIE(SRGSSRIE):
  18     IE_DESC = 'RTS.ch'
  19     _VALID_URL = r'rts:(?P<rts_id>\d+)|https?://(?:.+?\.)?rts\.ch/(?:[^/]+/){2,}(?P<id>[0-9]+)-(?P<display_id>.+?)\.html'
  20
  21     _TESTS = [
  22         {
  23             'url': 'http://www.rts.ch/archives/tv/divers/3449373-les-enfants-terribles.html',
  24             'md5': 'ff7f8450a90cf58dacb64e29707b4a8e',
  25             'info_dict': {
  26                 'id': '3449373',
  27                 'display_id': 'les-enfants-terribles',
  28                 'ext': 'mp4',
  29                 'duration': 1488,
  30                 'title': 'Les Enfants Terribles',
  31                 'description': 'France Pommier et sa soeur Luce Feral, les deux filles de ce groupe de 5.',
  32                 'uploader': 'Divers',
  33                 'upload_date': '19680921',
  34                 'timestamp': -40280400,
  35                 'thumbnail': r're:^https?://.*\.image',
  36                 'view_count': int,
  37             },
  38         },
  39         {
  40             'url': 'http://www.rts.ch/emissions/passe-moi-les-jumelles/5624067-entre-ciel-et-mer.html',
  41             'info_dict': {
  42                 'id': '5624065',
  43                 'title': 'Passe-moi les jumelles',
  44             },
  45             'playlist_mincount': 4,
  46         },
  47         {
  48             'url': 'http://www.rts.ch/video/sport/hockey/5745975-1-2-kloten-fribourg-5-2-second-but-pour-gotteron-par-kwiatowski.html',
  49             'info_dict': {
  50                 'id': '5745975',
  51                 'display_id': '1-2-kloten-fribourg-5-2-second-but-pour-gotteron-par-kwiatowski',
  52                 'ext': 'mp4',
  53                 'duration': 48,
  54                 'title': '1/2, Kloten - Fribourg (5-2): second but pour Gottéron par Kwiatowski',
  55                 'description': 'Hockey - Playoff',
  56                 'uploader': 'Hockey',
  57                 'upload_date': '20140403',
  58                 'timestamp': 1396556882,
  59                 'thumbnail': r're:^https?://.*\.image',
  60                 'view_count': int,
  61             },
  62             'params': {
  63                 # m3u8 download
  64                 'skip_download': True,
  65             },
  66             'skip': 'Blocked outside Switzerland',
  67         },
  68         {
  69             'url': 'http://www.rts.ch/video/info/journal-continu/5745356-londres-cachee-par-un-epais-smog.html',
  70             'md5': '1bae984fe7b1f78e94abc74e802ed99f',
  71             'info_dict': {
  72                 'id': '5745356',
  73                 'display_id': 'londres-cachee-par-un-epais-smog',
  74                 'ext': 'mp4',
  75                 'duration': 33,
  76                 'title': 'Londres cachée par un épais smog',
  77                 'description': 'Un important voile de smog recouvre Londres depuis mercredi, provoqué par la pollution et du sable du Sahara.',
  78                 'uploader': 'L\'actu en vidéo',
  79                 'upload_date': '20140403',
  80                 'timestamp': 1396537322,
  81                 'thumbnail': r're:^https?://.*\.image',
  82                 'view_count': int,
  83             },
  84         },
  85         {
  86             'url': 'http://www.rts.ch/audio/couleur3/programmes/la-belle-video-de-stephane-laurenceau/5706148-urban-hippie-de-damien-krisl-03-04-2014.html',
  87             'md5': 'dd8ef6a22dff163d063e2a52bc8adcae',
  88             'info_dict': {
  89                 'id': '5706148',
  90                 'display_id': 'urban-hippie-de-damien-krisl-03-04-2014',
  91                 'ext': 'mp3',
  92                 'duration': 123,
  93                 'title': '"Urban Hippie", de Damien Krisl',
  94                 'description': 'Des Hippies super glam.',
  95                 'upload_date': '20140403',
  96                 'timestamp': 1396551600,
  97             },
  98         },
  99         {
 100             # article with videos on rhs
 101             'url': 'http://www.rts.ch/sport/hockey/6693917-hockey-davos-decroche-son-31e-titre-de-champion-de-suisse.html',
 102             'info_dict': {
 103                 'id': '6693917',
 104                 'title': 'Hockey: Davos décroche son 31e titre de champion de Suisse',
 105             },
 106             'playlist_mincount': 5,
 107         },
 108         {
 109             'url': 'http://pages.rts.ch/emissions/passe-moi-les-jumelles/5624065-entre-ciel-et-mer.html',
 110             'only_matching': True,
 111         }
 112     ]
 113
 114     def _real_extract(self, url):
 115         m = re.match(self._VALID_URL, url)
 116         media_id = m.group('rts_id') or m.group('id')
 117         display_id = m.group('display_id') or media_id
 118
 119         def download_json(internal_id):
 120             return self._download_json(
 121                 'http://www.rts.ch/a/%s.html?f=json/article' % internal_id,
 122                 display_id)
 123
 124         all_info = download_json(media_id)
 125
 126         # media_id extracted out of URL is not always a real id
 127         if 'video' not in all_info and 'audio' not in all_info:
 128             entries = []
 129
 130             for item in all_info.get('items', []):
 131                 item_url = item.get('url')
 132                 if not item_url:
 133                     continue
 134                 entries.append(self.url_result(item_url, 'RTS'))
 135
 136             if not entries:
 137                 page, urlh = self._download_webpage_handle(url, display_id)
 138                 if re.match(self._VALID_URL, urlh.geturl()).group('id') != media_id:
 139                     return self.url_result(urlh.geturl(), 'RTS')
 140
 141                 # article with videos on rhs
 142                 videos = re.findall(
 143                     r'<article[^>]+class="content-item"[^>]*>\s*<a[^>]+data-video-urn="urn:([^"]+)"',
 144                     page)
 145                 if not videos:
 146                     videos = re.findall(
 147                         r'(?s)<iframe[^>]+class="srg-player"[^>]+src="[^"]+urn:([^"]+)"',
 148                         page)
 149                 if videos:
 150                     entries = [self.url_result('srgssr:%s' % video_urn, 'SRGSSR') for video_urn in videos]
 151
 152             if entries:
 153                 return self.playlist_result(entries, media_id, all_info.get('title'))
 154
 155             internal_id = self._html_search_regex(
 156                 r'<(?:video|audio) data-id="([0-9]+)"', page,
 157                 'internal video id')
 158             all_info = download_json(internal_id)
 159
 160         media_type = 'video' if 'video' in all_info else 'audio'
 161
 162         # check for errors
 163         self.get_media_data('rts', media_type, media_id)
 164
 165         info = all_info['video']['JSONinfo'] if 'video' in all_info else all_info['audio']
 166
 167         title = info['title']
 168
 169         def extract_bitrate(url):
 170             return int_or_none(self._search_regex(
 171                 r'-([0-9]+)k\.', url, 'bitrate', default=None))
 172
 173         formats = []
 174         streams = info.get('streams', {})
 175         for format_id, format_url in streams.items():
 176             if format_id == 'hds_sd' and 'hds' in streams:
 177                 continue
 178             if format_id == 'hls_sd' and 'hls' in streams:
 179                 continue
 180             ext = determine_ext(format_url)
 181             if ext in ('m3u8', 'f4m'):
 182                 format_url = self._get_tokenized_src(format_url, media_id, format_id)
 183                 if ext == 'f4m':
 184                     formats.extend(self._extract_f4m_formats(
 185                         format_url + ('?' if '?' not in format_url else '&') + 'hdcore=3.4.0',
 186                         media_id, f4m_id=format_id, fatal=False))
 187                 else:
 188                     formats.extend(self._extract_m3u8_formats(
 189                         format_url, media_id, 'mp4', 'm3u8_native', m3u8_id=format_id, fatal=False))
 190             else:
 191                 formats.append({
 192                     'format_id': format_id,
 193                     'url': format_url,
 194                     'tbr': extract_bitrate(format_url),
 195                 })
 196
 197         for media in info.get('media', []):
 198             media_url = media.get('url')
 199             if not media_url or re.match(r'https?://', media_url):
 200                 continue
 201             rate = media.get('rate')
 202             ext = media.get('ext') or determine_ext(media_url, 'mp4')
 203             format_id = ext
 204             if rate:
 205                 format_id += '-%dk' % rate
 206             formats.append({
 207                 'format_id': format_id,
 208                 'url': 'http://download-video.rts.ch/' + media_url,
 209                 'tbr': rate or extract_bitrate(media_url),
 210             })
 211
 212         self._check_formats(formats, media_id)
 213         self._sort_formats(formats)
 214
 215         duration = info.get('duration') or info.get('cutout') or info.get('cutduration')
 216         if isinstance(duration, compat_str):
 217             duration = parse_duration(duration)
 218
 219         return {
 220             'id': media_id,
 221             'display_id': display_id,
 222             'formats': formats,
 223             'title': title,
 224             'description': info.get('intro'),
 225             'duration': duration,
 226             'view_count': int_or_none(info.get('plays')),
 227             'uploader': info.get('programName'),
 228             'timestamp': parse_iso8601(info.get('broadcast_date')),
 229             'thumbnail': unescapeHTML(info.get('preview_image_url')),
 230         }