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

   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_urlparse,
   9     compat_urlparse,
  10 )
  11 from ..utils import (
  12     orderedSet,
  13     remove_end,
  14     extract_attributes,
  15     mimetype2ext,
  16     determine_ext,
  17     int_or_none,
  18     parse_iso8601,
  19 )
  20
  21
  22 class CondeNastIE(InfoExtractor):
  23     """
  24     Condé Nast is a media group, some of its sites use a custom HTML5 player
  25     that works the same in all of them.
  26     """
  27
  28     # The keys are the supported sites and the values are the name to be shown
  29     # to the user and in the extractor description.
  30     _SITES = {
  31         'allure': 'Allure',
  32         'architecturaldigest': 'Architectural Digest',
  33         'arstechnica': 'Ars Technica',
  34         'bonappetit': 'Bon Appétit',
  35         'brides': 'Brides',
  36         'cnevids': 'Condé Nast',
  37         'cntraveler': 'Condé Nast Traveler',
  38         'details': 'Details',
  39         'epicurious': 'Epicurious',
  40         'glamour': 'Glamour',
  41         'golfdigest': 'Golf Digest',
  42         'gq': 'GQ',
  43         'newyorker': 'The New Yorker',
  44         'self': 'SELF',
  45         'teenvogue': 'Teen Vogue',
  46         'vanityfair': 'Vanity Fair',
  47         'vogue': 'Vogue',
  48         'wired': 'WIRED',
  49         'wmagazine': 'W Magazine',
  50     }
  51
  52     _VALID_URL = r'https?://(?:video|www|player)\.(?P<site>%s)\.com/(?P<type>watch|series|video|embed(?:js)?)/(?P<id>[^/?#]+)' % '|'.join(_SITES.keys())
  53     IE_DESC = 'Condé Nast media group: %s' % ', '.join(sorted(_SITES.values()))
  54
  55     EMBED_URL = r'(?:https?:)?//player\.(?P<site>%s)\.com/(?P<type>embed(?:js)?)/.+?' % '|'.join(_SITES.keys())
  56
  57     _TESTS = [{
  58         'url': 'http://video.wired.com/watch/3d-printed-speakers-lit-with-led',
  59         'md5': '1921f713ed48aabd715691f774c451f7',
  60         'info_dict': {
  61             'id': '5171b343c2b4c00dd0c1ccb3',
  62             'ext': 'mp4',
  63             'title': '3D Printed Speakers Lit With LED',
  64             'description': 'Check out these beautiful 3D printed LED speakers.  You can\'t actually buy them, but LumiGeek is working on a board that will let you make you\'re own.',
  65             'uploader': 'wired',
  66             'upload_date': '20130314',
  67             'timestamp': 1363219200,
  68         }
  69     }, {
  70         # JS embed
  71         'url': 'http://player.cnevids.com/embedjs/55f9cf8b61646d1acf00000c/5511d76261646d5566020000.js',
  72         'md5': 'f1a6f9cafb7083bab74a710f65d08999',
  73         'info_dict': {
  74             'id': '55f9cf8b61646d1acf00000c',
  75             'ext': 'mp4',
  76             'title': '3D printed TSA Travel Sentry keys really do open TSA locks',
  77             'uploader': 'arstechnica',
  78             'upload_date': '20150916',
  79             'timestamp': 1442434955,
  80         }
  81     }]
  82
  83     def _extract_series(self, url, webpage):
  84         title = self._html_search_regex(
  85             r'(?s)<div class="cne-series-info">.*?<h1>(.+?)</h1>',
  86             webpage, 'series title')
  87         url_object = compat_urllib_parse_urlparse(url)
  88         base_url = '%s://%s' % (url_object.scheme, url_object.netloc)
  89         m_paths = re.finditer(
  90             r'(?s)<p class="cne-thumb-title">.*?<a href="(/watch/.+?)["\?]', webpage)
  91         paths = orderedSet(m.group(1) for m in m_paths)
  92         build_url = lambda path: compat_urlparse.urljoin(base_url, path)
  93         entries = [self.url_result(build_url(path), 'CondeNast') for path in paths]
  94         return self.playlist_result(entries, playlist_title=title)
  95
  96     def _extract_video(self, webpage, url_type):
  97         query = {}
  98         params = self._search_regex(
  99             r'(?s)var params = {(.+?)}[;,]', webpage, 'player params', default=None)
 100         if params:
 101             query.update({
 102                 'videoId': self._search_regex(r'videoId: [\'"](.+?)[\'"]', params, 'video id'),
 103                 'playerId': self._search_regex(r'playerId: [\'"](.+?)[\'"]', params, 'player id'),
 104                 'target': self._search_regex(r'target: [\'"](.+?)[\'"]', params, 'target'),
 105             })
 106         else:
 107             params = extract_attributes(self._search_regex(
 108                 r'(<[^>]+data-js="video-player"[^>]+>)',
 109                 webpage, 'player params element'))
 110             query.update({
 111                 'videoId': params['data-video'],
 112                 'playerId': params['data-player'],
 113                 'target': params['id'],
 114             })
 115         video_id = query['videoId']
 116         video_info = None
 117         info_page = self._download_webpage(
 118             'http://player.cnevids.com/player/video.js',
 119             video_id, 'Downloading video info', query=query, fatal=False)
 120         if info_page:
 121             video_info = self._parse_json(self._search_regex(
 122                 r'loadCallback\(({.+})\)', info_page, 'video info'), video_id)['video']
 123         else:
 124             info_page = self._download_webpage(
 125                 'http://player.cnevids.com/player/loader.js',
 126                 video_id, 'Downloading loader info', query=query)
 127             video_info = self._parse_json(self._search_regex(
 128                 r'var\s+video\s*=\s*({.+?});', info_page, 'video info'), video_id)
 129         title = video_info['title']
 130
 131         formats = []
 132         for fdata in video_info.get('sources', [{}])[0]:
 133             src = fdata.get('src')
 134             if not src:
 135                 continue
 136             ext = mimetype2ext(fdata.get('type')) or determine_ext(src)
 137             quality = fdata.get('quality')
 138             formats.append({
 139                 'format_id': ext + ('-%s' % quality if quality else ''),
 140                 'url': src,
 141                 'ext': ext,
 142                 'quality': 1 if quality == 'high' else 0,
 143             })
 144         self._sort_formats(formats)
 145
 146         info = self._search_json_ld(
 147             webpage, video_id, fatal=False) if url_type != 'embed' else {}
 148         info.update({
 149             'id': video_id,
 150             'formats': formats,
 151             'title': title,
 152             'thumbnail': video_info.get('poster_frame'),
 153             'uploader': video_info.get('brand'),
 154             'duration': int_or_none(video_info.get('duration')),
 155             'tags': video_info.get('tags'),
 156             'series': video_info.get('series_title'),
 157             'season': video_info.get('season_title'),
 158             'timestamp': parse_iso8601(video_info.get('premiere_date')),
 159         })
 160         return info
 161
 162     def _real_extract(self, url):
 163         site, url_type, item_id = re.match(self._VALID_URL, url).groups()
 164
 165         # Convert JS embed to regular embed
 166         if url_type == 'embedjs':
 167             parsed_url = compat_urlparse.urlparse(url)
 168             url = compat_urlparse.urlunparse(parsed_url._replace(
 169                 path=remove_end(parsed_url.path, '.js').replace('/embedjs/', '/embed/')))
 170             url_type = 'embed'
 171
 172         self.to_screen('Extracting from %s with the Condé Nast extractor' % self._SITES[site])
 173         webpage = self._download_webpage(url, item_id)
 174
 175         if url_type == 'series':
 176             return self._extract_series(url, webpage)
 177         else:
 178             return self._extract_video(webpage, url_type)