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,
   9     compat_urllib_parse_urlparse,
  10     compat_urlparse,
  11 )
  12 from ..utils import (
  13     orderedSet,
  14     remove_end,
  15 )
  16
  17
  18 class CondeNastIE(InfoExtractor):
  19     """
  20     Condé Nast is a media group, some of its sites use a custom HTML5 player
  21     that works the same in all of them.
  22     """
  23
  24     # The keys are the supported sites and the values are the name to be shown
  25     # to the user and in the extractor description.
  26     _SITES = {
  27         'allure': 'Allure',
  28         'architecturaldigest': 'Architectural Digest',
  29         'arstechnica': 'Ars Technica',
  30         'bonappetit': 'Bon Appétit',
  31         'brides': 'Brides',
  32         'cnevids': 'Condé Nast',
  33         'cntraveler': 'Condé Nast Traveler',
  34         'details': 'Details',
  35         'epicurious': 'Epicurious',
  36         'glamour': 'Glamour',
  37         'golfdigest': 'Golf Digest',
  38         'gq': 'GQ',
  39         'newyorker': 'The New Yorker',
  40         'self': 'SELF',
  41         'teenvogue': 'Teen Vogue',
  42         'vanityfair': 'Vanity Fair',
  43         'vogue': 'Vogue',
  44         'wired': 'WIRED',
  45         'wmagazine': 'W Magazine',
  46     }
  47
  48     _VALID_URL = r'http://(?:video|www|player)\.(?P<site>%s)\.com/(?P<type>watch|series|video|embed(?:js)?)/(?P<id>[^/?#]+)' % '|'.join(_SITES.keys())
  49     IE_DESC = 'Condé Nast media group: %s' % ', '.join(sorted(_SITES.values()))
  50
  51     EMBED_URL = r'(?:https?:)?//player\.(?P<site>%s)\.com/(?P<type>embed(?:js)?)/.+?' % '|'.join(_SITES.keys())
  52
  53     _TESTS = [{
  54         'url': 'http://video.wired.com/watch/3d-printed-speakers-lit-with-led',
  55         'md5': '1921f713ed48aabd715691f774c451f7',
  56         'info_dict': {
  57             'id': '5171b343c2b4c00dd0c1ccb3',
  58             'ext': 'mp4',
  59             'title': '3D Printed Speakers Lit With LED',
  60             '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.',
  61         }
  62     }, {
  63         # JS embed
  64         'url': 'http://player.cnevids.com/embedjs/55f9cf8b61646d1acf00000c/5511d76261646d5566020000.js',
  65         'md5': 'f1a6f9cafb7083bab74a710f65d08999',
  66         'info_dict': {
  67             'id': '55f9cf8b61646d1acf00000c',
  68             'ext': 'mp4',
  69             'title': '3D printed TSA Travel Sentry keys really do open TSA locks',
  70         }
  71     }]
  72
  73     def _extract_series(self, url, webpage):
  74         title = self._html_search_regex(r'<div class="cne-series-info">.*?<h1>(.+?)</h1>',
  75                                         webpage, 'series title', flags=re.DOTALL)
  76         url_object = compat_urllib_parse_urlparse(url)
  77         base_url = '%s://%s' % (url_object.scheme, url_object.netloc)
  78         m_paths = re.finditer(r'<p class="cne-thumb-title">.*?<a href="(/watch/.+?)["\?]',
  79                               webpage, flags=re.DOTALL)
  80         paths = orderedSet(m.group(1) for m in m_paths)
  81         build_url = lambda path: compat_urlparse.urljoin(base_url, path)
  82         entries = [self.url_result(build_url(path), 'CondeNast') for path in paths]
  83         return self.playlist_result(entries, playlist_title=title)
  84
  85     def _extract_video(self, webpage, url_type):
  86         if url_type != 'embed':
  87             description = self._html_search_regex(
  88                 [
  89                     r'<div class="cne-video-description">(.+?)</div>',
  90                     r'<div class="video-post-content">(.+?)</div>',
  91                 ],
  92                 webpage, 'description', fatal=False, flags=re.DOTALL)
  93         else:
  94             description = None
  95         params = self._search_regex(r'var params = {(.+?)}[;,]', webpage,
  96                                     'player params', flags=re.DOTALL)
  97         video_id = self._search_regex(r'videoId: [\'"](.+?)[\'"]', params, 'video id')
  98         player_id = self._search_regex(r'playerId: [\'"](.+?)[\'"]', params, 'player id')
  99         target = self._search_regex(r'target: [\'"](.+?)[\'"]', params, 'target')
 100         data = compat_urllib_parse.urlencode({'videoId': video_id,
 101                                               'playerId': player_id,
 102                                               'target': target,
 103                                               })
 104         base_info_url = self._search_regex(r'url = [\'"](.+?)[\'"][,;]',
 105                                            webpage, 'base info url',
 106                                            default='http://player.cnevids.com/player/loader.js?')
 107         info_url = base_info_url + data
 108         info_page = self._download_webpage(info_url, video_id,
 109                                            'Downloading video info')
 110         video_info = self._search_regex(r'var\s+video\s*=\s*({.+?});', info_page, 'video info')
 111         video_info = self._parse_json(video_info, video_id)
 112
 113         formats = [{
 114             'format_id': '%s-%s' % (fdata['type'].split('/')[-1], fdata['quality']),
 115             'url': fdata['src'],
 116             'ext': fdata['type'].split('/')[-1],
 117             'quality': 1 if fdata['quality'] == 'high' else 0,
 118         } for fdata in video_info['sources'][0]]
 119         self._sort_formats(formats)
 120
 121         return {
 122             'id': video_id,
 123             'formats': formats,
 124             'title': video_info['title'],
 125             'thumbnail': video_info['poster_frame'],
 126             'description': description,
 127         }
 128
 129     def _real_extract(self, url):
 130         mobj = re.match(self._VALID_URL, url)
 131         site = mobj.group('site')
 132         url_type = mobj.group('type')
 133         item_id = mobj.group('id')
 134
 135         # Convert JS embed to regular embed
 136         if url_type == 'embedjs':
 137             parsed_url = compat_urlparse.urlparse(url)
 138             url = compat_urlparse.urlunparse(parsed_url._replace(
 139                 path=remove_end(parsed_url.path, '.js').replace('/embedjs/', '/embed/')))
 140             url_type = 'embed'
 141
 142         self.to_screen('Extracting from %s with the Condé Nast extractor' % self._SITES[site])
 143         webpage = self._download_webpage(url, item_id)
 144
 145         if url_type == 'series':
 146             return self._extract_series(url, webpage)
 147         else:
 148             return self._extract_video(webpage, url_type)