yt_dlp/extractor/redtube.py

   1 from .common import InfoExtractor
   2 from ..utils import (
   3     determine_ext,
   4     ExtractorError,
   5     int_or_none,
   6     merge_dicts,
   7     str_to_int,
   8     unified_strdate,
   9     url_or_none,
  10 )
  11
  12
  13 class RedTubeIE(InfoExtractor):
  14     _VALID_URL = r'https?://(?:(?:\w+\.)?redtube\.com/|embed\.redtube\.com/\?.*?\bid=)(?P<id>[0-9]+)'
  15     _EMBED_REGEX = [r'<iframe[^>]+?src=["\'](?P<url>(?:https?:)?//embed\.redtube\.com/\?.*?\bid=\d+)']
  16     _TESTS = [{
  17         'url': 'https://www.redtube.com/38864951',
  18         'md5': '4fba70cbca3aefd25767ab4b523c9878',
  19         'info_dict': {
  20             'id': '38864951',
  21             'ext': 'mp4',
  22             'title': 'Public Sex on the Balcony in Freezing Paris! Amateur Couple LeoLulu',
  23             'description': 'Watch video Public Sex on the Balcony in Freezing Paris! Amateur Couple LeoLulu on Redtube, home of free Blowjob porn videos and Blonde sex movies online. Video length: (10:46) - Uploaded by leolulu - Verified User - Starring Pornstar: Leolulu',
  24             'upload_date': '20210111',
  25             'timestamp': 1610343109,
  26             'duration': 646,
  27             'view_count': int,
  28             'age_limit': 18,
  29             'thumbnail': r're:https://\wi-ph\.rdtcdn\.com/videos/.+/.+\.jpg',
  30         },
  31     }, {
  32         'url': 'http://embed.redtube.com/?bgcolor=000000&id=1443286',
  33         'only_matching': True,
  34     }, {
  35         'url': 'http://it.redtube.com/66418',
  36         'only_matching': True,
  37     }]
  38
  39     def _real_extract(self, url):
  40         video_id = self._match_id(url)
  41         webpage = self._download_webpage(
  42             'http://www.redtube.com/%s' % video_id, video_id)
  43
  44         ERRORS = (
  45             (('video-deleted-info', '>This video has been removed'), 'has been removed'),
  46             (('private_video_text', '>This video is private', '>Send a friend request to its owner to be able to view it'), 'is private'),
  47         )
  48
  49         for patterns, message in ERRORS:
  50             if any(p in webpage for p in patterns):
  51                 raise ExtractorError(
  52                     'Video %s %s' % (video_id, message), expected=True)
  53
  54         info = self._search_json_ld(webpage, video_id, default={})
  55
  56         if not info.get('title'):
  57             info['title'] = self._html_search_regex(
  58                 (r'<h(\d)[^>]+class="(?:video_title_text|videoTitle|video_title)[^"]*">(?P<title>(?:(?!\1).)+)</h\1>',
  59                  r'(?:videoTitle|title)\s*:\s*(["\'])(?P<title>(?:(?!\1).)+)\1',),
  60                 webpage, 'title', group='title',
  61                 default=None) or self._og_search_title(webpage)
  62
  63         formats = []
  64         sources = self._parse_json(
  65             self._search_regex(
  66                 r'sources\s*:\s*({.+?})', webpage, 'source', default='{}'),
  67             video_id, fatal=False)
  68         if sources and isinstance(sources, dict):
  69             for format_id, format_url in sources.items():
  70                 if format_url:
  71                     formats.append({
  72                         'url': format_url,
  73                         'format_id': format_id,
  74                         'height': int_or_none(format_id),
  75                     })
  76         medias = self._parse_json(
  77             self._search_regex(
  78                 r'mediaDefinition["\']?\s*:\s*(\[.+?}\s*\])', webpage,
  79                 'media definitions', default='{}'),
  80             video_id, fatal=False)
  81         for media in medias if isinstance(medias, list) else []:
  82             format_url = url_or_none(media.get('videoUrl'))
  83             if not format_url:
  84                 continue
  85             format_id = media.get('format')
  86             quality = media.get('quality')
  87             if format_id == 'hls' or (format_id == 'mp4' and not quality):
  88                 more_media = self._download_json(format_url, video_id, fatal=False)
  89             else:
  90                 more_media = [media]
  91             for media in more_media if isinstance(more_media, list) else []:
  92                 format_url = url_or_none(media.get('videoUrl'))
  93                 if not format_url:
  94                     continue
  95                 format_id = media.get('format')
  96                 if format_id == 'hls' or determine_ext(format_url) == 'm3u8':
  97                     formats.extend(self._extract_m3u8_formats(
  98                         format_url, video_id, 'mp4',
  99                         entry_protocol='m3u8_native', m3u8_id=format_id or 'hls',
 100                         fatal=False))
 101                     continue
 102                 format_id = media.get('quality')
 103                 formats.append({
 104                     'url': format_url,
 105                     'ext': 'mp4',
 106                     'format_id': format_id,
 107                     'height': int_or_none(format_id),
 108                 })
 109         if not formats:
 110             video_url = self._html_search_regex(
 111                 r'<source src="(.+?)" type="video/mp4">', webpage, 'video URL')
 112             formats.append({'url': video_url, 'ext': 'mp4'})
 113         self._sort_formats(formats)
 114
 115         thumbnail = self._og_search_thumbnail(webpage)
 116         upload_date = unified_strdate(self._search_regex(
 117             r'<span[^>]+>(?:ADDED|Published on) ([^<]+)<',
 118             webpage, 'upload date', default=None))
 119         duration = int_or_none(self._og_search_property(
 120             'video:duration', webpage, default=None) or self._search_regex(
 121                 r'videoDuration\s*:\s*(\d+)', webpage, 'duration', default=None))
 122         view_count = str_to_int(self._search_regex(
 123             (r'<div[^>]*>Views</div>\s*<div[^>]*>\s*([\d,.]+)',
 124              r'<span[^>]*>VIEWS</span>\s*</td>\s*<td>\s*([\d,.]+)',
 125              r'<span[^>]+\bclass=["\']video_view_count[^>]*>\s*([\d,.]+)'),
 126             webpage, 'view count', default=None))
 127
 128         # No self-labeling, but they describe themselves as
 129         # "Home of Videos Porno"
 130         age_limit = 18
 131
 132         return merge_dicts(info, {
 133             'id': video_id,
 134             'ext': 'mp4',
 135             'thumbnail': thumbnail,
 136             'upload_date': upload_date,
 137             'duration': duration,
 138             'view_count': view_count,
 139             'age_limit': age_limit,
 140             'formats': formats,
 141         })