yt_dlp/extractor/redtube.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     determine_ext,
   6     ExtractorError,
   7     int_or_none,
   8     merge_dicts,
   9     str_to_int,
  10     unified_strdate,
  11     url_or_none,
  12 )
  13
  14
  15 class RedTubeIE(InfoExtractor):
  16     _VALID_URL = r'https?://(?:(?:\w+\.)?redtube\.com/|embed\.redtube\.com/\?.*?\bid=)(?P<id>[0-9]+)'
  17     _TESTS = [{
  18         'url': 'https://www.redtube.com/38864951',
  19         'md5': '4fba70cbca3aefd25767ab4b523c9878',
  20         'info_dict': {
  21             'id': '38864951',
  22             'ext': 'mp4',
  23             'title': 'Public Sex on the Balcony in Freezing Paris! Amateur Couple LeoLulu',
  24             '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',
  25             'upload_date': '20210111',
  26             'timestamp': 1610343109,
  27             'duration': 646,
  28             'view_count': int,
  29             'age_limit': 18,
  30             'thumbnail': r're:https://\wi-ph\.rdtcdn\.com/videos/.+/.+\.jpg',
  31         },
  32     }, {
  33         'url': 'http://embed.redtube.com/?bgcolor=000000&id=1443286',
  34         'only_matching': True,
  35     }, {
  36         'url': 'http://it.redtube.com/66418',
  37         'only_matching': True,
  38     }]
  39
  40     @staticmethod
  41     def _extract_urls(webpage):
  42         return re.findall(
  43             r'<iframe[^>]+?src=["\'](?P<url>(?:https?:)?//embed\.redtube\.com/\?.*?\bid=\d+)',
  44             webpage)
  45
  46     def _real_extract(self, url):
  47         video_id = self._match_id(url)
  48         webpage = self._download_webpage(
  49             'http://www.redtube.com/%s' % video_id, video_id)
  50
  51         ERRORS = (
  52             (('video-deleted-info', '>This video has been removed'), 'has been removed'),
  53             (('private_video_text', '>This video is private', '>Send a friend request to its owner to be able to view it'), 'is private'),
  54         )
  55
  56         for patterns, message in ERRORS:
  57             if any(p in webpage for p in patterns):
  58                 raise ExtractorError(
  59                     'Video %s %s' % (video_id, message), expected=True)
  60
  61         info = self._search_json_ld(webpage, video_id, default={})
  62
  63         if not info.get('title'):
  64             info['title'] = self._html_search_regex(
  65                 (r'<h(\d)[^>]+class="(?:video_title_text|videoTitle|video_title)[^"]*">(?P<title>(?:(?!\1).)+)</h\1>',
  66                  r'(?:videoTitle|title)\s*:\s*(["\'])(?P<title>(?:(?!\1).)+)\1',),
  67                 webpage, 'title', group='title',
  68                 default=None) or self._og_search_title(webpage)
  69
  70         formats = []
  71         sources = self._parse_json(
  72             self._search_regex(
  73                 r'sources\s*:\s*({.+?})', webpage, 'source', default='{}'),
  74             video_id, fatal=False)
  75         if sources and isinstance(sources, dict):
  76             for format_id, format_url in sources.items():
  77                 if format_url:
  78                     formats.append({
  79                         'url': format_url,
  80                         'format_id': format_id,
  81                         'height': int_or_none(format_id),
  82                     })
  83         medias = self._parse_json(
  84             self._search_regex(
  85                 r'mediaDefinition["\']?\s*:\s*(\[.+?}\s*\])', webpage,
  86                 'media definitions', default='{}'),
  87             video_id, fatal=False)
  88         for media in medias if isinstance(medias, list) else []:
  89             format_url = url_or_none(media.get('videoUrl'))
  90             if not format_url:
  91                 continue
  92             format_id = media.get('format')
  93             quality = media.get('quality')
  94             if format_id == 'hls' or (format_id == 'mp4' and not quality):
  95                 more_media = self._download_json(format_url, video_id, fatal=False)
  96             else:
  97                 more_media = [media]
  98             for media in more_media if isinstance(more_media, list) else []:
  99                 format_url = url_or_none(media.get('videoUrl'))
 100                 if not format_url:
 101                     continue
 102                 format_id = media.get('format')
 103                 if format_id == 'hls' or determine_ext(format_url) == 'm3u8':
 104                     formats.extend(self._extract_m3u8_formats(
 105                         format_url, video_id, 'mp4',
 106                         entry_protocol='m3u8_native', m3u8_id=format_id or 'hls',
 107                         fatal=False))
 108                     continue
 109                 format_id = media.get('quality')
 110                 formats.append({
 111                     'url': format_url,
 112                     'ext': 'mp4',
 113                     'format_id': format_id,
 114                     'height': int_or_none(format_id),
 115                 })
 116         if not formats:
 117             video_url = self._html_search_regex(
 118                 r'<source src="(.+?)" type="video/mp4">', webpage, 'video URL')
 119             formats.append({'url': video_url, 'ext': 'mp4'})
 120         self._sort_formats(formats)
 121
 122         thumbnail = self._og_search_thumbnail(webpage)
 123         upload_date = unified_strdate(self._search_regex(
 124             r'<span[^>]+>(?:ADDED|Published on) ([^<]+)<',
 125             webpage, 'upload date', default=None))
 126         duration = int_or_none(self._og_search_property(
 127             'video:duration', webpage, default=None) or self._search_regex(
 128                 r'videoDuration\s*:\s*(\d+)', webpage, 'duration', default=None))
 129         view_count = str_to_int(self._search_regex(
 130             (r'<div[^>]*>Views</div>\s*<div[^>]*>\s*([\d,.]+)',
 131              r'<span[^>]*>VIEWS</span>\s*</td>\s*<td>\s*([\d,.]+)',
 132              r'<span[^>]+\bclass=["\']video_view_count[^>]*>\s*([\d,.]+)'),
 133             webpage, 'view count', default=None))
 134
 135         # No self-labeling, but they describe themselves as
 136         # "Home of Videos Porno"
 137         age_limit = 18
 138
 139         return merge_dicts(info, {
 140             'id': video_id,
 141             'ext': 'mp4',
 142             'thumbnail': thumbnail,
 143             'upload_date': upload_date,
 144             'duration': duration,
 145             'view_count': view_count,
 146             'age_limit': age_limit,
 147             'formats': formats,
 148         })