youtube_dl/extractor/redtube.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     ExtractorError,
   8     int_or_none,
   9     merge_dicts,
  10     str_to_int,
  11     unified_strdate,
  12     url_or_none,
  13 )
  14
  15
  16 class RedTubeIE(InfoExtractor):
  17     _VALID_URL = r'https?://(?:(?:www\.)?redtube\.com/|embed\.redtube\.com/\?.*?\bid=)(?P<id>[0-9]+)'
  18     _TESTS = [{
  19         'url': 'http://www.redtube.com/66418',
  20         'md5': 'fc08071233725f26b8f014dba9590005',
  21         'info_dict': {
  22             'id': '66418',
  23             'ext': 'mp4',
  24             'title': 'Sucked on a toilet',
  25             'upload_date': '20110811',
  26             'duration': 596,
  27             'view_count': int,
  28             'age_limit': 18,
  29         }
  30     }, {
  31         'url': 'http://embed.redtube.com/?bgcolor=000000&id=1443286',
  32         'only_matching': True,
  33     }]
  34
  35     @staticmethod
  36     def _extract_urls(webpage):
  37         return re.findall(
  38             r'<iframe[^>]+?src=["\'](?P<url>(?:https?:)?//embed\.redtube\.com/\?.*?\bid=\d+)',
  39             webpage)
  40
  41     def _real_extract(self, url):
  42         video_id = self._match_id(url)
  43         webpage = self._download_webpage(
  44             'http://www.redtube.com/%s' % video_id, video_id)
  45
  46         if any(s in webpage for s in ['video-deleted-info', '>This video has been removed']):
  47             raise ExtractorError('Video %s has been removed' % video_id, expected=True)
  48
  49         info = self._search_json_ld(webpage, video_id, default={})
  50
  51         if not info.get('title'):
  52             info['title'] = self._html_search_regex(
  53                 (r'<h(\d)[^>]+class="(?:video_title_text|videoTitle)[^"]*">(?P<title>(?:(?!\1).)+)</h\1>',
  54                  r'(?:videoTitle|title)\s*:\s*(["\'])(?P<title>(?:(?!\1).)+)\1',),
  55                 webpage, 'title', group='title',
  56                 default=None) or self._og_search_title(webpage)
  57
  58         formats = []
  59         sources = self._parse_json(
  60             self._search_regex(
  61                 r'sources\s*:\s*({.+?})', webpage, 'source', default='{}'),
  62             video_id, fatal=False)
  63         if sources and isinstance(sources, dict):
  64             for format_id, format_url in sources.items():
  65                 if format_url:
  66                     formats.append({
  67                         'url': format_url,
  68                         'format_id': format_id,
  69                         'height': int_or_none(format_id),
  70                     })
  71         medias = self._parse_json(
  72             self._search_regex(
  73                 r'mediaDefinition\s*:\s*(\[.+?\])', webpage,
  74                 'media definitions', default='{}'),
  75             video_id, fatal=False)
  76         if medias and isinstance(medias, list):
  77             for media in medias:
  78                 format_url = url_or_none(media.get('videoUrl'))
  79                 if not format_url:
  80                     continue
  81                 format_id = media.get('quality')
  82                 formats.append({
  83                     'url': format_url,
  84                     'format_id': format_id,
  85                     'height': int_or_none(format_id),
  86                 })
  87         if not formats:
  88             video_url = self._html_search_regex(
  89                 r'<source src="(.+?)" type="video/mp4">', webpage, 'video URL')
  90             formats.append({'url': video_url})
  91         self._sort_formats(formats)
  92
  93         thumbnail = self._og_search_thumbnail(webpage)
  94         upload_date = unified_strdate(self._search_regex(
  95             r'<span[^>]+>(?:ADDED|Published on) ([^<]+)<',
  96             webpage, 'upload date', default=None))
  97         duration = int_or_none(self._og_search_property(
  98             'video:duration', webpage, default=None) or self._search_regex(
  99                 r'videoDuration\s*:\s*(\d+)', webpage, 'duration', default=None))
 100         view_count = str_to_int(self._search_regex(
 101             (r'<div[^>]*>Views</div>\s*<div[^>]*>\s*([\d,.]+)',
 102              r'<span[^>]*>VIEWS</span>\s*</td>\s*<td>\s*([\d,.]+)',
 103              r'<span[^>]+\bclass=["\']video_view_count[^>]*>\s*([\d,.]+)'),
 104             webpage, 'view count', default=None))
 105
 106         # No self-labeling, but they describe themselves as
 107         # "Home of Videos Porno"
 108         age_limit = 18
 109
 110         return merge_dicts(info, {
 111             'id': video_id,
 112             'ext': 'mp4',
 113             'thumbnail': thumbnail,
 114             'upload_date': upload_date,
 115             'duration': duration,
 116             'view_count': view_count,
 117             'age_limit': age_limit,
 118             'formats': formats,
 119         })