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