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 ..compat import compat_str
   7 from ..utils import (
   8     ExtractorError,
   9     int_or_none,
  10     str_to_int,
  11     unified_strdate,
  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': '7b8c22b5e7098a3e1c09709df1126d2d',
  20         'info_dict': {
  21             'id': '66418',
  22             'ext': 'mp4',
  23             'title': 'Sucked on a toilet',
  24             'upload_date': '20120831',
  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'<h1 class="videoTitle[^"]*">(?P<title>.+?)</h1>',
  50              r'videoTitle\s*:\s*(["\'])(?P<title>)\1'),
  51             webpage, 'title', group='title')
  52
  53         formats = []
  54         sources = self._parse_json(
  55             self._search_regex(
  56                 r'sources\s*:\s*({.+?})', webpage, 'source', default='{}'),
  57             video_id, fatal=False)
  58         if sources and isinstance(sources, dict):
  59             for format_id, format_url in sources.items():
  60                 if format_url:
  61                     formats.append({
  62                         'url': format_url,
  63                         'format_id': format_id,
  64                         'height': int_or_none(format_id),
  65                     })
  66         medias = self._parse_json(
  67             self._search_regex(
  68                 r'mediaDefinition\s*:\s*(\[.+?\])', webpage,
  69                 'media definitions', default='{}'),
  70             video_id, fatal=False)
  71         if medias and isinstance(medias, list):
  72             for media in medias:
  73                 format_url = media.get('videoUrl')
  74                 if not format_url or not isinstance(format_url, compat_str):
  75                     continue
  76                 format_id = media.get('quality')
  77                 formats.append({
  78                     'url': format_url,
  79                     'format_id': format_id,
  80                     'height': int_or_none(format_id),
  81                 })
  82         if not formats:
  83             video_url = self._html_search_regex(
  84                 r'<source src="(.+?)" type="video/mp4">', webpage, 'video URL')
  85             formats.append({'url': video_url})
  86         self._sort_formats(formats)
  87
  88         thumbnail = self._og_search_thumbnail(webpage)
  89         upload_date = unified_strdate(self._search_regex(
  90             r'<span[^>]+class="added-time"[^>]*>ADDED ([^<]+)<',
  91             webpage, 'upload date', fatal=False))
  92         duration = int_or_none(self._search_regex(
  93             r'videoDuration\s*:\s*(\d+)', webpage, 'duration', default=None))
  94         view_count = str_to_int(self._search_regex(
  95             r'<span[^>]*>VIEWS</span></td>\s*<td>([\d,.]+)',
  96             webpage, 'view count', fatal=False))
  97
  98         # No self-labeling, but they describe themselves as
  99         # "Home of Videos Porno"
 100         age_limit = 18
 101
 102         return {
 103             'id': video_id,
 104             'ext': 'mp4',
 105             'title': title,
 106             'thumbnail': thumbnail,
 107             'upload_date': upload_date,
 108             'duration': duration,
 109             'view_count': view_count,
 110             'age_limit': age_limit,
 111             'formats': formats,
 112         }