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     determine_ext,
   8     ExtractorError,
   9     int_or_none,
  10     merge_dicts,
  11     str_to_int,
  12     unified_strdate,
  13     url_or_none,
  14 )
  15
  16
  17 class RedTubeIE(InfoExtractor):
  18     _VALID_URL = r'https?://(?:(?:www\.)?redtube\.com/|embed\.redtube\.com/\?.*?\bid=)(?P<id>[0-9]+)'
  19     _TESTS = [{
  20         'url': 'http://www.redtube.com/66418',
  21         'md5': 'fc08071233725f26b8f014dba9590005',
  22         'info_dict': {
  23             'id': '66418',
  24             'ext': 'mp4',
  25             'title': 'Sucked on a toilet',
  26             'upload_date': '20110811',
  27             'duration': 596,
  28             'view_count': int,
  29             'age_limit': 18,
  30         }
  31     }, {
  32         'url': 'http://embed.redtube.com/?bgcolor=000000&id=1443286',
  33         'only_matching': True,
  34     }]
  35
  36     @staticmethod
  37     def _extract_urls(webpage):
  38         return re.findall(
  39             r'<iframe[^>]+?src=["\'](?P<url>(?:https?:)?//embed\.redtube\.com/\?.*?\bid=\d+)',
  40             webpage)
  41
  42     def _real_extract(self, url):
  43         video_id = self._match_id(url)
  44         webpage = self._download_webpage(
  45             'http://www.redtube.com/%s' % video_id, video_id)
  46
  47         ERRORS = (
  48             (('video-deleted-info', '>This video has been removed'), 'has been removed'),
  49             (('private_video_text', '>This video is private', '>Send a friend request to its owner to be able to view it'), 'is private'),
  50         )
  51
  52         for patterns, message in ERRORS:
  53             if any(p in webpage for p in patterns):
  54                 raise ExtractorError(
  55                     'Video %s %s' % (video_id, message), expected=True)
  56
  57         info = self._search_json_ld(webpage, video_id, default={})
  58
  59         if not info.get('title'):
  60             info['title'] = self._html_search_regex(
  61                 (r'<h(\d)[^>]+class="(?:video_title_text|videoTitle|video_title)[^"]*">(?P<title>(?:(?!\1).)+)</h\1>',
  62                  r'(?:videoTitle|title)\s*:\s*(["\'])(?P<title>(?:(?!\1).)+)\1',),
  63                 webpage, 'title', group='title',
  64                 default=None) or self._og_search_title(webpage)
  65
  66         formats = []
  67         sources = self._parse_json(
  68             self._search_regex(
  69                 r'sources\s*:\s*({.+?})', webpage, 'source', default='{}'),
  70             video_id, fatal=False)
  71         if sources and isinstance(sources, dict):
  72             for format_id, format_url in sources.items():
  73                 if format_url:
  74                     formats.append({
  75                         'url': format_url,
  76                         'format_id': format_id,
  77                         'height': int_or_none(format_id),
  78                     })
  79         medias = self._parse_json(
  80             self._search_regex(
  81                 r'mediaDefinition["\']?\s*:\s*(\[.+?}\s*\])', webpage,
  82                 'media definitions', default='{}'),
  83             video_id, fatal=False)
  84         if medias and isinstance(medias, list):
  85             for media in medias:
  86                 format_url = url_or_none(media.get('videoUrl'))
  87                 if not format_url:
  88                     continue
  89                 if media.get('format') == 'hls' or determine_ext(format_url) == 'm3u8':
  90                     formats.extend(self._extract_m3u8_formats(
  91                         format_url, video_id, 'mp4',
  92                         entry_protocol='m3u8_native', m3u8_id='hls',
  93                         fatal=False))
  94                     continue
  95                 format_id = media.get('quality')
  96                 formats.append({
  97                     'url': format_url,
  98                     'format_id': format_id,
  99                     'height': int_or_none(format_id),
 100                 })
 101         if not formats:
 102             video_url = self._html_search_regex(
 103                 r'<source src="(.+?)" type="video/mp4">', webpage, 'video URL')
 104             formats.append({'url': video_url})
 105         self._sort_formats(formats)
 106
 107         thumbnail = self._og_search_thumbnail(webpage)
 108         upload_date = unified_strdate(self._search_regex(
 109             r'<span[^>]+>(?:ADDED|Published on) ([^<]+)<',
 110             webpage, 'upload date', default=None))
 111         duration = int_or_none(self._og_search_property(
 112             'video:duration', webpage, default=None) or self._search_regex(
 113                 r'videoDuration\s*:\s*(\d+)', webpage, 'duration', default=None))
 114         view_count = str_to_int(self._search_regex(
 115             (r'<div[^>]*>Views</div>\s*<div[^>]*>\s*([\d,.]+)',
 116              r'<span[^>]*>VIEWS</span>\s*</td>\s*<td>\s*([\d,.]+)',
 117              r'<span[^>]+\bclass=["\']video_view_count[^>]*>\s*([\d,.]+)'),
 118             webpage, 'view count', default=None))
 119
 120         # No self-labeling, but they describe themselves as
 121         # "Home of Videos Porno"
 122         age_limit = 18
 123
 124         return merge_dicts(info, {
 125             'id': video_id,
 126             'ext': 'mp4',
 127             'thumbnail': thumbnail,
 128             'upload_date': upload_date,
 129             'duration': duration,
 130             'view_count': view_count,
 131             'age_limit': age_limit,
 132             'formats': formats,
 133         })