_ Git - youtube-dl/blob - youtube_dl/extractor/xtube.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..compat import compat_urllib_parse_unquote
   7 from ..utils import (
   8     parse_duration,
   9     sanitized_Request,
  10     str_to_int,
  11 )
  12
  13
  14 class XTubeIE(InfoExtractor):
  15     _VALID_URL = r'https?://(?:www\.)?(?P<url>xtube\.com/watch\.php\?v=(?P<id>[^/?&#]+))'
  16     _TEST = {
  17         'url': 'http://www.xtube.com/watch.php?v=kVTUy_G222_',
  18         'md5': '092fbdd3cbe292c920ef6fc6a8a9cdab',
  19         'info_dict': {
  20             'id': 'kVTUy_G222_',
  21             'ext': 'mp4',
  22             'title': 'strange erotica',
  23             'description': 'contains:an ET kind of thing',
  24             'uploader': 'greenshowers',
  25             'duration': 450,
  26             'age_limit': 18,
  27         }
  28     }
  29
  30     def _real_extract(self, url):
  31         video_id = self._match_id(url)
  32
  33         req = sanitized_Request(url)
  34         req.add_header('Cookie', 'age_verified=1')
  35         webpage = self._download_webpage(req, video_id)
  36
  37         video_title = self._html_search_regex(
  38             r'<p class="title">([^<]+)', webpage, 'title')
  39         video_uploader = self._html_search_regex(
  40             [r"var\s+contentOwnerId\s*=\s*'([^']+)",
  41              r'By:\s*<a href="/community/profile\.php\?user=([^"]+)'],
  42             webpage, 'uploader', fatal=False)
  43         video_description = self._html_search_regex(
  44             r'<p class="fieldsDesc">([^<]+)',
  45             webpage, 'description', fatal=False)
  46         duration = parse_duration(self._html_search_regex(
  47             r'<span class="bold">Runtime:</span> ([^<]+)</p>',
  48             webpage, 'duration', fatal=False))
  49         view_count = str_to_int(self._html_search_regex(
  50             r'<span class="bold">Views:</span> ([\d,\.]+)</p>',
  51             webpage, 'view count', fatal=False))
  52         comment_count = str_to_int(self._html_search_regex(
  53             r'<div id="commentBar">([\d,\.]+) Comments</div>',
  54             webpage, 'comment count', fatal=False))
  55
  56         formats = []
  57         for format_id, video_url in re.findall(
  58                 r'flashvars\.quality_(.+?)\s*=\s*"([^"]+)"', webpage):
  59             fmt = {
  60                 'url': compat_urllib_parse_unquote(video_url),
  61                 'format_id': format_id,
  62             }
  63             m = re.search(r'^(?P<height>\d+)[pP]', format_id)
  64             if m:
  65                 fmt['height'] = int(m.group('height'))
  66             formats.append(fmt)
  67
  68         if not formats:
  69             video_url = compat_urllib_parse_unquote(self._search_regex(
  70                 r'flashvars\.video_url\s*=\s*"([^"]+)"',
  71                 webpage, 'video URL'))
  72             formats.append({'url': video_url})
  73
  74         self._sort_formats(formats)
  75
  76         return {
  77             'id': video_id,
  78             'title': video_title,
  79             'uploader': video_uploader,
  80             'description': video_description,
  81             'duration': duration,
  82             'view_count': view_count,
  83             'comment_count': comment_count,
  84             'formats': formats,
  85             'age_limit': 18,
  86         }
  87
  88
  89 class XTubeUserIE(InfoExtractor):
  90     IE_DESC = 'XTube user profile'
  91     _VALID_URL = r'https?://(?:www\.)?xtube\.com/community/profile\.php\?(.*?)user=(?P<username>[^&#]+)(?:$|[&#])'
  92     _TEST = {
  93         'url': 'http://www.xtube.com/community/profile.php?user=greenshowers',
  94         'info_dict': {
  95             'id': 'greenshowers',
  96             'age_limit': 18,
  97         },
  98         'playlist_mincount': 155,
  99     }
 100
 101     def _real_extract(self, url):
 102         mobj = re.match(self._VALID_URL, url)
 103         username = mobj.group('username')
 104
 105         profile_page = self._download_webpage(
 106             url, username, note='Retrieving profile page')
 107
 108         video_count = int(self._search_regex(
 109             r'<strong>%s\'s Videos \(([0-9]+)\)</strong>' % username, profile_page,
 110             'video count'))
 111
 112         PAGE_SIZE = 25
 113         urls = []
 114         page_count = (video_count + PAGE_SIZE + 1) // PAGE_SIZE
 115         for n in range(1, page_count + 1):
 116             lpage_url = 'http://www.xtube.com/user_videos.php?page=%d&u=%s' % (n, username)
 117             lpage = self._download_webpage(
 118                 lpage_url, username,
 119                 note='Downloading page %d/%d' % (n, page_count))
 120             urls.extend(
 121                 re.findall(r'addthis:url="([^"]+)"', lpage))
 122
 123         return {
 124             '_type': 'playlist',
 125             'id': username,
 126             'age_limit': 18,
 127             'entries': [{
 128                 '_type': 'url',
 129                 'url': eurl,
 130                 'ie_key': 'XTube',
 131             } for eurl in urls]
 132         }