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

   1 from __future__ import unicode_literals
   2
   3 import re
   4 import json
   5
   6 from .common import InfoExtractor
   7 from ..compat import (
   8     compat_urllib_request,
   9 )
  10 from ..utils import (
  11     parse_duration,
  12     str_to_int,
  13 )
  14
  15
  16 class XTubeIE(InfoExtractor):
  17     _VALID_URL = r'https?://(?:www\.)?(?P<url>xtube\.com/watch\.php\?v=(?P<id>[^/?&]+))'
  18     _TEST = {
  19         'url': 'http://www.xtube.com/watch.php?v=kVTUy_G222_',
  20         'md5': '092fbdd3cbe292c920ef6fc6a8a9cdab',
  21         'info_dict': {
  22             'id': 'kVTUy_G222_',
  23             'ext': 'mp4',
  24             'title': 'strange erotica',
  25             'description': 'http://www.xtube.com an ET kind of thing',
  26             'uploader': 'greenshowers',
  27             'duration': 450,
  28             'age_limit': 18,
  29         }
  30     }
  31
  32     def _real_extract(self, url):
  33         mobj = re.match(self._VALID_URL, url)
  34         video_id = mobj.group('id')
  35         url = 'http://www.' + mobj.group('url')
  36
  37         req = compat_urllib_request.Request(url)
  38         req.add_header('Cookie', 'age_verified=1')
  39         webpage = self._download_webpage(req, video_id)
  40
  41         video_title = self._html_search_regex(r'<p class="title">([^<]+)', webpage, 'title')
  42         video_uploader = self._html_search_regex(
  43             r'so_s\.addVariable\("owner_u", "([^"]+)', webpage, 'uploader', fatal=False)
  44         video_description = self._html_search_regex(
  45             r'<p class="fieldsDesc">([^<]+)', webpage, 'description', fatal=False)
  46         duration = parse_duration(self._html_search_regex(
  47             r'<span class="bold">Runtime:</span> ([^<]+)</p>', webpage, 'duration', fatal=False))
  48         view_count = self._html_search_regex(
  49             r'<span class="bold">Views:</span> ([\d,\.]+)</p>', webpage, 'view count', fatal=False)
  50         if view_count:
  51             view_count = str_to_int(view_count)
  52         comment_count = self._html_search_regex(
  53             r'<div id="commentBar">([\d,\.]+) Comments</div>', webpage, 'comment count', fatal=False)
  54         if comment_count:
  55             comment_count = str_to_int(comment_count)
  56
  57         player_quality_option = json.loads(self._html_search_regex(
  58             r'playerQualityOption = ({.+?});', webpage, 'player quality option'))
  59
  60         QUALITIES = ['3gp', 'mp4_normal', 'mp4_high', 'flv', 'mp4_ultra', 'mp4_720', 'mp4_1080']
  61         formats = [
  62             {
  63                 'url': furl,
  64                 'format_id': format_id,
  65                 'preference': QUALITIES.index(format_id) if format_id in QUALITIES else -1,
  66             } for format_id, furl in player_quality_option.items()
  67         ]
  68         self._sort_formats(formats)
  69
  70         return {
  71             'id': video_id,
  72             'title': video_title,
  73             'uploader': video_uploader,
  74             'description': video_description,
  75             'duration': duration,
  76             'view_count': view_count,
  77             'comment_count': comment_count,
  78             'formats': formats,
  79             'age_limit': 18,
  80         }
  81
  82
  83 class XTubeUserIE(InfoExtractor):
  84     IE_DESC = 'XTube user profile'
  85     _VALID_URL = r'https?://(?:www\.)?xtube\.com/community/profile\.php\?(.*?)user=(?P<username>[^&#]+)(?:$|[&#])'
  86     _TEST = {
  87         'url': 'http://www.xtube.com/community/profile.php?user=greenshowers',
  88         'info_dict': {
  89             'id': 'greenshowers',
  90         },
  91         'playlist_mincount': 155,
  92     }
  93
  94     def _real_extract(self, url):
  95         mobj = re.match(self._VALID_URL, url)
  96         username = mobj.group('username')
  97
  98         profile_page = self._download_webpage(
  99             url, username, note='Retrieving profile page')
 100
 101         video_count = int(self._search_regex(
 102             r'<strong>%s\'s Videos \(([0-9]+)\)</strong>' % username, profile_page,
 103             'video count'))
 104
 105         PAGE_SIZE = 25
 106         urls = []
 107         page_count = (video_count + PAGE_SIZE + 1) // PAGE_SIZE
 108         for n in range(1, page_count + 1):
 109             lpage_url = 'http://www.xtube.com/user_videos.php?page=%d&u=%s' % (n, username)
 110             lpage = self._download_webpage(
 111                 lpage_url, username,
 112                 note='Downloading page %d/%d' % (n, page_count))
 113             urls.extend(
 114                 re.findall(r'addthis:url="([^"]+)"', lpage))
 115
 116         return {
 117             '_type': 'playlist',
 118             'id': username,
 119             'entries': [{
 120                 '_type': 'url',
 121                 'url': eurl,
 122                 'ie_key': 'XTube',
 123             } for eurl in urls]
 124         }