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

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..compat import (
   7     compat_urllib_request,
   8 )
   9 from ..utils import (
  10     parse_duration,
  11     parse_iso8601,
  12     str_to_int,
  13 )
  14
  15
  16 class FourTubeIE(InfoExtractor):
  17     IE_NAME = '4tube'
  18     _VALID_URL = r'https?://(?:www\.)?4tube\.com/videos/(?P<id>\d+)'
  19
  20     _TEST = {
  21         'url': 'http://www.4tube.com/videos/209733/hot-babe-holly-michaels-gets-her-ass-stuffed-by-black',
  22         'md5': '6516c8ac63b03de06bc8eac14362db4f',
  23         'info_dict': {
  24             'id': '209733',
  25             'ext': 'mp4',
  26             'title': 'Hot Babe Holly Michaels gets her ass stuffed by black',
  27             'uploader': 'WCP Club',
  28             'uploader_id': 'wcp-club',
  29             'upload_date': '20131031',
  30             'timestamp': 1383263892,
  31             'duration': 583,
  32             'view_count': int,
  33             'like_count': int,
  34             'categories': list,
  35             'age_limit': 18,
  36         }
  37     }
  38
  39     def _real_extract(self, url):
  40         video_id = self._match_id(url)
  41         webpage = self._download_webpage(url, video_id)
  42
  43         title = self._html_search_meta('name', webpage)
  44         timestamp = parse_iso8601(self._html_search_meta(
  45             'uploadDate', webpage))
  46         thumbnail = self._html_search_meta('thumbnailUrl', webpage)
  47         uploader_id = self._html_search_regex(
  48             r'<a class="img-avatar" href="[^"]+/channels/([^/"]+)" title="Go to [^"]+ page">',
  49             webpage, 'uploader id', fatal=False)
  50         uploader = self._html_search_regex(
  51             r'<a class="img-avatar" href="[^"]+/channels/[^/"]+" title="Go to ([^"]+) page">',
  52             webpage, 'uploader', fatal=False)
  53
  54         categories_html = self._search_regex(
  55             r'(?s)><i class="icon icon-tag"></i>\s*Categories / Tags\s*.*?<ul class="list">(.*?)</ul>',
  56             webpage, 'categories', fatal=False)
  57         categories = None
  58         if categories_html:
  59             categories = [
  60                 c.strip() for c in re.findall(
  61                     r'(?s)<li><a.*?>(.*?)</a>', categories_html)]
  62
  63         view_count = str_to_int(self._search_regex(
  64             r'<meta itemprop="interactionCount" content="UserPlays:([0-9,]+)">',
  65             webpage, 'view count', fatal=False))
  66         like_count = str_to_int(self._search_regex(
  67             r'<meta itemprop="interactionCount" content="UserLikes:([0-9,]+)">',
  68             webpage, 'like count', fatal=False))
  69         duration = parse_duration(self._html_search_meta('duration', webpage))
  70
  71         media_id = self._search_regex(
  72             r'<button[^>]+data-id=(["\'])(?P<id>\d+)\1[^>]+data-quality=', webpage,
  73             'media id', default=None, group='id')
  74         sources = [
  75             quality
  76             for _, quality in re.findall(r'<button[^>]+data-quality=(["\'])(.+?)\1', webpage)]
  77         if not (media_id and sources):
  78             player_js = self._download_webpage(
  79                 self._search_regex(
  80                     r'<script[^>]id=(["\'])playerembed\1[^>]+src=(["\'])(?P<url>.+?)\2',
  81                     webpage, 'player JS', group='url'),
  82                 video_id, 'Downloading player JS')
  83             params_js = self._search_regex(
  84                 r'\$\.ajax\(url,\ opts\);\s*\}\s*\}\)\(([0-9,\[\] ]+)\)',
  85                 player_js, 'initialization parameters')
  86             params = self._parse_json('[%s]' % params_js, video_id)
  87             media_id = params[0]
  88             sources = ['%s' % p for p in params[2]]
  89
  90         token_url = 'http://tkn.4tube.com/{0}/desktop/{1}'.format(
  91             media_id, '+'.join(sources))
  92         headers = {
  93             b'Content-Type': b'application/x-www-form-urlencoded',
  94             b'Origin': b'http://www.4tube.com',
  95         }
  96         token_req = compat_urllib_request.Request(token_url, b'{}', headers)
  97         tokens = self._download_json(token_req, video_id)
  98         formats = [{
  99             'url': tokens[format]['token'],
 100             'format_id': format + 'p',
 101             'resolution': format + 'p',
 102             'quality': int(format),
 103         } for format in sources]
 104         self._sort_formats(formats)
 105
 106         return {
 107             'id': video_id,
 108             'title': title,
 109             'formats': formats,
 110             'categories': categories,
 111             'thumbnail': thumbnail,
 112             'uploader': uploader,
 113             'uploader_id': uploader_id,
 114             'timestamp': timestamp,
 115             'like_count': like_count,
 116             'view_count': view_count,
 117             'duration': duration,
 118             'age_limit': 18,
 119         }