youtube_dl/extractor/fourtube.py

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