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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import itertools
   5 import random
   6 import re
   7 import string
   8 import time
   9
  10 from .common import InfoExtractor
  11 from ..utils import (
  12     ExtractorError,
  13     get_element_by_class,
  14     js_to_json,
  15     str_or_none,
  16     strip_jsonp,
  17     urljoin,
  18 )
  19
  20
  21 class YoukuIE(InfoExtractor):
  22     IE_NAME = 'youku'
  23     IE_DESC = '优酷'
  24     _VALID_URL = r'''(?x)
  25         (?:
  26             https?://(
  27                 (?:v|player)\.youku\.com/(?:v_show/id_|player\.php/sid/)|
  28                 video\.tudou\.com/v/)|
  29             youku:)
  30         (?P<id>[A-Za-z0-9]+)(?:\.html|/v\.swf|)
  31     '''
  32
  33     _TESTS = [{
  34         # MD5 is unstable
  35         'url': 'http://v.youku.com/v_show/id_XMTc1ODE5Njcy.html',
  36         'info_dict': {
  37             'id': 'XMTc1ODE5Njcy',
  38             'title': '★Smile﹗♡ Git Fresh -Booty Music舞蹈.',
  39             'ext': 'mp4',
  40             'duration': 74.73,
  41             'thumbnail': r're:^https?://.*',
  42             'uploader': '。躲猫猫、',
  43             'uploader_id': '36017967',
  44             'uploader_url': 'http://i.youku.com/u/UMTQ0MDcxODY4',
  45             'tags': list,
  46         }
  47     }, {
  48         'url': 'http://player.youku.com/player.php/sid/XNDgyMDQ2NTQw/v.swf',
  49         'only_matching': True,
  50     }, {
  51         'url': 'http://v.youku.com/v_show/id_XODgxNjg1Mzk2_ev_1.html',
  52         'info_dict': {
  53             'id': 'XODgxNjg1Mzk2',
  54             'ext': 'mp4',
  55             'title': '武媚娘传奇 85',
  56             'duration': 1999.61,
  57             'thumbnail': r're:^https?://.*',
  58             'uploader': '疯狂豆花',
  59             'uploader_id': '62583473',
  60             'uploader_url': 'http://i.youku.com/u/UMjUwMzMzODky',
  61             'tags': list,
  62         },
  63     }, {
  64         'url': 'http://v.youku.com/v_show/id_XMTI1OTczNDM5Mg==.html',
  65         'info_dict': {
  66             'id': 'XMTI1OTczNDM5Mg',
  67             'ext': 'mp4',
  68             'title': '花千骨 04',
  69             'duration': 2363,
  70             'thumbnail': r're:^https?://.*',
  71             'uploader': '放剧场-花千骨',
  72             'uploader_id': '772849359',
  73             'uploader_url': 'http://i.youku.com/u/UMzA5MTM5NzQzNg==',
  74             'tags': list,
  75         },
  76     }, {
  77         'url': 'http://v.youku.com/v_show/id_XNjA1NzA2Njgw.html',
  78         'note': 'Video protected with password',
  79         'info_dict': {
  80             'id': 'XNjA1NzA2Njgw',
  81             'ext': 'mp4',
  82             'title': '邢義田复旦讲座之想象中的胡人—从“左衽孔子”说起',
  83             'duration': 7264.5,
  84             'thumbnail': r're:^https?://.*',
  85             'uploader': 'FoxJin1006',
  86             'uploader_id': '322014285',
  87             'uploader_url': 'http://i.youku.com/u/UMTI4ODA1NzE0MA==',
  88             'tags': list,
  89         },
  90         'params': {
  91             'videopassword': '100600',
  92         },
  93     }, {
  94         # /play/get.json contains streams with "channel_type":"tail"
  95         'url': 'http://v.youku.com/v_show/id_XOTUxMzg4NDMy.html',
  96         'info_dict': {
  97             'id': 'XOTUxMzg4NDMy',
  98             'ext': 'mp4',
  99             'title': '我的世界☆明月庄主☆车震猎杀☆杀人艺术Minecraft',
 100             'duration': 702.08,
 101             'thumbnail': r're:^https?://.*',
 102             'uploader': '明月庄主moon',
 103             'uploader_id': '38465621',
 104             'uploader_url': 'http://i.youku.com/u/UMTUzODYyNDg0',
 105             'tags': list,
 106         },
 107     }, {
 108         'url': 'http://video.tudou.com/v/XMjIyNzAzMTQ4NA==.html?f=46177805',
 109         'info_dict': {
 110             'id': 'XMjIyNzAzMTQ4NA',
 111             'ext': 'mp4',
 112             'title': '卡马乔国足开大脚长传冲吊集锦',
 113             'duration': 289,
 114             'thumbnail': r're:^https?://.*',
 115             'uploader': '阿卜杜拉之星',
 116             'uploader_id': '2382249',
 117             'uploader_url': 'http://i.youku.com/u/UOTUyODk5Ng==',
 118             'tags': list,
 119         },
 120     }, {
 121         'url': 'http://video.tudou.com/v/XMjE4ODI3OTg2MA==.html',
 122         'only_matching': True,
 123     }]
 124
 125     @staticmethod
 126     def get_ysuid():
 127         return '%d%s' % (int(time.time()), ''.join([
 128             random.choice(string.ascii_letters) for i in range(3)]))
 129
 130     def get_format_name(self, fm):
 131         _dict = {
 132             '3gp': 'h6',
 133             '3gphd': 'h5',
 134             'flv': 'h4',
 135             'flvhd': 'h4',
 136             'mp4': 'h3',
 137             'mp4hd': 'h3',
 138             'mp4hd2': 'h4',
 139             'mp4hd3': 'h4',
 140             'hd2': 'h2',
 141             'hd3': 'h1',
 142         }
 143         return _dict.get(fm)
 144
 145     def _real_extract(self, url):
 146         video_id = self._match_id(url)
 147
 148         self._set_cookie('youku.com', '__ysuid', self.get_ysuid())
 149         self._set_cookie('youku.com', 'xreferrer', 'http://www.youku.com')
 150
 151         _, urlh = self._download_webpage_handle(
 152             'https://log.mmstat.com/eg.js', video_id, 'Retrieving cna info')
 153         # The etag header is '"foobar"'; let's remove the double quotes
 154         cna = urlh.headers['etag'][1:-1]
 155
 156         # request basic data
 157         basic_data_params = {
 158             'vid': video_id,
 159             'ccode': '0402' if 'tudou.com' in url else '0401',
 160             'client_ip': '192.168.1.1',
 161             'utid': cna,
 162             'client_ts': time.time() / 1000,
 163         }
 164
 165         video_password = self._downloader.params.get('videopassword')
 166         if video_password:
 167             basic_data_params['password'] = video_password
 168
 169         headers = {
 170             'Referer': url,
 171         }
 172         headers.update(self.geo_verification_headers())
 173         data = self._download_json(
 174             'https://ups.youku.com/ups/get.json', video_id,
 175             'Downloading JSON metadata',
 176             query=basic_data_params, headers=headers)['data']
 177
 178         error = data.get('error')
 179         if error:
 180             error_note = error.get('note')
 181             if error_note is not None and '因版权原因无法观看此视频' in error_note:
 182                 raise ExtractorError(
 183                     'Youku said: Sorry, this video is available in China only', expected=True)
 184             elif error_note and '该视频被设为私密' in error_note:
 185                 raise ExtractorError(
 186                     'Youku said: Sorry, this video is private', expected=True)
 187             else:
 188                 msg = 'Youku server reported error %i' % error.get('code')
 189                 if error_note is not None:
 190                     msg += ': ' + error_note
 191                 raise ExtractorError(msg)
 192
 193         # get video title
 194         video_data = data['video']
 195         title = video_data['title']
 196
 197         formats = [{
 198             'url': stream['m3u8_url'],
 199             'format_id': self.get_format_name(stream.get('stream_type')),
 200             'ext': 'mp4',
 201             'protocol': 'm3u8_native',
 202             'filesize': int(stream.get('size')),
 203             'width': stream.get('width'),
 204             'height': stream.get('height'),
 205         } for stream in data['stream'] if stream.get('channel_type') != 'tail']
 206         self._sort_formats(formats)
 207
 208         return {
 209             'id': video_id,
 210             'title': title,
 211             'formats': formats,
 212             'duration': video_data.get('seconds'),
 213             'thumbnail': video_data.get('logo'),
 214             'uploader': video_data.get('username'),
 215             'uploader_id': str_or_none(video_data.get('userid')),
 216             'uploader_url': data.get('uploader', {}).get('homepage'),
 217             'tags': video_data.get('tags'),
 218         }
 219
 220
 221 class YoukuShowIE(InfoExtractor):
 222     _VALID_URL = r'https?://list\.youku\.com/show/id_(?P<id>[0-9a-z]+)\.html'
 223     IE_NAME = 'youku:show'
 224
 225     _TEST = {
 226         'url': 'http://list.youku.com/show/id_zc7c670be07ff11e48b3f.html',
 227         'info_dict': {
 228             'id': 'zc7c670be07ff11e48b3f',
 229             'title': '花千骨 未删减版',
 230             'description': 'md5:a1ae6f5618571bbeb5c9821f9c81b558',
 231         },
 232         'playlist_count': 50,
 233     }
 234
 235     _PAGE_SIZE = 40
 236
 237     def _real_extract(self, url):
 238         show_id = self._match_id(url)
 239         webpage = self._download_webpage(url, show_id)
 240
 241         entries = []
 242         page_config = self._parse_json(self._search_regex(
 243             r'var\s+PageConfig\s*=\s*({.+});', webpage, 'page config'),
 244             show_id, transform_source=js_to_json)
 245         for idx in itertools.count(0):
 246             if idx == 0:
 247                 playlist_data_url = 'http://list.youku.com/show/module'
 248                 query = {'id': page_config['showid'], 'tab': 'point'}
 249             else:
 250                 playlist_data_url = 'http://list.youku.com/show/point'
 251                 query = {
 252                     'id': page_config['showid'],
 253                     'stage': 'reload_%d' % (self._PAGE_SIZE * idx + 1),
 254                 }
 255             query['callback'] = 'cb'
 256             playlist_data = self._download_json(
 257                 playlist_data_url, show_id, query=query,
 258                 note='Downloading playlist data page %d' % (idx + 1),
 259                 transform_source=lambda s: js_to_json(strip_jsonp(s)))['html']
 260             video_urls = re.findall(
 261                 r'<div[^>]+class="p-thumb"[^<]+<a[^>]+href="([^"]+)"',
 262                 playlist_data)
 263             new_entries = [
 264                 self.url_result(urljoin(url, video_url), YoukuIE.ie_key())
 265                 for video_url in video_urls]
 266             entries.extend(new_entries)
 267             if len(new_entries) < self._PAGE_SIZE:
 268                 break
 269
 270         desc = self._html_search_meta('description', webpage, fatal=False)
 271         playlist_title = desc.split(',')[0] if desc else None
 272         detail_li = get_element_by_class('p-intro', webpage)
 273         playlist_description = get_element_by_class(
 274             'intro-more', detail_li) if detail_li else None
 275
 276         return self.playlist_result(
 277             entries, show_id, playlist_title, playlist_description)