yt_dlp/extractor/youku.py

   1 import random
   2 import re
   3 import string
   4 import time
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     ExtractorError,
   9     get_element_by_class,
  10     js_to_json,
  11     str_or_none,
  12     strip_jsonp,
  13 )
  14
  15
  16 class YoukuIE(InfoExtractor):
  17     IE_NAME = 'youku'
  18     IE_DESC = '优酷'
  19     _VALID_URL = r'''(?x)
  20         (?:
  21             https?://(
  22                 (?:v|player)\.youku\.com/(?:v_show/id_|player\.php/sid/)|
  23                 video\.tudou\.com/v/)|
  24             youku:)
  25         (?P<id>[A-Za-z0-9]+)(?:\.html|/v\.swf|)
  26     '''
  27
  28     _TESTS = [{
  29         # MD5 is unstable
  30         'url': 'http://v.youku.com/v_show/id_XMTc1ODE5Njcy.html',
  31         'info_dict': {
  32             'id': 'XMTc1ODE5Njcy',
  33             'title': '★Smile﹗♡ Git Fresh -Booty Music舞蹈.',
  34             'ext': 'mp4',
  35             'duration': 74.73,
  36             'thumbnail': r're:^https?://.*',
  37             'uploader': '。躲猫猫、',
  38             'uploader_id': '36017967',
  39             'uploader_url': 'http://i.youku.com/u/UMTQ0MDcxODY4',
  40             'tags': list,
  41         }
  42     }, {
  43         'url': 'http://player.youku.com/player.php/sid/XNDgyMDQ2NTQw/v.swf',
  44         'only_matching': True,
  45     }, {
  46         'url': 'http://v.youku.com/v_show/id_XODgxNjg1Mzk2_ev_1.html',
  47         'info_dict': {
  48             'id': 'XODgxNjg1Mzk2',
  49             'ext': 'mp4',
  50             'title': '武媚娘传奇 85',
  51             'duration': 1999.61,
  52             'thumbnail': r're:^https?://.*',
  53             'uploader': '疯狂豆花',
  54             'uploader_id': '62583473',
  55             'uploader_url': 'http://i.youku.com/u/UMjUwMzMzODky',
  56             'tags': list,
  57         },
  58     }, {
  59         'url': 'http://v.youku.com/v_show/id_XMTI1OTczNDM5Mg==.html',
  60         'info_dict': {
  61             'id': 'XMTI1OTczNDM5Mg',
  62             'ext': 'mp4',
  63             'title': '花千骨 04',
  64             'duration': 2363,
  65             'thumbnail': r're:^https?://.*',
  66             'uploader': '放剧场-花千骨',
  67             'uploader_id': '772849359',
  68             'uploader_url': 'http://i.youku.com/u/UMzA5MTM5NzQzNg==',
  69             'tags': list,
  70         },
  71     }, {
  72         'url': 'http://v.youku.com/v_show/id_XNjA1NzA2Njgw.html',
  73         'note': 'Video protected with password',
  74         'info_dict': {
  75             'id': 'XNjA1NzA2Njgw',
  76             'ext': 'mp4',
  77             'title': '邢義田复旦讲座之想象中的胡人—从“左衽孔子”说起',
  78             'duration': 7264.5,
  79             'thumbnail': r're:^https?://.*',
  80             'uploader': 'FoxJin1006',
  81             'uploader_id': '322014285',
  82             'uploader_url': 'http://i.youku.com/u/UMTI4ODA1NzE0MA==',
  83             'tags': list,
  84         },
  85         'params': {
  86             'videopassword': '100600',
  87         },
  88     }, {
  89         # /play/get.json contains streams with "channel_type":"tail"
  90         'url': 'http://v.youku.com/v_show/id_XOTUxMzg4NDMy.html',
  91         'info_dict': {
  92             'id': 'XOTUxMzg4NDMy',
  93             'ext': 'mp4',
  94             'title': '我的世界☆明月庄主☆车震猎杀☆杀人艺术Minecraft',
  95             'duration': 702.08,
  96             'thumbnail': r're:^https?://.*',
  97             'uploader': '明月庄主moon',
  98             'uploader_id': '38465621',
  99             'uploader_url': 'http://i.youku.com/u/UMTUzODYyNDg0',
 100             'tags': list,
 101         },
 102     }, {
 103         'url': 'http://video.tudou.com/v/XMjIyNzAzMTQ4NA==.html?f=46177805',
 104         'info_dict': {
 105             'id': 'XMjIyNzAzMTQ4NA',
 106             'ext': 'mp4',
 107             'title': '卡马乔国足开大脚长传冲吊集锦',
 108             'duration': 289,
 109             'thumbnail': r're:^https?://.*',
 110             'uploader': '阿卜杜拉之星',
 111             'uploader_id': '2382249',
 112             'uploader_url': 'http://i.youku.com/u/UOTUyODk5Ng==',
 113             'tags': list,
 114         },
 115     }, {
 116         'url': 'http://video.tudou.com/v/XMjE4ODI3OTg2MA==.html',
 117         'only_matching': True,
 118     }]
 119
 120     @staticmethod
 121     def get_ysuid():
 122         return '%d%s' % (int(time.time()), ''.join([
 123             random.choice(string.ascii_letters) for i in range(3)]))
 124
 125     def get_format_name(self, fm):
 126         _dict = {
 127             '3gp': 'h6',
 128             '3gphd': 'h5',
 129             'flv': 'h4',
 130             'flvhd': 'h4',
 131             'mp4': 'h3',
 132             'mp4hd': 'h3',
 133             'mp4hd2': 'h4',
 134             'mp4hd3': 'h4',
 135             'hd2': 'h2',
 136             'hd3': 'h1',
 137         }
 138         return _dict.get(fm)
 139
 140     def _real_extract(self, url):
 141         video_id = self._match_id(url)
 142
 143         self._set_cookie('youku.com', '__ysuid', self.get_ysuid())
 144         self._set_cookie('youku.com', 'xreferrer', 'http://www.youku.com')
 145
 146         _, urlh = self._download_webpage_handle(
 147             'https://log.mmstat.com/eg.js', video_id, 'Retrieving cna info')
 148         # The etag header is '"foobar"'; let's remove the double quotes
 149         cna = urlh.headers['etag'][1:-1]
 150
 151         # request basic data
 152         basic_data_params = {
 153             'vid': video_id,
 154             'ccode': '0532',
 155             'client_ip': '192.168.1.1',
 156             'utid': cna,
 157             'client_ts': time.time() / 1000,
 158         }
 159
 160         video_password = self.get_param('videopassword')
 161         if video_password:
 162             basic_data_params['password'] = video_password
 163
 164         headers = {
 165             'Referer': url,
 166         }
 167         headers.update(self.geo_verification_headers())
 168         data = self._download_json(
 169             'https://ups.youku.com/ups/get.json', video_id,
 170             'Downloading JSON metadata',
 171             query=basic_data_params, headers=headers)['data']
 172
 173         error = data.get('error')
 174         if error:
 175             error_note = error.get('note')
 176             if error_note is not None and '因版权原因无法观看此视频' in error_note:
 177                 raise ExtractorError(
 178                     'Youku said: Sorry, this video is available in China only', expected=True)
 179             elif error_note and '该视频被设为私密' in error_note:
 180                 raise ExtractorError(
 181                     'Youku said: Sorry, this video is private', expected=True)
 182             else:
 183                 msg = 'Youku server reported error %i' % error.get('code')
 184                 if error_note is not None:
 185                     msg += ': ' + error_note
 186                 raise ExtractorError(msg)
 187
 188         # get video title
 189         video_data = data['video']
 190         title = video_data['title']
 191
 192         formats = [{
 193             'url': stream['m3u8_url'],
 194             'format_id': self.get_format_name(stream.get('stream_type')),
 195             'ext': 'mp4',
 196             'protocol': 'm3u8_native',
 197             'filesize': int(stream.get('size')),
 198             'width': stream.get('width'),
 199             'height': stream.get('height'),
 200         } for stream in data['stream'] if stream.get('channel_type') != 'tail']
 201
 202         return {
 203             'id': video_id,
 204             'title': title,
 205             'formats': formats,
 206             'duration': video_data.get('seconds'),
 207             'thumbnail': video_data.get('logo'),
 208             'uploader': video_data.get('username'),
 209             'uploader_id': str_or_none(video_data.get('userid')),
 210             'uploader_url': data.get('uploader', {}).get('homepage'),
 211             'tags': video_data.get('tags'),
 212         }
 213
 214
 215 class YoukuShowIE(InfoExtractor):
 216     _VALID_URL = r'https?://list\.youku\.com/show/id_(?P<id>[0-9a-z]+)\.html'
 217     IE_NAME = 'youku:show'
 218
 219     _TESTS = [{
 220         'url': 'http://list.youku.com/show/id_zc7c670be07ff11e48b3f.html',
 221         'info_dict': {
 222             'id': 'zc7c670be07ff11e48b3f',
 223             'title': '花千骨 DVD版',
 224             'description': 'md5:a1ae6f5618571bbeb5c9821f9c81b558',
 225         },
 226         'playlist_count': 50,
 227     }, {
 228         # Episode number not starting from 1
 229         'url': 'http://list.youku.com/show/id_zefbfbd70efbfbd780bef.html',
 230         'info_dict': {
 231             'id': 'zefbfbd70efbfbd780bef',
 232             'title': '超级飞侠3',
 233             'description': 'md5:275715156abebe5ccc2a1992e9d56b98',
 234         },
 235         'playlist_count': 24,
 236     }, {
 237         # Ongoing playlist. The initial page is the last one
 238         'url': 'http://list.youku.com/show/id_za7c275ecd7b411e1a19e.html',
 239         'only_matching': True,
 240     }, {
 241         #  No data-id value.
 242         'url': 'http://list.youku.com/show/id_zefbfbd61237fefbfbdef.html',
 243         'only_matching': True,
 244     }, {
 245         #  Wrong number of reload_id.
 246         'url': 'http://list.youku.com/show/id_z20eb4acaf5c211e3b2ad.html',
 247         'only_matching': True,
 248     }]
 249
 250     def _extract_entries(self, playlist_data_url, show_id, note, query):
 251         query['callback'] = 'cb'
 252         playlist_data = self._download_json(
 253             playlist_data_url, show_id, query=query, note=note,
 254             transform_source=lambda s: js_to_json(strip_jsonp(s))).get('html')
 255         if playlist_data is None:
 256             return [None, None]
 257         drama_list = (get_element_by_class('p-drama-grid', playlist_data)
 258                       or get_element_by_class('p-drama-half-row', playlist_data))
 259         if drama_list is None:
 260             raise ExtractorError('No episodes found')
 261         video_urls = re.findall(r'<a[^>]+href="([^"]+)"', drama_list)
 262         return playlist_data, [
 263             self.url_result(self._proto_relative_url(video_url, 'http:'), YoukuIE.ie_key())
 264             for video_url in video_urls]
 265
 266     def _real_extract(self, url):
 267         show_id = self._match_id(url)
 268         webpage = self._download_webpage(url, show_id)
 269
 270         entries = []
 271         page_config = self._parse_json(self._search_regex(
 272             r'var\s+PageConfig\s*=\s*({.+});', webpage, 'page config'),
 273             show_id, transform_source=js_to_json)
 274         first_page, initial_entries = self._extract_entries(
 275             'http://list.youku.com/show/module', show_id,
 276             note='Downloading initial playlist data page',
 277             query={
 278                 'id': page_config['showid'],
 279                 'tab': 'showInfo',
 280             })
 281         first_page_reload_id = self._html_search_regex(
 282             r'<div[^>]+id="(reload_\d+)', first_page, 'first page reload id')
 283         # The first reload_id has the same items as first_page
 284         reload_ids = re.findall('<li[^>]+data-id="([^"]+)">', first_page)
 285         entries.extend(initial_entries)
 286         for idx, reload_id in enumerate(reload_ids):
 287             if reload_id == first_page_reload_id:
 288                 continue
 289             _, new_entries = self._extract_entries(
 290                 'http://list.youku.com/show/episode', show_id,
 291                 note='Downloading playlist data page %d' % (idx + 1),
 292                 query={
 293                     'id': page_config['showid'],
 294                     'stage': reload_id,
 295                 })
 296             if new_entries is not None:
 297                 entries.extend(new_entries)
 298         desc = self._html_search_meta('description', webpage, fatal=False)
 299         playlist_title = desc.split(',')[0] if desc else None
 300         detail_li = get_element_by_class('p-intro', webpage)
 301         playlist_description = get_element_by_class(
 302             'intro-more', detail_li) if detail_li else None
 303
 304         return self.playlist_result(
 305             entries, show_id, playlist_title, playlist_description)