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     clean_html,
  10     get_element_by_class,
  11     js_to_json,
  12     str_or_none,
  13     strip_jsonp,
  14 )
  15
  16
  17 class YoukuIE(InfoExtractor):
  18     IE_NAME = 'youku'
  19     IE_DESC = '优酷'
  20     _VALID_URL = r'''(?x)
  21         (?:
  22             https?://(
  23                 (?:v|player)\.youku\.com/(?:v_show/id_|player\.php/sid/)|
  24                 video\.tudou\.com/v/)|
  25             youku:)
  26         (?P<id>[A-Za-z0-9]+)(?:\.html|/v\.swf|)
  27     '''
  28
  29     _TESTS = [{
  30         'url': 'http://player.youku.com/player.php/sid/XNDgyMDQ2NTQw/v.swf',
  31         'only_matching': True,
  32     }, {
  33         'url': 'http://v.youku.com/v_show/id_XNjA1NzA2Njgw.html',
  34         'note': 'Video protected with password',
  35         'info_dict': {
  36             'id': 'XNjA1NzA2Njgw',
  37             'ext': 'mp4',
  38             'title': '邢義田复旦讲座之想象中的胡人—从“左衽孔子”说起',
  39             'duration': 7264.5,
  40             'thumbnail': r're:^https?://.*',
  41             'uploader': 'FoxJin1006',
  42             'uploader_id': '322014285',
  43             'uploader_url': 'http://i.youku.com/u/UMTI4ODA1NzE0MA==',
  44             'tags': list,
  45             'skip': '404',
  46         },
  47         'params': {
  48             'videopassword': '100600',
  49         },
  50     }, {
  51         # /play/get.json contains streams with "channel_type":"tail"
  52         'url': 'http://v.youku.com/v_show/id_XOTUxMzg4NDMy.html',
  53         'info_dict': {
  54             'id': 'XOTUxMzg4NDMy',
  55             'ext': 'mp4',
  56             'title': '我的世界☆明月庄主☆车震猎杀☆杀人艺术Minecraft',
  57             'duration': 702.08,
  58             'thumbnail': r're:^https?://.*',
  59             'uploader': '明月庄主moon',
  60             'uploader_id': '38465621',
  61             'uploader_url': 'https://www.youku.com/profile/index/?uid=UMTUzODYyNDg0',
  62             'tags': list,
  63         },
  64     }, {
  65         'url': 'https://v.youku.com/v_show/id_XNTA2NTA0MjA1Mg==.html',
  66         'info_dict': {
  67             'id': 'XNTA2NTA0MjA1Mg',
  68             'ext': 'mp4',
  69             'title': 'Minecraft我的世界：建造超大巨型航空飞机，菜鸟vs高手vs黑客',
  70             'duration': 542.13,
  71             'thumbnail': r're:^https?://.*',
  72             'uploader': '波哥游戏解说',
  73             'uploader_id': '156688084',
  74             'uploader_url': 'https://www.youku.com/profile/index/?uid=UNjI2NzUyMzM2',
  75             'tags': list,
  76         },
  77     }, {
  78         'url': 'https://v.youku.com/v_show/id_XNTE1MzczOTg4MA==.html',
  79         'info_dict': {
  80             'id': 'XNTE1MzczOTg4MA',
  81             'ext': 'mp4',
  82             'title': '国产超A特工片',
  83             'duration': 362.97,
  84             'thumbnail': r're:^https?://.*',
  85             'uploader': '陈晓娟说历史',
  86             'uploader_id': '1640913339',
  87             'uploader_url': 'https://www.youku.com/profile/index/?uid=UNjU2MzY1MzM1Ng==',
  88             'tags': list,
  89         },
  90     }]
  91
  92     @staticmethod
  93     def get_ysuid():
  94         return '%d%s' % (int(time.time()), ''.join(
  95             random.choices(string.ascii_letters, k=3)))
  96
  97     def get_format_name(self, fm):
  98         _dict = {
  99             '3gp': 'h6',
 100             '3gphd': 'h5',
 101             'flv': 'h4',
 102             'flvhd': 'h4',
 103             'mp4': 'h3',
 104             'mp4hd': 'h3',
 105             'mp4hd2': 'h4',
 106             'mp4hd3': 'h4',
 107             'hd2': 'h2',
 108             'hd3': 'h1',
 109         }
 110         return _dict.get(fm)
 111
 112     def _real_extract(self, url):
 113         video_id = self._match_id(url)
 114
 115         self._set_cookie('youku.com', '__ysuid', self.get_ysuid())
 116         self._set_cookie('youku.com', 'xreferrer', 'http://www.youku.com')
 117
 118         _, urlh = self._download_webpage_handle(
 119             'https://log.mmstat.com/eg.js', video_id, 'Retrieving cna info')
 120         # The etag header is '"foobar"'; let's remove the double quotes
 121         cna = urlh.headers['etag'][1:-1]
 122
 123         # request basic data
 124         basic_data_params = {
 125             'vid': video_id,
 126             'ccode': '0524',
 127             'client_ip': '192.168.1.1',
 128             'utid': cna,
 129             'client_ts': time.time() / 1000,
 130         }
 131
 132         video_password = self.get_param('videopassword')
 133         if video_password:
 134             basic_data_params['password'] = video_password
 135
 136         headers = {
 137             'Referer': url,
 138         }
 139         headers.update(self.geo_verification_headers())
 140         data = self._download_json(
 141             'https://ups.youku.com/ups/get.json', video_id,
 142             'Downloading JSON metadata',
 143             query=basic_data_params, headers=headers)['data']
 144
 145         error = data.get('error')
 146         if error:
 147             error_note = error.get('note')
 148             if error_note is not None and '因版权原因无法观看此视频' in error_note:
 149                 raise ExtractorError(
 150                     'Youku said: Sorry, this video is available in China only', expected=True)
 151             elif error_note and '该视频被设为私密' in error_note:
 152                 raise ExtractorError(
 153                     'Youku said: Sorry, this video is private', expected=True)
 154             else:
 155                 msg = 'Youku server reported error %i' % error.get('code')
 156                 if error_note is not None:
 157                     msg += ': ' + clean_html(error_note)
 158                 raise ExtractorError(msg)
 159
 160         # get video title
 161         video_data = data['video']
 162         title = video_data['title']
 163
 164         formats = [{
 165             'url': stream['m3u8_url'],
 166             'format_id': self.get_format_name(stream.get('stream_type')),
 167             'ext': 'mp4',
 168             'protocol': 'm3u8_native',
 169             'filesize': int(stream.get('size')),
 170             'width': stream.get('width'),
 171             'height': stream.get('height'),
 172         } for stream in data['stream'] if stream.get('channel_type') != 'tail']
 173
 174         return {
 175             'id': video_id,
 176             'title': title,
 177             'formats': formats,
 178             'duration': video_data.get('seconds'),
 179             'thumbnail': video_data.get('logo'),
 180             'uploader': video_data.get('username'),
 181             'uploader_id': str_or_none(video_data.get('userid')),
 182             'uploader_url': data.get('uploader', {}).get('homepage'),
 183             'tags': video_data.get('tags'),
 184         }
 185
 186
 187 class YoukuShowIE(InfoExtractor):
 188     _VALID_URL = r'https?://list\.youku\.com/show/id_(?P<id>[0-9a-z]+)\.html'
 189     IE_NAME = 'youku:show'
 190
 191     _TESTS = [{
 192         'url': 'http://list.youku.com/show/id_zc7c670be07ff11e48b3f.html',
 193         'info_dict': {
 194             'id': 'zc7c670be07ff11e48b3f',
 195             'title': '花千骨 DVD版',
 196             'description': 'md5:a1ae6f5618571bbeb5c9821f9c81b558',
 197         },
 198         'playlist_count': 50,
 199     }, {
 200         # Episode number not starting from 1
 201         'url': 'http://list.youku.com/show/id_zefbfbd70efbfbd780bef.html',
 202         'info_dict': {
 203             'id': 'zefbfbd70efbfbd780bef',
 204             'title': '超级飞侠3',
 205             'description': 'md5:275715156abebe5ccc2a1992e9d56b98',
 206         },
 207         'playlist_count': 24,
 208     }, {
 209         # Ongoing playlist. The initial page is the last one
 210         'url': 'http://list.youku.com/show/id_za7c275ecd7b411e1a19e.html',
 211         'only_matching': True,
 212     }, {
 213         #  No data-id value.
 214         'url': 'http://list.youku.com/show/id_zefbfbd61237fefbfbdef.html',
 215         'only_matching': True,
 216     }, {
 217         #  Wrong number of reload_id.
 218         'url': 'http://list.youku.com/show/id_z20eb4acaf5c211e3b2ad.html',
 219         'only_matching': True,
 220     }]
 221
 222     def _extract_entries(self, playlist_data_url, show_id, note, query):
 223         query['callback'] = 'cb'
 224         playlist_data = self._download_json(
 225             playlist_data_url, show_id, query=query, note=note,
 226             transform_source=lambda s: js_to_json(strip_jsonp(s))).get('html')
 227         if playlist_data is None:
 228             return [None, None]
 229         drama_list = (get_element_by_class('p-drama-grid', playlist_data)
 230                       or get_element_by_class('p-drama-half-row', playlist_data))
 231         if drama_list is None:
 232             raise ExtractorError('No episodes found')
 233         video_urls = re.findall(r'<a[^>]+href="([^"]+)"', drama_list)
 234         return playlist_data, [
 235             self.url_result(self._proto_relative_url(video_url, 'http:'), YoukuIE.ie_key())
 236             for video_url in video_urls]
 237
 238     def _real_extract(self, url):
 239         show_id = self._match_id(url)
 240         webpage = self._download_webpage(url, show_id)
 241
 242         entries = []
 243         page_config = self._parse_json(self._search_regex(
 244             r'var\s+PageConfig\s*=\s*({.+});', webpage, 'page config'),
 245             show_id, transform_source=js_to_json)
 246         first_page, initial_entries = self._extract_entries(
 247             'http://list.youku.com/show/module', show_id,
 248             note='Downloading initial playlist data page',
 249             query={
 250                 'id': page_config['showid'],
 251                 'tab': 'showInfo',
 252             })
 253         first_page_reload_id = self._html_search_regex(
 254             r'<div[^>]+id="(reload_\d+)', first_page, 'first page reload id')
 255         # The first reload_id has the same items as first_page
 256         reload_ids = re.findall('<li[^>]+data-id="([^"]+)">', first_page)
 257         entries.extend(initial_entries)
 258         for idx, reload_id in enumerate(reload_ids):
 259             if reload_id == first_page_reload_id:
 260                 continue
 261             _, new_entries = self._extract_entries(
 262                 'http://list.youku.com/show/episode', show_id,
 263                 note='Downloading playlist data page %d' % (idx + 1),
 264                 query={
 265                     'id': page_config['showid'],
 266                     'stage': reload_id,
 267                 })
 268             if new_entries is not None:
 269                 entries.extend(new_entries)
 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)