Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/youku.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import random
   5 import re
   6 import string
   7 import time
   8
   9 from .common import InfoExtractor
  10 from ..utils import (
  11     ExtractorError,
  12     get_element_by_class,
  13     js_to_json,
  14     str_or_none,
  15     strip_jsonp,
  16 )
  17
  18
  19 class YoukuIE(InfoExtractor):
  20     IE_NAME = 'youku'
  21     IE_DESC = '优酷'
  22     _VALID_URL = r'''(?x)
  23         (?:
  24             https?://(
  25                 (?:v|player)\.youku\.com/(?:v_show/id_|player\.php/sid/)|
  26                 video\.tudou\.com/v/)|
  27             youku:)
  28         (?P<id>[A-Za-z0-9]+)(?:\.html|/v\.swf|)
  29     '''
  30
  31     _TESTS = [{
  32         # MD5 is unstable
  33         'url': 'http://v.youku.com/v_show/id_XMTc1ODE5Njcy.html',
  34         'info_dict': {
  35             'id': 'XMTc1ODE5Njcy',
  36             'title': '★Smile﹗♡ Git Fresh -Booty Music舞蹈.',
  37             'ext': 'mp4',
  38             'duration': 74.73,
  39             'thumbnail': r're:^https?://.*',
  40             'uploader': '。躲猫猫、',
  41             'uploader_id': '36017967',
  42             'uploader_url': 'http://i.youku.com/u/UMTQ0MDcxODY4',
  43             'tags': list,
  44         }
  45     }, {
  46         'url': 'http://player.youku.com/player.php/sid/XNDgyMDQ2NTQw/v.swf',
  47         'only_matching': True,
  48     }, {
  49         'url': 'http://v.youku.com/v_show/id_XODgxNjg1Mzk2_ev_1.html',
  50         'info_dict': {
  51             'id': 'XODgxNjg1Mzk2',
  52             'ext': 'mp4',
  53             'title': '武媚娘传奇 85',
  54             'duration': 1999.61,
  55             'thumbnail': r're:^https?://.*',
  56             'uploader': '疯狂豆花',
  57             'uploader_id': '62583473',
  58             'uploader_url': 'http://i.youku.com/u/UMjUwMzMzODky',
  59             'tags': list,
  60         },
  61     }, {
  62         'url': 'http://v.youku.com/v_show/id_XMTI1OTczNDM5Mg==.html',
  63         'info_dict': {
  64             'id': 'XMTI1OTczNDM5Mg',
  65             'ext': 'mp4',
  66             'title': '花千骨 04',
  67             'duration': 2363,
  68             'thumbnail': r're:^https?://.*',
  69             'uploader': '放剧场-花千骨',
  70             'uploader_id': '772849359',
  71             'uploader_url': 'http://i.youku.com/u/UMzA5MTM5NzQzNg==',
  72             'tags': list,
  73         },
  74     }, {
  75         'url': 'http://v.youku.com/v_show/id_XNjA1NzA2Njgw.html',
  76         'note': 'Video protected with password',
  77         'info_dict': {
  78             'id': 'XNjA1NzA2Njgw',
  79             'ext': 'mp4',
  80             'title': '邢義田复旦讲座之想象中的胡人—从“左衽孔子”说起',
  81             'duration': 7264.5,
  82             'thumbnail': r're:^https?://.*',
  83             'uploader': 'FoxJin1006',
  84             'uploader_id': '322014285',
  85             'uploader_url': 'http://i.youku.com/u/UMTI4ODA1NzE0MA==',
  86             'tags': list,
  87         },
  88         'params': {
  89             'videopassword': '100600',
  90         },
  91     }, {
  92         # /play/get.json contains streams with "channel_type":"tail"
  93         'url': 'http://v.youku.com/v_show/id_XOTUxMzg4NDMy.html',
  94         'info_dict': {
  95             'id': 'XOTUxMzg4NDMy',
  96             'ext': 'mp4',
  97             'title': '我的世界☆明月庄主☆车震猎杀☆杀人艺术Minecraft',
  98             'duration': 702.08,
  99             'thumbnail': r're:^https?://.*',
 100             'uploader': '明月庄主moon',
 101             'uploader_id': '38465621',
 102             'uploader_url': 'http://i.youku.com/u/UMTUzODYyNDg0',
 103             'tags': list,
 104         },
 105     }, {
 106         'url': 'http://video.tudou.com/v/XMjIyNzAzMTQ4NA==.html?f=46177805',
 107         'info_dict': {
 108             'id': 'XMjIyNzAzMTQ4NA',
 109             'ext': 'mp4',
 110             'title': '卡马乔国足开大脚长传冲吊集锦',
 111             'duration': 289,
 112             'thumbnail': r're:^https?://.*',
 113             'uploader': '阿卜杜拉之星',
 114             'uploader_id': '2382249',
 115             'uploader_url': 'http://i.youku.com/u/UOTUyODk5Ng==',
 116             'tags': list,
 117         },
 118     }, {
 119         'url': 'http://video.tudou.com/v/XMjE4ODI3OTg2MA==.html',
 120         'only_matching': True,
 121     }]
 122
 123     @staticmethod
 124     def get_ysuid():
 125         return '%d%s' % (int(time.time()), ''.join([
 126             random.choice(string.ascii_letters) for i in range(3)]))
 127
 128     def get_format_name(self, fm):
 129         _dict = {
 130             '3gp': 'h6',
 131             '3gphd': 'h5',
 132             'flv': 'h4',
 133             'flvhd': 'h4',
 134             'mp4': 'h3',
 135             'mp4hd': 'h3',
 136             'mp4hd2': 'h4',
 137             'mp4hd3': 'h4',
 138             'hd2': 'h2',
 139             'hd3': 'h1',
 140         }
 141         return _dict.get(fm)
 142
 143     def _real_extract(self, url):
 144         video_id = self._match_id(url)
 145
 146         self._set_cookie('youku.com', '__ysuid', self.get_ysuid())
 147         self._set_cookie('youku.com', 'xreferrer', 'http://www.youku.com')
 148
 149         _, urlh = self._download_webpage_handle(
 150             'https://log.mmstat.com/eg.js', video_id, 'Retrieving cna info')
 151         # The etag header is '"foobar"'; let's remove the double quotes
 152         cna = urlh.headers['etag'][1:-1]
 153
 154         # request basic data
 155         basic_data_params = {
 156             'vid': video_id,
 157             'ccode': '0590',
 158             'client_ip': '192.168.1.1',
 159             'utid': cna,
 160             'client_ts': time.time() / 1000,
 161         }
 162
 163         video_password = self._downloader.params.get('videopassword')
 164         if video_password:
 165             basic_data_params['password'] = video_password
 166
 167         headers = {
 168             'Referer': url,
 169         }
 170         headers.update(self.geo_verification_headers())
 171         data = self._download_json(
 172             'https://ups.youku.com/ups/get.json', video_id,
 173             'Downloading JSON metadata',
 174             query=basic_data_params, headers=headers)['data']
 175
 176         error = data.get('error')
 177         if error:
 178             error_note = error.get('note')
 179             if error_note is not None and '因版权原因无法观看此视频' in error_note:
 180                 raise ExtractorError(
 181                     'Youku said: Sorry, this video is available in China only', expected=True)
 182             elif error_note and '该视频被设为私密' in error_note:
 183                 raise ExtractorError(
 184                     'Youku said: Sorry, this video is private', expected=True)
 185             else:
 186                 msg = 'Youku server reported error %i' % error.get('code')
 187                 if error_note is not None:
 188                     msg += ': ' + error_note
 189                 raise ExtractorError(msg)
 190
 191         # get video title
 192         video_data = data['video']
 193         title = video_data['title']
 194
 195         formats = [{
 196             'url': stream['m3u8_url'],
 197             'format_id': self.get_format_name(stream.get('stream_type')),
 198             'ext': 'mp4',
 199             'protocol': 'm3u8_native',
 200             'filesize': int(stream.get('size')),
 201             'width': stream.get('width'),
 202             'height': stream.get('height'),
 203         } for stream in data['stream'] if stream.get('channel_type') != 'tail']
 204         self._sort_formats(formats)
 205
 206         return {
 207             'id': video_id,
 208             'title': title,
 209             'formats': formats,
 210             'duration': video_data.get('seconds'),
 211             'thumbnail': video_data.get('logo'),
 212             'uploader': video_data.get('username'),
 213             'uploader_id': str_or_none(video_data.get('userid')),
 214             'uploader_url': data.get('uploader', {}).get('homepage'),
 215             'tags': video_data.get('tags'),
 216         }
 217
 218
 219 class YoukuShowIE(InfoExtractor):
 220     _VALID_URL = r'https?://list\.youku\.com/show/id_(?P<id>[0-9a-z]+)\.html'
 221     IE_NAME = 'youku:show'
 222
 223     _TESTS = [{
 224         'url': 'http://list.youku.com/show/id_zc7c670be07ff11e48b3f.html',
 225         'info_dict': {
 226             'id': 'zc7c670be07ff11e48b3f',
 227             'title': '花千骨 DVD版',
 228             'description': 'md5:a1ae6f5618571bbeb5c9821f9c81b558',
 229         },
 230         'playlist_count': 50,
 231     }, {
 232         # Episode number not starting from 1
 233         'url': 'http://list.youku.com/show/id_zefbfbd70efbfbd780bef.html',
 234         'info_dict': {
 235             'id': 'zefbfbd70efbfbd780bef',
 236             'title': '超级飞侠3',
 237             'description': 'md5:275715156abebe5ccc2a1992e9d56b98',
 238         },
 239         'playlist_count': 24,
 240     }, {
 241         # Ongoing playlist. The initial page is the last one
 242         'url': 'http://list.youku.com/show/id_za7c275ecd7b411e1a19e.html',
 243         'only_matching': True,
 244     }, {
 245         #  No data-id value.
 246         'url': 'http://list.youku.com/show/id_zefbfbd61237fefbfbdef.html',
 247         'only_matching': True,
 248     }, {
 249         #  Wrong number of reload_id.
 250         'url': 'http://list.youku.com/show/id_z20eb4acaf5c211e3b2ad.html',
 251         'only_matching': True,
 252     }]
 253
 254     def _extract_entries(self, playlist_data_url, show_id, note, query):
 255         query['callback'] = 'cb'
 256         playlist_data = self._download_json(
 257             playlist_data_url, show_id, query=query, note=note,
 258             transform_source=lambda s: js_to_json(strip_jsonp(s))).get('html')
 259         if playlist_data is None:
 260             return [None, None]
 261         drama_list = (get_element_by_class('p-drama-grid', playlist_data)
 262                       or get_element_by_class('p-drama-half-row', playlist_data))
 263         if drama_list is None:
 264             raise ExtractorError('No episodes found')
 265         video_urls = re.findall(r'<a[^>]+href="([^"]+)"', drama_list)
 266         return playlist_data, [
 267             self.url_result(self._proto_relative_url(video_url, 'http:'), YoukuIE.ie_key())
 268             for video_url in video_urls]
 269
 270     def _real_extract(self, url):
 271         show_id = self._match_id(url)
 272         webpage = self._download_webpage(url, show_id)
 273
 274         entries = []
 275         page_config = self._parse_json(self._search_regex(
 276             r'var\s+PageConfig\s*=\s*({.+});', webpage, 'page config'),
 277             show_id, transform_source=js_to_json)
 278         first_page, initial_entries = self._extract_entries(
 279             'http://list.youku.com/show/module', show_id,
 280             note='Downloading initial playlist data page',
 281             query={
 282                 'id': page_config['showid'],
 283                 'tab': 'showInfo',
 284             })
 285         first_page_reload_id = self._html_search_regex(
 286             r'<div[^>]+id="(reload_\d+)', first_page, 'first page reload id')
 287         # The first reload_id has the same items as first_page
 288         reload_ids = re.findall('<li[^>]+data-id="([^"]+)">', first_page)
 289         entries.extend(initial_entries)
 290         for idx, reload_id in enumerate(reload_ids):
 291             if reload_id == first_page_reload_id:
 292                 continue
 293             _, new_entries = self._extract_entries(
 294                 'http://list.youku.com/show/episode', show_id,
 295                 note='Downloading playlist data page %d' % (idx + 1),
 296                 query={
 297                     'id': page_config['showid'],
 298                     'stage': reload_id,
 299                 })
 300             if new_entries is not None:
 301                 entries.extend(new_entries)
 302         desc = self._html_search_meta('description', webpage, fatal=False)
 303         playlist_title = desc.split(',')[0] if desc else None
 304         detail_li = get_element_by_class('p-intro', webpage)
 305         playlist_description = get_element_by_class(
 306             'intro-more', detail_li) if detail_li else None
 307
 308         return self.playlist_result(
 309             entries, show_id, playlist_title, playlist_description)