yt_dlp/extractor/sohu.py

   1 import base64
   2 import re
   3 import urllib.parse
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     ExtractorError,
   8     float_or_none,
   9     int_or_none,
  10     traverse_obj,
  11     try_get,
  12     unified_timestamp,
  13     url_or_none,
  14     urljoin,
  15 )
  16
  17
  18 class SohuIE(InfoExtractor):
  19     _VALID_URL = r'https?://(?P<mytv>my\.)?tv\.sohu\.com/.+?/(?(mytv)|n)(?P<id>\d+)\.shtml.*?'
  20
  21     # Sohu videos give different MD5 sums on Travis CI and my machine
  22     _TESTS = [{
  23         'note': 'This video is available only in Mainland China',
  24         'url': 'http://tv.sohu.com/20130724/n382479172.shtml#super',
  25         'info_dict': {
  26             'id': '382479172',
  27             'ext': 'mp4',
  28             'title': 'MV：Far East Movement《The Illest》',
  29         },
  30         'skip': 'On available in China',
  31     }, {
  32         'url': 'http://tv.sohu.com/20150305/n409385080.shtml',
  33         'info_dict': {
  34             'id': '409385080',
  35             'ext': 'mp4',
  36             'title': '《2015湖南卫视羊年元宵晚会》唐嫣《花好月圆》',
  37         },
  38         'skip': 'no longer available',
  39     }, {
  40         'url': 'http://my.tv.sohu.com/us/232799889/78693464.shtml',
  41         'info_dict': {
  42             'id': '78693464',
  43             'ext': 'mp4',
  44             'title': '【爱范品】第31期：MWC见不到的奇葩手机',
  45             'uploader': '爱范儿视频',
  46             'duration': 213,
  47             'timestamp': 1425519600,
  48             'upload_date': '20150305',
  49             'thumbnail': 'http://e3f49eaa46b57.cdn.sohucs.com//group1/M10/83/FA/MTAuMTAuODguODA=/6_14cbccdde5eg104SysCutcloud_78693464_7_0b.jpg',
  50             'tags': ['爱范儿', '爱范品', 'MWC', '手机'],
  51         },
  52     }, {
  53         'note': 'Multipart video',
  54         'url': 'http://my.tv.sohu.com/pl/8384802/78910339.shtml',
  55         'info_dict': {
  56             'id': '78910339',
  57             'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆',
  58             'uploader': '小苍cany',
  59             'duration': 744.0,
  60             'timestamp': 1426269360,
  61             'upload_date': '20150313',
  62             'thumbnail': 'http://e3f49eaa46b57.cdn.sohucs.com//group1/M11/89/57/MTAuMTAuODguODA=/6_14cea022a1dg102SysCutcloud_78910339_8_0b.jpg',
  63             'tags': ['小苍MM', '英雄联盟', '实战秘籍'],
  64         },
  65         'playlist': [{
  66             'info_dict': {
  67                 'id': '78910339_part1',
  68                 'ext': 'mp4',
  69                 'duration': 294,
  70                 'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆',
  71             },
  72         }, {
  73             'info_dict': {
  74                 'id': '78910339_part2',
  75                 'ext': 'mp4',
  76                 'duration': 300,
  77                 'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆',
  78             },
  79         }, {
  80             'info_dict': {
  81                 'id': '78910339_part3',
  82                 'ext': 'mp4',
  83                 'duration': 150,
  84                 'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆',
  85             },
  86         }],
  87     }, {
  88         'note': 'Video with title containing dash',
  89         'url': 'http://my.tv.sohu.com/us/249884221/78932792.shtml',
  90         'info_dict': {
  91             'id': '78932792',
  92             'ext': 'mp4',
  93             'title': 'youtube-dl testing video',
  94             'duration': 360,
  95             'timestamp': 1426348620,
  96             'upload_date': '20150314',
  97             'thumbnail': 'http://e3f49eaa46b57.cdn.sohucs.com//group1/M02/8A/00/MTAuMTAuODguNzk=/6_14cee1be192g102SysCutcloud_78932792_7_7b.jpg',
  98             'tags': [],
  99         },
 100         'params': {
 101             'skip_download': True,
 102         },
 103     }]
 104
 105     def _real_extract(self, url):
 106
 107         def _fetch_data(vid_id, mytv=False):
 108             if mytv:
 109                 base_data_url = 'http://my.tv.sohu.com/play/videonew.do?vid='
 110             else:
 111                 base_data_url = 'http://hot.vrs.sohu.com/vrs_flash.action?vid='
 112
 113             return self._download_json(
 114                 base_data_url + vid_id, video_id,
 115                 f'Downloading JSON data for {vid_id}',
 116                 headers=self.geo_verification_headers())
 117
 118         mobj = self._match_valid_url(url)
 119         video_id = mobj.group('id')
 120         mytv = mobj.group('mytv') is not None
 121
 122         webpage = self._download_webpage(url, video_id)
 123
 124         title = re.sub(r'( - 高清正版在线观看)? - 搜狐视频$', '', self._og_search_title(webpage))
 125
 126         vid = self._html_search_regex(
 127             r'var vid ?= ?["\'](\d+)["\']',
 128             webpage, 'video path')
 129         vid_data = _fetch_data(vid, mytv)
 130         if vid_data['play'] != 1:
 131             if vid_data.get('status') == 12:
 132                 raise ExtractorError(
 133                     f'{self.IE_NAME} said: There\'s something wrong in the video.',
 134                     expected=True)
 135             else:
 136                 self.raise_geo_restricted(
 137                     f'{self.IE_NAME} said: The video is only licensed to users in Mainland China.')
 138
 139         formats_json = {}
 140         for format_id in ('nor', 'high', 'super', 'ori', 'h2644k', 'h2654k'):
 141             vid_id = vid_data['data'].get(f'{format_id}Vid')
 142             if not vid_id:
 143                 continue
 144             vid_id = str(vid_id)
 145             formats_json[format_id] = vid_data if vid == vid_id else _fetch_data(vid_id, mytv)
 146
 147         part_count = vid_data['data']['totalBlocks']
 148
 149         playlist = []
 150         for i in range(part_count):
 151             formats = []
 152             for format_id, format_data in formats_json.items():
 153                 allot = format_data['allot']
 154
 155                 data = format_data['data']
 156                 clip_url = traverse_obj(data, (('clipsURL', 'mp4PlayUrl'), i, {url_or_none}), get_all=False)
 157                 if not clip_url:
 158                     raise ExtractorError(f'Unable to extract url for clip {i}')
 159                 su = data['su']
 160
 161                 video_url = 'newflv.sohu.ccgslb.net'
 162                 cdn_id = None
 163                 retries = 0
 164
 165                 while 'newflv.sohu.ccgslb.net' in video_url:
 166                     params = {
 167                         'prot': 9,
 168                         'file': clip_url,
 169                         'new': su[i],
 170                         'prod': 'h5n',
 171                         'rb': 1,
 172                     }
 173
 174                     if cdn_id is not None:
 175                         params['idc'] = cdn_id
 176
 177                     download_note = f'Downloading {format_id} video URL part {i + 1} of {part_count}'
 178
 179                     if retries > 0:
 180                         download_note += f' (retry #{retries})'
 181                     part_info = self._parse_json(self._download_webpage(
 182                         f'http://{allot}/?{urllib.parse.urlencode(params)}',
 183                         video_id, download_note), video_id)
 184
 185                     video_url = part_info['url']
 186                     cdn_id = part_info.get('nid')
 187
 188                     retries += 1
 189                     if retries > 5:
 190                         raise ExtractorError('Failed to get video URL')
 191
 192                 formats.append({
 193                     'url': video_url,
 194                     'format_id': format_id,
 195                     'filesize': int_or_none(
 196                         try_get(data, lambda x: x['clipsBytes'][i])),
 197                     'width': int_or_none(data.get('width')),
 198                     'height': int_or_none(data.get('height')),
 199                     'fps': int_or_none(data.get('fps')),
 200                 })
 201
 202             playlist.append({
 203                 'id': f'{video_id}_part{i + 1}',
 204                 'title': title,
 205                 'duration': vid_data['data']['clipsDuration'][i],
 206                 'formats': formats,
 207             })
 208
 209         if len(playlist) == 1:
 210             info = playlist[0]
 211             info['id'] = video_id
 212         else:
 213             info = {
 214                 '_type': 'multi_video',
 215                 'entries': playlist,
 216                 'id': video_id,
 217                 'title': title,
 218                 'duration': traverse_obj(vid_data, ('data', 'totalDuration', {float_or_none})),
 219             }
 220
 221         if mytv:
 222             publish_time = unified_timestamp(self._search_regex(
 223                 r'publishTime:\s*["\'](\d+-\d+-\d+ \d+:\d+)["\']', webpage, 'publish time', fatal=False))
 224         else:
 225             publish_time = traverse_obj(vid_data, ('tv_application_time', {unified_timestamp}))
 226
 227         return {
 228             'timestamp': publish_time - 8 * 3600 if publish_time else None,
 229             **traverse_obj(vid_data, {
 230                 'alt_title': ('data', 'subName', {str}),
 231                 'uploader': ('wm_data', 'wm_username', {str}),
 232                 'thumbnail': ('data', 'coverImg', {url_or_none}),
 233                 'tags': ('data', 'tag', {str.split}),
 234             }),
 235             **info,
 236         }
 237
 238
 239 class SohuVIE(InfoExtractor):
 240     _VALID_URL = r'https?://tv\.sohu\.com/v/(?P<id>[\w=-]+)\.html(?:$|[#?])'
 241
 242     _TESTS = [{
 243         'note': 'Multipart video',
 244         'url': 'https://tv.sohu.com/v/MjAyMzA2MTQvbjYwMTMxNTE5Mi5zaHRtbA==.html',
 245         'info_dict': {
 246             'id': '601315192',
 247             'title': '《淬火丹心》第1集',
 248             'alt_title': '“点天灯”发生事故',
 249             'duration': 2701.692,
 250             'timestamp': 1686758040,
 251             'upload_date': '20230614',
 252             'thumbnail': 'http://photocdn.tv.sohu.com/img/20230614/vrsa_hor_1686738763256_454010551.jpg',
 253         },
 254         'playlist_mincount': 9,
 255         'skip': 'Only available in China',
 256     }, {
 257         'url': 'https://tv.sohu.com/v/dXMvMjMyNzk5ODg5Lzc4NjkzNDY0LnNodG1s.html',
 258         'info_dict': {
 259             'id': '78693464',
 260             'ext': 'mp4',
 261             'title': '【爱范品】第31期：MWC见不到的奇葩手机',
 262             'uploader': '爱范儿视频',
 263             'duration': 213,
 264             'timestamp': 1425519600,
 265             'upload_date': '20150305',
 266             'thumbnail': 'http://e3f49eaa46b57.cdn.sohucs.com//group1/M10/83/FA/MTAuMTAuODguODA=/6_14cbccdde5eg104SysCutcloud_78693464_7_0b.jpg',
 267             'tags': ['爱范儿', '爱范品', 'MWC', '手机'],
 268         },
 269     }, {
 270         'note': 'Multipart video',
 271         'url': 'https://tv.sohu.com/v/dXMvMjQyNTYyMTYzLzc4OTEwMzM5LnNodG1s.html?src=pl',
 272         'info_dict': {
 273             'id': '78910339',
 274             'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆',
 275             'uploader': '小苍cany',
 276             'duration': 744.0,
 277             'timestamp': 1426269360,
 278             'upload_date': '20150313',
 279             'thumbnail': 'http://e3f49eaa46b57.cdn.sohucs.com//group1/M11/89/57/MTAuMTAuODguODA=/6_14cea022a1dg102SysCutcloud_78910339_8_0b.jpg',
 280             'tags': ['小苍MM', '英雄联盟', '实战秘籍'],
 281         },
 282         'playlist_mincount': 3,
 283     }]
 284
 285     def _real_extract(self, url):
 286         encoded_id = self._match_id(url)
 287         path = base64.urlsafe_b64decode(encoded_id).decode()
 288         subdomain = 'tv' if re.match(r'\d+/n\d+\.shtml', path) else 'my.tv'
 289         return self.url_result(urljoin(f'http://{subdomain}.sohu.com/', path), SohuIE)