yt_dlp/extractor/sohu.py

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