yt_dlp/extractor/xiaohongshu.py

   1
   2 from .common import InfoExtractor
   3 from ..utils import (
   4     float_or_none,
   5     int_or_none,
   6     js_to_json,
   7     url_or_none,
   8 )
   9 from ..utils.traversal import traverse_obj
  10
  11
  12 class XiaoHongShuIE(InfoExtractor):
  13     _VALID_URL = r'https?://www\.xiaohongshu\.com/explore/(?P<id>[\da-f]+)'
  14     IE_DESC = '小红书'
  15     _TESTS = [{
  16         'url': 'https://www.xiaohongshu.com/explore/6411cf99000000001300b6d9',
  17         'md5': '2a87a77ddbedcaeeda8d7eae61b61228',
  18         'info_dict': {
  19             'id': '6411cf99000000001300b6d9',
  20             'ext': 'mp4',
  21             'uploader_id': '5c31698d0000000007018a31',
  22             'description': '#今日快乐今日发[话题]# #吃货薯看这里[话题]# #香妃蛋糕[话题]# #小五卷蛋糕[话题]# #新手蛋糕卷[话题]#',
  23             'title': '香妃蛋糕也太香了吧🔥不需要卷❗️绝对的友好',
  24             'tags': ['今日快乐今日发', '吃货薯看这里', '香妃蛋糕', '小五卷蛋糕', '新手蛋糕卷'],
  25             'duration': 101.726,
  26             'thumbnail': r're:https?://sns-webpic-qc\.xhscdn\.com/\d+/[a-z0-9]+/[\w]+',
  27         },
  28     }]
  29
  30     def _real_extract(self, url):
  31         display_id = self._match_id(url)
  32         webpage = self._download_webpage(url, display_id)
  33         initial_state = self._search_json(
  34             r'window\.__INITIAL_STATE__\s*=', webpage, 'initial state', display_id, transform_source=js_to_json)
  35
  36         note_info = traverse_obj(initial_state, ('note', 'noteDetailMap', display_id, 'note'))
  37         video_info = traverse_obj(note_info, ('video', 'media', 'stream', ('h264', 'av1', 'h265'), ...))
  38
  39         formats = []
  40         for info in video_info:
  41             format_info = traverse_obj(info, {
  42                 'fps': ('fps', {int_or_none}),
  43                 'width': ('width', {int_or_none}),
  44                 'height': ('height', {int_or_none}),
  45                 'vcodec': ('videoCodec', {str}),
  46                 'acodec': ('audioCodec', {str}),
  47                 'abr': ('audioBitrate', {int_or_none}),
  48                 'vbr': ('videoBitrate', {int_or_none}),
  49                 'audio_channels': ('audioChannels', {int_or_none}),
  50                 'tbr': ('avgBitrate', {int_or_none}),
  51                 'format': ('qualityType', {str}),
  52                 'filesize': ('size', {int_or_none}),
  53                 'duration': ('duration', {float_or_none(scale=1000)}),
  54             })
  55
  56             formats.extend(traverse_obj(info, (('mediaUrl', ('backupUrls', ...)), {
  57                 lambda u: url_or_none(u) and {'url': u, **format_info}})))
  58
  59         thumbnails = []
  60         for image_info in traverse_obj(note_info, ('imageList', ...)):
  61             thumbnail_info = traverse_obj(image_info, {
  62                 'height': ('height', {int_or_none}),
  63                 'width': ('width', {int_or_none}),
  64             })
  65             for thumb_url in traverse_obj(image_info, (('urlDefault', 'urlPre'), {url_or_none})):
  66                 thumbnails.append({
  67                     'url': thumb_url,
  68                     **thumbnail_info,
  69                 })
  70
  71         return {
  72             'id': display_id,
  73             'formats': formats,
  74             'thumbnails': thumbnails,
  75             'title': self._html_search_meta(['og:title'], webpage, default=None),
  76             **traverse_obj(note_info, {
  77                 'title': ('title', {str}),
  78                 'description': ('desc', {str}),
  79                 'tags': ('tagList', ..., 'name', {str}),
  80                 'uploader_id': ('user', 'userId', {str}),
  81             }),
  82         }