yt_dlp/extractor/douyin.py

   1 # coding: utf-8
   2
   3 from ..utils import (
   4     int_or_none,
   5     traverse_obj,
   6     url_or_none,
   7 )
   8 from .common import (
   9     InfoExtractor,
  10     compat_urllib_parse_unquote,
  11 )
  12
  13
  14 class DouyinIE(InfoExtractor):
  15     _VALID_URL = r'https?://(?:www\.)?douyin\.com/video/(?P<id>[0-9]+)'
  16     _TESTS = [{
  17         'url': 'https://www.douyin.com/video/6961737553342991651',
  18         'md5': '10523312c8b8100f353620ac9dc8f067',
  19         'info_dict': {
  20             'id': '6961737553342991651',
  21             'ext': 'mp4',
  22             'title': '#杨超越  小小水手带你去远航❤️',
  23             'uploader': '杨超越',
  24             'upload_date': '20210513',
  25             'timestamp': 1620905839,
  26             'uploader_id': '110403406559',
  27             'view_count': int,
  28             'like_count': int,
  29             'repost_count': int,
  30             'comment_count': int,
  31         }
  32     }, {
  33         'url': 'https://www.douyin.com/video/6982497745948921092',
  34         'md5': 'd78408c984b9b5102904cf6b6bc2d712',
  35         'info_dict': {
  36             'id': '6982497745948921092',
  37             'ext': 'mp4',
  38             'title': '这个夏日和小羊@杨超越 一起遇见白色幻想',
  39             'uploader': '杨超越工作室',
  40             'upload_date': '20210708',
  41             'timestamp': 1625739481,
  42             'uploader_id': '408654318141572',
  43             'view_count': int,
  44             'like_count': int,
  45             'repost_count': int,
  46             'comment_count': int,
  47         }
  48     }, {
  49         'url': 'https://www.douyin.com/video/6953975910773099811',
  50         'md5': '72e882e24f75064c218b76c8b713c185',
  51         'info_dict': {
  52             'id': '6953975910773099811',
  53             'ext': 'mp4',
  54             'title': '#一起看海  出现在你的夏日里',
  55             'uploader': '杨超越',
  56             'upload_date': '20210422',
  57             'timestamp': 1619098692,
  58             'uploader_id': '110403406559',
  59             'view_count': int,
  60             'like_count': int,
  61             'repost_count': int,
  62             'comment_count': int,
  63         }
  64     }, {
  65         'url': 'https://www.douyin.com/video/6950251282489675042',
  66         'md5': 'b4db86aec367ef810ddd38b1737d2fed',
  67         'info_dict': {
  68             'id': '6950251282489675042',
  69             'ext': 'mp4',
  70             'title': '哈哈哈，成功了哈哈哈哈哈哈',
  71             'uploader': '杨超越',
  72             'upload_date': '20210412',
  73             'timestamp': 1618231483,
  74             'uploader_id': '110403406559',
  75             'view_count': int,
  76             'like_count': int,
  77             'repost_count': int,
  78             'comment_count': int,
  79         }
  80     }, {
  81         'url': 'https://www.douyin.com/video/6963263655114722595',
  82         'md5': '1abe1c477d05ee62efb40bf2329957cf',
  83         'info_dict': {
  84             'id': '6963263655114722595',
  85             'ext': 'mp4',
  86             'title': '#哪个爱豆的105度最甜 换个角度看看我哈哈',
  87             'uploader': '杨超越',
  88             'upload_date': '20210517',
  89             'timestamp': 1621261163,
  90             'uploader_id': '110403406559',
  91             'view_count': int,
  92             'like_count': int,
  93             'repost_count': int,
  94             'comment_count': int,
  95         }
  96     }]
  97
  98     def _real_extract(self, url):
  99         video_id = self._match_id(url)
 100         webpage = self._download_webpage(url, video_id)
 101         render_data = self._parse_json(
 102             self._search_regex(
 103                 r'<script [^>]*\bid=[\'"]RENDER_DATA[\'"][^>]*>(%7B.+%7D)</script>',
 104                 webpage, 'render data'),
 105             video_id, transform_source=compat_urllib_parse_unquote)
 106         details = traverse_obj(render_data, (..., 'aweme', 'detail'), get_all=False)
 107
 108         thumbnails = [{'url': self._proto_relative_url(url)} for url in traverse_obj(
 109             details, ('video', ('cover', 'dynamicCover', 'originCover')), expected_type=url_or_none, default=[])]
 110
 111         common = {
 112             'width': traverse_obj(details, ('video', 'width'), expected_type=int),
 113             'height': traverse_obj(details, ('video', 'height'), expected_type=int),
 114             'ext': 'mp4',
 115         }
 116         formats = [{**common, 'url': self._proto_relative_url(url)} for url in traverse_obj(
 117             details, ('video', 'playAddr', ..., 'src'), expected_type=url_or_none, default=[]) if url]
 118         self._remove_duplicate_formats(formats)
 119
 120         download_url = traverse_obj(details, ('download', 'url'), expected_type=url_or_none)
 121         if download_url:
 122             formats.append({
 123                 **common,
 124                 'format_id': 'download',
 125                 'url': self._proto_relative_url(download_url),
 126                 'quality': 1,
 127             })
 128         self._sort_formats(formats)
 129
 130         return {
 131             'id': video_id,
 132             'title': details.get('desc') or self._html_search_meta('title', webpage),
 133             'formats': formats,
 134             'thumbnails': thumbnails,
 135             'uploader': traverse_obj(details, ('authorInfo', 'nickname'), expected_type=str),
 136             'uploader_id': traverse_obj(details, ('authorInfo', 'uid'), expected_type=str),
 137             'uploader_url': 'https://www.douyin.com/user/%s' % traverse_obj(
 138                 details, ('authorInfo', 'secUid'), expected_type=str),
 139             'timestamp': int_or_none(details.get('createTime')),
 140             'duration': traverse_obj(details, ('video', 'duration'), expected_type=int),
 141             'view_count': traverse_obj(details, ('stats', 'playCount'), expected_type=int),
 142             'like_count': traverse_obj(details, ('stats', 'diggCount'), expected_type=int),
 143             'repost_count': traverse_obj(details, ('stats', 'shareCount'), expected_type=int),
 144             'comment_count': traverse_obj(details, ('stats', 'commentCount'), expected_type=int),
 145         }