_ Git - youtube-dl/blob - youtube_dl/extractor/tiktok.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     compat_str,
   7     ExtractorError,
   8     int_or_none,
   9     str_or_none,
  10     try_get,
  11     url_or_none,
  12 )
  13
  14
  15 class TikTokBaseIE(InfoExtractor):
  16     def _extract_aweme(self, data):
  17         video = data['video']
  18         description = str_or_none(try_get(data, lambda x: x['desc']))
  19         width = int_or_none(try_get(data, lambda x: video['width']))
  20         height = int_or_none(try_get(data, lambda x: video['height']))
  21
  22         format_urls = set()
  23         formats = []
  24         for format_id in (
  25                 'play_addr_lowbr', 'play_addr', 'play_addr_h264',
  26                 'download_addr'):
  27             for format in try_get(
  28                     video, lambda x: x[format_id]['url_list'], list) or []:
  29                 format_url = url_or_none(format)
  30                 if not format_url:
  31                     continue
  32                 if format_url in format_urls:
  33                     continue
  34                 format_urls.add(format_url)
  35                 formats.append({
  36                     'url': format_url,
  37                     'ext': 'mp4',
  38                     'height': height,
  39                     'width': width,
  40                 })
  41         self._sort_formats(formats)
  42
  43         thumbnail = url_or_none(try_get(
  44             video, lambda x: x['cover']['url_list'][0], compat_str))
  45         uploader = try_get(data, lambda x: x['author']['nickname'], compat_str)
  46         timestamp = int_or_none(data.get('create_time'))
  47         comment_count = int_or_none(data.get('comment_count')) or int_or_none(
  48             try_get(data, lambda x: x['statistics']['comment_count']))
  49         repost_count = int_or_none(try_get(
  50             data, lambda x: x['statistics']['share_count']))
  51
  52         aweme_id = data['aweme_id']
  53
  54         return {
  55             'id': aweme_id,
  56             'title': uploader or aweme_id,
  57             'description': description,
  58             'thumbnail': thumbnail,
  59             'uploader': uploader,
  60             'timestamp': timestamp,
  61             'comment_count': comment_count,
  62             'repost_count': repost_count,
  63             'formats': formats,
  64         }
  65
  66
  67 class TikTokIE(TikTokBaseIE):
  68     _VALID_URL = r'''(?x)
  69                         https?://
  70                             (?:
  71                                 (?:m\.)?tiktok\.com/v|
  72                                 (?:www\.)?tiktok\.com/share/video
  73                             )
  74                             /(?P<id>\d+)
  75                     '''
  76     _TESTS = [{
  77         'url': 'https://m.tiktok.com/v/6606727368545406213.html',
  78         'md5': 'd584b572e92fcd48888051f238022420',
  79         'info_dict': {
  80             'id': '6606727368545406213',
  81             'ext': 'mp4',
  82             'title': 'Zureeal',
  83             'description': '#bowsette#mario#cosplay#uk#lgbt#gaming#asian#bowsettecosplay',
  84             'thumbnail': r're:^https?://.*~noop.image',
  85             'uploader': 'Zureeal',
  86             'timestamp': 1538248586,
  87             'upload_date': '20180929',
  88             'comment_count': int,
  89             'repost_count': int,
  90         }
  91     }, {
  92         'url': 'https://www.tiktok.com/share/video/6606727368545406213',
  93         'only_matching': True,
  94     }]
  95
  96     def _real_extract(self, url):
  97         video_id = self._match_id(url)
  98         webpage = self._download_webpage(
  99             'https://m.tiktok.com/v/%s.html' % video_id, video_id)
 100         data = self._parse_json(self._search_regex(
 101             r'\bdata\s*=\s*({.+?})\s*;', webpage, 'data'), video_id)
 102         return self._extract_aweme(data)
 103
 104
 105 class TikTokUserIE(TikTokBaseIE):
 106     _VALID_URL = r'''(?x)
 107                         https?://
 108                             (?:
 109                                 (?:m\.)?tiktok\.com/h5/share/usr|
 110                                 (?:www\.)?tiktok\.com/share/user
 111                             )
 112                             /(?P<id>\d+)
 113                     '''
 114     _TESTS = [{
 115         'url': 'https://m.tiktok.com/h5/share/usr/188294915489964032.html',
 116         'info_dict': {
 117             'id': '188294915489964032',
 118         },
 119         'playlist_mincount': 24,
 120     }, {
 121         'url': 'https://www.tiktok.com/share/user/188294915489964032',
 122         'only_matching': True,
 123     }]
 124
 125     def _real_extract(self, url):
 126         user_id = self._match_id(url)
 127         data = self._download_json(
 128             'https://m.tiktok.com/h5/share/usr/list/%s/' % user_id, user_id,
 129             query={'_signature': '_'})
 130         entries = []
 131         for aweme in data['aweme_list']:
 132             try:
 133                 entry = self._extract_aweme(aweme)
 134             except ExtractorError:
 135                 continue
 136             entry['extractor_key'] = TikTokIE.ie_key()
 137             entries.append(entry)
 138         return self.playlist_result(entries, user_id)