youtube_dlc/extractor/tiktok.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3 from datetime import datetime
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     ExtractorError,
   8     int_or_none,
   9     str_or_none,
  10     try_get
  11 )
  12
  13
  14 class TikTokBaseIE(InfoExtractor):
  15     def _extract_aweme(self, video_data, webpage):
  16         video_info = try_get(
  17             video_data, lambda x: x['videoData']['itemInfos'], dict)
  18         author_info = try_get(
  19             video_data, lambda x: x['videoData']['authorInfos'], dict)
  20         share_info = try_get(video_data, lambda x: x['shareMeta'], dict)
  21
  22         unique_id = str_or_none(author_info.get('uniqueId'))
  23         timestamp = try_get(video_info, lambda x: int(x['createTime']), int)
  24         date = datetime.fromtimestamp(timestamp).strftime('%Y%m%d')
  25
  26         height = try_get(video_info, lambda x: x['video']['videoMeta']['height'], int)
  27         width = try_get(video_info, lambda x: x['video']['videoMeta']['width'], int)
  28         thumbnails = []
  29         thumbnails.append({
  30             'url': video_info.get('thumbnail') or self._og_search_thumbnail(webpage),
  31             'width': width,
  32             'height': height
  33         })
  34
  35         formats = []
  36         formats.append({
  37             'url': try_get(video_info, lambda x: x['video']['urls'][0]),
  38             'ext': 'mp4',
  39             'height': height,
  40             'width': width
  41         })
  42
  43         return {
  44             'comment_count': int_or_none(video_info.get('commentCount')),
  45             'duration': try_get(video_info, lambda x: x['video']['videoMeta']['duration'], int),
  46             'height': height,
  47             'id': str_or_none(video_info.get('id')),
  48             'like_count': int_or_none(video_info.get('diggCount')),
  49             'repost_count': int_or_none(video_info.get('shareCount')),
  50             'thumbnail': try_get(video_info, lambda x: x['covers'][0]),
  51             'timestamp': timestamp,
  52             'width': width,
  53             'title': str_or_none(share_info.get('title')) or self._og_search_title(webpage),
  54             'creator': str_or_none(author_info.get('nickName')),
  55             'uploader': unique_id,
  56             'uploader_id': str_or_none(author_info.get('userId')),
  57             'uploader_url': 'https://www.tiktok.com/@' + unique_id,
  58             'thumbnails': thumbnails,
  59             'upload_date': date,
  60             'webpage_url': self._og_search_url(webpage),
  61             'description': str_or_none(video_info.get('text')) or str_or_none(share_info.get('desc')),
  62             'ext': 'mp4',
  63             'formats': formats
  64         }
  65
  66
  67 class TikTokIE(TikTokBaseIE):
  68     _VALID_URL = r'https?://www\.tiktok\.com/@[\w\._]+/video/(?P<id>\d+)'
  69
  70     _TESTS = [{
  71         'url': 'https://www.tiktok.com/@leenabhushan/video/6748451240264420610',
  72         'md5': '34a7543afd5a151b0840ba6736fb633b',
  73         'info_dict': {
  74             'comment_count': int,
  75             'creator': 'facestoriesbyleenabh',
  76             'description': 'md5:a9f6c0c44a1ff2249cae610372d0ae95',
  77             'duration': 13,
  78             'ext': 'mp4',
  79             'formats': list,
  80             'height': 1280,
  81             'id': '6748451240264420610',
  82             'like_count': int,
  83             'repost_count': int,
  84             'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
  85             'thumbnails': list,
  86             'timestamp': 1571246252,
  87             'title': 'facestoriesbyleenabh on TikTok',
  88             'upload_date': '20191016',
  89             'uploader': 'leenabhushan',
  90             'uploader_id': '6691488002098119685',
  91             'uploader_url': r're:https://www.tiktok.com/@leenabhushan',
  92             'webpage_url': r're:https://www.tiktok.com/@leenabhushan/(video/)?6748451240264420610',
  93             'width': 720,
  94         }
  95     }, {
  96         'url': 'https://www.tiktok.com/@patroxofficial/video/6742501081818877190?langCountry=en',
  97         'md5': '06b9800d47d5fe51a19e322dd86e61c9',
  98         'info_dict': {
  99             'comment_count': int,
 100             'creator': 'patroX',
 101             'description': 'md5:5e2a23877420bb85ce6521dbee39ba94',
 102             'duration': 27,
 103             'ext': 'mp4',
 104             'formats': list,
 105             'height': 960,
 106             'id': '6742501081818877190',
 107             'like_count': int,
 108             'repost_count': int,
 109             'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
 110             'thumbnails': list,
 111             'timestamp': 1569860870,
 112             'title': 'patroX on TikTok',
 113             'upload_date': '20190930',
 114             'uploader': 'patroxofficial',
 115             'uploader_id': '18702747',
 116             'uploader_url': r're:https://www.tiktok.com/@patroxofficial',
 117             'webpage_url': r're:https://www.tiktok.com/@patroxofficial/(video/)?6742501081818877190',
 118             'width': 540,
 119         }
 120     }]
 121
 122     def _real_extract(self, url):
 123         video_id = self._match_id(url)
 124
 125         webpage = self._download_webpage(url, video_id, note='Downloading video webpage')
 126         json_string = self._search_regex(
 127             r'id=\"__NEXT_DATA__\"\s+type=\"application\/json\"\s*[^>]+>\s*(?P<json_string_ld>[^<]+)',
 128             webpage, 'json_string', group='json_string_ld')
 129         json_data = self._parse_json(json_string, video_id)
 130         video_data = try_get(json_data, lambda x: x['props']['pageProps'], expected_type=dict)
 131
 132         # Chech statusCode for success
 133         if video_data.get('statusCode') == 0:
 134             return self._extract_aweme(video_data, webpage)
 135
 136         raise ExtractorError('Video not available', video_id=video_id)