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, props_data, webpage, url):
  16         video_data = try_get(props_data, lambda x: x['pageProps'], expected_type=dict)
  17         video_info = try_get(
  18             video_data, lambda x: x['itemInfo']['itemStruct'], dict)
  19         author_info = try_get(
  20             video_data, lambda x: x['itemInfo']['itemStruct']['author'], dict)
  21         share_info = try_get(video_data, lambda x: x['itemInfo']['shareMeta'], dict)
  22
  23         unique_id = str_or_none(author_info.get('uniqueId'))
  24         timestamp = try_get(video_info, lambda x: int(x['createTime']), int)
  25         date = datetime.fromtimestamp(timestamp).strftime('%Y%m%d')
  26
  27         height = try_get(video_info, lambda x: x['video']['height'], int)
  28         width = try_get(video_info, lambda x: x['video']['width'], int)
  29         thumbnails = []
  30         thumbnails.append({
  31             'url': video_info.get('thumbnail') or self._og_search_thumbnail(webpage),
  32             'width': width,
  33             'height': height
  34         })
  35
  36         url = ''
  37         if not url:
  38             url = try_get(video_info, lambda x: x['video']['playAddr'])
  39         if not url:
  40             url = try_get(video_info, lambda x: x['video']['downloadAddr'])
  41         formats = []
  42         formats.append({
  43             'url': url,
  44             'ext': 'mp4',
  45             'height': height,
  46             'width': width
  47         })
  48
  49         tracker = try_get(props_data, lambda x: x['initialProps']['$wid'])
  50         return {
  51             'comment_count': int_or_none(video_info.get('commentCount')),
  52             'duration': try_get(video_info, lambda x: x['video']['videoMeta']['duration'], int),
  53             'height': height,
  54             'id': str_or_none(video_info.get('id')),
  55             'like_count': int_or_none(video_info.get('diggCount')),
  56             'repost_count': int_or_none(video_info.get('shareCount')),
  57             'thumbnail': try_get(video_info, lambda x: x['covers'][0]),
  58             'timestamp': timestamp,
  59             'width': width,
  60             'title': str_or_none(share_info.get('title')) or self._og_search_title(webpage),
  61             'creator': str_or_none(author_info.get('nickName')),
  62             'uploader': unique_id,
  63             'uploader_id': str_or_none(author_info.get('userId')),
  64             'uploader_url': 'https://www.tiktok.com/@' + unique_id,
  65             'thumbnails': thumbnails,
  66             'upload_date': date,
  67             'webpage_url': self._og_search_url(webpage),
  68             'description': str_or_none(video_info.get('text')) or str_or_none(share_info.get('desc')),
  69             'ext': 'mp4',
  70             'formats': formats,
  71             'http_headers': {
  72                 'Referer': url,
  73                 'Cookie': 'tt_webid=%s; tt_webid_v2=%s' % (tracker, tracker),
  74             }
  75         }
  76
  77
  78 class TikTokIE(TikTokBaseIE):
  79     _VALID_URL = r'https?://www\.tiktok\.com/@[\w\._]+/video/(?P<id>\d+)'
  80
  81     _TESTS = [{
  82         'url': 'https://www.tiktok.com/@leenabhushan/video/6748451240264420610',
  83         'md5': '34a7543afd5a151b0840ba6736fb633b',
  84         'info_dict': {
  85             'comment_count': int,
  86             'creator': 'facestoriesbyleenabh',
  87             'description': 'md5:a9f6c0c44a1ff2249cae610372d0ae95',
  88             'duration': 13,
  89             'ext': 'mp4',
  90             'formats': list,
  91             'height': 1280,
  92             'id': '6748451240264420610',
  93             'like_count': int,
  94             'repost_count': int,
  95             'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
  96             'thumbnails': list,
  97             'timestamp': 1571246252,
  98             'title': 'facestoriesbyleenabh on TikTok',
  99             'upload_date': '20191016',
 100             'uploader': 'leenabhushan',
 101             'uploader_id': '6691488002098119685',
 102             'uploader_url': r're:https://www.tiktok.com/@leenabhushan',
 103             'webpage_url': r're:https://www.tiktok.com/@leenabhushan/(video/)?6748451240264420610',
 104             'width': 720,
 105         }
 106     }, {
 107         'url': 'https://www.tiktok.com/@patroxofficial/video/6742501081818877190?langCountry=en',
 108         'md5': '06b9800d47d5fe51a19e322dd86e61c9',
 109         'info_dict': {
 110             'comment_count': int,
 111             'creator': 'patroX',
 112             'description': 'md5:5e2a23877420bb85ce6521dbee39ba94',
 113             'duration': 27,
 114             'ext': 'mp4',
 115             'formats': list,
 116             'height': 960,
 117             'id': '6742501081818877190',
 118             'like_count': int,
 119             'repost_count': int,
 120             'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
 121             'thumbnails': list,
 122             'timestamp': 1569860870,
 123             'title': 'patroX on TikTok',
 124             'upload_date': '20190930',
 125             'uploader': 'patroxofficial',
 126             'uploader_id': '18702747',
 127             'uploader_url': r're:https://www.tiktok.com/@patroxofficial',
 128             'webpage_url': r're:https://www.tiktok.com/@patroxofficial/(video/)?6742501081818877190',
 129             'width': 540,
 130         }
 131     }]
 132
 133     def _real_extract(self, url):
 134         video_id = self._match_id(url)
 135
 136         # If we only call once, we get a 403 when downlaoding the video.
 137         webpage = self._download_webpage(url, video_id, note='Downloading video webpage')
 138         webpage = self._download_webpage(url, video_id, note='Downloading video webpage')
 139         json_string = self._search_regex(
 140             r'id=\"__NEXT_DATA__\"\s+type=\"application\/json\"\s*[^>]+>\s*(?P<json_string_ld>[^<]+)',
 141             webpage, 'json_string', group='json_string_ld')
 142         json_data = self._parse_json(json_string, video_id)
 143         props_data = try_get(json_data, lambda x: x['props'], expected_type=dict)
 144
 145         # Chech statusCode for success
 146         if props_data.get('pageProps').get('statusCode') == 0:
 147             return self._extract_aweme(props_data, webpage, url)
 148
 149         raise ExtractorError('Video not available', video_id=video_id)