Merge branch 'merval-merval/fix_tiktok'

This commit is contained in:
Tom-Oliver Heidel 2020-10-26 02:16:24 +01:00
commit ff3c7af428

View file

@ -12,19 +12,20 @@
class TikTokBaseIE(InfoExtractor): class TikTokBaseIE(InfoExtractor):
def _extract_aweme(self, video_data, webpage, url): def _extract_aweme(self, props_data, webpage, url):
video_data = try_get(props_data, lambda x: x['pageProps'], expected_type=dict)
video_info = try_get( video_info = try_get(
video_data, lambda x: x['videoData']['itemInfos'], dict) video_data, lambda x: x['itemInfo']['itemStruct'], dict)
author_info = try_get( author_info = try_get(
video_data, lambda x: x['videoData']['authorInfos'], dict) video_data, lambda x: x['itemInfo']['itemStruct']['author'], dict)
share_info = try_get(video_data, lambda x: x['shareMeta'], dict) share_info = try_get(video_data, lambda x: x['itemInfo']['shareMeta'], dict)
unique_id = str_or_none(author_info.get('uniqueId')) unique_id = str_or_none(author_info.get('uniqueId'))
timestamp = try_get(video_info, lambda x: int(x['createTime']), int) timestamp = try_get(video_info, lambda x: int(x['createTime']), int)
date = datetime.fromtimestamp(timestamp).strftime('%Y%m%d') date = datetime.fromtimestamp(timestamp).strftime('%Y%m%d')
height = try_get(video_info, lambda x: x['video']['videoMeta']['height'], int) height = try_get(video_info, lambda x: x['video']['height'], int)
width = try_get(video_info, lambda x: x['video']['videoMeta']['width'], int) width = try_get(video_info, lambda x: x['video']['width'], int)
thumbnails = [] thumbnails = []
thumbnails.append({ thumbnails.append({
'url': video_info.get('thumbnail') or self._og_search_thumbnail(webpage), 'url': video_info.get('thumbnail') or self._og_search_thumbnail(webpage),
@ -32,14 +33,20 @@ def _extract_aweme(self, video_data, webpage, url):
'height': height 'height': height
}) })
url = ''
if not url:
url = try_get(video_info, lambda x: x['video']['playAddr'])
if not url:
url = try_get(video_info, lambda x: x['video']['downloadAddr'])
formats = [] formats = []
formats.append({ formats.append({
'url': try_get(video_info, lambda x: x['video']['urls'][0]), 'url': url,
'ext': 'mp4', 'ext': 'mp4',
'height': height, 'height': height,
'width': width 'width': width
}) })
tracker = try_get(props_data, lambda x: x['initialProps']['$wid'])
return { return {
'comment_count': int_or_none(video_info.get('commentCount')), 'comment_count': int_or_none(video_info.get('commentCount')),
'duration': try_get(video_info, lambda x: x['video']['videoMeta']['duration'], int), 'duration': try_get(video_info, lambda x: x['video']['videoMeta']['duration'], int),
@ -63,6 +70,7 @@ def _extract_aweme(self, video_data, webpage, url):
'formats': formats, 'formats': formats,
'http_headers': { 'http_headers': {
'Referer': url, 'Referer': url,
'Cookie': 'tt_webid=%s; tt_webid_v2=%s' % (tracker, tracker),
} }
} }
@ -130,10 +138,10 @@ def _real_extract(self, url):
r'id=\"__NEXT_DATA__\"\s+type=\"application\/json\"\s*[^>]+>\s*(?P<json_string_ld>[^<]+)', r'id=\"__NEXT_DATA__\"\s+type=\"application\/json\"\s*[^>]+>\s*(?P<json_string_ld>[^<]+)',
webpage, 'json_string', group='json_string_ld') webpage, 'json_string', group='json_string_ld')
json_data = self._parse_json(json_string, video_id) json_data = self._parse_json(json_string, video_id)
video_data = try_get(json_data, lambda x: x['props']['pageProps'], expected_type=dict) props_data = try_get(json_data, lambda x: x['props'], expected_type=dict)
# Chech statusCode for success # Chech statusCode for success
if video_data.get('statusCode') == 0: if props_data.get('pageProps').get('statusCode') == 0:
return self._extract_aweme(video_data, webpage, url) return self._extract_aweme(props_data, webpage, url)
raise ExtractorError('Video not available', video_id=video_id) raise ExtractorError('Video not available', video_id=video_id)