- title = video_info.find('.//title').text
- extension = video_info.find('.//movie_type').text
- video_format = extension.upper()
- thumbnail = video_info.find('.//thumbnail_url').text
- description = video_info.find('.//description').text
- upload_date = unified_strdate(video_info.find('.//first_retrieve').text.split('+')[0])
- view_count = int_or_none(video_info.find('.//view_counter').text)
- comment_count = int_or_none(video_info.find('.//comment_num').text)
- duration = parse_duration(video_info.find('.//length').text)
- webpage_url = video_info.find('.//watch_url').text
-
- if video_info.find('.//ch_id') is not None:
- uploader_id = video_info.find('.//ch_id').text
- uploader = video_info.find('.//ch_name').text
- elif video_info.find('.//user_id') is not None:
- uploader_id = video_info.find('.//user_id').text
- uploader = video_info.find('.//user_nickname').text
- else:
- uploader_id = uploader = None
+ title = get_video_info('title')
+ if not title:
+ title = self._og_search_title(webpage, default=None)
+ if not title:
+ title = self._html_search_regex(
+ r'<span[^>]+class="videoHeaderTitle"[^>]*>([^<]+)</span>',
+ webpage, 'video title')
+
+ watch_api_data_string = self._html_search_regex(
+ r'<div[^>]+id="watchAPIDataContainer"[^>]+>([^<]+)</div>',
+ webpage, 'watch api data', default=None)
+ watch_api_data = self._parse_json(watch_api_data_string, video_id) if watch_api_data_string else {}
+ video_detail = watch_api_data.get('videoDetail', {})
+
+ thumbnail = (
+ get_video_info(['thumbnail_url', 'thumbnailURL']) or
+ self._html_search_meta('image', webpage, 'thumbnail', default=None) or
+ video_detail.get('thumbnail'))
+
+ description = get_video_info('description')
+
+ timestamp = (parse_iso8601(get_video_info('first_retrieve')) or
+ unified_timestamp(get_video_info('postedDateTime')))
+ if not timestamp:
+ match = self._html_search_meta('datePublished', webpage, 'date published', default=None)
+ if match:
+ timestamp = parse_iso8601(match.replace('+', ':00+'))
+ if not timestamp and video_detail.get('postedAt'):
+ timestamp = parse_iso8601(
+ video_detail['postedAt'].replace('/', '-'),
+ delimiter=' ', timezone=datetime.timedelta(hours=9))
+
+ view_count = int_or_none(get_video_info(['view_counter', 'viewCount']))
+ if not view_count:
+ match = self._html_search_regex(
+ r'>Views: <strong[^>]*>([^<]+)</strong>',
+ webpage, 'view count', default=None)
+ if match:
+ view_count = int_or_none(match.replace(',', ''))
+ view_count = view_count or video_detail.get('viewCount')
+
+ comment_count = (int_or_none(get_video_info('comment_num')) or
+ video_detail.get('commentCount') or
+ try_get(api_data, lambda x: x['thread']['commentCount']))
+ if not comment_count:
+ match = self._html_search_regex(
+ r'>Comments: <strong[^>]*>([^<]+)</strong>',
+ webpage, 'comment count', default=None)
+ if match:
+ comment_count = int_or_none(match.replace(',', ''))
+
+ duration = (parse_duration(
+ get_video_info('length') or
+ self._html_search_meta(
+ 'video:duration', webpage, 'video duration', default=None)) or
+ video_detail.get('length') or
+ get_video_info('duration'))
+
+ webpage_url = get_video_info('watch_url') or url
+
+ # Note: cannot use api_data.get('owner', {}) because owner may be set to "null"
+ # in the JSON, which will cause None to be returned instead of {}.
+ owner = try_get(api_data, lambda x: x.get('owner'), dict) or {}
+ uploader_id = get_video_info(['ch_id', 'user_id']) or owner.get('id')
+ uploader = get_video_info(['ch_name', 'user_nickname']) or owner.get('nickname')
Rapsys Git / youtubedl / blobdiff