X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/00368b4c3a5d4e909e1b7ecfc4030bf28da020f3..8593410c28e395b68b410169356663541005c3ae:/youtube_dl/extractor/niconico.py diff --git a/youtube_dl/extractor/niconico.py b/youtube_dl/extractor/niconico.py index 026329d..eb07ca7 100644 --- a/youtube_dl/extractor/niconico.py +++ b/youtube_dl/extractor/niconico.py @@ -40,7 +40,7 @@ class NiconicoIE(InfoExtractor): 'uploader': 'takuya0301', 'uploader_id': '2698420', 'upload_date': '20131123', - 'timestamp': 1385182762, + 'timestamp': int, # timestamp is unstable 'description': '(c) copyright 2008, Blender Foundation / www.bigbuckbunny.org', 'duration': 33, 'view_count': int, @@ -115,8 +115,8 @@ class NiconicoIE(InfoExtractor): 'skip': 'Requires an account', }, { # "New" HTML5 video + # md5 is unstable 'url': 'http://www.nicovideo.jp/watch/sm31464864', - 'md5': '351647b4917660986dc0fa8864085135', 'info_dict': { 'id': 'sm31464864', 'ext': 'mp4', @@ -124,7 +124,7 @@ class NiconicoIE(InfoExtractor): 'description': 'md5:e52974af9a96e739196b2c1ca72b5feb', 'timestamp': 1498514060, 'upload_date': '20170626', - 'uploader': 'ゲス', + 'uploader': 'ゲスト', 'uploader_id': '40826363', 'thumbnail': r're:https?://.*', 'duration': 198, @@ -132,6 +132,25 @@ class NiconicoIE(InfoExtractor): 'comment_count': int, }, 'skip': 'Requires an account', + }, { + # Video without owner + 'url': 'http://www.nicovideo.jp/watch/sm18238488', + 'md5': 'd265680a1f92bdcbbd2a507fc9e78a9e', + 'info_dict': { + 'id': 'sm18238488', + 'ext': 'mp4', + 'title': '【実写版】ミュータントタートルズ', + 'description': 'md5:15df8988e47a86f9e978af2064bf6d8e', + 'timestamp': 1341160408, + 'upload_date': '20120701', + 'uploader': None, + 'uploader_id': None, + 'thumbnail': r're:https?://.*', + 'duration': 5271, + 'view_count': int, + 'comment_count': int, + }, + 'skip': 'Requires an account', }, { 'url': 'http://sp.nicovideo.jp/watch/sm28964488?ss_pos=1&cp_in=wt_tg', 'only_matching': True, @@ -144,7 +163,7 @@ class NiconicoIE(InfoExtractor): self._login() def _login(self): - (username, password) = self._get_login_info() + username, password = self._get_login_info() # No authentication to be performed if not username: return True @@ -233,7 +252,7 @@ class NiconicoIE(InfoExtractor): }, 'timing_constraint': 'unlimited' } - })) + }).encode()) resolution = video_quality.get('resolution', {}) @@ -350,14 +369,14 @@ class NiconicoIE(InfoExtractor): 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')) + 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'))) + 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: @@ -376,9 +395,9 @@ class NiconicoIE(InfoExtractor): 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'])) + 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: ]*>([^<]+)', @@ -387,15 +406,17 @@ class NiconicoIE(InfoExtractor): 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')) + 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 - owner = api_data.get('owner', {}) + # 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')