dict_get,
ExtractorError,
int_or_none,
+ float_or_none,
parse_duration,
parse_iso8601,
+ remove_start,
try_get,
unified_timestamp,
urlencode_postdata,
'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,
},
'skip': 'The viewing period of the video you were searching for has expired.',
}, {
- # video not available via `getflv`
+ # video not available via `getflv`; "old" HTML5 video
'url': 'http://www.nicovideo.jp/watch/sm1151009',
+ 'md5': '8fa81c364eb619d4085354eab075598a',
'info_dict': {
'id': 'sm1151009',
- 'ext': 'flv',
+ 'ext': 'mp4',
'title': 'マスターシステム本体内蔵のスペハリのメインテーマ(PSG版)',
'description': 'md5:6ee077e0581ff5019773e2e714cdd0b7',
'thumbnail': r're:https?://.*',
'comment_count': int,
},
'skip': 'Requires an account',
+ }, {
+ # "New" HTML5 video
+ # md5 is unstable
+ 'url': 'http://www.nicovideo.jp/watch/sm31464864',
+ 'info_dict': {
+ 'id': 'sm31464864',
+ 'ext': 'mp4',
+ 'title': '新作TVアニメ「戦姫絶唱シンフォギアAXZ」PV 最高画質',
+ 'description': 'md5:e52974af9a96e739196b2c1ca72b5feb',
+ 'timestamp': 1498514060,
+ 'upload_date': '20170626',
+ 'uploader': 'ゲスト',
+ 'uploader_id': '40826363',
+ 'thumbnail': r're:https?://.*',
+ 'duration': 198,
+ 'view_count': int,
+ '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,
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
self._downloader.report_warning('unable to log in: bad username or password')
return login_ok
+ def _extract_format_for_quality(self, api_data, video_id, audio_quality, video_quality):
+ def yesno(boolean):
+ return 'yes' if boolean else 'no'
+
+ session_api_data = api_data['video']['dmcInfo']['session_api']
+ session_api_endpoint = session_api_data['urls'][0]
+
+ format_id = '-'.join(map(lambda s: remove_start(s['id'], 'archive_'), [video_quality, audio_quality]))
+
+ session_response = self._download_json(
+ session_api_endpoint['url'], video_id,
+ query={'_format': 'json'},
+ headers={'Content-Type': 'application/json'},
+ note='Downloading JSON metadata for %s' % format_id,
+ data=json.dumps({
+ 'session': {
+ 'client_info': {
+ 'player_id': session_api_data['player_id'],
+ },
+ 'content_auth': {
+ 'auth_type': session_api_data['auth_types'][session_api_data['protocols'][0]],
+ 'content_key_timeout': session_api_data['content_key_timeout'],
+ 'service_id': 'nicovideo',
+ 'service_user_id': session_api_data['service_user_id']
+ },
+ 'content_id': session_api_data['content_id'],
+ 'content_src_id_sets': [{
+ 'content_src_ids': [{
+ 'src_id_to_mux': {
+ 'audio_src_ids': [audio_quality['id']],
+ 'video_src_ids': [video_quality['id']],
+ }
+ }]
+ }],
+ 'content_type': 'movie',
+ 'content_uri': '',
+ 'keep_method': {
+ 'heartbeat': {
+ 'lifetime': session_api_data['heartbeat_lifetime']
+ }
+ },
+ 'priority': session_api_data['priority'],
+ 'protocol': {
+ 'name': 'http',
+ 'parameters': {
+ 'http_parameters': {
+ 'parameters': {
+ 'http_output_download_parameters': {
+ 'use_ssl': yesno(session_api_endpoint['is_ssl']),
+ 'use_well_known_port': yesno(session_api_endpoint['is_well_known_port']),
+ }
+ }
+ }
+ }
+ },
+ 'recipe_id': session_api_data['recipe_id'],
+ 'session_operation_auth': {
+ 'session_operation_auth_by_signature': {
+ 'signature': session_api_data['signature'],
+ 'token': session_api_data['token'],
+ }
+ },
+ 'timing_constraint': 'unlimited'
+ }
+ }).encode())
+
+ resolution = video_quality.get('resolution', {})
+
+ return {
+ 'url': session_response['data']['session']['content_uri'],
+ 'format_id': format_id,
+ 'ext': 'mp4', # Session API are used in HTML5, which always serves mp4
+ 'abr': float_or_none(audio_quality.get('bitrate'), 1000),
+ 'vbr': float_or_none(video_quality.get('bitrate'), 1000),
+ 'height': resolution.get('height'),
+ 'width': resolution.get('width'),
+ }
+
def _real_extract(self, url):
video_id = self._match_id(url)
api_data = self._parse_json(self._html_search_regex(
'data-api-data="([^"]+)"', webpage,
'API data', default='{}'), video_id)
- video_real_url = try_get(
- api_data, lambda x: x['video']['smileInfo']['url'])
- if video_real_url:
- def get_video_info(items):
- return dict_get(api_data['video'], items)
- else:
+ def _format_id_from_url(video_url):
+ return 'economy' if video_real_url.endswith('low') else 'normal'
+
+ try:
+ video_real_url = api_data['video']['smileInfo']['url']
+ except KeyError: # Flash videos
# Get flv info
flv_info_webpage = self._download_webpage(
'http://flapi.nicovideo.jp/api/getflv/' + video_id + '?as3=1',
else:
raise ExtractorError('Unable to find video URL')
- video_real_url = flv_info['url'][0]
-
video_info_xml = self._download_xml(
'http://ext.nicovideo.jp/api/getthumbinfo/' + video_id,
video_id, note='Downloading video info page')
if ret:
return ret
+ video_real_url = flv_info['url'][0]
+
+ extension = get_video_info('movie_type')
+ if not extension:
+ extension = determine_ext(video_real_url)
+
+ formats = [{
+ 'url': video_real_url,
+ 'ext': extension,
+ 'format_id': _format_id_from_url(video_real_url),
+ }]
+ else:
+ formats = []
+
+ dmc_info = api_data['video'].get('dmcInfo')
+ if dmc_info: # "New" HTML5 videos
+ quality_info = dmc_info['quality']
+ for audio_quality in quality_info['audios']:
+ for video_quality in quality_info['videos']:
+ if not audio_quality['available'] or not video_quality['available']:
+ continue
+ formats.append(self._extract_format_for_quality(
+ api_data, video_id, audio_quality, video_quality))
+
+ self._sort_formats(formats)
+ else: # "Old" HTML5 videos
+ formats = [{
+ 'url': video_real_url,
+ 'ext': 'mp4',
+ 'format_id': _format_id_from_url(video_real_url),
+ }]
+
+ def get_video_info(items):
+ return dict_get(api_data['video'], items)
+
# Start extracting information
title = get_video_info('title')
if not title:
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', {})
- extension = get_video_info(['movie_type', 'movieType'])
- if not extension:
- extension = determine_ext(video_real_url)
-
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:
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: <strong[^>]*>([^<]+)</strong>',
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')
return {
'id': video_id,
- 'url': video_real_url,
'title': title,
- 'ext': extension,
- 'format_id': 'economy' if video_real_url.endswith('low') else 'normal',
+ 'formats': formats,
'thumbnail': thumbnail,
'description': description,
'uploader': uploader,