mirror of
https://github.com/yt-dlp/yt-dlp.git
synced 2024-11-26 02:55:17 +00:00
[niconico] Partial fix
TODO:
Smile formats don't work
Extraction code that no longer works needs to be removed
Adapted from:
b5eff52dd9
by animelover1984
Related:
https://github.com/yt-dlp/yt-dlp/issues/171
https://github.com/animelover1984/youtube-dl/issues/29
https://github.com/ytdl-org/youtube-dl/issues/28470
This commit is contained in:
pukkandan
parent
605d299f83
commit
7978e172f3
1 changed files with 97 additions and 86 deletions
|
|
@ -8,6 +8,7 @@
|
|||
from .common import InfoExtractor
|
||||
from ..postprocessor.ffmpeg import FFmpegPostProcessor
|
||||
from ..compat import (
|
||||
compat_str,
|
||||
compat_parse_qs,
|
||||
compat_urllib_parse_urlparse,
|
||||
)
|
||||
|
|
@ -20,6 +21,7 @@
|
|||
parse_duration,
|
||||
parse_iso8601,
|
||||
PostProcessingError,
|
||||
str_or_none,
|
||||
remove_start,
|
||||
try_get,
|
||||
unified_timestamp,
|
||||
|
|
@ -34,7 +36,7 @@ class NiconicoIE(InfoExtractor):
|
|||
|
||||
_TESTS = [{
|
||||
'url': 'http://www.nicovideo.jp/watch/sm22312215',
|
||||
'md5': 'd1a75c0823e2f629128c43e1212760f9',
|
||||
'md5': 'a5bad06f1347452102953f323c69da34s',
|
||||
'info_dict': {
|
||||
'id': 'sm22312215',
|
||||
'ext': 'mp4',
|
||||
|
|
@ -203,7 +205,7 @@ def _get_heartbeat_info(self, info_dict):
|
|||
'data-api-data="([^"]+)"', webpage,
|
||||
'API data', default='{}'), video_id)
|
||||
|
||||
session_api_data = try_get(api_data, lambda x: x['video']['dmcInfo']['session_api'])
|
||||
session_api_data = try_get(api_data, lambda x: x['media']['delivery']['movie']['session'])
|
||||
session_api_endpoint = try_get(session_api_data, lambda x: x['urls'][0])
|
||||
|
||||
# ping
|
||||
|
|
@ -220,7 +222,7 @@ def _get_heartbeat_info(self, info_dict):
|
|||
yesno = lambda x: 'yes' if x else 'no'
|
||||
|
||||
# m3u8 (encryption)
|
||||
if 'encryption' in try_get(api_data, lambda x: x['video']['dmcInfo']) or {}:
|
||||
if 'encryption' in (try_get(api_data, lambda x: x['media']['delivery']['movie']) or {}):
|
||||
protocol = 'm3u8'
|
||||
session_api_http_parameters = {
|
||||
'parameters': {
|
||||
|
|
@ -244,8 +246,8 @@ def _get_heartbeat_info(self, info_dict):
|
|||
session_api_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']),
|
||||
'use_ssl': yesno(session_api_endpoint['isSsl']),
|
||||
'use_well_known_port': yesno(session_api_endpoint['isWellKnownPort']),
|
||||
}
|
||||
}
|
||||
}
|
||||
|
|
@ -258,15 +260,15 @@ def _get_heartbeat_info(self, info_dict):
|
|||
data=json.dumps({
|
||||
'session': {
|
||||
'client_info': {
|
||||
'player_id': session_api_data.get('player_id'),
|
||||
'player_id': session_api_data.get('playerId'),
|
||||
},
|
||||
'content_auth': {
|
||||
'auth_type': try_get(session_api_data, lambda x: x['auth_types'][session_api_data['protocols'][0]]),
|
||||
'content_key_timeout': session_api_data.get('content_key_timeout'),
|
||||
'auth_type': try_get(session_api_data, lambda x: x['authTypes'][session_api_data['protocols'][0]]),
|
||||
'content_key_timeout': session_api_data.get('contentKeyTimeout'),
|
||||
'service_id': 'nicovideo',
|
||||
'service_user_id': session_api_data.get('service_user_id')
|
||||
'service_user_id': session_api_data.get('serviceUserId')
|
||||
},
|
||||
'content_id': session_api_data.get('content_id'),
|
||||
'content_id': session_api_data.get('contentId'),
|
||||
'content_src_id_sets': [{
|
||||
'content_src_ids': [{
|
||||
'src_id_to_mux': {
|
||||
|
|
@ -279,7 +281,7 @@ def _get_heartbeat_info(self, info_dict):
|
|||
'content_uri': '',
|
||||
'keep_method': {
|
||||
'heartbeat': {
|
||||
'lifetime': session_api_data.get('heartbeat_lifetime')
|
||||
'lifetime': session_api_data.get('heartbeatLifetime')
|
||||
}
|
||||
},
|
||||
'priority': session_api_data.get('priority'),
|
||||
|
|
@ -289,7 +291,7 @@ def _get_heartbeat_info(self, info_dict):
|
|||
'http_parameters': session_api_http_parameters
|
||||
}
|
||||
},
|
||||
'recipe_id': session_api_data.get('recipe_id'),
|
||||
'recipe_id': session_api_data.get('recipeId'),
|
||||
'session_operation_auth': {
|
||||
'session_operation_auth_by_signature': {
|
||||
'signature': session_api_data.get('signature'),
|
||||
|
|
@ -308,7 +310,7 @@ def _get_heartbeat_info(self, info_dict):
|
|||
'url': session_api_endpoint['url'] + '/' + session_response['data']['session']['id'] + '?_format=json&_method=PUT',
|
||||
'data': json.dumps(session_response['data']),
|
||||
# interval, convert milliseconds to seconds, then halve to make a buffer.
|
||||
'interval': float_or_none(session_api_data.get('heartbeat_lifetime'), scale=2000),
|
||||
'interval': float_or_none(session_api_data.get('heartbeatLifetime'), scale=2000),
|
||||
}
|
||||
|
||||
return info_dict, heartbeat_info_dict
|
||||
|
|
@ -327,15 +329,17 @@ def parse_format_id(id_code):
|
|||
format_id = '-'.join(map(lambda s: remove_start(s['id'], 'archive_'), [video_quality, audio_quality]))
|
||||
vdict = parse_format_id(video_quality['id'])
|
||||
adict = parse_format_id(audio_quality['id'])
|
||||
resolution = video_quality.get('resolution', {'height': vdict.get('res')})
|
||||
resolution = try_get(video_quality, lambda x: x['metadata']['resolution'], dict) or {'height': vdict.get('res')}
|
||||
vbr = try_get(video_quality, lambda x: x['metadata']['bitrate'], float)
|
||||
|
||||
return {
|
||||
'url': '%s:%s/%s/%s' % (protocol, video_id, video_quality['id'], audio_quality['id']),
|
||||
'format_id': format_id,
|
||||
'format_note': 'DMC %s' % try_get(video_quality, lambda x: x['metadata']['label'], compat_str),
|
||||
'ext': 'mp4', # Session API are used in HTML5, which always serves mp4
|
||||
'vcodec': vdict.get('codec'),
|
||||
'acodec': adict.get('codec'),
|
||||
'vbr': float_or_none(video_quality.get('bitrate'), 1000) or float_or_none(vdict.get('br')),
|
||||
'vbr': float_or_none(vbr, 1000) or float_or_none(vdict.get('br')),
|
||||
'abr': float_or_none(audio_quality.get('bitrate'), 1000) or float_or_none(adict.get('br')),
|
||||
'height': int_or_none(resolution.get('height', vdict.get('res'))),
|
||||
'width': int_or_none(resolution.get('width')),
|
||||
|
|
@ -394,22 +398,23 @@ def get_video_info_xml(items):
|
|||
formats = []
|
||||
|
||||
# Get HTML5 videos info
|
||||
try:
|
||||
dmc_info = api_data['video']['dmcInfo']
|
||||
except KeyError:
|
||||
raise ExtractorError('The video can\'t downloaded.',
|
||||
expected=True)
|
||||
quality_info = try_get(api_data, lambda x: x['media']['delivery']['movie'])
|
||||
if not quality_info:
|
||||
raise ExtractorError('The video can\'t downloaded.', expected=True)
|
||||
|
||||
quality_info = dmc_info.get('quality')
|
||||
for audio_quality in quality_info.get('audios') or {}:
|
||||
for video_quality in quality_info.get('videos') or {}:
|
||||
if not audio_quality.get('available') or not video_quality.get('available'):
|
||||
if not audio_quality.get('isAvailable') or not video_quality.get('isAvailable'):
|
||||
continue
|
||||
formats.append(self._extract_format_for_quality(
|
||||
api_data, video_id, audio_quality, video_quality))
|
||||
|
||||
# Get flv/swf info
|
||||
timestamp = None
|
||||
video_real_url = try_get(api_data, lambda x: x['video']['smileInfo']['url'])
|
||||
if not video_real_url:
|
||||
self.report_warning('Unable to obtain smile video information')
|
||||
else:
|
||||
is_economy = video_real_url.endswith('low')
|
||||
|
||||
if is_economy:
|
||||
|
|
@ -517,6 +522,7 @@ def get_video_info_xml(items):
|
|||
timestamp = parse_iso8601(
|
||||
video_detail['postedAt'].replace('/', '-'),
|
||||
delimiter=' ', timezone=datetime.timedelta(hours=9))
|
||||
timestamp = timestamp or try_get(api_data, lambda x: parse_iso8601(x['video']['registeredAt']))
|
||||
|
||||
view_count = int_or_none(get_video_info_web(['view_counter', 'viewCount']))
|
||||
if not view_count:
|
||||
|
|
@ -525,11 +531,16 @@ def get_video_info_xml(items):
|
|||
webpage, 'view count', default=None)
|
||||
if match:
|
||||
view_count = int_or_none(match.replace(',', ''))
|
||||
view_count = view_count or video_detail.get('viewCount')
|
||||
view_count = (
|
||||
view_count
|
||||
or video_detail.get('viewCount')
|
||||
or try_get(api_data, lambda x: x['video']['count']['view']))
|
||||
|
||||
comment_count = (int_or_none(get_video_info_web('comment_num'))
|
||||
comment_count = (
|
||||
int_or_none(get_video_info_web('comment_num'))
|
||||
or video_detail.get('commentCount')
|
||||
or try_get(api_data, lambda x: x['thread']['commentCount']))
|
||||
or try_get(api_data, lambda x: x['video']['count']['comment']))
|
||||
|
||||
if not comment_count:
|
||||
match = self._html_search_regex(
|
||||
r'>Comments: <strong[^>]*>([^<]+)</strong>',
|
||||
|
|
@ -559,7 +570,7 @@ def get_video_info_xml(items):
|
|||
# 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 = (
|
||||
uploader_id = str_or_none(
|
||||
get_video_info_web(['ch_id', 'user_id'])
|
||||
or owner.get('id')
|
||||
or channel_id
|
||||
|
|
|
|||
Loading…
Reference in a new issue