[extractor/crunchyroll] Rework with support for movies, music and artists (#6237)

This adds `CrunchyrollMusicIE` and `CrunchyrollArtistIE` extractors using the new, reworked base class and expands the `CrunchyrollBetaIE` with support for movies and movie listings and more complete metadata extraction

Authored by: Grub4K
pull/7125/head
Simon Sawicki 1 year ago committed by GitHub
parent 8417f26b8a
commit 032de83ea9
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23

@ -406,6 +406,8 @@ from .crowdbunker import (
from .crunchyroll import ( from .crunchyroll import (
CrunchyrollBetaIE, CrunchyrollBetaIE,
CrunchyrollBetaShowIE, CrunchyrollBetaShowIE,
CrunchyrollMusicIE,
CrunchyrollArtistIE,
) )
from .cspan import CSpanIE, CSpanCongressIE from .cspan import CSpanIE, CSpanCongressIE
from .ctsnews import CtsNewsIE from .ctsnews import CtsNewsIE

@ -1,28 +1,37 @@
import base64 import base64
import urllib.parse import urllib.error
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import ( from ..utils import (
ExtractorError, ExtractorError,
float_or_none, float_or_none,
format_field, format_field,
int_or_none,
join_nonempty, join_nonempty,
parse_age_limit,
parse_count,
parse_iso8601, parse_iso8601,
qualities, qualities,
remove_start,
time_seconds,
traverse_obj, traverse_obj,
try_get, url_or_none,
urlencode_postdata,
) )
class CrunchyrollBaseIE(InfoExtractor): class CrunchyrollBaseIE(InfoExtractor):
_LOGIN_URL = 'https://www.crunchyroll.com/welcome/login' _BASE_URL = 'https://www.crunchyroll.com'
_API_BASE = 'https://api.crunchyroll.com' _API_BASE = 'https://api.crunchyroll.com'
_NETRC_MACHINE = 'crunchyroll' _NETRC_MACHINE = 'crunchyroll'
params = None _AUTH_HEADERS = None
_API_ENDPOINT = None
_BASIC_AUTH = None
_QUERY = {}
@property @property
def is_logged_in(self): def is_logged_in(self):
return self._get_cookies(self._LOGIN_URL).get('etp_rt') return self._get_cookies(self._BASE_URL).get('etp_rt')
def _perform_login(self, username, password): def _perform_login(self, username, password):
if self.is_logged_in: if self.is_logged_in:
@ -35,7 +44,7 @@ class CrunchyrollBaseIE(InfoExtractor):
'device_id': 'whatvalueshouldbeforweb', 'device_id': 'whatvalueshouldbeforweb',
'device_type': 'com.crunchyroll.static', 'device_type': 'com.crunchyroll.static',
'access_token': 'giKq5eY27ny3cqz', 'access_token': 'giKq5eY27ny3cqz',
'referer': self._LOGIN_URL 'referer': f'{self._BASE_URL}/welcome/login'
}) })
if upsell_response['code'] != 'ok': if upsell_response['code'] != 'ok':
raise ExtractorError('Could not get session id') raise ExtractorError('Could not get session id')
@ -43,66 +52,164 @@ class CrunchyrollBaseIE(InfoExtractor):
login_response = self._download_json( login_response = self._download_json(
f'{self._API_BASE}/login.1.json', None, 'Logging in', f'{self._API_BASE}/login.1.json', None, 'Logging in',
data=urllib.parse.urlencode({ data=urlencode_postdata({
'account': username, 'account': username,
'password': password, 'password': password,
'session_id': session_id 'session_id': session_id
}).encode('ascii')) }))
if login_response['code'] != 'ok': if login_response['code'] != 'ok':
raise ExtractorError('Login failed. Server message: %s' % login_response['message'], expected=True) raise ExtractorError('Login failed. Server message: %s' % login_response['message'], expected=True)
if not self.is_logged_in: if not self.is_logged_in:
raise ExtractorError('Login succeeded but did not set etp_rt cookie') raise ExtractorError('Login succeeded but did not set etp_rt cookie')
def _get_embedded_json(self, webpage, display_id): def _update_query(self, lang):
initial_state = self._parse_json(self._search_regex( if lang in CrunchyrollBaseIE._QUERY:
r'__INITIAL_STATE__\s*=\s*({.+?})\s*;', webpage, 'initial state'), display_id) return
app_config = self._parse_json(self._search_regex(
r'__APP_CONFIG__\s*=\s*({.+?})\s*;', webpage, 'app config'), display_id) webpage = self._download_webpage(
return initial_state, app_config f'{self._BASE_URL}/{lang}', None, note=f'Retrieving main page (lang={lang or None})')
def _get_params(self, lang): initial_state = self._search_json(r'__INITIAL_STATE__\s*=', webpage, 'initial state', None)
if not CrunchyrollBaseIE.params: CrunchyrollBaseIE._QUERY[lang] = traverse_obj(initial_state, {
if self._get_cookies(f'https://www.crunchyroll.com/{lang}').get('etp_rt'): 'locale': ('localization', 'locale'),
grant_type, key = 'etp_rt_cookie', 'accountAuthClientId' }) or None
else:
grant_type, key = 'client_id', 'anonClientId' if CrunchyrollBaseIE._BASIC_AUTH:
return
initial_state, app_config = self._get_embedded_json(self._download_webpage( app_config = self._search_json(r'__APP_CONFIG__\s*=', webpage, 'app config', None)
f'https://www.crunchyroll.com/{lang}', None, note='Retrieving main page'), None) cx_api_param = app_config['cxApiParams']['accountAuthClientId' if self.is_logged_in else 'anonClientId']
api_domain = app_config['cxApiParams']['apiDomain'].replace('beta.crunchyroll.com', 'www.crunchyroll.com') self.write_debug(f'Using cxApiParam={cx_api_param}')
CrunchyrollBaseIE._BASIC_AUTH = 'Basic ' + base64.b64encode(f'{cx_api_param}:'.encode()).decode()
def _update_auth(self):
if CrunchyrollBaseIE._AUTH_HEADERS and CrunchyrollBaseIE._AUTH_REFRESH > time_seconds():
return
assert CrunchyrollBaseIE._BASIC_AUTH, '_update_query needs to be called at least one time beforehand'
grant_type = 'etp_rt_cookie' if self.is_logged_in else 'client_id'
auth_response = self._download_json( auth_response = self._download_json(
f'{api_domain}/auth/v1/token', None, note=f'Authenticating with grant_type={grant_type}', f'{self._BASE_URL}/auth/v1/token', None, note=f'Authenticating with grant_type={grant_type}',
headers={ headers={'Authorization': CrunchyrollBaseIE._BASIC_AUTH}, data=f'grant_type={grant_type}'.encode())
'Authorization': 'Basic ' + str(base64.b64encode(('%s:' % app_config['cxApiParams'][key]).encode('ascii')), 'ascii')
}, data=f'grant_type={grant_type}'.encode('ascii')) CrunchyrollBaseIE._AUTH_HEADERS = {'Authorization': auth_response['token_type'] + ' ' + auth_response['access_token']}
policy_response = self._download_json( CrunchyrollBaseIE._AUTH_REFRESH = time_seconds(seconds=traverse_obj(auth_response, ('expires_in', {float_or_none}), default=300) - 10)
f'{api_domain}/index/v2', None, note='Retrieving signed policy',
headers={ def _call_base_api(self, endpoint, internal_id, lang, note=None, query={}):
'Authorization': auth_response['token_type'] + ' ' + auth_response['access_token'] self._update_query(lang)
}) self._update_auth()
cms = policy_response.get('cms_web')
bucket = cms['bucket'] if not endpoint.startswith('/'):
params = { endpoint = f'/{endpoint}'
'Policy': cms['policy'],
'Signature': cms['signature'], return self._download_json(
'Key-Pair-Id': cms['key_pair_id'] f'{self._BASE_URL}{endpoint}', internal_id, note or f'Calling API: {endpoint}',
headers=CrunchyrollBaseIE._AUTH_HEADERS, query={**CrunchyrollBaseIE._QUERY[lang], **query})
def _call_api(self, path, internal_id, lang, note='api', query={}):
if not path.startswith(f'/content/v2/{self._API_ENDPOINT}/'):
path = f'/content/v2/{self._API_ENDPOINT}/{path}'
try:
result = self._call_base_api(
path, internal_id, lang, f'Downloading {note} JSON ({self._API_ENDPOINT})', query=query)
except ExtractorError as error:
if isinstance(error.cause, urllib.error.HTTPError) and error.cause.code == 404:
return None
raise
if not result:
raise ExtractorError(f'Unexpected response when downloading {note} JSON')
return result
def _extract_formats(self, stream_response, display_id=None):
requested_formats = self._configuration_arg('format') or ['adaptive_hls']
available_formats = {}
for stream_type, streams in traverse_obj(
stream_response, (('streams', ('data', 0)), {dict.items}, ...)):
if stream_type not in requested_formats:
continue
for stream in traverse_obj(streams, lambda _, v: v['url']):
hardsub_lang = stream.get('hardsub_locale') or ''
format_id = join_nonempty(stream_type, format_field(stream, 'hardsub_locale', 'hardsub-%s'))
available_formats[hardsub_lang] = (stream_type, format_id, hardsub_lang, stream['url'])
requested_hardsubs = [('' if val == 'none' else val) for val in (self._configuration_arg('hardsub') or ['none'])]
if '' in available_formats and 'all' not in requested_hardsubs:
full_format_langs = set(requested_hardsubs)
self.to_screen(
'To get all formats of a hardsub language, use '
'"--extractor-args crunchyrollbeta:hardsub=<language_code or all>". '
'See https://github.com/yt-dlp/yt-dlp#crunchyrollbeta-crunchyroll for more info',
only_once=True)
else:
full_format_langs = set(map(str.lower, available_formats))
audio_locale = traverse_obj(stream_response, ((None, 'meta'), 'audio_locale'), get_all=False)
hardsub_preference = qualities(requested_hardsubs[::-1])
formats = []
for stream_type, format_id, hardsub_lang, stream_url in available_formats.values():
if stream_type.endswith('hls'):
if hardsub_lang.lower() in full_format_langs:
adaptive_formats = self._extract_m3u8_formats(
stream_url, display_id, 'mp4', m3u8_id=format_id,
fatal=False, note=f'Downloading {format_id} HLS manifest')
else:
adaptive_formats = (self._m3u8_meta_format(stream_url, ext='mp4', m3u8_id=format_id),)
elif stream_type.endswith('dash'):
adaptive_formats = self._extract_mpd_formats(
stream_url, display_id, mpd_id=format_id,
fatal=False, note=f'Downloading {format_id} MPD manifest')
else:
self.report_warning(f'Encountered unknown stream_type: {stream_type!r}', display_id, only_once=True)
continue
for f in adaptive_formats:
if f.get('acodec') != 'none':
f['language'] = audio_locale
f['quality'] = hardsub_preference(hardsub_lang.lower())
formats.extend(adaptive_formats)
return formats
def _extract_subtitles(self, data):
subtitles = {}
for locale, subtitle in traverse_obj(data, ((None, 'meta'), 'subtitles', {dict.items}, ...)):
subtitles[locale] = [traverse_obj(subtitle, {'url': 'url', 'ext': 'format'})]
return subtitles
class CrunchyrollCmsBaseIE(CrunchyrollBaseIE):
_API_ENDPOINT = 'cms'
_CMS_EXPIRY = None
def _call_cms_api_signed(self, path, internal_id, lang, note='api'):
if not CrunchyrollCmsBaseIE._CMS_EXPIRY or CrunchyrollCmsBaseIE._CMS_EXPIRY <= time_seconds():
response = self._call_base_api('index/v2', None, lang, 'Retrieving signed policy')['cms_web']
CrunchyrollCmsBaseIE._CMS_QUERY = {
'Policy': response['policy'],
'Signature': response['signature'],
'Key-Pair-Id': response['key_pair_id'],
} }
locale = traverse_obj(initial_state, ('localization', 'locale')) CrunchyrollCmsBaseIE._CMS_BUCKET = response['bucket']
if locale: CrunchyrollCmsBaseIE._CMS_EXPIRY = parse_iso8601(response['expires']) - 10
params['locale'] = locale
CrunchyrollBaseIE.params = (api_domain, bucket, params) if not path.startswith('/cms/v2'):
return CrunchyrollBaseIE.params path = f'/cms/v2{CrunchyrollCmsBaseIE._CMS_BUCKET}/{path}'
return self._call_base_api(
path, internal_id, lang, f'Downloading {note} JSON (signed cms)', query=CrunchyrollCmsBaseIE._CMS_QUERY)
class CrunchyrollBetaIE(CrunchyrollBaseIE): class CrunchyrollBetaIE(CrunchyrollCmsBaseIE):
IE_NAME = 'crunchyroll' IE_NAME = 'crunchyroll'
_VALID_URL = r'''(?x) _VALID_URL = r'''(?x)
https?://(?:beta|www)\.crunchyroll\.com/ https?://(?:beta\.|www\.)?crunchyroll\.com/
(?P<lang>(?:\w{2}(?:-\w{2})?/)?) (?P<lang>(?:\w{2}(?:-\w{2})?/)?)
watch/(?P<id>\w+) watch/(?!concert|musicvideo)(?P<id>\w+)'''
(?:/(?P<display_id>[\w-]+))?/?(?:[?#]|$)'''
_TESTS = [{ _TESTS = [{
# Premium only
'url': 'https://www.crunchyroll.com/watch/GY2P1Q98Y/to-the-future', 'url': 'https://www.crunchyroll.com/watch/GY2P1Q98Y/to-the-future',
'info_dict': { 'info_dict': {
'id': 'GY2P1Q98Y', 'id': 'GY2P1Q98Y',
@ -119,11 +226,15 @@ class CrunchyrollBetaIE(CrunchyrollBaseIE):
'season_number': 1, 'season_number': 1,
'episode': 'To the Future', 'episode': 'To the Future',
'episode_number': 73, 'episode_number': 73,
'thumbnail': r're:^https://www.crunchyroll.com/imgsrv/.*\.jpeg$', 'thumbnail': r're:^https://www.crunchyroll.com/imgsrv/.*\.jpeg?$',
'chapters': 'count:2', 'chapters': 'count:2',
'age_limit': 14,
'like_count': int,
'dislike_count': int,
}, },
'params': {'skip_download': 'm3u8', 'format': 'all[format_id~=hardsub]'}, 'params': {'skip_download': 'm3u8', 'format': 'all[format_id~=hardsub]'},
}, { }, {
# Premium only
'url': 'https://www.crunchyroll.com/watch/GYE5WKQGR', 'url': 'https://www.crunchyroll.com/watch/GYE5WKQGR',
'info_dict': { 'info_dict': {
'id': 'GYE5WKQGR', 'id': 'GYE5WKQGR',
@ -131,7 +242,7 @@ class CrunchyrollBetaIE(CrunchyrollBaseIE):
'duration': 366.459, 'duration': 366.459,
'timestamp': 1476788400, 'timestamp': 1476788400,
'description': 'md5:74b67283ffddd75f6e224ca7dc031e76', 'description': 'md5:74b67283ffddd75f6e224ca7dc031e76',
'title': 'SHELTER Episode Porter Robinson presents Shelter the Animation', 'title': 'SHELTER Porter Robinson presents Shelter the Animation',
'upload_date': '20161018', 'upload_date': '20161018',
'series': 'SHELTER', 'series': 'SHELTER',
'series_id': 'GYGG09WWY', 'series_id': 'GYGG09WWY',
@ -140,11 +251,58 @@ class CrunchyrollBetaIE(CrunchyrollBaseIE):
'season_number': 1, 'season_number': 1,
'episode': 'Porter Robinson presents Shelter the Animation', 'episode': 'Porter Robinson presents Shelter the Animation',
'episode_number': 0, 'episode_number': 0,
'thumbnail': r're:^https://www.crunchyroll.com/imgsrv/.*\.jpeg$', 'thumbnail': r're:^https://www.crunchyroll.com/imgsrv/.*\.jpeg?$',
'chapters': 'count:0', 'age_limit': 14,
'like_count': int,
'dislike_count': int,
}, },
'params': {'skip_download': True}, 'params': {'skip_download': True},
'skip': 'Video is Premium only', }, {
'url': 'https://www.crunchyroll.com/watch/GJWU2VKK3/cherry-blossom-meeting-and-a-coming-blizzard',
'info_dict': {
'id': 'GJWU2VKK3',
'ext': 'mp4',
'duration': 1420.054,
'description': 'md5:2d1c67c0ec6ae514d9c30b0b99a625cd',
'title': 'The Ice Guy and His Cool Female Colleague Episode 1 Cherry Blossom Meeting and a Coming Blizzard',
'series': 'The Ice Guy and His Cool Female Colleague',
'series_id': 'GW4HM75NP',
'season': 'The Ice Guy and His Cool Female Colleague',
'season_id': 'GY9PC21VE',
'season_number': 1,
'episode': 'Cherry Blossom Meeting and a Coming Blizzard',
'episode_number': 1,
'chapters': 'count:2',
'thumbnail': r're:^https://www.crunchyroll.com/imgsrv/.*\.jpeg?$',
'timestamp': 1672839000,
'upload_date': '20230104',
'age_limit': 14,
'like_count': int,
'dislike_count': int,
},
'params': {'skip_download': 'm3u8'},
}, {
'url': 'https://www.crunchyroll.com/watch/GM8F313NQ',
'info_dict': {
'id': 'GM8F313NQ',
'ext': 'mp4',
'title': 'Garakowa -Restore the World-',
'description': 'md5:8d2f8b6b9dd77d87810882e7d2ee5608',
'duration': 3996.104,
'age_limit': 13,
'thumbnail': r're:^https://www.crunchyroll.com/imgsrv/.*\.jpeg?$',
},
'params': {'skip_download': 'm3u8'},
}, {
'url': 'https://www.crunchyroll.com/watch/G62PEZ2E6',
'info_dict': {
'id': 'G62PEZ2E6',
'description': 'md5:8d2f8b6b9dd77d87810882e7d2ee5608',
'age_limit': 13,
'duration': 65.138,
'title': 'Garakowa -Restore the World-',
},
'playlist_mincount': 5,
}, { }, {
'url': 'https://www.crunchyroll.com/watch/GY2P1Q98Y', 'url': 'https://www.crunchyroll.com/watch/GY2P1Q98Y',
'only_matching': True, 'only_matching': True,
@ -152,125 +310,147 @@ class CrunchyrollBetaIE(CrunchyrollBaseIE):
'url': 'https://beta.crunchyroll.com/pt-br/watch/G8WUN8VKP/the-ruler-of-conspiracy', 'url': 'https://beta.crunchyroll.com/pt-br/watch/G8WUN8VKP/the-ruler-of-conspiracy',
'only_matching': True, 'only_matching': True,
}] }]
# We want to support lazy playlist filtering and movie listings cannot be inside a playlist
_RETURN_TYPE = 'video'
def _real_extract(self, url): def _real_extract(self, url):
lang, internal_id, display_id = self._match_valid_url(url).group('lang', 'id', 'display_id') lang, internal_id = self._match_valid_url(url).group('lang', 'id')
api_domain, bucket, params = self._get_params(lang)
episode_response = self._download_json( # We need to use unsigned API call to allow ratings query string
f'{api_domain}/cms/v2{bucket}/episodes/{internal_id}', display_id, response = traverse_obj(self._call_api(
note='Retrieving episode metadata', query=params) f'objects/{internal_id}', internal_id, lang, 'object info', {'ratings': 'true'}), ('data', 0, {dict}))
if episode_response.get('is_premium_only') and not bucket.endswith('crunchyroll'): if not response:
if self.is_logged_in: raise ExtractorError(f'No video with id {internal_id} could be found (possibly region locked?)', expected=True)
raise ExtractorError('This video is for premium members only', expected=True)
else:
self.raise_login_required('This video is for premium members only')
stream_response = self._download_json( object_type = response.get('type')
f'{api_domain}{episode_response["__links__"]["streams"]["href"]}', display_id, if object_type == 'episode':
note='Retrieving stream info', query=params) result = self._transform_episode_response(response)
get_streams = lambda name: (traverse_obj(stream_response, name) or {}).items()
requested_hardsubs = [('' if val == 'none' else val) for val in (self._configuration_arg('hardsub') or ['none'])] elif object_type == 'movie':
hardsub_preference = qualities(requested_hardsubs[::-1]) result = self._transform_movie_response(response)
requested_formats = self._configuration_arg('format') or ['adaptive_hls']
available_formats = {} elif object_type == 'movie_listing':
for stream_type, streams in get_streams('streams'): first_movie_id = traverse_obj(response, ('movie_listing_metadata', 'first_movie_id'))
if stream_type not in requested_formats: if not self._yes_playlist(internal_id, first_movie_id):
continue return self.url_result(f'{self._BASE_URL}/{lang}watch/{first_movie_id}', CrunchyrollBetaIE, first_movie_id)
for stream in streams.values():
if not stream.get('url'):
continue
hardsub_lang = stream.get('hardsub_locale') or ''
format_id = join_nonempty(stream_type, format_field(stream, 'hardsub_locale', 'hardsub-%s'))
available_formats[hardsub_lang] = (stream_type, format_id, hardsub_lang, stream['url'])
if '' in available_formats and 'all' not in requested_hardsubs: def entries():
full_format_langs = set(requested_hardsubs) movies = self._call_api(f'movie_listings/{internal_id}/movies', internal_id, lang, 'movie list')
self.to_screen( for movie_response in traverse_obj(movies, ('data', ...)):
'To get all formats of a hardsub language, use ' yield self.url_result(
'"--extractor-args crunchyrollbeta:hardsub=<language_code or all>". ' f'{self._BASE_URL}/{lang}watch/{movie_response["id"]}',
'See https://github.com/yt-dlp/yt-dlp#crunchyrollbeta-crunchyroll for more info', CrunchyrollBetaIE, **self._transform_movie_response(movie_response))
only_once=True)
else: return self.playlist_result(entries(), **self._transform_movie_response(response))
full_format_langs = set(map(str.lower, available_formats))
formats = []
for stream_type, format_id, hardsub_lang, stream_url in available_formats.values():
if stream_type.endswith('hls'):
if hardsub_lang.lower() in full_format_langs:
adaptive_formats = self._extract_m3u8_formats(
stream_url, display_id, 'mp4', m3u8_id=format_id,
fatal=False, note=f'Downloading {format_id} HLS manifest')
else:
adaptive_formats = (self._m3u8_meta_format(stream_url, ext='mp4', m3u8_id=format_id),)
elif stream_type.endswith('dash'):
adaptive_formats = self._extract_mpd_formats(
stream_url, display_id, mpd_id=format_id,
fatal=False, note=f'Downloading {format_id} MPD manifest')
else: else:
self.report_warning(f'Encountered unknown stream_type: {stream_type!r}', display_id, only_once=True) raise ExtractorError(f'Unknown object type {object_type}')
continue
for f in adaptive_formats: # There might be multiple audio languages for one object (`<object>_metadata.versions`),
if f.get('acodec') != 'none': # so we need to get the id from `streams_link` instead or we dont know which language to choose
f['language'] = stream_response.get('audio_locale') streams_link = response.get('streams_link')
f['quality'] = hardsub_preference(hardsub_lang.lower()) if not streams_link and traverse_obj(response, (f'{object_type}_metadata', 'is_premium_only')):
formats.extend(adaptive_formats) message = f'This {object_type} is for premium members only'
if self.is_logged_in:
raise ExtractorError(message, expected=True)
self.raise_login_required(message)
# We need go from unsigned to signed api to avoid getting soft banned
stream_response = self._call_cms_api_signed(remove_start(
streams_link, '/content/v2/cms/'), internal_id, lang, 'stream info')
result['formats'] = self._extract_formats(stream_response, internal_id)
result['subtitles'] = self._extract_subtitles(stream_response)
chapters = None
# if no intro chapter is available, a 403 without usable data is returned # if no intro chapter is available, a 403 without usable data is returned
intro_chapter = self._download_json(f'https://static.crunchyroll.com/datalab-intro-v2/{internal_id}.json', intro_chapter = self._download_json(
display_id, fatal=False, errnote=False) f'https://static.crunchyroll.com/datalab-intro-v2/{internal_id}.json',
internal_id, note='Downloading chapter info', fatal=False, errnote=False)
if isinstance(intro_chapter, dict): if isinstance(intro_chapter, dict):
chapters = [{ result['chapters'] = [{
'title': 'Intro', 'title': 'Intro',
'start_time': float_or_none(intro_chapter.get('startTime')), 'start_time': float_or_none(intro_chapter.get('startTime')),
'end_time': float_or_none(intro_chapter.get('endTime')) 'end_time': float_or_none(intro_chapter.get('endTime')),
}] }]
def calculate_count(item):
return parse_count(''.join((item['displayed'], item.get('unit') or '')))
result.update(traverse_obj(response, ('rating', {
'like_count': ('up', {calculate_count}),
'dislike_count': ('down', {calculate_count}),
})))
return result
@staticmethod
def _transform_episode_response(data):
metadata = traverse_obj(data, (('episode_metadata', None), {dict}), get_all=False) or {}
return { return {
'id': internal_id, 'id': data['id'],
'title': '%s Episode %s %s' % ( 'title': ' \u2013 '.join((
episode_response.get('season_title'), episode_response.get('episode'), episode_response.get('title')), ('%s%s' % (
'description': try_get(episode_response, lambda x: x['description'].replace(r'\r\n', '\n')), format_field(metadata, 'season_title'),
'duration': float_or_none(episode_response.get('duration_ms'), 1000), format_field(metadata, 'episode', ' Episode %s'))),
'timestamp': parse_iso8601(episode_response.get('upload_date')), format_field(data, 'title'))),
'series': episode_response.get('series_title'), **traverse_obj(data, {
'series_id': episode_response.get('series_id'), 'episode': ('title', {str}),
'season': episode_response.get('season_title'), 'description': ('description', {str}, {lambda x: x.replace(r'\r\n', '\n')}),
'season_id': episode_response.get('season_id'), 'thumbnails': ('images', 'thumbnail', ..., ..., {
'season_number': episode_response.get('season_number'), 'url': ('source', {url_or_none}),
'episode': episode_response.get('title'), 'width': ('width', {int_or_none}),
'episode_number': episode_response.get('sequence_number'), 'height': ('height', {int_or_none}),
'formats': formats, }),
'thumbnails': [{ }),
'url': thumb.get('source'), **traverse_obj(metadata, {
'width': thumb.get('width'), 'duration': ('duration_ms', {lambda x: float_or_none(x, 1000)}),
'height': thumb.get('height'), 'timestamp': ('upload_date', {parse_iso8601}),
} for thumb in traverse_obj(episode_response, ('images', 'thumbnail', ..., ...)) or []], 'series': ('series_title', {str}),
'subtitles': { 'series_id': ('series_id', {str}),
lang: [{ 'season': ('season_title', {str}),
'url': subtitle_data.get('url'), 'season_id': ('season_id', {str}),
'ext': subtitle_data.get('format') 'season_number': ('season_number', ({int}, {float_or_none})),
}] for lang, subtitle_data in get_streams('subtitles') 'episode_number': ('sequence_number', ({int}, {float_or_none})),
}, 'age_limit': ('maturity_ratings', -1, {parse_age_limit}),
'chapters': chapters 'language': ('audio_locale', {str}),
}, get_all=False),
} }
@staticmethod
def _transform_movie_response(data):
metadata = traverse_obj(data, (('movie_metadata', 'movie_listing_metadata', None), {dict}), get_all=False) or {}
return {
'id': data['id'],
**traverse_obj(data, {
'title': ('title', {str}),
'description': ('description', {str}, {lambda x: x.replace(r'\r\n', '\n')}),
'thumbnails': ('images', 'thumbnail', ..., ..., {
'url': ('source', {url_or_none}),
'width': ('width', {int_or_none}),
'height': ('height', {int_or_none}),
}),
}),
**traverse_obj(metadata, {
'duration': ('duration_ms', {lambda x: float_or_none(x, 1000)}),
'age_limit': ('maturity_ratings', -1, {parse_age_limit}),
}),
}
class CrunchyrollBetaShowIE(CrunchyrollBaseIE):
class CrunchyrollBetaShowIE(CrunchyrollCmsBaseIE):
IE_NAME = 'crunchyroll:playlist' IE_NAME = 'crunchyroll:playlist'
_VALID_URL = r'''(?x) _VALID_URL = r'''(?x)
https?://(?:beta|www)\.crunchyroll\.com/ https?://(?:beta\.|www\.)?crunchyroll\.com/
(?P<lang>(?:\w{2}(?:-\w{2})?/)?) (?P<lang>(?:\w{2}(?:-\w{2})?/)?)
series/(?P<id>\w+) series/(?P<id>\w+)'''
(?:/(?P<display_id>[\w-]+))?/?(?:[?#]|$)'''
_TESTS = [{ _TESTS = [{
'url': 'https://www.crunchyroll.com/series/GY19NQ2QR/Girl-Friend-BETA', 'url': 'https://www.crunchyroll.com/series/GY19NQ2QR/Girl-Friend-BETA',
'info_dict': { 'info_dict': {
'id': 'GY19NQ2QR', 'id': 'GY19NQ2QR',
'title': 'Girl Friend BETA', 'title': 'Girl Friend BETA',
'description': 'md5:99c1b22ee30a74b536a8277ced8eb750',
# XXX: `thumbnail` does not get set from `thumbnails` in playlist
# 'thumbnail': r're:^https://www.crunchyroll.com/imgsrv/.*\.jpeg?$',
'age_limit': 14,
}, },
'playlist_mincount': 10, 'playlist_mincount': 10,
}, { }, {
@ -279,41 +459,163 @@ class CrunchyrollBetaShowIE(CrunchyrollBaseIE):
}] }]
def _real_extract(self, url): def _real_extract(self, url):
lang, internal_id, display_id = self._match_valid_url(url).group('lang', 'id', 'display_id') lang, internal_id = self._match_valid_url(url).group('lang', 'id')
api_domain, bucket, params = self._get_params(lang)
series_response = self._download_json( def entries():
f'{api_domain}/cms/v2{bucket}/series/{internal_id}', display_id, seasons_response = self._call_cms_api_signed(f'seasons?series_id={internal_id}', internal_id, lang, 'seasons')
note='Retrieving series metadata', query=params) for season in traverse_obj(seasons_response, ('items', ..., {dict})):
episodes_response = self._call_cms_api_signed(
f'episodes?season_id={season["id"]}', season["id"], lang, 'episode list')
for episode_response in traverse_obj(episodes_response, ('items', ..., {dict})):
yield self.url_result(
f'{self._BASE_URL}/{lang}watch/{episode_response["id"]}',
CrunchyrollBetaIE, **CrunchyrollBetaIE._transform_episode_response(episode_response))
seasons_response = self._download_json( return self.playlist_result(
f'{api_domain}/cms/v2{bucket}/seasons?series_id={internal_id}', display_id, entries(), internal_id,
note='Retrieving season list', query=params) **traverse_obj(self._call_api(f'series/{internal_id}', internal_id, lang, 'series'), ('data', 0, {
'title': ('title', {str}),
'description': ('description', {lambda x: x.replace(r'\r\n', '\n')}),
'age_limit': ('maturity_ratings', -1, {parse_age_limit}),
'thumbnails': ('images', ..., ..., ..., {
'url': ('source', {url_or_none}),
'width': ('width', {int_or_none}),
'height': ('height', {int_or_none}),
})
})))
def entries():
for season in seasons_response['items']: class CrunchyrollMusicIE(CrunchyrollBaseIE):
episodes_response = self._download_json( IE_NAME = 'crunchyroll:music'
f'{api_domain}/cms/v2{bucket}/episodes?season_id={season["id"]}', display_id, _VALID_URL = r'''(?x)
note=f'Retrieving episode list for {season.get("slug_title")}', query=params) https?://(?:www\.)?crunchyroll\.com/
for episode in episodes_response['items']: (?P<lang>(?:\w{2}(?:-\w{2})?/)?)
episode_id = episode['id'] watch/(?P<type>concert|musicvideo)/(?P<id>\w{10})'''
episode_display_id = episode['slug_title'] _TESTS = [{
yield { 'url': 'https://www.crunchyroll.com/watch/musicvideo/MV88BB7F2C',
'_type': 'url', 'info_dict': {
'url': f'https://www.crunchyroll.com/{lang}watch/{episode_id}/{episode_display_id}', 'ext': 'mp4',
'ie_key': CrunchyrollBetaIE.ie_key(), 'id': 'MV88BB7F2C',
'id': episode_id, 'display_id': 'crossing-field',
'title': '%s Episode %s %s' % (episode.get('season_title'), episode.get('episode'), episode.get('title')), 'title': 'Crossing Field',
'description': try_get(episode, lambda x: x['description'].replace(r'\r\n', '\n')), 'track': 'Crossing Field',
'duration': float_or_none(episode.get('duration_ms'), 1000), 'artist': 'LiSA',
'series': episode.get('series_title'), 'thumbnail': r're:(?i)^https://www.crunchyroll.com/imgsrv/.*\.jpeg?$',
'series_id': episode.get('series_id'), 'genre': ['Anime'],
'season': episode.get('season_title'), },
'season_id': episode.get('season_id'), 'params': {'skip_download': 'm3u8'},
'season_number': episode.get('season_number'), }, {
'episode': episode.get('title'), 'url': 'https://www.crunchyroll.com/watch/concert/MC2E2AC135',
'episode_number': episode.get('sequence_number'), 'info_dict': {
'language': episode.get('audio_locale'), 'ext': 'mp4',
'id': 'MC2E2AC135',
'display_id': 'live-is-smile-always-364joker-at-yokohama-arena',
'title': 'LiVE is Smile Always-364+JOKER- at YOKOHAMA ARENA',
'track': 'LiVE is Smile Always-364+JOKER- at YOKOHAMA ARENA',
'artist': 'LiSA',
'thumbnail': r're:(?i)^https://www.crunchyroll.com/imgsrv/.*\.jpeg?$',
'description': 'md5:747444e7e6300907b7a43f0a0503072e',
'genre': ['J-Pop'],
},
'params': {'skip_download': 'm3u8'},
}, {
'url': 'https://www.crunchyroll.com/watch/musicvideo/MV88BB7F2C/crossing-field',
'only_matching': True,
}, {
'url': 'https://www.crunchyroll.com/watch/concert/MC2E2AC135/live-is-smile-always-364joker-at-yokohama-arena',
'only_matching': True,
}]
_API_ENDPOINT = 'music'
def _real_extract(self, url):
lang, internal_id, object_type = self._match_valid_url(url).group('lang', 'id', 'type')
path, name = {
'concert': ('concerts', 'concert info'),
'musicvideo': ('music_videos', 'music video info'),
}[object_type]
response = traverse_obj(self._call_api(f'{path}/{internal_id}', internal_id, lang, name), ('data', 0, {dict}))
if not response:
raise ExtractorError(f'No video with id {internal_id} could be found (possibly region locked?)', expected=True)
streams_link = response.get('streams_link')
if not streams_link and response.get('isPremiumOnly'):
message = f'This {response.get("type") or "media"} is for premium members only'
if self.is_logged_in:
raise ExtractorError(message, expected=True)
self.raise_login_required(message)
result = self._transform_music_response(response)
stream_response = self._call_api(streams_link, internal_id, lang, 'stream info')
result['formats'] = self._extract_formats(stream_response, internal_id)
return result
@staticmethod
def _transform_music_response(data):
return {
'id': data['id'],
**traverse_obj(data, {
'display_id': 'slug',
'title': 'title',
'track': 'title',
'artist': ('artist', 'name'),
'description': ('description', {str}, {lambda x: x.replace(r'\r\n', '\n') or None}),
'thumbnails': ('images', ..., ..., {
'url': ('source', {url_or_none}),
'width': ('width', {int_or_none}),
'height': ('height', {int_or_none}),
}),
'genre': ('genres', ..., 'displayValue'),
'age_limit': ('maturity_ratings', -1, {parse_age_limit}),
}),
} }
return self.playlist_result(entries(), internal_id, series_response.get('title'))
class CrunchyrollArtistIE(CrunchyrollBaseIE):
IE_NAME = 'crunchyroll:artist'
_VALID_URL = r'''(?x)
https?://(?:www\.)?crunchyroll\.com/
(?P<lang>(?:\w{2}(?:-\w{2})?/)?)
artist/(?P<id>\w{10})'''
_TESTS = [{
'url': 'https://www.crunchyroll.com/artist/MA179CB50D',
'info_dict': {
'id': 'MA179CB50D',
'title': 'LiSA',
'genre': ['J-Pop', 'Anime', 'Rock'],
'description': 'md5:16d87de61a55c3f7d6c454b73285938e',
},
'playlist_mincount': 83,
}, {
'url': 'https://www.crunchyroll.com/artist/MA179CB50D/lisa',
'only_matching': True,
}]
_API_ENDPOINT = 'music'
def _real_extract(self, url):
lang, internal_id = self._match_valid_url(url).group('lang', 'id')
response = traverse_obj(self._call_api(
f'artists/{internal_id}', internal_id, lang, 'artist info'), ('data', 0))
def entries():
for attribute, path in [('concerts', 'concert'), ('videos', 'musicvideo')]:
for internal_id in traverse_obj(response, (attribute, ...)):
yield self.url_result(f'{self._BASE_URL}/watch/{path}/{internal_id}', CrunchyrollMusicIE, internal_id)
return self.playlist_result(entries(), **self._transform_artist_response(response))
@staticmethod
def _transform_artist_response(data):
return {
'id': data['id'],
**traverse_obj(data, {
'title': 'name',
'description': ('description', {str}, {lambda x: x.replace(r'\r\n', '\n')}),
'thumbnails': ('images', ..., ..., {
'url': ('source', {url_or_none}),
'width': ('width', {int_or_none}),
'height': ('height', {int_or_none}),
}),
'genre': ('genres', ..., 'displayValue'),
}),
}

Loading…
Cancel
Save