Upgrade yt_dlp and download script

This commit is contained in:
2025-05-02 16:11:08 -05:00
parent 3a2e8eeb08
commit d68d9ce4f9
1194 changed files with 60099 additions and 44436 deletions

View File

@@ -1,17 +1,23 @@
import json
import re
import urllib.parse
import time
import uuid
from .common import InfoExtractor
from ..networking.exceptions import HTTPError
from ..utils import (
ExtractorError,
determine_ext,
int_or_none,
join_nonempty,
jwt_decode_hs256,
parse_duration,
parse_iso8601,
traverse_obj,
try_get,
url_or_none,
urlencode_postdata,
)
from ..utils.traversal import traverse_obj
class MLBBaseIE(InfoExtractor):
@@ -203,7 +209,7 @@ class MLBIE(MLBBaseIE):
def _download_video_data(self, display_id):
return self._download_json(
'http://content.mlb.com/mlb/item/id/v1/%s/details/web-v1.json' % display_id,
f'http://content.mlb.com/mlb/item/id/v1/{display_id}/details/web-v1.json',
display_id)
@@ -227,7 +233,7 @@ class MLBVideoIE(MLBBaseIE):
@classmethod
def suitable(cls, url):
return False if MLBIE.suitable(url) else super(MLBVideoIE, cls).suitable(url)
return False if MLBIE.suitable(url) else super().suitable(url)
@staticmethod
def _get_feed(video):
@@ -268,83 +274,232 @@ class MLBVideoIE(MLBBaseIE):
timestamp
title
}
}''' % display_id,
}''' % display_id, # noqa: UP031
})['data']['mediaPlayback'][0]
class MLBTVIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?mlb\.com/tv/g(?P<id>\d{6})'
_NETRC_MACHINE = 'mlb'
_TESTS = [{
'url': 'https://www.mlb.com/tv/g661581/vee2eff5f-a7df-4c20-bdb4-7b926fa12638',
'info_dict': {
'id': '661581',
'ext': 'mp4',
'title': '2022-07-02 - St. Louis Cardinals @ Philadelphia Phillies',
'release_date': '20220702',
'release_timestamp': 1656792300,
},
'params': {
'skip_download': True,
'params': {'skip_download': 'm3u8'},
}, {
# makeup game: has multiple dates, need to avoid games with 'rescheduleDate'
'url': 'https://www.mlb.com/tv/g747039/vd22541c4-5a29-45f7-822b-635ec041cf5e',
'info_dict': {
'id': '747039',
'ext': 'mp4',
'title': '2024-07-29 - Toronto Blue Jays @ Baltimore Orioles',
'release_date': '20240729',
'release_timestamp': 1722280200,
},
'params': {'skip_download': 'm3u8'},
}]
_GRAPHQL_INIT_QUERY = '''\
mutation initSession($device: InitSessionInput!, $clientType: ClientType!, $experience: ExperienceTypeInput) {
initSession(device: $device, clientType: $clientType, experience: $experience) {
deviceId
sessionId
entitlements {
code
}
location {
countryCode
regionName
zipCode
latitude
longitude
}
clientExperience
features
}
}'''
_GRAPHQL_PLAYBACK_QUERY = '''\
mutation initPlaybackSession(
$adCapabilities: [AdExperienceType]
$mediaId: String!
$deviceId: String!
$sessionId: String!
$quality: PlaybackQuality
) {
initPlaybackSession(
adCapabilities: $adCapabilities
mediaId: $mediaId
deviceId: $deviceId
sessionId: $sessionId
quality: $quality
) {
playbackSessionId
playback {
url
token
expiration
cdn
}
}
}'''
_APP_VERSION = '7.8.2'
_device_id = None
_session_id = None
_access_token = None
_token_expiry = 0
@property
def _api_headers(self):
if (self._token_expiry - 120) <= time.time():
self.write_debug('Access token has expired; re-logging in')
self._perform_login(*self._get_login_info())
return {'Authorization': f'Bearer {self._access_token}'}
def _real_initialize(self):
if not self._access_token:
self.raise_login_required(
'All videos are only available to registered users', method='password')
def _set_device_id(self, username):
if self._device_id:
return
device_id_cache = self.cache.load(self._NETRC_MACHINE, 'device_ids', default={})
self._device_id = device_id_cache.get(username)
if self._device_id:
return
self._device_id = str(uuid.uuid4())
device_id_cache[username] = self._device_id
self.cache.store(self._NETRC_MACHINE, 'device_ids', device_id_cache)
def _perform_login(self, username, password):
data = f'grant_type=password&username={urllib.parse.quote(username)}&password={urllib.parse.quote(password)}&scope=openid offline_access&client_id=0oa3e1nutA1HLzAKG356'
access_token = self._download_json(
'https://ids.mlb.com/oauth2/aus1m088yK07noBfh356/v1/token', None,
headers={
'User-Agent': 'okhttp/3.12.1',
'Content-Type': 'application/x-www-form-urlencoded'
}, data=data.encode())['access_token']
try:
self._access_token = self._download_json(
'https://ids.mlb.com/oauth2/aus1m088yK07noBfh356/v1/token', None,
'Logging in', 'Unable to log in', headers={
'User-Agent': 'okhttp/3.12.1',
'Content-Type': 'application/x-www-form-urlencoded',
}, data=urlencode_postdata({
'grant_type': 'password',
'username': username,
'password': password,
'scope': 'openid offline_access',
'client_id': '0oa3e1nutA1HLzAKG356',
}))['access_token']
except ExtractorError as error:
if isinstance(error.cause, HTTPError) and error.cause.status == 400:
raise ExtractorError('Invalid username or password', expected=True)
raise
entitlement = self._download_webpage(
f'https://media-entitlement.mlb.com/api/v3/jwt?os=Android&appname=AtBat&did={str(uuid.uuid4())}', None,
headers={
'User-Agent': 'okhttp/3.12.1',
'Authorization': f'Bearer {access_token}'
})
self._token_expiry = traverse_obj(self._access_token, ({jwt_decode_hs256}, 'exp', {int})) or 0
self._set_device_id(username)
data = f'grant_type=urn:ietf:params:oauth:grant-type:token-exchange&subject_token={entitlement}&subject_token_type=urn:ietf:params:oauth:token-type:jwt&platform=android-tv'
self._access_token = self._download_json(
'https://us.edge.bamgrid.com/token', None,
self._session_id = self._call_api({
'operationName': 'initSession',
'query': self._GRAPHQL_INIT_QUERY,
'variables': {
'device': {
'appVersion': self._APP_VERSION,
'deviceFamily': 'desktop',
'knownDeviceId': self._device_id,
'languagePreference': 'ENGLISH',
'manufacturer': '',
'model': '',
'os': '',
'osVersion': '',
},
'clientType': 'WEB',
},
}, None, 'session ID')['data']['initSession']['sessionId']
def _call_api(self, data, video_id, description='GraphQL JSON', fatal=True):
return self._download_json(
'https://media-gateway.mlb.com/graphql', video_id,
f'Downloading {description}', f'Unable to download {description}', fatal=fatal,
headers={
**self._api_headers,
'Accept': 'application/json',
'Authorization': 'Bearer bWxidHYmYW5kcm9pZCYxLjAuMA.6LZMbH2r--rbXcgEabaDdIslpo4RyZrlVfWZhsAgXIk',
'Content-Type': 'application/x-www-form-urlencoded'
}, data=data.encode())['access_token']
'Content-Type': 'application/json',
'x-client-name': 'WEB',
'x-client-version': self._APP_VERSION,
}, data=json.dumps(data, separators=(',', ':')).encode())
def _extract_formats_and_subtitles(self, broadcast, video_id):
feed = traverse_obj(broadcast, ('homeAway', {str.title}))
medium = traverse_obj(broadcast, ('type', {str}))
language = traverse_obj(broadcast, ('language', {str.lower}))
format_id = join_nonempty(feed, medium, language)
response = self._call_api({
'operationName': 'initPlaybackSession',
'query': self._GRAPHQL_PLAYBACK_QUERY,
'variables': {
'adCapabilities': ['GOOGLE_STANDALONE_AD_PODS'],
'deviceId': self._device_id,
'mediaId': broadcast['mediaId'],
'quality': 'PLACEHOLDER',
'sessionId': self._session_id,
},
}, video_id, f'{format_id} broadcast JSON', fatal=False)
playback = traverse_obj(response, ('data', 'initPlaybackSession', 'playback', {dict}))
m3u8_url = traverse_obj(playback, ('url', {url_or_none}))
token = traverse_obj(playback, ('token', {str}))
if not (m3u8_url and token):
errors = '; '.join(traverse_obj(response, ('errors', ..., 'message', {str})))
if errors: # Only warn when 'blacked out' or 'not entitled'; radio formats may be available
self.report_warning(f'API returned errors for {format_id}: {errors}')
else:
self.report_warning(f'No formats available for {format_id} broadcast; skipping')
return [], {}
cdn_headers = {'x-cdn-token': token}
fmts, subs = self._extract_m3u8_formats_and_subtitles(
m3u8_url.replace(f'/{token}/', '/'), video_id, 'mp4',
m3u8_id=format_id, fatal=False, headers=cdn_headers)
for fmt in fmts:
fmt['http_headers'] = cdn_headers
fmt.setdefault('format_note', join_nonempty(feed, medium, delim=' '))
fmt.setdefault('language', language)
if fmt.get('vcodec') == 'none' and fmt['language'] == 'en':
fmt['source_preference'] = 10
return fmts, subs
def _real_extract(self, url):
video_id = self._match_id(url)
airings = self._download_json(
f'https://search-api-mlbtv.mlb.com/svc/search/v2/graphql/persisted/query/core/Airings?variables=%7B%22partnerProgramIds%22%3A%5B%22{video_id}%22%5D%2C%22applyEsniMediaRightsLabels%22%3Atrue%7D',
video_id)['data']['Airings']
data = self._download_json(
'https://statsapi.mlb.com/api/v1/schedule', video_id, query={
'gamePk': video_id,
'hydrate': 'broadcasts(all),statusFlags',
})
metadata = traverse_obj(data, (
'dates', ..., 'games',
lambda _, v: str(v['gamePk']) == video_id and not v.get('rescheduleDate'), any))
broadcasts = traverse_obj(metadata, (
'broadcasts', lambda _, v: v['mediaId'] and v['mediaState']['mediaStateCode'] != 'MEDIA_OFF'))
formats, subtitles = [], {}
for airing in airings:
m3u8_url = self._download_json(
airing['playbackUrls'][0]['href'].format(scenario='browser~csai'), video_id,
headers={
'Authorization': self._access_token,
'Accept': 'application/vnd.media-service+json; version=2'
})['stream']['complete']
f, s = self._extract_m3u8_formats_and_subtitles(
m3u8_url, video_id, 'mp4', m3u8_id=join_nonempty(airing.get('feedType'), airing.get('feedLanguage')))
formats.extend(f)
self._merge_subtitles(s, target=subtitles)
for broadcast in broadcasts:
fmts, subs = self._extract_formats_and_subtitles(broadcast, video_id)
formats.extend(fmts)
self._merge_subtitles(subs, target=subtitles)
return {
'id': video_id,
'title': traverse_obj(airings, (..., 'titles', 0, 'episodeName'), get_all=False),
'is_live': traverse_obj(airings, (..., 'mediaConfig', 'productType'), get_all=False) == 'LIVE',
'title': join_nonempty(
traverse_obj(metadata, ('officialDate', {str})),
traverse_obj(metadata, ('teams', ('away', 'home'), 'team', 'name', {str}, all, {' @ '.join})),
delim=' - '),
'is_live': traverse_obj(broadcasts, (..., 'mediaState', 'mediaStateCode', {str}, any)) == 'MEDIA_ON',
'release_timestamp': traverse_obj(metadata, ('gameDate', {parse_iso8601})),
'formats': formats,
'subtitles': subtitles,
'http_headers': {'Authorization': f'Bearer {self._access_token}'},
}
@@ -355,11 +510,11 @@ class MLBArticleIE(InfoExtractor):
'info_dict': {
'id': '36db7394-343c-4ea3-b8ca-ead2e61bca9a',
'title': 'Machado\'s grab draws hilarious irate reaction',
'modified_timestamp': 1650130737,
'modified_timestamp': 1675888370,
'description': 'md5:a19d4eb0487b2cb304e9a176f6b67676',
'modified_date': '20220416',
'modified_date': '20230208',
},
'playlist_count': 2,
'playlist_mincount': 2,
}]
def _real_extract(self, url):
@@ -367,15 +522,13 @@ class MLBArticleIE(InfoExtractor):
webpage = self._download_webpage(url, display_id)
apollo_cache_json = self._search_json(r'window\.initState\s*=', webpage, 'window.initState', display_id)['apolloCache']
content_data_id = traverse_obj(
apollo_cache_json, ('ROOT_QUERY', lambda k, _: k.startswith('getForgeContent'), 'id'), get_all=False)
content_real_info = apollo_cache_json[content_data_id]
content_real_info = traverse_obj(
apollo_cache_json, ('ROOT_QUERY', lambda k, _: k.startswith('getArticle')), get_all=False)
return self.playlist_from_matches(
traverse_obj(content_real_info, ('parts', lambda _, v: v['typename'] == 'Video', 'id')),
getter=lambda x: f'https://www.mlb.com/video/{apollo_cache_json[x]["slug"]}',
ie=MLBVideoIE, playlist_id=content_real_info.get('_translationId'),
traverse_obj(content_real_info, ('parts', lambda _, v: v['__typename'] == 'Video' or v['type'] == 'video')),
getter=lambda x: f'https://www.mlb.com/video/{x["slug"]}',
ie=MLBVideoIE, playlist_id=content_real_info.get('translationId'),
title=self._html_search_meta('og:title', webpage),
description=content_real_info.get('summary'),
modified_timestamp=parse_iso8601(content_real_info.get('lastUpdatedDate')))