kopia lustrzana https://github.com/yt-dlp/yt-dlp
Porównaj commity
9 Commity
64d4c4ba72
...
6c7bfb5c7c
Autor | SHA1 | Data |
---|---|---|
bashonly | 6c7bfb5c7c | |
bashonly | 805f40a001 | |
bashonly | b77ccbc67b | |
bashonly | d5d83395ab | |
bashonly | f2f108b754 | |
bashonly | d66fa6a118 | |
bashonly | 3b48105b8c | |
bashonly | 89f535e265 | |
bashonly | ff38a011d5 |
|
@ -53,7 +53,7 @@ jobs:
|
|||
with:
|
||||
python-version: ${{ matrix.python-version }}
|
||||
- name: Install test requirements
|
||||
run: python3 ./devscripts/install_deps.py --include dev --include curl_cffi
|
||||
run: python3 ./devscripts/install_deps.py --include dev --include curl-cffi
|
||||
- name: Run tests
|
||||
continue-on-error: False
|
||||
run: |
|
||||
|
|
|
@ -785,6 +785,25 @@ class TestHTTPImpersonateRequestHandler(TestRequestHandlerBase):
|
|||
assert res.status == 200
|
||||
assert std_headers['user-agent'].lower() not in res.read().decode().lower()
|
||||
|
||||
def test_response_extensions(self, handler):
|
||||
with handler() as rh:
|
||||
for target in rh.supported_targets:
|
||||
request = Request(
|
||||
f'http://127.0.0.1:{self.http_port}/gen_200', extensions={'impersonate': target})
|
||||
res = validate_and_send(rh, request)
|
||||
assert res.extensions['impersonate'] == rh._get_request_target(request)
|
||||
|
||||
def test_http_error_response_extensions(self, handler):
|
||||
with handler() as rh:
|
||||
for target in rh.supported_targets:
|
||||
request = Request(
|
||||
f'http://127.0.0.1:{self.http_port}/gen_404', extensions={'impersonate': target})
|
||||
try:
|
||||
validate_and_send(rh, request)
|
||||
except HTTPError as e:
|
||||
res = e.response
|
||||
assert res.extensions['impersonate'] == rh._get_request_target(request)
|
||||
|
||||
|
||||
class TestRequestHandlerMisc:
|
||||
"""Misc generic tests for request handlers, not related to request or validation testing"""
|
||||
|
|
|
@ -24,11 +24,15 @@ class CrunchyrollBaseIE(InfoExtractor):
|
|||
_BASE_URL = 'https://www.crunchyroll.com'
|
||||
_API_BASE = 'https://api.crunchyroll.com'
|
||||
_NETRC_MACHINE = 'crunchyroll'
|
||||
_REFRESH_TOKEN = None
|
||||
_AUTH_HEADERS = None
|
||||
_AUTH_EXPIRY = None
|
||||
_API_ENDPOINT = None
|
||||
_BASIC_AUTH = None
|
||||
_BASIC_AUTH = 'Basic ' + base64.b64encode(':'.join((
|
||||
't-kdgp2h8c3jub8fn0fq',
|
||||
'yfLDfMfrYvKXh4JXS1LEI2cCqu1v5Wan',
|
||||
)).encode()).decode()
|
||||
_IS_PREMIUM = None
|
||||
_CLIENT_ID = ('cr_web', 'noaihdevm_6iyg0a8l0q')
|
||||
_LOCALE_LOOKUP = {
|
||||
'ar': 'ar-SA',
|
||||
'de': 'de-DE',
|
||||
|
@ -43,69 +47,74 @@ class CrunchyrollBaseIE(InfoExtractor):
|
|||
'hi': 'hi-IN',
|
||||
}
|
||||
|
||||
@property
|
||||
def is_logged_in(self):
|
||||
return bool(self._get_cookies(self._BASE_URL).get('etp_rt'))
|
||||
def _set_auth_info(self, response):
|
||||
CrunchyrollBaseIE._IS_PREMIUM = 'cr_premium' in traverse_obj(response, ('access_token', {jwt_decode_hs256}, 'benefits', ...))
|
||||
CrunchyrollBaseIE._AUTH_HEADERS = {'Authorization': response['token_type'] + ' ' + response['access_token']}
|
||||
CrunchyrollBaseIE._AUTH_EXPIRY = time_seconds(seconds=traverse_obj(response, ('expires_in', {float_or_none}), default=300) - 10)
|
||||
|
||||
def _request_token(self, headers, data, note='Requesting token', errnote='Failed to request token'):
|
||||
try: # TODO: Add impersonation support here
|
||||
return self._download_json(
|
||||
f'{self._BASE_URL}/auth/v1/token', None, note=note, errnote=errnote,
|
||||
headers=headers, data=urlencode_postdata(data))
|
||||
except ExtractorError as error:
|
||||
if not isinstance(error.cause, HTTPError) or error.cause.status != 403:
|
||||
raise
|
||||
raise ExtractorError(
|
||||
'Request blocked by Cloudflare; navigate to Crunchyroll in your browser, '
|
||||
'then pass the fresh cookies (with --cookies-from-browser or --cookies) '
|
||||
'and your browser\'s User-Agent (with --user-agent)', expected=True)
|
||||
|
||||
def _perform_login(self, username, password):
|
||||
if self.is_logged_in:
|
||||
if not CrunchyrollBaseIE._REFRESH_TOKEN:
|
||||
CrunchyrollBaseIE._REFRESH_TOKEN = self.cache.load(self._NETRC_MACHINE, username)
|
||||
if CrunchyrollBaseIE._REFRESH_TOKEN:
|
||||
return
|
||||
|
||||
upsell_response = self._download_json(
|
||||
f'{self._API_BASE}/get_upsell_data.0.json', None, 'Getting session id',
|
||||
query={
|
||||
'sess_id': 1,
|
||||
'device_id': 'whatvalueshouldbeforweb',
|
||||
'device_type': 'com.crunchyroll.static',
|
||||
'access_token': 'giKq5eY27ny3cqz',
|
||||
'referer': f'{self._BASE_URL}/welcome/login'
|
||||
})
|
||||
if upsell_response['code'] != 'ok':
|
||||
raise ExtractorError('Could not get session id')
|
||||
session_id = upsell_response['data']['session_id']
|
||||
|
||||
login_response = self._download_json(
|
||||
f'{self._API_BASE}/login.1.json', None, 'Logging in',
|
||||
data=urlencode_postdata({
|
||||
'account': username,
|
||||
'password': password,
|
||||
'session_id': session_id
|
||||
}))
|
||||
if login_response['code'] != 'ok':
|
||||
raise ExtractorError('Login failed. Server message: %s' % login_response['message'], expected=True)
|
||||
if not self.is_logged_in:
|
||||
raise ExtractorError('Login succeeded but did not set etp_rt cookie')
|
||||
|
||||
def _update_auth(self):
|
||||
if CrunchyrollBaseIE._AUTH_HEADERS and CrunchyrollBaseIE._AUTH_REFRESH > time_seconds():
|
||||
return
|
||||
|
||||
if not CrunchyrollBaseIE._BASIC_AUTH:
|
||||
cx_api_param = self._CLIENT_ID[self.is_logged_in]
|
||||
self.write_debug(f'Using cxApiParam={cx_api_param}')
|
||||
CrunchyrollBaseIE._BASIC_AUTH = 'Basic ' + base64.b64encode(f'{cx_api_param}:'.encode()).decode()
|
||||
|
||||
auth_headers = {'Authorization': CrunchyrollBaseIE._BASIC_AUTH}
|
||||
if self.is_logged_in:
|
||||
grant_type = 'etp_rt_cookie'
|
||||
else:
|
||||
grant_type = 'client_id'
|
||||
auth_headers['ETP-Anonymous-ID'] = uuid.uuid4()
|
||||
try:
|
||||
auth_response = self._download_json(
|
||||
f'{self._BASE_URL}/auth/v1/token', None, note=f'Authenticating with grant_type={grant_type}',
|
||||
headers=auth_headers, data=f'grant_type={grant_type}'.encode())
|
||||
login_response = self._request_token(
|
||||
headers={'Authorization': self._BASIC_AUTH}, data={
|
||||
'username': username,
|
||||
'password': password,
|
||||
'grant_type': 'password',
|
||||
'scope': 'offline_access',
|
||||
}, note='Logging in', errnote='Failed to log in')
|
||||
except ExtractorError as error:
|
||||
if isinstance(error.cause, HTTPError) and error.cause.status == 403:
|
||||
raise ExtractorError(
|
||||
'Request blocked by Cloudflare; navigate to Crunchyroll in your browser, '
|
||||
'then pass the fresh cookies (with --cookies-from-browser or --cookies) '
|
||||
'and your browser\'s User-Agent (with --user-agent)', expected=True)
|
||||
if isinstance(error.cause, HTTPError) and error.cause.status == 401:
|
||||
raise ExtractorError('Invalid username and/or password', expected=True)
|
||||
raise
|
||||
|
||||
CrunchyrollBaseIE._IS_PREMIUM = 'cr_premium' in traverse_obj(auth_response, ('access_token', {jwt_decode_hs256}, 'benefits', ...))
|
||||
CrunchyrollBaseIE._AUTH_HEADERS = {'Authorization': auth_response['token_type'] + ' ' + auth_response['access_token']}
|
||||
CrunchyrollBaseIE._AUTH_REFRESH = time_seconds(seconds=traverse_obj(auth_response, ('expires_in', {float_or_none}), default=300) - 10)
|
||||
CrunchyrollBaseIE._REFRESH_TOKEN = login_response['refresh_token']
|
||||
self.cache.store(self._NETRC_MACHINE, username, CrunchyrollBaseIE._REFRESH_TOKEN)
|
||||
self._set_auth_info(login_response)
|
||||
|
||||
def _update_auth(self):
|
||||
if CrunchyrollBaseIE._AUTH_HEADERS and CrunchyrollBaseIE._AUTH_EXPIRY > time_seconds():
|
||||
return
|
||||
|
||||
auth_headers = {'Authorization': self._BASIC_AUTH}
|
||||
if CrunchyrollBaseIE._REFRESH_TOKEN:
|
||||
data = {
|
||||
'refresh_token': CrunchyrollBaseIE._REFRESH_TOKEN,
|
||||
'grant_type': 'refresh_token',
|
||||
'scope': 'offline_access',
|
||||
}
|
||||
else:
|
||||
data = {'grant_type': 'client_id'}
|
||||
auth_headers['ETP-Anonymous-ID'] = uuid.uuid4()
|
||||
try:
|
||||
auth_response = self._request_token(auth_headers, data)
|
||||
except ExtractorError as error:
|
||||
username, password = self._get_login_info()
|
||||
if not username or not isinstance(error.cause, HTTPError) or error.cause.status != 400:
|
||||
raise
|
||||
self.to_screen('Refresh token has expired. Re-logging in')
|
||||
CrunchyrollBaseIE._REFRESH_TOKEN = None
|
||||
self.cache.store(self._NETRC_MACHINE, username, None)
|
||||
self._perform_login(username, password)
|
||||
return
|
||||
|
||||
self._set_auth_info(auth_response)
|
||||
|
||||
def _locale_from_language(self, language):
|
||||
config_locale = self._configuration_arg('metadata', ie_key=CrunchyrollBetaIE, casesense=True)
|
||||
|
@ -168,7 +177,8 @@ class CrunchyrollBaseIE(InfoExtractor):
|
|||
self._update_auth()
|
||||
stream_response = self._download_json(
|
||||
f'https://cr-play-service.prd.crunchyrollsvc.com/v1/{identifier}/console/switch/play',
|
||||
display_id, note='Downloading stream info', headers=CrunchyrollBaseIE._AUTH_HEADERS)
|
||||
display_id, note='Downloading stream info', errnote='Failed to download stream info',
|
||||
headers=CrunchyrollBaseIE._AUTH_HEADERS)
|
||||
|
||||
available_formats = {'': ('', '', stream_response['url'])}
|
||||
for hardsub_lang, stream in traverse_obj(stream_response, ('hardSubs', {dict.items}, lambda _, v: v[1]['url'])):
|
||||
|
@ -383,9 +393,9 @@ class CrunchyrollBetaIE(CrunchyrollCmsBaseIE):
|
|||
|
||||
if not self._IS_PREMIUM and traverse_obj(response, (f'{object_type}_metadata', 'is_premium_only')):
|
||||
message = f'This {object_type} is for premium members only'
|
||||
if self.is_logged_in:
|
||||
if CrunchyrollBaseIE._REFRESH_TOKEN:
|
||||
raise ExtractorError(message, expected=True)
|
||||
self.raise_login_required(message)
|
||||
self.raise_login_required(message, method='password')
|
||||
|
||||
result['formats'], result['subtitles'] = self._extract_stream(internal_id)
|
||||
|
||||
|
@ -575,9 +585,9 @@ class CrunchyrollMusicIE(CrunchyrollBaseIE):
|
|||
|
||||
if not self._IS_PREMIUM and response.get('isPremiumOnly'):
|
||||
message = f'This {response.get("type") or "media"} is for premium members only'
|
||||
if self.is_logged_in:
|
||||
if CrunchyrollBaseIE._REFRESH_TOKEN:
|
||||
raise ExtractorError(message, expected=True)
|
||||
self.raise_login_required(message)
|
||||
self.raise_login_required(message, method='password')
|
||||
|
||||
result = self._transform_music_response(response)
|
||||
result['formats'], _ = self._extract_stream(f'music/{internal_id}', internal_id)
|
||||
|
|
|
@ -5,7 +5,13 @@ import math
|
|||
import urllib.parse
|
||||
|
||||
from ._helper import InstanceStoreMixin, select_proxy
|
||||
from .common import Features, Request, register_preference, register_rh
|
||||
from .common import (
|
||||
Features,
|
||||
Request,
|
||||
Response,
|
||||
register_preference,
|
||||
register_rh,
|
||||
)
|
||||
from .exceptions import (
|
||||
CertificateVerifyError,
|
||||
HTTPError,
|
||||
|
@ -14,11 +20,7 @@ from .exceptions import (
|
|||
SSLError,
|
||||
TransportError,
|
||||
)
|
||||
from .impersonate import (
|
||||
ImpersonateRequestHandler,
|
||||
ImpersonateResponse,
|
||||
ImpersonateTarget,
|
||||
)
|
||||
from .impersonate import ImpersonateRequestHandler, ImpersonateTarget
|
||||
from ..dependencies import curl_cffi
|
||||
from ..utils import int_or_none
|
||||
|
||||
|
@ -78,16 +80,15 @@ class CurlCFFIResponseReader(io.IOBase):
|
|||
super().close()
|
||||
|
||||
|
||||
class CurlCFFIResponseAdapter(ImpersonateResponse):
|
||||
class CurlCFFIResponseAdapter(Response):
|
||||
fp: CurlCFFIResponseReader
|
||||
|
||||
def __init__(self, response: curl_cffi.requests.Response, impersonate: ImpersonateTarget):
|
||||
def __init__(self, response: curl_cffi.requests.Response):
|
||||
super().__init__(
|
||||
fp=CurlCFFIResponseReader(response),
|
||||
headers=response.headers,
|
||||
url=response.url,
|
||||
status=response.status_code,
|
||||
impersonate=impersonate)
|
||||
status=response.status_code)
|
||||
|
||||
def read(self, amt=None):
|
||||
try:
|
||||
|
@ -131,6 +132,15 @@ class CurlCFFIRH(ImpersonateRequestHandler, InstanceStoreMixin):
|
|||
extensions.pop('cookiejar', None)
|
||||
extensions.pop('timeout', None)
|
||||
|
||||
def send(self, request: Request) -> Response:
|
||||
try:
|
||||
response = super().send(request)
|
||||
except HTTPError as e:
|
||||
e.response.extensions['impersonate'] = self._get_request_target(request)
|
||||
raise
|
||||
response.extensions['impersonate'] = self._get_request_target(request)
|
||||
return response
|
||||
|
||||
def _send(self, request: Request):
|
||||
max_redirects_exceeded = False
|
||||
session: curl_cffi.requests.Session = self._get_instance(
|
||||
|
@ -177,8 +187,6 @@ class CurlCFFIRH(ImpersonateRequestHandler, InstanceStoreMixin):
|
|||
session.curl.setopt(CurlOpt.LOW_SPEED_LIMIT, 1) # 1 byte per second
|
||||
session.curl.setopt(CurlOpt.LOW_SPEED_TIME, math.ceil(timeout))
|
||||
|
||||
impersonate_target = self._get_request_target(request)
|
||||
|
||||
try:
|
||||
curl_response = session.request(
|
||||
method=request.method,
|
||||
|
@ -188,7 +196,8 @@ class CurlCFFIRH(ImpersonateRequestHandler, InstanceStoreMixin):
|
|||
verify=self.verify,
|
||||
max_redirects=5,
|
||||
timeout=timeout,
|
||||
impersonate=self._SUPPORTED_IMPERSONATE_TARGET_MAP.get(impersonate_target),
|
||||
impersonate=self._SUPPORTED_IMPERSONATE_TARGET_MAP.get(
|
||||
self._get_request_target(request)),
|
||||
interface=self.source_address,
|
||||
stream=True
|
||||
)
|
||||
|
@ -208,7 +217,7 @@ class CurlCFFIRH(ImpersonateRequestHandler, InstanceStoreMixin):
|
|||
else:
|
||||
raise TransportError(cause=e) from e
|
||||
|
||||
response = CurlCFFIResponseAdapter(curl_response, impersonate_target)
|
||||
response = CurlCFFIResponseAdapter(curl_response)
|
||||
|
||||
if not 200 <= response.status < 300:
|
||||
raise HTTPError(response, redirect_loop=max_redirects_exceeded)
|
||||
|
|
|
@ -517,6 +517,7 @@ class Response(io.IOBase):
|
|||
self.reason = reason or HTTPStatus(status).phrase
|
||||
except ValueError:
|
||||
self.reason = None
|
||||
self.extensions = {}
|
||||
|
||||
def readable(self):
|
||||
return self.fp.readable()
|
||||
|
|
|
@ -5,7 +5,7 @@ from abc import ABC
|
|||
from dataclasses import dataclass
|
||||
from typing import Any
|
||||
|
||||
from .common import RequestHandler, Response, register_preference
|
||||
from .common import RequestHandler, register_preference
|
||||
from .exceptions import UnsupportedRequest
|
||||
from ..compat.types import NoneType
|
||||
from ..utils import classproperty, join_nonempty
|
||||
|
@ -58,18 +58,6 @@ class ImpersonateTarget:
|
|||
return cls(**mobj.groupdict())
|
||||
|
||||
|
||||
class ImpersonateResponse(Response):
|
||||
"""
|
||||
Wrapper class for a Response to an impersonated Request.
|
||||
|
||||
Parameters:
|
||||
@param impersonate: the ImpersonateTarget used in the originating Request.
|
||||
"""
|
||||
def __init__(self, *args, impersonate: ImpersonateTarget, **kwargs):
|
||||
super().__init__(*args, **kwargs)
|
||||
self.impersonate = impersonate
|
||||
|
||||
|
||||
class ImpersonateRequestHandler(RequestHandler, ABC):
|
||||
"""
|
||||
Base class for request handlers that support browser impersonation.
|
||||
|
|
Ładowanie…
Reference in New Issue