Merge branch 'yt-dlp:master' into main

.github/workflows/build.yml

@@ -468,7 +468,7 @@ jobs:
       - name: Install Requirements
         run: |
           python devscripts/install_deps.py -o --include build
-          python devscripts/install_deps.py
+          python devscripts/install_deps.py --include curl-cffi
           python -m pip install -U "https://yt-dlp.github.io/Pyinstaller-Builds/i686/pyinstaller-6.7.0-py3-none-any.whl"
 
       - name: Prepare

README.md

@@ -202,7 +202,7 @@ The following provide support for impersonating browser requests. This may be re
 
 * [**curl_cffi**](https://github.com/yifeikong/curl_cffi) (recommended) - Python binding for [curl-impersonate](https://github.com/lwthiker/curl-impersonate). Provides impersonation targets for Chrome, Edge and Safari. Licensed under [MIT](https://github.com/yifeikong/curl_cffi/blob/main/LICENSE)
   * Can be installed with the `curl-cffi` group, e.g. `pip install "yt-dlp[default,curl-cffi]"`
-  * Currently only included in `yt-dlp.exe` and `yt-dlp_macos` builds
+  * Currently included in `yt-dlp.exe`, `yt-dlp_x86.exe`, `yt-dlp_linux` and `yt-dlp_macos` builds
 
 
 ### Metadata

bundle/docker/static/entrypoint.sh

@@ -2,7 +2,7 @@
 set -e
 
 source ~/.local/share/pipx/venvs/pyinstaller/bin/activate
-python -m devscripts.install_deps --include secretstorage
+python -m devscripts.install_deps --include secretstorage --include curl-cffi
 python -m devscripts.make_lazy_extractors
 python devscripts/update-version.py -c "${channel}" -r "${origin}" "${version}"
 python -m bundle.pyinstaller

pyproject.toml

@@ -53,7 +53,9 @@ dependencies = [
 
 [project.optional-dependencies]
 default = []
-curl-cffi = ["curl-cffi==0.5.10; implementation_name=='cpython'"]
+curl-cffi = [
+    "curl-cffi>=0.5.10,!=0.6.*,<0.8; implementation_name=='cpython'",
+]
 secretstorage = [
     "cffi",
     "secretstorage",

test/test_networking.py

@@ -914,7 +914,6 @@ def mock_close(*args, **kwargs):
 class TestCurlCFFIRequestHandler(TestRequestHandlerBase):
 
     @pytest.mark.parametrize('params,extensions', [
-        ({}, {'impersonate': ImpersonateTarget('chrome')}),
         ({'impersonate': ImpersonateTarget('chrome', '110')}, {}),
         ({'impersonate': ImpersonateTarget('chrome', '99')}, {'impersonate': ImpersonateTarget('chrome', '110')}),
     ])

yt_dlp/extractor/box.py

@@ -12,7 +12,7 @@
 
 
 class BoxIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:[^.]+\.)?app\.box\.com/s/(?P<shared_name>[^/?#]+)(?:/file/(?P<id>\d+))?'
+    _VALID_URL = r'https?://(?:[^.]+\.)?(?P<service>app|ent)\.box\.com/s/(?P<shared_name>[^/?#]+)(?:/file/(?P<id>\d+))?'
     _TESTS = [{
         'url': 'https://mlssoccer.app.box.com/s/0evd2o3e08l60lr4ygukepvnkord1o1x/file/510727257538',
         'md5': '1f81b2fd3960f38a40a3b8823e5fcd43',
@@ -38,10 +38,22 @@ class BoxIE(InfoExtractor):
             'uploader_id': '239068974',
         },
         'params': {'skip_download': 'dash fragment too small'},
+    }, {
+        'url': 'https://thejacksonlaboratory.ent.box.com/s/2x09dm6vcg6y28o0oox1so4l0t8wzt6l/file/1536173056065',
+        'info_dict': {
+            'id': '1536173056065',
+            'ext': 'mp4',
+            'uploader_id': '18523128264',
+            'uploader': 'Lexi Hennigan',
+            'title': 'iPSC Symposium recording part 1.mp4',
+            'timestamp': 1716228343,
+            'upload_date': '20240520',
+        },
+        'params': {'skip_download': 'dash fragment too small'},
     }]
 
     def _real_extract(self, url):
-        shared_name, file_id = self._match_valid_url(url).groups()
+        shared_name, file_id, service = self._match_valid_url(url).group('shared_name', 'id', 'service')
         webpage = self._download_webpage(url, file_id or shared_name)
 
         if not file_id:
@@ -57,14 +69,14 @@ def _real_extract(self, url):
         request_token = self._search_json(
             r'Box\.config\s*=', webpage, 'Box config', file_id)['requestToken']
         access_token = self._download_json(
-            'https://app.box.com/app-api/enduserapp/elements/tokens', file_id,
+            f'https://{service}.box.com/app-api/enduserapp/elements/tokens', file_id,
             'Downloading token JSON metadata',
             data=json.dumps({'fileIDs': [file_id]}).encode(), headers={
                 'Content-Type': 'application/json',
                 'X-Request-Token': request_token,
                 'X-Box-EndUser-API': 'sharedName=' + shared_name,
             })[file_id]['read']
-        shared_link = 'https://app.box.com/s/' + shared_name
+        shared_link = f'https://{service}.box.com/s/{shared_name}'
         f = self._download_json(
             'https://api.box.com/2.0/files/' + file_id, file_id,
             'Downloading file JSON metadata', headers={

yt_dlp/extractor/tiktok.py

@@ -1458,9 +1458,11 @@ def _real_extract(self, url):
 
         if webpage:
             data = self._get_sigi_state(webpage, uploader or room_id)
-            room_id = (traverse_obj(data, ('UserModule', 'users', ..., 'roomId', {str_or_none}), get_all=False)
-                       or self._search_regex(r'snssdk\d*://live\?room_id=(\d+)', webpage, 'room ID', default=None)
-                       or room_id)
+            room_id = (
+                traverse_obj(data, ((
+                    ('LiveRoom', 'liveRoomUserInfo', 'user'),
+                    ('UserModule', 'users', ...)), 'roomId', {str}, any))
+                or self._search_regex(r'snssdk\d*://live\?room_id=(\d+)', webpage, 'room ID', default=room_id))
             uploader = uploader or traverse_obj(
                 data, ('LiveRoom', 'liveRoomUserInfo', 'user', 'uniqueId'),
                 ('UserModule', 'users', ..., 'uniqueId'), get_all=False, expected_type=str)

yt_dlp/extractor/tv5mondeplus.py

@@ -96,7 +96,7 @@ def _extract_subtitles(data_captions):
 
     def _real_extract(self, url):
         display_id = self._match_id(url)
-        webpage = self._download_webpage(url, display_id)
+        webpage = self._download_webpage(url, display_id, impersonate=True)
 
         if ">Ce programme n'est malheureusement pas disponible pour votre zone géographique.<" in webpage:
             self.raise_geo_restricted(countries=['FR'])
@@ -122,8 +122,9 @@ def process_video_files(v):
                     if not token:
                         continue
                     deferred_json = self._download_json(
-                        f'https://api.tv5monde.com/player/asset/{d_param}/resolve?condenseKS=true', display_id,
-                        note='Downloading deferred info', headers={'Authorization': f'Bearer {token}'}, fatal=False)
+                        f'https://api.tv5monde.com/player/asset/{d_param}/resolve?condenseKS=true',
+                        display_id, 'Downloading deferred info', fatal=False, impersonate=True,
+                        headers={'Authorization': f'Bearer {token}'})
                     v_url = traverse_obj(deferred_json, (0, 'url', {url_or_none}))
                     if not v_url:
                         continue

yt_dlp/extractor/youtube.py

@@ -3142,7 +3142,7 @@ def _extract_n_function_name(self, jscode):
 
     def _extract_n_function_code(self, video_id, player_url):
         player_id = self._extract_player_info(player_url)
-        func_code = self.cache.load('youtube-nsig', player_id, min_ver='2022.09.1')
+        func_code = self.cache.load('youtube-nsig', player_id, min_ver='2024.07.09')
         jscode = func_code or self._load_player(video_id, player_url)
         jsi = JSInterpreter(jscode)
 

yt_dlp/networking/_curlcffi.py

@@ -2,6 +2,7 @@
 
 import io
 import math
+import re
 import urllib.parse
 
 from ._helper import InstanceStoreMixin, select_proxy
@@ -27,11 +28,12 @@
 if curl_cffi is None:
     raise ImportError('curl_cffi is not installed')
 
-curl_cffi_version = tuple(int_or_none(x, default=0) for x in curl_cffi.__version__.split('.'))
 
-if curl_cffi_version != (0, 5, 10):
+curl_cffi_version = tuple(map(int, re.split(r'[^\d]+', curl_cffi.__version__)[:3]))
+
+if curl_cffi_version != (0, 5, 10) and not ((0, 7, 0) <= curl_cffi_version < (0, 8, 0)):
     curl_cffi._yt_dlp__version = f'{curl_cffi.__version__} (unsupported)'
-    raise ImportError('Only curl_cffi 0.5.10 is supported')
+    raise ImportError('Only curl_cffi versions 0.5.10, 0.7.X are supported')
 
 import curl_cffi.requests
 from curl_cffi.const import CurlECode, CurlOpt
@@ -110,6 +112,13 @@ class CurlCFFIRH(ImpersonateRequestHandler, InstanceStoreMixin):
     _SUPPORTED_FEATURES = (Features.NO_PROXY, Features.ALL_PROXY)
     _SUPPORTED_PROXY_SCHEMES = ('http', 'https', 'socks4', 'socks4a', 'socks5', 'socks5h')
     _SUPPORTED_IMPERSONATE_TARGET_MAP = {
+        **({
+            ImpersonateTarget('chrome', '124', 'macos', '14'): curl_cffi.requests.BrowserType.chrome124,
+            ImpersonateTarget('chrome', '123', 'macos', '14'): curl_cffi.requests.BrowserType.chrome123,
+            ImpersonateTarget('chrome', '120', 'macos', '14'): curl_cffi.requests.BrowserType.chrome120,
+            ImpersonateTarget('chrome', '119', 'macos', '14'): curl_cffi.requests.BrowserType.chrome119,
+            ImpersonateTarget('chrome', '116', 'windows', '10'): curl_cffi.requests.BrowserType.chrome116,
+        } if curl_cffi_version >= (0, 7, 0) else {}),
         ImpersonateTarget('chrome', '110', 'windows', '10'): curl_cffi.requests.BrowserType.chrome110,
         ImpersonateTarget('chrome', '107', 'windows', '10'): curl_cffi.requests.BrowserType.chrome107,
         ImpersonateTarget('chrome', '104', 'windows', '10'): curl_cffi.requests.BrowserType.chrome104,
@@ -118,9 +127,15 @@ class CurlCFFIRH(ImpersonateRequestHandler, InstanceStoreMixin):
         ImpersonateTarget('chrome', '99', 'windows', '10'): curl_cffi.requests.BrowserType.chrome99,
         ImpersonateTarget('edge', '101', 'windows', '10'): curl_cffi.requests.BrowserType.edge101,
         ImpersonateTarget('edge', '99', 'windows', '10'): curl_cffi.requests.BrowserType.edge99,
+        **({
+            ImpersonateTarget('safari', '17.0', 'macos', '14'): curl_cffi.requests.BrowserType.safari17_0,
+        } if curl_cffi_version >= (0, 7, 0) else {}),
         ImpersonateTarget('safari', '15.5', 'macos', '12'): curl_cffi.requests.BrowserType.safari15_5,
         ImpersonateTarget('safari', '15.3', 'macos', '11'): curl_cffi.requests.BrowserType.safari15_3,
         ImpersonateTarget('chrome', '99', 'android', '12'): curl_cffi.requests.BrowserType.chrome99_android,
+        **({
+            ImpersonateTarget('safari', '17.2', 'ios', '17.2'): curl_cffi.requests.BrowserType.safari17_2_ios,
+        } if curl_cffi_version >= (0, 7, 0) else {}),
     }
 
     def _create_instance(self, cookiejar=None):
@@ -187,7 +202,7 @@ def _send(self, request: Request):
         timeout = self._calculate_timeout(request)
 
         # set CURLOPT_LOW_SPEED_LIMIT and CURLOPT_LOW_SPEED_TIME to act as a read timeout. [1]
-        # curl_cffi does not currently do this. [2]
+        # This is required only for 0.5.10 [2]
         # Note: CURLOPT_LOW_SPEED_TIME is in seconds, so we need to round up to the nearest second. [3]
         # [1] https://unix.stackexchange.com/a/305311
         # [2] https://github.com/yifeikong/curl_cffi/issues/156
@@ -203,7 +218,7 @@ def _send(self, request: Request):
                 data=request.data,
                 verify=self.verify,
                 max_redirects=5,
-                timeout=timeout,
+                timeout=(timeout, timeout),
                 impersonate=self._SUPPORTED_IMPERSONATE_TARGET_MAP.get(
                     self._get_request_target(request)),
                 interface=self.source_address,
@@ -222,7 +237,7 @@ def _send(self, request: Request):
 
             elif (
                 e.code == CurlECode.PROXY
-                or (e.code == CurlECode.RECV_ERROR and 'Received HTTP code 407 from proxy after CONNECT' in str(e))
+                or (e.code == CurlECode.RECV_ERROR and 'CONNECT' in str(e))
             ):
                 raise ProxyError(cause=e) from e
             else: