Merge branch 'yt-dlp:master' into pr/live-sections

.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

Changelog.md

@@ -4,6 +4,23 @@
 # To create a release, dispatch the https://github.com/yt-dlp/yt-dlp/actions/workflows/release.yml workflow on master
 -->
 
+### 2024.07.09
+
+#### Core changes
+- [Do not alter default format selection when simulated](https://github.com/yt-dlp/yt-dlp/commit/0b570f2a90ce2363ba06089217514d644e7be2e0) ([#9862](https://github.com/yt-dlp/yt-dlp/issues/9862)) by [seproDev](https://github.com/seproDev)
+
+#### Extractor changes
+- **youtube**: [Remove broken `n` function extraction fallback](https://github.com/yt-dlp/yt-dlp/commit/7ead7332af69422cee931aec3faa277288e9e212) ([#10396](https://github.com/yt-dlp/yt-dlp/issues/10396)) by [pukkandan](https://github.com/pukkandan), [seproDev](https://github.com/seproDev)
+
+### 2024.07.08
+
+#### Core changes
+- **jsinterp**: [Implement `Function.prototype` resolving for `call` and `apply`](https://github.com/yt-dlp/yt-dlp/commit/6c056ea7aeb03660281653a9668547f2548f194f) ([#10392](https://github.com/yt-dlp/yt-dlp/issues/10392)) by [Grub4K](https://github.com/Grub4K)
+
+#### Extractor changes
+- **soundcloud**: [Fix rate-limit handling](https://github.com/yt-dlp/yt-dlp/commit/4b50b292cc98534fb8c7cdf0ae5cb85862f7ebfc) ([#10389](https://github.com/yt-dlp/yt-dlp/issues/10389)) by [bashonly](https://github.com/bashonly)
+- **youtube**: [Fix JS `n` function name extraction](https://github.com/yt-dlp/yt-dlp/commit/297b0a379282a15c80d82d51f3757c961db2dae1) ([#10390](https://github.com/yt-dlp/yt-dlp/issues/10390)) by [bashonly](https://github.com/bashonly), [seproDev](https://github.com/seproDev)
+
 ### 2024.07.07
 
 #### Important changes

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
@@ -2219,6 +2219,7 @@ Some of yt-dlp's default options are different from that of youtube-dl and youtu
 * yt-dlp versions between 2021.11.10 and 2023.06.21 estimated `filesize_approx` values for fragmented/manifest formats. This was added for convenience in [f2fe69](https://github.com/yt-dlp/yt-dlp/commit/f2fe69c7b0d208bdb1f6292b4ae92bc1e1a7444a), but was reverted in [0dff8e](https://github.com/yt-dlp/yt-dlp/commit/0dff8e4d1e6e9fb938f4256ea9af7d81f42fd54f) due to the potentially extreme inaccuracy of the estimated values. Use `--compat-options manifest-filesize-approx` to keep extracting the estimated values
 * yt-dlp uses modern http client backends such as `requests`. Use `--compat-options prefer-legacy-http-handler` to prefer the legacy http handler (`urllib`) to be used for standard http requests.
 * The sub-modules `swfinterp`, `casefold` are removed.
+* Passing `--simulate` (or calling `extract_info` with `download=False`) no longer alters the default format selection. See [#9843](https://github.com/yt-dlp/yt-dlp/issues/9843) for details.
 
 For ease of use, a few more compat options are available:
 

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_YoutubeDL.py

@@ -4,6 +4,7 @@
 import os
 import sys
 import unittest
+from unittest.mock import patch
 
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
@@ -520,7 +521,33 @@ class TestFormatSelection(unittest.TestCase):
             ydl.process_ie_result(info_dict)
         self.assertEqual(ydl.downloaded_info_dicts, [])
 
-    def test_default_format_spec(self):
+    @patch('yt_dlp.postprocessor.ffmpeg.FFmpegMergerPP.available', False)
+    def test_default_format_spec_without_ffmpeg(self):
+        ydl = YDL({})
+        self.assertEqual(ydl._default_format_spec({}), 'best/bestvideo+bestaudio')
+
+        ydl = YDL({'simulate': True})
+        self.assertEqual(ydl._default_format_spec({}), 'best/bestvideo+bestaudio')
+
+        ydl = YDL({})
+        self.assertEqual(ydl._default_format_spec({'is_live': True}), 'best/bestvideo+bestaudio')
+
+        ydl = YDL({'simulate': True})
+        self.assertEqual(ydl._default_format_spec({'is_live': True}), 'best/bestvideo+bestaudio')
+
+        ydl = YDL({'outtmpl': '-'})
+        self.assertEqual(ydl._default_format_spec({}), 'best/bestvideo+bestaudio')
+
+        ydl = YDL({})
+        self.assertEqual(ydl._default_format_spec({}), 'best/bestvideo+bestaudio')
+        self.assertEqual(ydl._default_format_spec({'is_live': True}), 'best/bestvideo+bestaudio')
+
+    @patch('yt_dlp.postprocessor.ffmpeg.FFmpegMergerPP.available', True)
+    @patch('yt_dlp.postprocessor.ffmpeg.FFmpegMergerPP.can_merge', lambda _: True)
+    def test_default_format_spec_with_ffmpeg(self):
+        ydl = YDL({})
+        self.assertEqual(ydl._default_format_spec({}), 'bestvideo*+bestaudio/best')
+
         ydl = YDL({'simulate': True})
         self.assertEqual(ydl._default_format_spec({}), 'bestvideo*+bestaudio/best')
 
@@ -528,13 +555,13 @@ class TestFormatSelection(unittest.TestCase):
         self.assertEqual(ydl._default_format_spec({'is_live': True}), 'best/bestvideo+bestaudio')
 
         ydl = YDL({'simulate': True})
-        self.assertEqual(ydl._default_format_spec({'is_live': True}), 'bestvideo*+bestaudio/best')
+        self.assertEqual(ydl._default_format_spec({'is_live': True}), 'best/bestvideo+bestaudio')
 
         ydl = YDL({'outtmpl': '-'})
         self.assertEqual(ydl._default_format_spec({}), 'best/bestvideo+bestaudio')
 
         ydl = YDL({})
-        self.assertEqual(ydl._default_format_spec({}, download=False), 'bestvideo*+bestaudio/best')
+        self.assertEqual(ydl._default_format_spec({}), 'bestvideo*+bestaudio/best')
         self.assertEqual(ydl._default_format_spec({'is_live': True}), 'best/bestvideo+bestaudio')
 
 

test/test_jsinterp.py

@@ -376,6 +376,33 @@ class TestJSInterpreter(unittest.TestCase):
         jsi = JSInterpreter('''function f(p,a,c,k,e,d){while(c--)if(k[c])p=p.replace(new RegExp('\\b'+c.toString(a)+'\\b','g'),k[c]);return p}''')
         self.assertEqual(jsi.call_function('f', '''h 7=g("1j");7.7h({7g:[{33:"w://7f-7e-7d-7c.v.7b/7a/79/78/77/76.74?t=73&s=2s&e=72&f=2t&71=70.0.0.1&6z=6y&6x=6w"}],6v:"w://32.v.u/6u.31",16:"r%",15:"r%",6t:"6s",6r:"",6q:"l",6p:"l",6o:"6n",6m:\'6l\',6k:"6j",9:[{33:"/2u?b=6i&n=50&6h=w://32.v.u/6g.31",6f:"6e"}],1y:{6d:1,6c:\'#6b\',6a:\'#69\',68:"67",66:30,65:r,},"64":{63:"%62 2m%m%61%5z%5y%5x.u%5w%5v%5u.2y%22 2k%m%1o%22 5t%m%1o%22 5s%m%1o%22 2j%m%5r%22 16%m%5q%22 15%m%5p%22 5o%2z%5n%5m%2z",5l:"w://v.u/d/1k/5k.2y",5j:[]},\'5i\':{"5h":"5g"},5f:"5e",5d:"w://v.u",5c:{},5b:l,1x:[0.25,0.50,0.75,1,1.25,1.5,2]});h 1m,1n,5a;h 59=0,58=0;h 7=g("1j");h 2x=0,57=0,56=0;$.55({54:{\'53-52\':\'2i-51\'}});7.j(\'4z\',6(x){c(5>0&&x.1l>=5&&1n!=1){1n=1;$(\'q.4y\').4x(\'4w\')}});7.j(\'13\',6(x){2x=x.1l});7.j(\'2g\',6(x){2w(x)});7.j(\'4v\',6(){$(\'q.2v\').4u()});6 2w(x){$(\'q.2v\').4t();c(1m)19;1m=1;17=0;c(4s.4r===l){17=1}$.4q(\'/2u?b=4p&2l=1k&4o=2t-4n-4m-2s-4l&4k=&4j=&4i=&17=\'+17,6(2r){$(\'#4h\').4g(2r)});$(\'.3-8-4f-4e:4d("4c")\').2h(6(e){2q();g().4b(0);g().4a(l)});6 2q(){h $14=$("<q />").2p({1l:"49",16:"r%",15:"r%",48:0,2n:0,2o:47,46:"45(10%, 10%, 10%, 0.4)","44-43":"42"});$("<41 />").2p({16:"60%",15:"60%",2o:40,"3z-2n":"3y"}).3x({\'2m\':\'/?b=3w&2l=1k\',\'2k\':\'0\',\'2j\':\'2i\'}).2f($14);$14.2h(6(){$(3v).3u();g().2g()});$14.2f($(\'#1j\'))}g().13(0);}6 3t(){h 9=7.1b(2e);2d.2c(9);c(9.n>1){1r(i=0;i<9.n;i++){c(9[i].1a==2e){2d.2c(\'!!=\'+i);7.1p(i)}}}}7.j(\'3s\',6(){g().1h("/2a/3r.29","3q 10 28",6(){g().13(g().27()+10)},"2b");$("q[26=2b]").23().21(\'.3-20-1z\');g().1h("/2a/3p.29","3o 10 28",6(){h 12=g().27()-10;c(12<0)12=0;g().13(12)},"24");$("q[26=24]").23().21(\'.3-20-1z\');});6 1i(){}7.j(\'3n\',6(){1i()});7.j(\'3m\',6(){1i()});7.j("k",6(y){h 9=7.1b();c(9.n<2)19;$(\'.3-8-3l-3k\').3j(6(){$(\'#3-8-a-k\').1e(\'3-8-a-z\');$(\'.3-a-k\').p(\'o-1f\',\'11\')});7.1h("/3i/3h.3g","3f 3e",6(){$(\'.3-1w\').3d(\'3-8-1v\');$(\'.3-8-1y, .3-8-1x\').p(\'o-1g\',\'11\');c($(\'.3-1w\').3c(\'3-8-1v\')){$(\'.3-a-k\').p(\'o-1g\',\'l\');$(\'.3-a-k\').p(\'o-1f\',\'l\');$(\'.3-8-a\').1e(\'3-8-a-z\');$(\'.3-8-a:1u\').3b(\'3-8-a-z\')}3a{$(\'.3-a-k\').p(\'o-1g\',\'11\');$(\'.3-a-k\').p(\'o-1f\',\'11\');$(\'.3-8-a:1u\').1e(\'3-8-a-z\')}},"39");7.j("38",6(y){1d.37(\'1c\',y.9[y.36].1a)});c(1d.1t(\'1c\')){35("1s(1d.1t(\'1c\'));",34)}});h 18;6 1s(1q){h 9=7.1b();c(9.n>1){1r(i=0;i<9.n;i++){c(9[i].1a==1q){c(i==18){19}18=i;7.1p(i)}}}}',36,270,'|||jw|||function|player|settings|tracks|submenu||if||||jwplayer|var||on|audioTracks|true|3D|length|aria|attr|div|100|||sx|filemoon|https||event|active||false|tt|seek|dd|height|width|adb|current_audio|return|name|getAudioTracks|default_audio|localStorage|removeClass|expanded|checked|addButton|callMeMaybe|vplayer|0fxcyc2ajhp1|position|vvplay|vvad|220|setCurrentAudioTrack|audio_name|for|audio_set|getItem|last|open|controls|playbackRates|captions|rewind|icon|insertAfter||detach|ff00||button|getPosition|sec|png|player8|ff11|log|console|track_name|appendTo|play|click|no|scrolling|frameborder|file_code|src|top|zIndex|css|showCCform|data|1662367683|383371|dl|video_ad|doPlay|prevt|mp4|3E||jpg|thumbs|file|300|setTimeout|currentTrack|setItem|audioTrackChanged|dualSound|else|addClass|hasClass|toggleClass|Track|Audio|svg|dualy|images|mousedown|buttons|topbar|playAttemptFailed|beforePlay|Rewind|fr|Forward|ff|ready|set_audio_track|remove|this|upload_srt|prop|50px|margin|1000001|iframe|center|align|text|rgba|background|1000000|left|absolute|pause|setCurrentCaptions|Upload|contains|item|content|html|fviews|referer|prem|embed|3e57249ef633e0d03bf76ceb8d8a4b65|216|83|hash|view|get|TokenZir|window|hide|show|complete|slow|fadeIn|video_ad_fadein|time||cache|Cache|Content|headers|ajaxSetup|v2done|tott|vastdone2|vastdone1|vvbefore|playbackRateControls|cast|aboutlink|FileMoon|abouttext|UHD|1870|qualityLabels|sites|GNOME_POWER|link|2Fiframe|3C|allowfullscreen|22360|22640|22no|marginheight|marginwidth|2FGNOME_POWER|2F0fxcyc2ajhp1|2Fe|2Ffilemoon|2F|3A||22https|3Ciframe|code|sharing|fontOpacity|backgroundOpacity|Tahoma|fontFamily|303030|backgroundColor|FFFFFF|color|userFontScale|thumbnails|kind|0fxcyc2ajhp10000|url|get_slides|start|startparam|none|preload|html5|primary|hlshtml|androidhls|duration|uniform|stretching|0fxcyc2ajhp1_xt|image|2048|sp|6871|asn|127|srv|43200|_g3XlBcu2lmD9oDexD2NLWSmah2Nu3XcDrl93m9PwXY|m3u8||master|0fxcyc2ajhp1_x|00076|01|hls2|to|s01|delivery|storage|moon|sources|setup'''.split('|')))
 
+    def test_join(self):
+        test_input = list('test')
+        tests = [
+            'function f(a, b){return a.join(b)}',
+            'function f(a, b){return Array.prototype.join.call(a, b)}',
+            'function f(a, b){return Array.prototype.join.apply(a, [b])}',
+        ]
+        for test in tests:
+            jsi = JSInterpreter(test)
+            self._test(jsi, 'test', args=[test_input, ''])
+            self._test(jsi, 't-e-s-t', args=[test_input, '-'])
+            self._test(jsi, '', args=[[], '-'])
+
+    def test_split(self):
+        test_result = list('test')
+        tests = [
+            'function f(a, b){return a.split(b)}',
+            'function f(a, b){return String.prototype.split.call(a, b)}',
+            'function f(a, b){return String.prototype.split.apply(a, [b])}',
+        ]
+        for test in tests:
+            jsi = JSInterpreter(test)
+            self._test(jsi, test_result, args=['test', ''])
+            self._test(jsi, test_result, args=['t-e-s-t', '-'])
+            self._test(jsi, [''], args=['', '-'])
+            self._test(jsi, [], args=['', ''])
+
 
 if __name__ == '__main__':
     unittest.main()

test/test_networking.py

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

test/test_youtube_signature.py

@@ -167,6 +167,10 @@ _NSIG_TESTS = [
         'https://www.youtube.com/s/player/590f65a6/player_ias.vflset/en_US/base.js',
         '1tm7-g_A9zsI8_Lay_', 'xI4Vem4Put_rOg',
     ),
+    (
+        'https://www.youtube.com/s/player/b22ef6e7/player_ias.vflset/en_US/base.js',
+        'b6HcntHGkvBLk_FRf', 'kNPW6A7FyP2l8A',
+    ),
 ]
 
 

yt_dlp/YoutubeDL.py

@@ -2195,9 +2195,8 @@ class YoutubeDL:
                                    or all(f.get('acodec') == 'none' for f in formats)),  # OR, No formats with audio
         }))
 
-    def _default_format_spec(self, info_dict, download=True):
-        download = download and not self.params.get('simulate')
-        prefer_best = download and (
+    def _default_format_spec(self, info_dict):
+        prefer_best = (
             self.params['outtmpl']['default'] == '-'
             or info_dict.get('is_live') and not self.params.get('live_from_start'))
 
@@ -2205,7 +2204,7 @@ class YoutubeDL:
             merger = FFmpegMergerPP(self)
             return merger.available and merger.can_merge()
 
-        if not prefer_best and download and not can_merge():
+        if not prefer_best and not can_merge():
             prefer_best = True
             formats = self._get_formats(info_dict)
             evaluate_formats = lambda spec: self._select_formats(formats, self.build_format_selector(spec))
@@ -2964,7 +2963,7 @@ class YoutubeDL:
                     continue
 
             if format_selector is None:
-                req_format = self._default_format_spec(info_dict, download=download)
+                req_format = self._default_format_spec(info_dict)
                 self.write_debug(f'Default format spec: {req_format}')
                 format_selector = self.build_format_selector(req_format)
 

yt_dlp/extractor/box.py

@@ -12,7 +12,7 @@ from ..utils.traversal import traverse_obj
 
 
 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 @@ class BoxIE(InfoExtractor):
         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/soundcloud.py

@@ -314,23 +314,11 @@ class SoundcloudBaseIE(InfoExtractor):
                 self.write_debug(f'"{identifier}" is not a requested format, skipping')
                 continue
 
-            stream = None
-            for retry in self.RetryManager(fatal=False):
-                try:
-                    stream = self._call_api(
-                        format_url, track_id, f'Downloading {identifier} format info JSON',
-                        query=query, headers=self._HEADERS)
-                except ExtractorError as e:
-                    if isinstance(e.cause, HTTPError) and e.cause.status == 429:
-                        self.report_warning(
-                            'You have reached the API rate limit, which is ~600 requests per '
-                            '10 minutes. Use the --extractor-retries and --retry-sleep options '
-                            'to configure an appropriate retry count and wait time', only_once=True)
-                        retry.error = e.cause
-                    else:
-                        self.report_warning(e.msg)
-
-            stream_url = traverse_obj(stream, ('url', {url_or_none}))
+            # XXX: if not extract_flat, 429 error must be caught where _extract_info_dict is called
+            stream_url = traverse_obj(self._call_api(
+                format_url, track_id, f'Downloading {identifier} format info JSON',
+                query=query, headers=self._HEADERS), ('url', {url_or_none}))
+
             if invalid_url(stream_url):
                 continue
             format_urls.add(stream_url)
@@ -647,7 +635,17 @@ class SoundcloudIE(SoundcloudBaseIE):
         info = self._call_api(
             info_json_url, full_title, 'Downloading info JSON', query=query, headers=self._HEADERS)
 
-        return self._extract_info_dict(info, full_title, token)
+        for retry in self.RetryManager():
+            try:
+                return self._extract_info_dict(info, full_title, token)
+            except ExtractorError as e:
+                if not isinstance(e.cause, HTTPError) or not e.cause.status == 429:
+                    raise
+                self.report_warning(
+                    'You have reached the API rate limit, which is ~600 requests per '
+                    '10 minutes. Use the --extractor-retries and --retry-sleep options '
+                    'to configure an appropriate retry count and wait time', only_once=True)
+                retry.error = e.cause
 
 
 class SoundcloudPlaylistBaseIE(SoundcloudBaseIE):

yt_dlp/extractor/tiktok.py

@@ -1458,9 +1458,11 @@ class TikTokLiveIE(TikTokBaseIE):
 
         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 @@ class TV5MondePlusIE(InfoExtractor):
 
     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'])

yt_dlp/extractor/youtube.py

@@ -3159,7 +3159,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
 
     def _extract_n_function_name(self, jscode):
         funcname, idx = self._search_regex(
-            r'\.get\("n"\)\)&&\(b=(?P<nfunc>[a-zA-Z0-9$]+)(?:\[(?P<idx>\d+)\])?\([a-zA-Z0-9]\)',
+            r'''(?x)(?:\.get\("n"\)\)&&\(b=|b=String\.fromCharCode\(110\),c=a\.get\(b\)\)&&\(c=)
+            (?P<nfunc>[a-zA-Z0-9$]+)(?:\[(?P<idx>\d+)\])?\([a-zA-Z0-9]\)''',
             jscode, 'Initial JS player n function name', group=('nfunc', 'idx'))
         if not idx:
             return funcname
@@ -3170,7 +3171,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
 
     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)
 
@@ -3179,17 +3180,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
 
         func_name = self._extract_n_function_name(jscode)
 
-        # For redundancy
-        func_code = self._search_regex(
-            rf'''(?xs){func_name}\s*=\s*function\s*\((?P<var>[\w$]+)\)\s*
-                     # NB: The end of the regex is intentionally kept strict
-                     {{(?P<code>.+?}}\s*return\ [\w$]+.join\(""\))}};''',
-            jscode, 'nsig function', group=('var', 'code'), default=None)
-        if func_code:
-            func_code = ([func_code[0]], func_code[1])
-        else:
-            self.write_debug('Extracting nsig function with jsinterp')
-            func_code = jsi.extract_function_code(func_name)
+        func_code = jsi.extract_function_code(func_name)
 
         self.cache.store('youtube-nsig', player_id, func_code)
         return jsi, player_id, func_code

yt_dlp/jsinterp.py

@@ -636,6 +636,8 @@ class JSInterpreter:
                     raise self.Exception(f'{member} {msg}', expr)
 
             def eval_method():
+                nonlocal member
+
                 if (variable, member) == ('console', 'debug'):
                     if Debugger.ENABLED:
                         Debugger.write(self.interpret_expression(f'[{arg_str}]', local_vars, allow_recursion))
@@ -644,6 +646,7 @@ class JSInterpreter:
                 types = {
                     'String': str,
                     'Math': float,
+                    'Array': list,
                 }
                 obj = local_vars.get(variable, types.get(variable, NO_DEFAULT))
                 if obj is NO_DEFAULT:
@@ -667,6 +670,21 @@ class JSInterpreter:
                     self.interpret_expression(v, local_vars, allow_recursion)
                     for v in self._separate(arg_str)]
 
+                # Fixup prototype call
+                if isinstance(obj, type) and member.startswith('prototype.'):
+                    new_member, _, func_prototype = member.partition('.')[2].partition('.')
+                    assertion(argvals, 'takes one or more arguments')
+                    assertion(isinstance(argvals[0], obj), f'needs binding to type {obj}')
+                    if func_prototype == 'call':
+                        obj, *argvals = argvals
+                    elif func_prototype == 'apply':
+                        assertion(len(argvals) == 2, 'takes two arguments')
+                        obj, argvals = argvals
+                        assertion(isinstance(argvals, list), 'second argument needs to be a list')
+                    else:
+                        raise self.Exception(f'Unsupported Function method {func_prototype}', expr)
+                    member = new_member
+
                 if obj is str:
                     if member == 'fromCharCode':
                         assertion(argvals, 'takes one or more arguments')

yt_dlp/networking/_curlcffi.py

@@ -2,6 +2,7 @@ from __future__ import annotations
 
 import io
 import math
+import re
 import urllib.parse
 
 from ._helper import InstanceStoreMixin, select_proxy
@@ -27,11 +28,12 @@ from ..utils import int_or_none
 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 @@ class CurlCFFIRH(ImpersonateRequestHandler, InstanceStoreMixin):
         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 @@ class CurlCFFIRH(ImpersonateRequestHandler, InstanceStoreMixin):
                 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 @@ class CurlCFFIRH(ImpersonateRequestHandler, InstanceStoreMixin):
 
             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:

yt_dlp/version.py

@@ -1,8 +1,8 @@
 # Autogenerated by devscripts/update-version.py
 
-__version__ = '2024.07.07'
+__version__ = '2024.07.09'
 
-RELEASE_GIT_HEAD = 'b337d2989ce0614651d363383f6f743d977248ef'
+RELEASE_GIT_HEAD = '7ead7332af69422cee931aec3faa277288e9e212'
 
 VARIANT = None
 
@@ -12,4 +12,4 @@ CHANNEL = 'stable'
 
 ORIGIN = 'yt-dlp/yt-dlp'
 
-_pkg_version = '2024.07.07'
+_pkg_version = '2024.07.09'