Raise minimum recommended Python version to 3.10 (#13859)

Ref: https://github.com/yt-dlp/yt-dlp/issues/13858

Authored by: bashonly

.github/workflows/build.yml

@@ -242,7 +242,7 @@ jobs:
     permissions:
       contents: read
       actions: write  # For cleaning up cache
-    runs-on: macos-13
+    runs-on: macos-14
 
     steps:
       - uses: actions/checkout@v4
@@ -261,6 +261,8 @@ jobs:
       - name: Install Requirements
         run: |
           brew install coreutils
+          # We need to use system Python in order to roll our own universal2 curl_cffi wheel
+          brew uninstall --ignore-dependencies python3
           python3 -m venv ~/yt-dlp-build-venv
           source ~/yt-dlp-build-venv/bin/activate
           python3 devscripts/install_deps.py -o --include build

bundle/pyinstaller.py

@@ -62,16 +62,22 @@ def parse_options():
 
 def exe(onedir):
     """@returns (name, path)"""
+    platform_name, machine, extension = {
+        'win32': (None, MACHINE, '.exe'),
+        'darwin': ('macos', None, None),
+    }.get(OS_NAME, (OS_NAME, MACHINE, None))
+
     name = '_'.join(filter(None, (
         'yt-dlp',
-        {'win32': '', 'darwin': 'macos'}.get(OS_NAME, OS_NAME),
-        MACHINE,
+        platform_name,
+        machine,
     )))
+
     return name, ''.join(filter(None, (
         'dist/',
         onedir and f'{name}/',
         name,
-        OS_NAME == 'win32' and '.exe',
+        extension,
     )))
 
 

yt_dlp/downloader/hls.py

@@ -205,7 +205,7 @@ class HlsFD(FragmentFD):
             line = line.strip()
             if line:
                 if not line.startswith('#'):
-                    if format_index and discontinuity_count != format_index:
+                    if format_index is not None and discontinuity_count != format_index:
                         continue
                     if ad_frag_next:
                         continue
@@ -231,7 +231,7 @@ class HlsFD(FragmentFD):
                         byte_range = {}
 
                 elif line.startswith('#EXT-X-MAP'):
-                    if format_index and discontinuity_count != format_index:
+                    if format_index is not None and discontinuity_count != format_index:
                         continue
                     if frag_index > 0:
                         self.report_error(

yt_dlp/extractor/_extractors.py

@@ -571,10 +571,6 @@ from .dw import (
     DWIE,
     DWArticleIE,
 )
-from .eagleplatform import (
-    ClipYouEmbedIE,
-    EaglePlatformIE,
-)
 from .ebaumsworld import EbaumsWorldIE
 from .ebay import EbayIE
 from .egghead import (

yt_dlp/extractor/eagleplatform.py

@@ -1,215 +0,0 @@
-import functools
-import re
-
-from .common import InfoExtractor
-from ..networking.exceptions import HTTPError
-from ..utils import (
-    ExtractorError,
-    int_or_none,
-    smuggle_url,
-    unsmuggle_url,
-    url_or_none,
-)
-
-
-class EaglePlatformIE(InfoExtractor):
-    _VALID_URL = r'''(?x)
-                    (?:
-                        eagleplatform:(?P<custom_host>[^/]+):|
-                        https?://(?P<host>.+?\.media\.eagleplatform\.com)/index/player\?.*\brecord_id=
-                    )
-                    (?P<id>\d+)
-                '''
-    _EMBED_REGEX = [r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//.+?\.media\.eagleplatform\.com/index/player\?.+?)\1']
-    _TESTS = [{
-        # http://lenta.ru/news/2015/03/06/navalny/
-        'url': 'http://lentaru.media.eagleplatform.com/index/player?player=new&record_id=227304&player_template_id=5201',
-        # Not checking MD5 as sometimes the direct HTTP link results in 404 and HLS is used
-        'info_dict': {
-            'id': '227304',
-            'ext': 'mp4',
-            'title': 'Навальный вышел на свободу',
-            'description': 'md5:d97861ac9ae77377f3f20eaf9d04b4f5',
-            'thumbnail': r're:^https?://.*\.jpg$',
-            'duration': 87,
-            'view_count': int,
-            'age_limit': 0,
-        },
-    }, {
-        # http://muz-tv.ru/play/7129/
-        # http://media.clipyou.ru/index/player?record_id=12820&width=730&height=415&autoplay=true
-        'url': 'eagleplatform:media.clipyou.ru:12820',
-        'md5': '358597369cf8ba56675c1df15e7af624',
-        'info_dict': {
-            'id': '12820',
-            'ext': 'mp4',
-            'title': "'O Sole Mio",
-            'thumbnail': r're:^https?://.*\.jpg$',
-            'duration': 216,
-            'view_count': int,
-        },
-        'skip': 'Georestricted',
-    }, {
-        # referrer protected video (https://tvrain.ru/lite/teleshow/kak_vse_nachinalos/namin-418921/)
-        'url': 'eagleplatform:tvrainru.media.eagleplatform.com:582306',
-        'only_matching': True,
-    }]
-
-    @classmethod
-    def _extract_embed_urls(cls, url, webpage):
-        add_referer = functools.partial(smuggle_url, data={'referrer': url})
-
-        res = tuple(super()._extract_embed_urls(url, webpage))
-        if res:
-            return map(add_referer, res)
-
-        PLAYER_JS_RE = r'''
-                        <script[^>]+
-                            src=(?P<qjs>["\'])(?:https?:)?//(?P<host>(?:(?!(?P=qjs)).)+\.media\.eagleplatform\.com)/player/player\.js(?P=qjs)
-                        .+?
-                    '''
-        # "Basic usage" embedding (see http://dultonmedia.github.io/eplayer/)
-        mobj = re.search(
-            rf'''(?xs)
-                    {PLAYER_JS_RE}
-                    <div[^>]+
-                        class=(?P<qclass>["\'])eagleplayer(?P=qclass)[^>]+
-                        data-id=["\'](?P<id>\d+)
-            ''', webpage)
-        if mobj is not None:
-            return [add_referer('eagleplatform:{host}:{id}'.format(**mobj.groupdict()))]
-        # Generalization of "Javascript code usage", "Combined usage" and
-        # "Usage without attaching to DOM" embeddings (see
-        # http://dultonmedia.github.io/eplayer/)
-        mobj = re.search(
-            r'''(?xs)
-                    %s
-                    <script>
-                    .+?
-                    new\s+EaglePlayer\(
-                        (?:[^,]+\s*,\s*)?
-                        {
-                            .+?
-                            \bid\s*:\s*["\']?(?P<id>\d+)
-                            .+?
-                        }
-                    \s*\)
-                    .+?
-                    </script>
-            ''' % PLAYER_JS_RE, webpage)  # noqa: UP031
-        if mobj is not None:
-            return [add_referer('eagleplatform:{host}:{id}'.format(**mobj.groupdict()))]
-
-    @staticmethod
-    def _handle_error(response):
-        status = int_or_none(response.get('status', 200))
-        if status != 200:
-            raise ExtractorError(' '.join(response['errors']), expected=True)
-
-    def _download_json(self, url_or_request, video_id, *args, **kwargs):
-        try:
-            response = super()._download_json(
-                url_or_request, video_id, *args, **kwargs)
-        except ExtractorError as ee:
-            if isinstance(ee.cause, HTTPError):
-                response = self._parse_json(ee.cause.response.read().decode('utf-8'), video_id)
-                self._handle_error(response)
-            raise
-        return response
-
-    def _get_video_url(self, url_or_request, video_id, note='Downloading JSON metadata'):
-        return self._download_json(url_or_request, video_id, note)['data'][0]
-
-    def _real_extract(self, url):
-        url, smuggled_data = unsmuggle_url(url, {})
-
-        mobj = self._match_valid_url(url)
-        host, video_id = mobj.group('custom_host') or mobj.group('host'), mobj.group('id')
-
-        headers = {}
-        query = {
-            'id': video_id,
-        }
-
-        referrer = smuggled_data.get('referrer')
-        if referrer:
-            headers['Referer'] = referrer
-            query['referrer'] = referrer
-
-        player_data = self._download_json(
-            f'http://{host}/api/player_data', video_id,
-            headers=headers, query=query)
-
-        media = player_data['data']['playlist']['viewports'][0]['medialist'][0]
-
-        title = media['title']
-        description = media.get('description')
-        thumbnail = self._proto_relative_url(media.get('snapshot'), 'http:')
-        duration = int_or_none(media.get('duration'))
-        view_count = int_or_none(media.get('views'))
-
-        age_restriction = media.get('age_restriction')
-        age_limit = None
-        if age_restriction:
-            age_limit = 0 if age_restriction == 'allow_all' else 18
-
-        secure_m3u8 = self._proto_relative_url(media['sources']['secure_m3u8']['auto'], 'http:')
-
-        formats = []
-
-        m3u8_url = self._get_video_url(secure_m3u8, video_id, 'Downloading m3u8 JSON')
-        m3u8_formats = self._extract_m3u8_formats(
-            m3u8_url, video_id, 'mp4', entry_protocol='m3u8_native',
-            m3u8_id='hls', fatal=False)
-        formats.extend(m3u8_formats)
-
-        m3u8_formats_dict = {}
-        for f in m3u8_formats:
-            if f.get('height') is not None:
-                m3u8_formats_dict[f['height']] = f
-
-        mp4_data = self._download_json(
-            # Secure mp4 URL is constructed according to Player.prototype.mp4 from
-            # http://lentaru.media.eagleplatform.com/player/player.js
-            re.sub(r'm3u8|hlsvod|hls|f4m', 'mp4s', secure_m3u8),
-            video_id, 'Downloading mp4 JSON', fatal=False)
-        if mp4_data:
-            for format_id, format_url in mp4_data.get('data', {}).items():
-                if not url_or_none(format_url):
-                    continue
-                height = int_or_none(format_id)
-                if height is not None and m3u8_formats_dict.get(height):
-                    f = m3u8_formats_dict[height].copy()
-                    f.update({
-                        'format_id': f['format_id'].replace('hls', 'http'),
-                        'protocol': 'http',
-                    })
-                else:
-                    f = {
-                        'format_id': f'http-{format_id}',
-                        'height': int_or_none(format_id),
-                    }
-                f['url'] = format_url
-                formats.append(f)
-
-        return {
-            'id': video_id,
-            'title': title,
-            'description': description,
-            'thumbnail': thumbnail,
-            'duration': duration,
-            'view_count': view_count,
-            'age_limit': age_limit,
-            'formats': formats,
-        }
-
-
-class ClipYouEmbedIE(InfoExtractor):
-    _VALID_URL = False
-
-    @classmethod
-    def _extract_embed_urls(cls, url, webpage):
-        mobj = re.search(
-            r'<iframe[^>]+src="https?://(?P<host>media\.clipyou\.ru)/index/player\?.*\brecord_id=(?P<id>\d+).*"', webpage)
-        if mobj is not None:
-            yield smuggle_url('eagleplatform:{host}:{id}'.format(**mobj.groupdict()), {'referrer': url})

yt_dlp/extractor/faulio.py

@@ -9,6 +9,7 @@ from ..utils.traversal import traverse_obj
 class FaulioLiveIE(InfoExtractor):
     _DOMAINS = (
         'aloula.sba.sa',
+        'bahry.com',
         'maraya.sba.net.ae',
         'sat7plus.org',
     )
@@ -25,6 +26,18 @@ class FaulioLiveIE(InfoExtractor):
         'params': {
             'skip_download': 'Livestream',
         },
+    }, {
+        'url': 'https://bahry.com/live/1',
+        'info_dict': {
+            'id': 'bahry.faulio.com_1',
+            'title': str,
+            'description': str,
+            'ext': 'mp4',
+            'live_status': 'is_live',
+        },
+        'params': {
+            'skip_download': 'Livestream',
+        },
     }, {
         'url': 'https://maraya.sba.net.ae/live/1',
         'info_dict': {

yt_dlp/extractor/generic.py

@@ -1010,38 +1010,6 @@ class GenericIE(InfoExtractor):
             },
             'add_ie': ['Kaltura'],
         },
-        # referrer protected EaglePlatform embed
-        {
-            'url': 'https://tvrain.ru/lite/teleshow/kak_vse_nachinalos/namin-418921/',
-            'info_dict': {
-                'id': '582306',
-                'ext': 'mp4',
-                'title': 'Стас Намин: «Мы нарушили девственность Кремля»',
-                'thumbnail': r're:^https?://.*\.jpg$',
-                'duration': 3382,
-                'view_count': int,
-            },
-            'params': {
-                'skip_download': True,
-            },
-        },
-        # ClipYou (EaglePlatform) embed (custom URL)
-        {
-            'url': 'http://muz-tv.ru/play/7129/',
-            # Not checking MD5 as sometimes the direct HTTP link results in 404 and HLS is used
-            'info_dict': {
-                'id': '12820',
-                'ext': 'mp4',
-                'title': "'O Sole Mio",
-                'thumbnail': r're:^https?://.*\.jpg$',
-                'duration': 216,
-                'view_count': int,
-            },
-            'params': {
-                'skip_download': True,
-            },
-            'skip': 'This video is unavailable.',
-        },
         # Pladform embed
         {
             'url': 'http://muz-tv.ru/kinozal/view/7400/',

yt_dlp/extractor/livejournal.py

@@ -3,6 +3,7 @@ from ..utils import int_or_none
 
 
 class LiveJournalIE(InfoExtractor):
+    _WORKING = False
     _VALID_URL = r'https?://(?:[^.]+\.)?livejournal\.com/video/album/\d+.+?\bid=(?P<id>\d+)'
     _TEST = {
         'url': 'https://andrei-bt.livejournal.com/video/album/407/?mode=view&id=51272',

yt_dlp/extractor/neteasemusic.py

@@ -34,7 +34,6 @@ class NetEaseMusicBaseIE(InfoExtractor):
         'sky',       # SVIP tier; 沉浸环绕声 (Surround Audio); flac
     )
     _API_BASE = 'http://music.163.com/api/'
-    _GEO_BYPASS = False
 
     def _create_eapi_cipher(self, api_path, query_body, cookies):
         request_text = json.dumps({**query_body, 'header': cookies}, separators=(',', ':'))
@@ -64,6 +63,8 @@ class NetEaseMusicBaseIE(InfoExtractor):
                 'MUSIC_U': ('MUSIC_U', {lambda i: i.value}),
             }),
         }
+        if self._x_forwarded_for_ip:
+            headers.setdefault('X-Real-IP', self._x_forwarded_for_ip)
         return self._download_json(
             urljoin('https://interface3.music.163.com/', f'/eapi{path}'), video_id,
             data=self._create_eapi_cipher(f'/api{path}', query_body, cookies), headers={

yt_dlp/update.py

@@ -139,7 +139,18 @@ def _get_binary_name():
 
 
 def _get_system_deprecation():
-    MIN_SUPPORTED, MIN_RECOMMENDED = (3, 9), (3, 9)
+    MIN_SUPPORTED, MIN_RECOMMENDED = (3, 9), (3, 10)
+
+    EXE_MSG_TMPL = ('Support for {} has been deprecated. '
+                    'See  https://github.com/yt-dlp/yt-dlp/{}  for details.\n{}')
+    STOP_MSG = 'You may stop receiving updates on this version at any time!'
+    variant = detect_variant()
+
+    # Temporary until macos_legacy executable builds are discontinued
+    if variant == 'darwin_legacy_exe':
+        return EXE_MSG_TMPL.format(
+            f'{variant} (the PyInstaller-bundled executable for macOS versions older than 10.15)',
+            'issues/13856', STOP_MSG)
 
     if sys.version_info > MIN_RECOMMENDED:
         return None
@@ -150,6 +161,13 @@ def _get_system_deprecation():
     if sys.version_info < MIN_SUPPORTED:
         return f'Python version {major}.{minor} is no longer supported! {PYTHON_MSG}'
 
+    # Temporary until aarch64/armv7l build flow is bumped to Ubuntu 22.04 and Python 3.10
+    if variant in ('linux_aarch64_exe', 'linux_armv7l_exe'):
+        libc_ver = version_tuple(os.confstr('CS_GNU_LIBC_VERSION').partition(' ')[2])
+        if libc_ver < (2, 35):
+            return EXE_MSG_TMPL.format('system glibc version < 2.35', 'issues/13858', STOP_MSG)
+        return None
+
     return f'Support for Python version {major}.{minor} has been deprecated. {PYTHON_MSG}'