Bump Python requirement to 3.8

Authored by: bashonly

.github/workflows/release.yml

@@ -266,7 +266,7 @@ jobs:
         run: |
           python devscripts/update-version.py -c "${{ env.channel }}" -r "${{ env.target_repo }}" -s "${{ env.suffix }}" "${{ env.version }}"
           python devscripts/make_lazy_extractors.py
-          sed -i -E "s/(name=')[^']+(',  # package name)/\1${{ env.pypi_project }}\2/" setup.py
+          sed -i -E '0,/(name = ")[^"]+(")/s//\1${{ env.pypi_project }}\2/' pyproject.toml
 
       - name: Build
         run: |

README.md

@@ -280,7 +280,7 @@ While all the other dependencies are optional, `ffmpeg` and `ffprobe` are highly
 
 * [**ffmpeg** and **ffprobe**](https://www.ffmpeg.org) - Required for [merging separate video and audio files](#format-selection) as well as for various [post-processing](#post-processing-options) tasks. License [depends on the build](https://www.ffmpeg.org/legal.html)
 
-    There are bugs in ffmpeg that causes various issues when used alongside yt-dlp. Since ffmpeg is such an important dependency, we provide [custom builds](https://github.com/yt-dlp/FFmpeg-Builds#ffmpeg-static-auto-builds) with patches for some of these issues at [yt-dlp/FFmpeg-Builds](https://github.com/yt-dlp/FFmpeg-Builds). See [the readme](https://github.com/yt-dlp/FFmpeg-Builds#patches-applied) for details on the specific issues solved by these builds
+    There are bugs in ffmpeg that cause various issues when used alongside yt-dlp. Since ffmpeg is such an important dependency, we provide [custom builds](https://github.com/yt-dlp/FFmpeg-Builds#ffmpeg-static-auto-builds) with patches for some of these issues at [yt-dlp/FFmpeg-Builds](https://github.com/yt-dlp/FFmpeg-Builds). See [the readme](https://github.com/yt-dlp/FFmpeg-Builds#patches-applied) for details on the specific issues solved by these builds
     
     **Important**: What you need is ffmpeg *binary*, **NOT** [the python package of the same name](https://pypi.org/project/ffmpeg)
 

pyproject.toml

@@ -11,7 +11,7 @@ maintainers = [
     {name = "Grub4K", email = "contact@grub4k.xyz"},
 ]
 description = "A youtube-dl fork with additional features and patches"
-requires-python = ">=3.7"
+requires-python = ">=3.8"
 keywords = [
     "youtube-dl",
     "video-downloader",
@@ -26,7 +26,6 @@ classifiers = [
     "Development Status :: 5 - Production/Stable",
     "Environment :: Console",
     "Programming Language :: Python",
-    "Programming Language :: Python :: 3.7",
     "Programming Language :: Python :: 3.8",
     "Programming Language :: Python :: 3.9",
     "Programming Language :: Python :: 3.10",
@@ -41,20 +40,18 @@ classifiers = [
 dynamic = ["version", "readme"]
 
 dependencies = [
-    "pycryptodomex",
-    "websockets",
     "brotli; implementation_name=='cpython'",
     "brotlicffi; implementation_name!='cpython'",
     "certifi",
-    "secretstorage; sys_platform=='linux' and (implementation_name!='pypy' or python_version>='3.10')",
+    "pycryptodomex",
+    "requests>=2.31.0,<3",
+    "urllib3>=1.26.17,<3",
+    "websockets>=12.0",
 ]
 
 [project.optional-dependencies]
-requests = [
-    "requests>=2.31.0,<3",
-    "urllib3>=1.26.17,<3",
-]
 mutagen = ["mutagen"]
+secretstorage = ["secretstorage"]
 
 [project.urls]
 Documentation = "https://github.com/yt-dlp/yt-dlp#readme"
@@ -66,7 +63,7 @@ Funding = "https://github.com/yt-dlp/yt-dlp/blob/master/Collaborators.md#collabo
 packages = ["yt_dlp"]
 
 [tool.setuptools.dynamic]
-version = {attr = "yt_dlp.version.__version__"}
+version = {attr = "yt_dlp.version._pkg_version"}
 readme = {file = ["README.md"], content-type = "text/markdown"}
 
 [project.scripts]

test/test_utils.py

@@ -2340,6 +2340,58 @@ Line 1
         self.assertEqual(traverse_obj(mobj, lambda k, _: k in (0, 'group')), ['0123', '3'],
                          msg='function on a `re.Match` should give group name as well')
 
+        # Test xml.etree.ElementTree.Element as input obj
+        etree = xml.etree.ElementTree.fromstring('''<?xml version="1.0"?>
+        <data>
+            <country name="Liechtenstein">
+                <rank>1</rank>
+                <year>2008</year>
+                <gdppc>141100</gdppc>
+                <neighbor name="Austria" direction="E"/>
+                <neighbor name="Switzerland" direction="W"/>
+            </country>
+            <country name="Singapore">
+                <rank>4</rank>
+                <year>2011</year>
+                <gdppc>59900</gdppc>
+                <neighbor name="Malaysia" direction="N"/>
+            </country>
+            <country name="Panama">
+                <rank>68</rank>
+                <year>2011</year>
+                <gdppc>13600</gdppc>
+                <neighbor name="Costa Rica" direction="W"/>
+                <neighbor name="Colombia" direction="E"/>
+            </country>
+        </data>''')
+        self.assertEqual(traverse_obj(etree, ''), etree,
+                         msg='empty str key should return the element itself')
+        self.assertEqual(traverse_obj(etree, 'country'), list(etree),
+                         msg='str key should lead all children with that tag name')
+        self.assertEqual(traverse_obj(etree, ...), list(etree),
+                         msg='`...` as key should return all children')
+        self.assertEqual(traverse_obj(etree, lambda _, x: x[0].text == '4'), [etree[1]],
+                         msg='function as key should get element as value')
+        self.assertEqual(traverse_obj(etree, lambda i, _: i == 1), [etree[1]],
+                         msg='function as key should get index as key')
+        self.assertEqual(traverse_obj(etree, 0), etree[0],
+                         msg='int key should return the nth child')
+        self.assertEqual(traverse_obj(etree, './/neighbor/@name'),
+                         ['Austria', 'Switzerland', 'Malaysia', 'Costa Rica', 'Colombia'],
+                         msg='`@<attribute>` at end of path should give that attribute')
+        self.assertEqual(traverse_obj(etree, '//neighbor/@fail'), [None, None, None, None, None],
+                         msg='`@<nonexistant>` at end of path should give `None`')
+        self.assertEqual(traverse_obj(etree, ('//neighbor/@', 2)), {'name': 'Malaysia', 'direction': 'N'},
+                         msg='`@` should give the full attribute dict')
+        self.assertEqual(traverse_obj(etree, '//year/text()'), ['2008', '2011', '2011'],
+                         msg='`text()` at end of path should give the inner text')
+        self.assertEqual(traverse_obj(etree, '//*[@direction]/@direction'), ['E', 'W', 'N', 'W', 'E'],
+                         msg='full python xpath features should be supported')
+        self.assertEqual(traverse_obj(etree, (0, '@name')), 'Liechtenstein',
+                         msg='special transformations should act on current element')
+        self.assertEqual(traverse_obj(etree, ('country', 0, ..., 'text()', {int_or_none})), [1, 2008, 141100],
+                         msg='special transformations should act on current element')
+
     def test_http_header_dict(self):
         headers = HTTPHeaderDict()
         headers['ytdl-test'] = b'0'

yt_dlp/cookies.py

@@ -186,7 +186,7 @@ def _firefox_browser_dir():
     if sys.platform in ('cygwin', 'win32'):
         return os.path.expandvars(R'%APPDATA%\Mozilla\Firefox\Profiles')
     elif sys.platform == 'darwin':
-        return os.path.expanduser('~/Library/Application Support/Firefox')
+        return os.path.expanduser('~/Library/Application Support/Firefox/Profiles')
     return os.path.expanduser('~/.mozilla/firefox')
 
 

yt_dlp/extractor/_extractors.py

@@ -2019,7 +2019,6 @@ from .tunein import (
     TuneInPodcastEpisodeIE,
     TuneInShortenerIE,
 )
-from .turbo import TurboIE
 from .tv2 import (
     TV2IE,
     TV2ArticleIE,
@@ -2223,6 +2222,7 @@ from .viki import (
     VikiIE,
     VikiChannelIE,
 )
+from .viously import ViouslyIE
 from .viqeo import ViqeoIE
 from .viu import (
     ViuIE,

yt_dlp/extractor/ard.py

@@ -4,6 +4,7 @@ from functools import partial
 from .common import InfoExtractor
 from ..utils import (
     OnDemandPagedList,
+    bug_reports_message,
     determine_ext,
     int_or_none,
     join_nonempty,
@@ -233,7 +234,7 @@ class ARDBetaMediathekIE(InfoExtractor):
         (?:(?:beta|www)\.)?ardmediathek\.de/
         (?:[^/]+/)?
         (?:player|live|video)/
-        (?:(?P<display_id>[^?#]+)/)?
+        (?:[^?#]+/)?
         (?P<id>[a-zA-Z0-9]+)
         /?(?:[?#]|$)'''
     _GEO_COUNTRIES = ['DE']
@@ -242,8 +243,8 @@ class ARDBetaMediathekIE(InfoExtractor):
         'url': 'https://www.ardmediathek.de/video/filme-im-mdr/liebe-auf-vier-pfoten/mdr-fernsehen/Y3JpZDovL21kci5kZS9zZW5kdW5nLzI4MjA0MC80MjIwOTEtNDAyNTM0',
         'md5': 'b6e8ab03f2bcc6e1f9e6cef25fcc03c4',
         'info_dict': {
-            'display_id': 'filme-im-mdr/liebe-auf-vier-pfoten/mdr-fernsehen',
+            'display_id': 'Y3JpZDovL21kci5kZS9zZW5kdW5nLzI4MjA0MC80MjIwOTEtNDAyNTM0',
-            'id': 'Y3JpZDovL21kci5kZS9zZW5kdW5nLzI4MjA0MC80MjIwOTEtNDAyNTM0',
+            'id': '12939099',
             'title': 'Liebe auf vier Pfoten',
             'description': r're:^Claudia Schmitt, Anwältin in Salzburg',
             'duration': 5222,
@@ -255,7 +256,7 @@ class ARDBetaMediathekIE(InfoExtractor):
             'series': 'Filme im MDR',
             'age_limit': 0,
             'channel': 'MDR',
-            '_old_archive_ids': ['ardbetamediathek 12939099'],
+            '_old_archive_ids': ['ardbetamediathek Y3JpZDovL21kci5kZS9zZW5kdW5nLzI4MjA0MC80MjIwOTEtNDAyNTM0'],
         },
     }, {
         'url': 'https://www.ardmediathek.de/mdr/video/die-robuste-roswita/Y3JpZDovL21kci5kZS9iZWl0cmFnL2Ntcy84MWMxN2MzZC0wMjkxLTRmMzUtODk4ZS0wYzhlOWQxODE2NGI/',
@@ -276,37 +277,37 @@ class ARDBetaMediathekIE(InfoExtractor):
         'url': 'https://www.ardmediathek.de/video/tagesschau-oder-tagesschau-20-00-uhr/das-erste/Y3JpZDovL2Rhc2Vyc3RlLmRlL3RhZ2Vzc2NoYXUvZmM4ZDUxMjgtOTE0ZC00Y2MzLTgzNzAtNDZkNGNiZWJkOTll',
         'md5': '1e73ded21cb79bac065117e80c81dc88',
         'info_dict': {
-            'id': 'Y3JpZDovL2Rhc2Vyc3RlLmRlL3RhZ2Vzc2NoYXUvZmM4ZDUxMjgtOTE0ZC00Y2MzLTgzNzAtNDZkNGNiZWJkOTll',
+            'id': '10049223',
             'ext': 'mp4',
             'title': 'tagesschau, 20:00 Uhr',
             'timestamp': 1636398000,
             'description': 'md5:39578c7b96c9fe50afdf5674ad985e6b',
             'upload_date': '20211108',
-            'display_id': 'tagesschau-oder-tagesschau-20-00-uhr/das-erste',
+            'display_id': 'Y3JpZDovL2Rhc2Vyc3RlLmRlL3RhZ2Vzc2NoYXUvZmM4ZDUxMjgtOTE0ZC00Y2MzLTgzNzAtNDZkNGNiZWJkOTll',
             'duration': 915,
             'episode': 'tagesschau, 20:00 Uhr',
             'series': 'tagesschau',
             'thumbnail': 'https://api.ardmediathek.de/image-service/images/urn:ard:image:fbb21142783b0a49?w=960&ch=ee69108ae344f678',
             'channel': 'ARD-Aktuell',
-            '_old_archive_ids': ['ardbetamediathek 10049223'],
+            '_old_archive_ids': ['ardbetamediathek Y3JpZDovL2Rhc2Vyc3RlLmRlL3RhZ2Vzc2NoYXUvZmM4ZDUxMjgtOTE0ZC00Y2MzLTgzNzAtNDZkNGNiZWJkOTll'],
         },
     }, {
         'url': 'https://www.ardmediathek.de/video/7-tage/7-tage-unter-harten-jungs/hr-fernsehen/N2I2YmM5MzgtNWFlOS00ZGFlLTg2NzMtYzNjM2JlNjk4MDg3',
         'md5': 'c428b9effff18ff624d4f903bda26315',
         'info_dict': {
-            'id': 'N2I2YmM5MzgtNWFlOS00ZGFlLTg2NzMtYzNjM2JlNjk4MDg3',
+            'id': '94834686',
             'ext': 'mp4',
             'duration': 2700,
             'episode': '7 Tage ... unter harten Jungs',
             'description': 'md5:0f215470dcd2b02f59f4bd10c963f072',
             'upload_date': '20231005',
             'timestamp': 1696491171,
-            'display_id': '7-tage/7-tage-unter-harten-jungs/hr-fernsehen',
+            'display_id': 'N2I2YmM5MzgtNWFlOS00ZGFlLTg2NzMtYzNjM2JlNjk4MDg3',
             'series': '7 Tage ...',
             'channel': 'HR',
             'thumbnail': 'https://api.ardmediathek.de/image-service/images/urn:ard:image:f6e6d5ffac41925c?w=960&ch=fa32ba69bc87989a',
             'title': '7 Tage ... unter harten Jungs',
-            '_old_archive_ids': ['ardbetamediathek 94834686'],
+            '_old_archive_ids': ['ardbetamediathek N2I2YmM5MzgtNWFlOS00ZGFlLTg2NzMtYzNjM2JlNjk4MDg3'],
         },
     }, {
         'url': 'https://beta.ardmediathek.de/ard/video/Y3JpZDovL2Rhc2Vyc3RlLmRlL3RhdG9ydC9mYmM4NGM1NC0xNzU4LTRmZGYtYWFhZS0wYzcyZTIxNGEyMDE',
@@ -357,14 +358,25 @@ class ARDBetaMediathekIE(InfoExtractor):
         }), get_all=False)
 
     def _real_extract(self, url):
-        video_id, display_id = self._match_valid_url(url).group('id', 'display_id')
+        display_id = self._match_id(url)
 
         page_data = self._download_json(
-            f'https://api.ardmediathek.de/page-gateway/pages/ard/item/{video_id}', video_id, query={
+            f'https://api.ardmediathek.de/page-gateway/pages/ard/item/{display_id}', display_id, query={
                 'embedded': 'false',
                 'mcV6': 'true',
             })
 
+        # For user convenience we use the old contentId instead of the longer crid
+        # Ref: https://github.com/yt-dlp/yt-dlp/issues/8731#issuecomment-1874398283
+        old_id = traverse_obj(page_data, ('tracking', 'atiCustomVars', 'contentId', {int}))
+        if old_id is not None:
+            video_id = str(old_id)
+            archive_ids = [make_archive_id(ARDBetaMediathekIE, display_id)]
+        else:
+            self.report_warning(f'Could not extract contentId{bug_reports_message()}')
+            video_id = display_id
+            archive_ids = None
+
         player_data = traverse_obj(
             page_data, ('widgets', lambda _, v: v['type'] in ('player_ondemand', 'player_live'), {dict}), get_all=False)
         is_live = player_data.get('type') == 'player_live'
@@ -419,8 +431,6 @@ class ARDBetaMediathekIE(InfoExtractor):
                 })
 
         age_limit = traverse_obj(page_data, ('fskRating', {lambda x: remove_start(x, 'FSK')}, {int_or_none}))
-        old_id = traverse_obj(page_data, ('tracking', 'atiCustomVars', 'contentId'))
-
         return {
             'id': video_id,
             'display_id': display_id,
@@ -438,7 +448,7 @@ class ARDBetaMediathekIE(InfoExtractor):
                 'channel': 'clipSourceName',
             })),
             **self._extract_episode_info(page_data.get('title')),
-            '_old_archive_ids': [make_archive_id(ARDBetaMediathekIE, old_id)],
+            '_old_archive_ids': archive_ids,
         }
 
 

yt_dlp/extractor/bigo.py

@@ -29,7 +29,8 @@ class BigoIE(InfoExtractor):
 
         info_raw = self._download_json(
             'https://ta.bigo.tv/official_website/studio/getInternalStudioInfo',
-            user_id, data=urlencode_postdata({'siteId': user_id}))
+            user_id, data=urlencode_postdata({'siteId': user_id}),
+            headers={'Accept': 'application/json'})
 
         if not isinstance(info_raw, dict):
             raise ExtractorError('Received invalid JSON data')

yt_dlp/extractor/nhk.py

@@ -665,7 +665,7 @@ class NhkRadiruLiveIE(InfoExtractor):
 
         noa_info = self._download_json(
             f'https:{config.find(".//url_program_noa").text}'.format(area=data.find('areakey').text),
-            station, note=f'Downloading {area} station metadata')
+            station, note=f'Downloading {area} station metadata', fatal=False)
         present_info = traverse_obj(noa_info, ('nowonair_list', self._NOA_STATION_IDS.get(station), 'present'))
 
         return {

yt_dlp/extractor/turbo.py

@@ -1,64 +0,0 @@
-import re
-
-from .common import InfoExtractor
-from ..compat import compat_str
-from ..utils import (
-    ExtractorError,
-    int_or_none,
-    qualities,
-    xpath_text,
-)
-
-
-class TurboIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?turbo\.fr/videos-voiture/(?P<id>[0-9]+)-'
-    _API_URL = 'http://www.turbo.fr/api/tv/xml.php?player_generique=player_generique&id={0:}'
-    _TEST = {
-        'url': 'http://www.turbo.fr/videos-voiture/454443-turbo-du-07-09-2014-renault-twingo-3-bentley-continental-gt-speed-ces-guide-achat-dacia.html',
-        'md5': '33f4b91099b36b5d5a91f84b5bcba600',
-        'info_dict': {
-            'id': '454443',
-            'ext': 'mp4',
-            'duration': 3715,
-            'title': 'Turbo du 07/09/2014 : Renault Twingo 3, Bentley Continental GT Speed, CES, Guide Achat Dacia... ',
-            'description': 'Turbo du 07/09/2014 : Renault Twingo 3, Bentley Continental GT Speed, CES, Guide Achat Dacia...',
-            'thumbnail': r're:^https?://.*\.jpg$',
-        }
-    }
-
-    def _real_extract(self, url):
-        mobj = self._match_valid_url(url)
-        video_id = mobj.group('id')
-
-        webpage = self._download_webpage(url, video_id)
-
-        playlist = self._download_xml(self._API_URL.format(video_id), video_id)
-        item = playlist.find('./channel/item')
-        if item is None:
-            raise ExtractorError('Playlist item was not found', expected=True)
-
-        title = xpath_text(item, './title', 'title')
-        duration = int_or_none(xpath_text(item, './durate', 'duration'))
-        thumbnail = xpath_text(item, './visuel_clip', 'thumbnail')
-        description = self._html_search_meta('description', webpage)
-
-        formats = []
-        get_quality = qualities(['3g', 'sd', 'hq'])
-        for child in item:
-            m = re.search(r'url_video_(?P<quality>.+)', child.tag)
-            if m:
-                quality = compat_str(m.group('quality'))
-                formats.append({
-                    'format_id': quality,
-                    'url': child.text,
-                    'quality': get_quality(quality),
-                })
-
-        return {
-            'id': video_id,
-            'title': title,
-            'duration': duration,
-            'thumbnail': thumbnail,
-            'description': description,
-            'formats': formats,
-        }

yt_dlp/extractor/twitch.py

@@ -8,7 +8,6 @@ from .common import InfoExtractor
 from ..compat import (
     compat_parse_qs,
     compat_str,
-    compat_urllib_parse_urlencode,
     compat_urllib_parse_urlparse,
 )
 from ..utils import (
@@ -191,6 +190,20 @@ class TwitchBaseIE(InfoExtractor):
             'url': thumbnail,
         }] if thumbnail else None
 
+    def _extract_twitch_m3u8_formats(self, video_id, token, signature):
+        """Subclasses must define _M3U8_PATH"""
+        return self._extract_m3u8_formats(
+            f'{self._USHER_BASE}/{self._M3U8_PATH}/{video_id}.m3u8', video_id, 'mp4', query={
+                'allow_source': 'true',
+                'allow_audio_only': 'true',
+                'allow_spectre': 'true',
+                'p': random.randint(1000000, 10000000),
+                'player': 'twitchweb',
+                'playlist_include_framerate': 'true',
+                'sig': signature,
+                'token': token,
+            })
+
 
 class TwitchVodIE(TwitchBaseIE):
     IE_NAME = 'twitch:vod'
@@ -203,6 +216,7 @@ class TwitchVodIE(TwitchBaseIE):
                         )
                         (?P<id>\d+)
                     '''
+    _M3U8_PATH = 'vod'
 
     _TESTS = [{
         'url': 'http://www.twitch.tv/riotgames/v/6528877?t=5m10s',
@@ -532,20 +546,8 @@ class TwitchVodIE(TwitchBaseIE):
         info = self._extract_info_gql(video, vod_id)
         access_token = self._download_access_token(vod_id, 'video', 'id')
 
-        formats = self._extract_m3u8_formats(
+        formats = self._extract_twitch_m3u8_formats(
-            '%s/vod/%s.m3u8?%s' % (
+            vod_id, access_token['value'], access_token['signature'])
-                self._USHER_BASE, vod_id,
-                compat_urllib_parse_urlencode({
-                    'allow_source': 'true',
-                    'allow_audio_only': 'true',
-                    'allow_spectre': 'true',
-                    'player': 'twitchweb',
-                    'playlist_include_framerate': 'true',
-                    'nauth': access_token['value'],
-                    'nauthsig': access_token['signature'],
-                })),
-            vod_id, 'mp4', entry_protocol='m3u8_native')
-
         formats.extend(self._extract_storyboard(vod_id, video.get('storyboard'), info.get('duration')))
 
         self._prefer_source(formats)
@@ -924,6 +926,7 @@ class TwitchStreamIE(TwitchBaseIE):
                         )
                         (?P<id>[^/#?]+)
                     '''
+    _M3U8_PATH = 'api/channel/hls'
 
     _TESTS = [{
         'url': 'http://www.twitch.tv/shroomztv',
@@ -1026,23 +1029,10 @@ class TwitchStreamIE(TwitchBaseIE):
 
         access_token = self._download_access_token(
             channel_name, 'stream', 'channelName')
-        token = access_token['value']
 
         stream_id = stream.get('id') or channel_name
-        query = {
+        formats = self._extract_twitch_m3u8_formats(
-            'allow_source': 'true',
+            channel_name, access_token['value'], access_token['signature'])
-            'allow_audio_only': 'true',
-            'allow_spectre': 'true',
-            'p': random.randint(1000000, 10000000),
-            'player': 'twitchweb',
-            'playlist_include_framerate': 'true',
-            'segment_preference': '4',
-            'sig': access_token['signature'].encode('utf-8'),
-            'token': token.encode('utf-8'),
-        }
-        formats = self._extract_m3u8_formats(
-            '%s/api/channel/hls/%s.m3u8' % (self._USHER_BASE, channel_name),
-            stream_id, 'mp4', query=query)
         self._prefer_source(formats)
 
         view_count = stream.get('viewers')

yt_dlp/extractor/viously.py

@@ -0,0 +1,60 @@
+import base64
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    extract_attributes,
+    int_or_none,
+    parse_iso8601,
+)
+from ..utils.traversal import traverse_obj
+
+
+class ViouslyIE(InfoExtractor):
+    _VALID_URL = False
+    _WEBPAGE_TESTS = [{
+        'url': 'http://www.turbo.fr/videos-voiture/454443-turbo-du-07-09-2014-renault-twingo-3-bentley-continental-gt-speed-ces-guide-achat-dacia.html',
+        'md5': '37a6c3381599381ff53a7e1e0575c0bc',
+        'info_dict': {
+            'id': 'F_xQzS2jwb3',
+            'ext': 'mp4',
+            'title': 'Turbo du 07/09/2014\xa0: Renault Twingo 3, Bentley Continental GT Speed, CES, Guide Achat Dacia...',
+            'description': 'Turbo du 07/09/2014\xa0: Renault Twingo 3, Bentley Continental GT Speed, CES, Guide Achat Dacia...',
+            'age_limit': 0,
+            'upload_date': '20230328',
+            'timestamp': 1680037507,
+            'duration': 3716,
+            'categories': ['motors'],
+        }
+    }]
+
+    def _extract_from_webpage(self, url, webpage):
+        viously_players = re.findall(r'<div[^>]*class="(?:[^"]*\s)?v(?:iou)?sly-player(?:\s[^"]*)?"[^>]*>', webpage)
+        if not viously_players:
+            return
+
+        def custom_decode(text):
+            STANDARD_ALPHABET = 'ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/='
+            CUSTOM_ALPHABET = 'VIOUSLYABCDEFGHJKMNPQRTWXZviouslyabcdefghjkmnpqrtwxz9876543210+/='
+            data = base64.b64decode(text.translate(str.maketrans(CUSTOM_ALPHABET, STANDARD_ALPHABET)))
+            return data.decode('utf-8').strip('\x00')
+
+        for video_id in traverse_obj(viously_players, (..., {extract_attributes}, 'id')):
+            formats = self._extract_m3u8_formats(
+                f'https://www.viously.com/video/hls/{video_id}/index.m3u8', video_id, fatal=False)
+            if not formats:
+                continue
+            data = self._download_json(
+                f'https://www.viously.com/export/json/{video_id}', video_id,
+                transform_source=custom_decode, fatal=False)
+            yield {
+                'id': video_id,
+                'formats': formats,
+                **traverse_obj(data, ('video', {
+                    'title': ('title', {str}),
+                    'description': ('description', {str}),
+                    'duration': ('duration', {int_or_none}),
+                    'timestamp': ('iso_date', {parse_iso8601}),
+                    'categories': ('category', 'name', {str}, {lambda x: [x] if x else None}),
+                })),
+            }

yt_dlp/utils/traversal.py

@@ -3,6 +3,7 @@ import contextlib
 import inspect
 import itertools
 import re
+import xml.etree.ElementTree
 
 from ._utils import (
     IDENTITY,
@@ -118,7 +119,7 @@ def traverse_obj(
             branching = True
             if isinstance(obj, collections.abc.Mapping):
                 result = obj.values()
-            elif is_iterable_like(obj):
+            elif is_iterable_like(obj) or isinstance(obj, xml.etree.ElementTree.Element):
                 result = obj
             elif isinstance(obj, re.Match):
                 result = obj.groups()
@@ -132,7 +133,7 @@ def traverse_obj(
             branching = True
             if isinstance(obj, collections.abc.Mapping):
                 iter_obj = obj.items()
-            elif is_iterable_like(obj):
+            elif is_iterable_like(obj) or isinstance(obj, xml.etree.ElementTree.Element):
                 iter_obj = enumerate(obj)
             elif isinstance(obj, re.Match):
                 iter_obj = itertools.chain(
@@ -168,7 +169,7 @@ def traverse_obj(
                 result = next((v for k, v in obj.groupdict().items() if casefold(k) == key), None)
 
         elif isinstance(key, (int, slice)):
-            if is_iterable_like(obj, collections.abc.Sequence):
+            if is_iterable_like(obj, (collections.abc.Sequence, xml.etree.ElementTree.Element)):
                 branching = isinstance(key, slice)
                 with contextlib.suppress(IndexError):
                     result = obj[key]
@@ -176,6 +177,34 @@ def traverse_obj(
                 with contextlib.suppress(IndexError):
                     result = str(obj)[key]
 
+        elif isinstance(obj, xml.etree.ElementTree.Element) and isinstance(key, str):
+            xpath, _, special = key.rpartition('/')
+            if not special.startswith('@') and special != 'text()':
+                xpath = key
+                special = None
+
+            # Allow abbreviations of relative paths, absolute paths error
+            if xpath.startswith('/'):
+                xpath = f'.{xpath}'
+            elif xpath and not xpath.startswith('./'):
+                xpath = f'./{xpath}'
+
+            def apply_specials(element):
+                if special is None:
+                    return element
+                if special == '@':
+                    return element.attrib
+                if special.startswith('@'):
+                    return try_call(element.attrib.get, args=(special[1:],))
+                if special == 'text()':
+                    return element.text
+                assert False, f'apply_specials is missing case for {special!r}'
+
+            if xpath:
+                result = list(map(apply_specials, obj.iterfind(xpath)))
+            else:
+                result = apply_specials(obj)
+
         return branching, result if branching else (result,)
 
     def lazy_last(iterable):