Change more extractors to list form

Was already added in other PR

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)
 
@@ -1307,7 +1307,8 @@ The available fields are:
  - `uploader_id` (string): Nickname or id of the video uploader
  - `uploader_url` (string): URL to the video uploader's profile
  - `license` (string): License name the video is licensed under
- - `creator` (string): The creator of the video
+ - `creators` (list): The creators of the video
+ - `creator` (string): The creators of the video; comma-separated
  - `timestamp` (numeric): UNIX timestamp of the moment the video became available
  - `upload_date` (string): Video upload date in UTC (YYYYMMDD)
  - `release_timestamp` (numeric): UNIX timestamp of the moment the video was released
@@ -1385,13 +1386,17 @@ Available for the media that is a track or a part of a music album:
  - `track` (string): Title of the track
  - `track_number` (numeric): Number of the track within an album or a disc
  - `track_id` (string): Id of the track
- - `artist` (string): Artist(s) of the track
+ - `artists` (list): Artist(s) of the track
- - `genre` (string): Genre(s) of the track
+ - `artist` (string): Artist(s) of the track; comma-separated
+ - `genres` (list): Genre(s) of the track
+ - `genre` (string): Genre(s) of the track; comma-separated
+ - `composers` (list): Composer(s) of the piece
+ - `composer` (string): Composer(s) of the piece; comma-separated
  - `album` (string): Title of the album the track belongs to
  - `album_type` (string): Type of the album
- - `album_artist` (string): List of all artists appeared on the album
+ - `album_artists` (list): All artists appeared on the album
+ - `album_artist` (string): All artists appeared on the album; comma-separated
  - `disc_number` (numeric): Number of the disc or other physical medium the track belongs to
- - `composer` (string): Name of the composer
 
 Available only when using `--download-sections` and for `chapter:` prefix when using `--split-chapters` for videos with internal chapters:
 
@@ -1768,10 +1773,11 @@ Metadata fields            | From
 `description`,  `synopsis` | `description`
 `purl`, `comment`          | `webpage_url`
 `track`                    | `track_number`
-`artist`                   | `artist`, `creator`, `uploader` or `uploader_id`
+`artist`                   | `artist`, `artists`, `creator`, `creators`, `uploader` or `uploader_id`
-`genre`                    | `genre`
+`composer`                 | `composer` or `composers`
+`genre`                    | `genre` or `genres`
 `album`                    | `album`
-`album_artist`             | `album_artist`
+`album_artist`             | `album_artist` or `album_artists`
 `disc`                     | `disc_number`
 `show`                     | `series`
 `season_number`            | `season_number`

test/helper.py

@@ -223,6 +223,10 @@ def sanitize_got_info_dict(got_dict):
     if test_info_dict.get('display_id') == test_info_dict.get('id'):
         test_info_dict.pop('display_id')
 
+    # Remove deprecated fields
+    for old in YoutubeDL._deprecated_multivalue_fields.keys():
+        test_info_dict.pop(old, None)
+
     # release_year may be generated from release_date
     if try_call(lambda: test_info_dict['release_year'] == int(test_info_dict['release_date'][:4])):
         test_info_dict.pop('release_year')

test/test_YoutubeDL.py

@@ -941,7 +941,7 @@ class TestYoutubeDL(unittest.TestCase):
         def get_videos(filter_=None):
             ydl = YDL({'match_filter': filter_, 'simulate': True})
             for v in videos:
-                ydl.process_ie_result(v, download=True)
+                ydl.process_ie_result(v.copy(), download=True)
             return [v['id'] for v in ydl.downloaded_info_dicts]
 
         res = get_videos()

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/YoutubeDL.py

@@ -581,6 +581,13 @@ class YoutubeDL:
         'http_headers', 'stretched_ratio', 'no_resume', 'has_drm', 'extra_param_to_segment_url', 'hls_aes', 'downloader_options',
         'page_url', 'app', 'play_path', 'tc_url', 'flash_version', 'rtmp_live', 'rtmp_conn', 'rtmp_protocol', 'rtmp_real_time'
     }
+    _deprecated_multivalue_fields = {
+        'album_artist': 'album_artists',
+        'artist': 'artists',
+        'composer': 'composers',
+        'creator': 'creators',
+        'genre': 'genres',
+    }
     _format_selection_exts = {
         'audio': set(MEDIA_EXTENSIONS.common_audio),
         'video': set(MEDIA_EXTENSIONS.common_video + ('3gp', )),
@@ -2641,6 +2648,14 @@ class YoutubeDL:
             if final and info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
                 info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
 
+        for old_key, new_key in self._deprecated_multivalue_fields.items():
+            if new_key in info_dict and old_key in info_dict:
+                self.deprecation_warning(f'Do not return {old_key!r} when {new_key!r} is present')
+            elif old_value := info_dict.get(old_key):
+                info_dict[new_key] = old_value.split(', ')
+            elif new_value := info_dict.get(new_key):
+                info_dict[old_key] = ', '.join(v.replace(',', '\N{FULLWIDTH COMMA}') for v in new_value)
+
     def _raise_pending_errors(self, info):
         err = info.pop('__pending_error', None)
         if err:

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/archiveorg.py

@@ -31,6 +31,7 @@ from ..utils import (
     unified_timestamp,
     url_or_none,
     urlhandle_detect_ext,
+    variadic,
 )
 
 
@@ -49,7 +50,7 @@ class ArchiveOrgIE(InfoExtractor):
             'release_date': '19681210',
             'timestamp': 1268695290,
             'upload_date': '20100315',
-            'creator': 'SRI International',
+            'creators': ['SRI International'],
             'uploader': 'laura@archive.org',
             'thumbnail': r're:https://archive\.org/download/.*\.jpg',
             'display_id': 'XD300-23_68HighlightsAResearchCntAugHumanIntellect.cdr',
@@ -109,7 +110,7 @@ class ArchiveOrgIE(InfoExtractor):
             'title': 'Turning',
             'ext': 'flac',
             'track': 'Turning',
-            'creator': 'Grateful Dead',
+            'creators': ['Grateful Dead'],
             'display_id': 'gd1977-05-08d01t01.flac',
             'track_number': 1,
             'album': '1977-05-08 - Barton Hall - Cornell University',
@@ -129,7 +130,7 @@ class ArchiveOrgIE(InfoExtractor):
             'location': 'Barton Hall - Cornell University',
             'duration': 438.68,
             'track': 'Deal',
-            'creator': 'Grateful Dead',
+            'creators': ['Grateful Dead'],
             'album': '1977-05-08 - Barton Hall - Cornell University',
             'release_date': '19770508',
             'display_id': 'gd1977-05-08d01t07.flac',
@@ -167,7 +168,7 @@ class ArchiveOrgIE(InfoExtractor):
             'upload_date': '20160610',
             'description': 'md5:f70956a156645a658a0dc9513d9e78b7',
             'uploader': 'dimitrios@archive.org',
-            'creator': 'British Broadcasting Corporation, Time-Life Films',
+            'creators': ['British Broadcasting Corporation', 'Time-Life Films'],
             'timestamp': 1465594947,
         },
         'playlist': [
@@ -257,8 +258,7 @@ class ArchiveOrgIE(InfoExtractor):
             'title': m['title'],
             'description': clean_html(m.get('description')),
             'uploader': dict_get(m, ['uploader', 'adder']),
-            'creator': traverse_obj(m, (
+            'creators': traverse_obj(m, ('creator', {variadic}, {lambda x: x[0] and list(x)})),
-                'creator', (({list}, {lambda x: join_nonempty(*x, delim=', ')}), {str})), get_all=False) or None,
             'license': m.get('licenseurl'),
             'release_date': unified_strdate(m.get('date')),
             'timestamp': unified_timestamp(dict_get(m, ['publicdate', 'addeddate'])),
@@ -273,8 +273,7 @@ class ArchiveOrgIE(InfoExtractor):
                     'title': f.get('title') or f['name'],
                     'display_id': f['name'],
                     'description': clean_html(f.get('description')),
-                    'creator': traverse_obj(f, (
+                    'creators': traverse_obj(f, ('creator', {variadic}, {lambda x: x[0] and list(x)})),
-                        'creator', (({list}, {lambda x: join_nonempty(*x, delim=', ')}), {str})), get_all=False) or None,
                     'duration': parse_duration(f.get('length')),
                     'track_number': int_or_none(f.get('track')),
                     'album': f.get('album'),

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/common.py

@@ -278,7 +278,7 @@ class InfoExtractor:
     description:    Full video description.
     uploader:       Full name of the video uploader.
     license:        License name the video is licensed under.
-    creator:        The creator of the video.
+    creators:       List of creators of the video.
     timestamp:      UNIX timestamp of the moment the video was uploaded
     upload_date:    Video upload date in UTC (YYYYMMDD).
                     If not explicitly set, calculated from timestamp
@@ -422,16 +422,16 @@ class InfoExtractor:
     track_number:   Number of the track within an album or a disc, as an integer.
     track_id:       Id of the track (useful in case of custom indexing, e.g. 6.iii),
                     as a unicode string.
-    artist:         Artist(s) of the track.
+    artists:        List of artists of the track.
-    genre:          Genre(s) of the track.
+    composers:      List of composers of the piece.
+    genres:         List of genres of the track.
     album:          Title of the album the track belongs to.
     album_type:     Type of the album (e.g. "Demo", "Full-length", "Split", "Compilation", etc).
-    album_artist:   List of all artists appeared on the album (e.g.
+    album_artists:  List of all artists appeared on the album.
-                    "Ash Borer / Fell Voices" or "Various Artists", useful for splits
+                    E.g. ["Ash Borer", "Fell Voices"] or ["Various Artists"].
-                    and compilations).
+                    Useful for splits and compilations.
     disc_number:    Number of the disc or other physical medium the track belongs to,
                     as an integer.
-    composer:       Composer of the piece
 
     The following fields should only be set for clips that should be cut from the original video:
 
@@ -442,6 +442,18 @@ class InfoExtractor:
     rows:           Number of rows in each storyboard fragment, as an integer
     columns:        Number of columns in each storyboard fragment, as an integer
 
+    The following fields are deprecated and should not be set by new code:
+    composer:       Use "composers" instead.
+                    Composer(s) of the piece, comma-separated.
+    artist:         Use "artists" instead.
+                    Artist(s) of the track, comma-separated.
+    genre:          Use "genres" instead.
+                    Genre(s) of the track, comma-separated.
+    album_artist:   Use "album_artists" instead.
+                    All artists appeared on the album, comma-separated.
+    creator:        Use "creators" instead.
+                    The creator of the video.
+
     Unless mentioned otherwise, the fields should be Unicode strings.
 
     Unless mentioned otherwise, None is equivalent to absence of information.

yt_dlp/extractor/crunchyroll.py

@@ -514,7 +514,7 @@ class CrunchyrollMusicIE(CrunchyrollBaseIE):
             'track': 'Egaono Hana',
             'artist': 'Goose house',
             'thumbnail': r're:(?i)^https://www.crunchyroll.com/imgsrv/.*\.jpeg?$',
-            'genre': ['J-Pop'],
+            'genres': ['J-Pop'],
         },
         'params': {'skip_download': 'm3u8'},
     }, {
@@ -527,7 +527,7 @@ class CrunchyrollMusicIE(CrunchyrollBaseIE):
             'track': 'Crossing Field',
             'artist': 'LiSA',
             'thumbnail': r're:(?i)^https://www.crunchyroll.com/imgsrv/.*\.jpeg?$',
-            'genre': ['Anime'],
+            'genres': ['Anime'],
         },
         'params': {'skip_download': 'm3u8'},
     }, {
@@ -541,7 +541,7 @@ class CrunchyrollMusicIE(CrunchyrollBaseIE):
             'artist': 'LiSA',
             'thumbnail': r're:(?i)^https://www.crunchyroll.com/imgsrv/.*\.jpeg?$',
             'description': 'md5:747444e7e6300907b7a43f0a0503072e',
-            'genre': ['J-Pop'],
+            'genres': ['J-Pop'],
         },
         'params': {'skip_download': 'm3u8'},
     }, {
@@ -594,7 +594,7 @@ class CrunchyrollMusicIE(CrunchyrollBaseIE):
                     'width': ('width', {int_or_none}),
                     'height': ('height', {int_or_none}),
                 }),
-                'genre': ('genres', ..., 'displayValue'),
+                'genres': ('genres', ..., 'displayValue'),
                 'age_limit': ('maturity_ratings', -1, {parse_age_limit}),
             }),
         }
@@ -611,7 +611,7 @@ class CrunchyrollArtistIE(CrunchyrollBaseIE):
         'info_dict': {
             'id': 'MA179CB50D',
             'title': 'LiSA',
-            'genre': ['J-Pop', 'Anime', 'Rock'],
+            'genres': ['J-Pop', 'Anime', 'Rock'],
             'description': 'md5:16d87de61a55c3f7d6c454b73285938e',
         },
         'playlist_mincount': 83,
@@ -645,6 +645,6 @@ class CrunchyrollArtistIE(CrunchyrollBaseIE):
                     'width': ('width', {int_or_none}),
                     'height': ('height', {int_or_none}),
                 }),
-                'genre': ('genres', ..., 'displayValue'),
+                'genres': ('genres', ..., 'displayValue'),
             }),
         }

yt_dlp/extractor/hypergryph.py

@@ -9,7 +9,7 @@ class MonsterSirenHypergryphMusicIE(InfoExtractor):
         'info_dict': {
             'id': '514562',
             'ext': 'wav',
-            'artist': ['塞壬唱片-MSR'],
+            'artists': ['塞壬唱片-MSR'],
             'album': 'Flame Shadow',
             'title': 'Flame Shadow',
         }
@@ -27,6 +27,6 @@ class MonsterSirenHypergryphMusicIE(InfoExtractor):
             'url': traverse_obj(json_data, ('player', 'songDetail', 'sourceUrl')),
             'ext': 'wav',
             'vcodec': 'none',
-            'artist': traverse_obj(json_data, ('player', 'songDetail', 'artists')),
+            'artists': traverse_obj(json_data, ('player', 'songDetail', 'artists', ...)),
             'album': traverse_obj(json_data, ('musicPlay', 'albumDetail', 'name'))
         }

yt_dlp/extractor/musicdex.py

@@ -17,11 +17,11 @@ class MusicdexBaseIE(InfoExtractor):
             'track_number': track_json.get('number'),
             'url': format_field(track_json, 'url', 'https://www.musicdex.org/%s'),
             'duration': track_json.get('duration'),
-            'genre': [genre.get('name') for genre in track_json.get('genres') or []],
+            'genres': [genre.get('name') for genre in track_json.get('genres') or []],
             'like_count': track_json.get('likes_count'),
             'view_count': track_json.get('plays'),
-            'artist': [artist.get('name') for artist in track_json.get('artists') or []],
+            'artists': [artist.get('name') for artist in track_json.get('artists') or []],
-            'album_artist': [artist.get('name') for artist in album_json.get('artists') or []],
+            'album_artists': [artist.get('name') for artist in album_json.get('artists') or []],
             'thumbnail': format_field(album_json, 'image', 'https://www.musicdex.org/%s'),
             'album': album_json.get('name'),
             'release_year': try_get(album_json, lambda x: date_from_str(unified_strdate(x['release_date'])).year),
@@ -43,11 +43,11 @@ class MusicdexSongIE(MusicdexBaseIE):
             'track': 'dual existence',
             'track_number': 1,
             'duration': 266000,
-            'genre': ['Anime'],
+            'genres': ['Anime'],
             'like_count': int,
             'view_count': int,
-            'artist': ['fripSide'],
+            'artists': ['fripSide'],
-            'album_artist': ['fripSide'],
+            'album_artists': ['fripSide'],
             'thumbnail': 'https://www.musicdex.org/storage/album/9iDIam1DHTVqUG4UclFIEq1WAFGXfPW4y0TtZa91.png',
             'album': 'To Aru Kagaku no Railgun T OP2 Single - dual existence',
             'release_year': 2020
@@ -69,9 +69,9 @@ class MusicdexAlbumIE(MusicdexBaseIE):
         'playlist_mincount': 28,
         'info_dict': {
             'id': '56',
-            'genre': ['OST'],
+            'genres': ['OST'],
             'view_count': int,
-            'artist': ['TENMON & Eiichiro Yanagi / minori'],
+            'artists': ['TENMON & Eiichiro Yanagi / minori'],
             'title': 'ef - a tale of memories Original Soundtrack 2 ~fortissimo~',
             'release_year': 2008,
             'thumbnail': 'https://www.musicdex.org/storage/album/2rSHkyYBYfB7sbvElpEyTMcUn6toY7AohOgJuDlE.jpg',
@@ -88,9 +88,9 @@ class MusicdexAlbumIE(MusicdexBaseIE):
             'id': id,
             'title': data_json.get('name'),
             'description': data_json.get('description'),
-            'genre': [genre.get('name') for genre in data_json.get('genres') or []],
+            'genres': [genre.get('name') for genre in data_json.get('genres') or []],
             'view_count': data_json.get('plays'),
-            'artist': [artist.get('name') for artist in data_json.get('artists') or []],
+            'artists': [artist.get('name') for artist in data_json.get('artists') or []],
             'thumbnail': format_field(data_json, 'image', 'https://www.musicdex.org/%s'),
             'release_year': try_get(data_json, lambda x: date_from_str(unified_strdate(x['release_date'])).year),
             'entries': entries,

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/stageplus.py

@@ -21,7 +21,7 @@ class StagePlusVODConcertIE(InfoExtractor):
             'id': 'vod_concert_APNM8GRFDPHMASJKBSPJACG',
             'title': 'Yuja Wang plays Rachmaninoff\'s Piano Concerto No. 2 – from Odeonsplatz',
             'description': 'md5:50f78ec180518c9bdb876bac550996fc',
-            'artist': ['Yuja Wang', 'Lorenzo Viotti'],
+            'artists': ['Yuja Wang', 'Lorenzo Viotti'],
             'upload_date': '20230331',
             'timestamp': 1680249600,
             'release_date': '20210709',
@@ -40,10 +40,10 @@ class StagePlusVODConcertIE(InfoExtractor):
                 'release_timestamp': 1625788800,
                 'duration': 2207,
                 'chapters': 'count:5',
-                'artist': ['Yuja Wang'],
+                'artists': ['Yuja Wang'],
-                'composer': ['Sergei Rachmaninoff'],
+                'composers': ['Sergei Rachmaninoff'],
                 'album': 'Yuja Wang plays Rachmaninoff\'s Piano Concerto No. 2 – from Odeonsplatz',
-                'album_artist': ['Yuja Wang', 'Lorenzo Viotti'],
+                'album_artists': ['Yuja Wang', 'Lorenzo Viotti'],
                 'track': 'Piano Concerto No. 2 in C Minor, Op. 18',
                 'track_number': 1,
                 'genre': 'Instrumental Concerto',
@@ -474,7 +474,7 @@ fragment BannerFields on Banner {
         metadata = traverse_obj(data, {
             'title': 'title',
             'description': ('shortDescription', {str}),
-            'artist': ('artists', 'edges', ..., 'node', 'name'),
+            'artists': ('artists', 'edges', ..., 'node', 'name'),
             'timestamp': ('archiveReleaseDate', {unified_timestamp}),
             'release_timestamp': ('productionDate', {unified_timestamp}),
         })
@@ -494,7 +494,7 @@ fragment BannerFields on Banner {
                 'formats': formats,
                 'subtitles': subtitles,
                 'album': metadata.get('title'),
-                'album_artist': metadata.get('artist'),
+                'album_artists': metadata.get('artist'),
                 'track_number': idx,
                 **metadata,
                 **traverse_obj(video, {
@@ -506,8 +506,8 @@ fragment BannerFields on Banner {
                             'title': 'title',
                             'start_time': ('mark', {float_or_none}),
                         }),
-                    'artist': ('artists', 'edges', ..., 'node', 'name'),
+                    'artists': ('artists', 'edges', ..., 'node', 'name'),
-                    'composer': ('work', 'composers', ..., 'name'),
+                    'composers': ('work', 'composers', ..., 'name'),
                     'genre': ('work', 'genre', 'title'),
                 }),
             })

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/extractor/youtube.py

@@ -2068,7 +2068,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'title': 'Voyeur Girl',
                 'description': 'md5:7ae382a65843d6df2685993e90a8628f',
                 'upload_date': '20190312',
-                'artist': 'Stephen',
+                'artists': ['Stephen'],
+                'creators': ['Stephen'],
                 'track': 'Voyeur Girl',
                 'album': 'it\'s too much love to know my dear',
                 'release_date': '20190313',
@@ -2081,7 +2082,6 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'channel': 'Stephen',  # TODO: should be "Stephen - Topic"
                 'uploader': 'Stephen',
                 'availability': 'public',
-                'creator': 'Stephen',
                 'duration': 169,
                 'thumbnail': 'https://i.ytimg.com/vi_webp/MgNrAu2pzNs/maxresdefault.webp',
                 'age_limit': 0,
@@ -4386,7 +4386,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                         release_year = release_date[:4]
                 info.update({
                     'album': mobj.group('album'.strip()),
-                    'artist': mobj.group('clean_artist') or ', '.join(a.strip() for a in mobj.group('artist').split('·')),
+                    'artists': ([a] if (a := mobj.group('clean_artist'))
+                                else [a.strip() for a in mobj.group('artist').split('·')]),
                     'track': mobj.group('track').strip(),
                     'release_date': release_date,
                     'release_year': int_or_none(release_year),
@@ -4532,7 +4533,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                     if mrr_title == 'Album':
                         info['album'] = mrr_contents_text
                     elif mrr_title == 'Artist':
-                        info['artist'] = mrr_contents_text
+                        info['artists'] = [mrr_contents_text]
                     elif mrr_title == 'Song':
                         info['track'] = mrr_contents_text
             owner_badges = self._extract_badges(traverse_obj(vsir, ('owner', 'videoOwnerRenderer', 'badges')))
@@ -4566,7 +4567,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                     if fmt.get('protocol') == 'm3u8_native':
                         fmt['__needs_testing'] = True
 
-        for s_k, d_k in [('artist', 'creator'), ('track', 'alt_title')]:
+        for s_k, d_k in [('artists', 'creators'), ('track', 'alt_title')]:
             v = info.get(s_k)
             if v:
                 info[d_k] = v

yt_dlp/postprocessor/ffmpeg.py

@@ -738,9 +738,10 @@ class FFmpegMetadataPP(FFmpegPostProcessor):
 
         def add(meta_list, info_list=None):
             value = next((
-                str(info[key]) for key in [f'{meta_prefix}_'] + list(variadic(info_list or meta_list))
+                info[key] for key in [f'{meta_prefix}_'] + list(variadic(info_list or meta_list))
                 if info.get(key) is not None), None)
             if value not in ('', None):
+                value = ', '.join(map(str, variadic(value)))
                 value = value.replace('\0', '')  # nul character cannot be passed in command line
                 metadata['common'].update({meta_f: value for meta_f in variadic(meta_list)})
 
@@ -754,10 +755,11 @@ class FFmpegMetadataPP(FFmpegPostProcessor):
         add(('description', 'synopsis'), 'description')
         add(('purl', 'comment'), 'webpage_url')
         add('track', 'track_number')
-        add('artist', ('artist', 'creator', 'uploader', 'uploader_id'))
+        add('artist', ('artist', 'artists', 'creator', 'creators', 'uploader', 'uploader_id'))
-        add('genre')
+        add('composer', ('composer', 'composers'))
+        add('genre', ('genre', 'genres'))
         add('album')
-        add('album_artist')
+        add('album_artist', ('album_artist', 'album_artists'))
         add('disc', 'disc_number')
         add('show', 'series')
         add('season_number')

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):