Merge 96800222076e707e1f98364024aa4fcedb980810 into da7223d4aa42ff9fc680b0951d043dd03cec2d30

* improve extraction of global "useful data" Array from player JS
* also handle tv-player and add tests: thx seproDev (yt-dlp/yt-dlp#12684)

Co-Authored-By: sepro <sepro@sepr0.com>

test/test_youtube_signature.py

@@ -232,8 +232,32 @@ _NSIG_TESTS = [
         'W9HJZKktxuYoDTqW', 'jHbbkcaxm54',
     ),
     (
-        'https://www.youtube.com/s/player/91201489/player_ias_tce.vflset/en_US/base.js',
+        'https://www.youtube.com/s/player/643afba4/player_ias.vflset/en_US/base.js',
-        'W9HJZKktxuYoDTqW', 'U48vOZHaeYS6vO',
+        'W9HJZKktxuYoDTqW', 'larxUlagTRAcSw',
+    ),
+    (
+        'https://www.youtube.com/s/player/e7567ecf/player_ias_tce.vflset/en_US/base.js',
+        'Sy4aDGc0VpYRR9ew_', '5UPOT1VhoZxNLQ',
+    ),
+    (
+        'https://www.youtube.com/s/player/d50f54ef/player_ias_tce.vflset/en_US/base.js',
+        'Ha7507LzRmH3Utygtj', 'XFTb2HoeOE5MHg',
+    ),
+    (
+        'https://www.youtube.com/s/player/074a8365/player_ias_tce.vflset/en_US/base.js',
+        'Ha7507LzRmH3Utygtj', 'ufTsrE0IVYrkl8v',
+    ),
+    (
+        'https://www.youtube.com/s/player/643afba4/player_ias.vflset/en_US/base.js',
+        'N5uAlLqm0eg1GyHO', 'dCBQOejdq5s-ww',
+    ),
+    (
+        'https://www.youtube.com/s/player/69f581a5/tv-player-ias.vflset/tv-player-ias.js',
+        '-qIP447rVlTTwaZjY', 'KNcGOksBAvwqQg',
+    ),
+    (
+        'https://www.youtube.com/s/player/643afba4/tv-player-ias.vflset/tv-player-ias.js',
+        'ir9-V6cdbCiyKxhr', '2PL7ZDYAALMfmA',
     ),
 ]
 

youtube_dl/extractor/extractors.py

@@ -1078,6 +1078,10 @@ from .rutube import (
     RutubePersonIE,
     RutubePlaylistIE,
 )
+from .glomex import (
+    GlomexIE,
+    GlomexEmbedIE,
+)
 from .rutv import RUTVIE
 from .ruutu import RuutuIE
 from .ruv import RuvIE

youtube_dl/extractor/generic.py

@@ -102,6 +102,7 @@ from .ustream import UstreamIE
 from .arte import ArteTVEmbedIE
 from .videopress import VideoPressIE
 from .rutube import RutubeIE
+from .glomex import GlomexEmbedIE
 from .limelight import LimelightBaseIE
 from .anvato import AnvatoIE
 from .washingtonpost import WashingtonPostIE
@@ -3400,6 +3401,12 @@ class GenericIE(InfoExtractor):
             return self.playlist_from_matches(
                 rutube_urls, video_id, video_title, ie=RutubeIE.ie_key())
 
+        # Look for Glomex embeds
+        glomex_urls = list(GlomexEmbedIE._extract_urls(webpage, url))
+        if glomex_urls:
+            return self.playlist_from_matches(
+                glomex_urls, video_id, video_title, ie=GlomexEmbedIE.ie_key())
+
         # Look for WashingtonPost embeds
         wapo_urls = WashingtonPostIE._extract_urls(webpage)
         if wapo_urls:

youtube_dl/extractor/glomex.py

@@ -0,0 +1,279 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..compat import (
+    compat_urllib_parse_urlparse,
+    compat_urllib_parse_urlencode,
+)
+from ..utils import (
+    ExtractorError,
+    determine_ext,
+    int_or_none,
+    try_get,
+    smuggle_url,
+    unsmuggle_url,
+    unescapeHTML,
+)
+
+
+class GlomexBaseIE(InfoExtractor):
+    _DEFAULT_ORIGIN_URL = 'https://player.glomex.com/'
+    _API_URL = 'https://integration-cloudfront-eu-west-1.mes.glomex.cloud/'
+
+    @staticmethod
+    def _smuggle_origin_url(url, origin_url):
+        return smuggle_url(url, {'origin': origin_url})
+
+    @classmethod
+    def _unsmuggle_origin_url(cls, url, fallback_origin_url=None):
+        defaults = {'origin': fallback_origin_url or cls._DEFAULT_ORIGIN_URL}
+        unsmuggled_url, data = unsmuggle_url(url, default=defaults)
+        return unsmuggled_url, data['origin']
+
+    def _get_videoid_type(self, video_id):
+        _VIDEOID_TYPES = {
+            'v': 'video',
+            'pl': 'playlist',
+            'rl': 'related videos playlist',
+            'cl': 'curated playlist',
+        }
+        prefix = video_id.split('-')[0]
+        return _VIDEOID_TYPES.get(prefix, 'unknown type')
+
+    def _download_api_data(self, video_id, integration, current_url=None):
+        query = {
+            'integration_id': integration,
+            'playlist_id': video_id,
+            'current_url': current_url or self._DEFAULT_ORIGIN_URL,
+        }
+        video_id_type = self._get_videoid_type(video_id)
+        return self._download_json(
+            self._API_URL,
+            video_id, 'Downloading %s JSON' % video_id_type,
+            'Unable to download %s JSON' % video_id_type,
+            query=query)
+
+    def _download_and_extract_api_data(self, video_id, integration, current_url):
+        api_data = self._download_api_data(video_id, integration, current_url)
+        videos = api_data['videos']
+        if not videos:
+            raise ExtractorError('no videos found for %s' % video_id)
+        if len(videos) == 1:
+            return self._extract_api_data(videos[0], video_id)
+        # assume some kind of playlist
+        videos = [
+            self._extract_api_data(video, video_id)
+            for video in videos
+        ]
+        return self.playlist_result(videos, video_id)
+
+    def _extract_api_data(self, video, video_id):
+        if video.get('error_code') == 'contentGeoblocked':
+            self.raise_geo_restricted(countries=video['geo_locations'])
+        info = self._extract_info(video, video_id)
+        info['formats'] = self._extract_formats(video, video_id)
+        return info
+
+    @staticmethod
+    def _extract_info(video, video_id=None, require_title=True):
+        title = video['title'] if require_title else video.get('title')
+
+        def append_image_url(url, default='profile:player-960x540'):
+            if url:
+                return '%s/%s' % (url, default)
+        thumbnail = append_image_url(try_get(video,
+                                             lambda x: x['image']['url']))
+        thumbnails = [
+            dict(width=960, height=540,
+                 **{k: append_image_url(v) if k == 'url' else v
+                    for k, v in image.items() if k in ('id', 'url')})
+            for image in video.get('images', [])
+        ] or None
+
+        return {
+            'id': video.get('clip_id') or video_id,
+            'title': title,
+            'description': video.get('description'),
+            'thumbnail': thumbnail,
+            'thumbnails': thumbnails,
+            'duration': int_or_none(video.get('clip_duration')),
+            'timestamp': video.get('created_at'),
+        }
+
+    def _extract_formats(self, options, video_id):
+        formats = []
+        for format_id, format_url in options['source'].items():
+            ext = determine_ext(format_url)
+            if ext == 'm3u8':
+                formats.extend(self._extract_m3u8_formats(
+                    format_url, video_id, 'mp4', m3u8_id=format_id,
+                    fatal=False))
+            else:
+                formats.append({
+                    'url': format_url,
+                    'format_id': format_id,
+                })
+        if options.get('language'):
+            for format in formats:
+                format['language'] = options.get('language')
+        self._sort_formats(formats)
+        return formats
+
+
+class GlomexIE(GlomexBaseIE):
+    IE_NAME = 'glomex'
+    IE_DESC = 'Glomex videos'
+    _VALID_URL = r'https?://video\.glomex\.com/[^/]+/(?P<id>v-[^-]+)'
+    # Hard-coded integration ID for video.glomex.com
+    _INTEGRATION_ID = '19syy24xjn1oqlpc'
+
+    _TEST = {
+        'url': 'https://video.glomex.com/sport/v-cb24uwg77hgh-nach-2-0-sieg-guardiola-mit-mancity-vor-naechstem-titel',
+        'md5': 'cec33a943c4240c9cb33abea8c26242e',
+        'info_dict': {
+            'id': 'v-cb24uwg77hgh',
+            'ext': 'mp4',
+            'title': 'md5:38a90cedcfadd72982c81acf13556e0c',
+            'description': 'md5:1ea6b6caff1443fcbbba159e432eedb8',
+            'duration': 29600,
+            'timestamp': 1619895017,
+            'upload_date': '20210501',
+            'age_limit': None,
+        },
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        # Defer to glomex:embed IE: Build and return a player URL using the
+        # matched video ID and the hard-coded integration ID
+        return self.url_result(
+            GlomexEmbedIE.build_player_url(video_id, self._INTEGRATION_ID,
+                                           url),
+            GlomexEmbedIE.ie_key(),
+            video_id
+        )
+
+
+class GlomexEmbedIE(GlomexBaseIE):
+    IE_NAME = 'glomex:embed'
+    IE_DESC = 'Glomex embedded videos'
+    _BASE_PLAYER_URL = 'https://player.glomex.com/integration/1/iframe-player.html'
+    _VALID_URL = r'''(?x)https?://player\.glomex\.com/integration/[^/]+/iframe-player\.html
+        \?(?:(?:integrationId=(?P<integration>[^&#]+)|playlistId=(?P<id>[^&#]+)|[^&=#]+=[^&#]+)&?)+'''
+
+    _TESTS = [{
+        'url': 'https://player.glomex.com/integration/1/iframe-player.html?integrationId=4059a013k56vb2yd&playlistId=v-cfa6lye0dkdd-sf',
+        'info_dict': {
+            'id': 'v-cfa6lye0dkdd-sf',
+            'ext': 'mp4',
+            'timestamp': 1635337199,
+            'duration': 133080,
+            'upload_date': '20211027',
+            'description': 'md5:e741185fc309310ff5d0c789b437be66',
+            'title': 'md5:35647293513a6c92363817a0fb0a7961',
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        'url': 'https://player.glomex.com/integration/1/iframe-player.html?origin=fullpage&integrationId=19syy24xjn1oqlpc&playlistId=rl-vcb49w1fb592p&playlistIndex=0',
+        'info_dict': {
+            'id': 'rl-vcb49w1fb592p',
+        },
+        'playlist_count': 100,
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        'url': 'https://player.glomex.com/integration/1/iframe-player.html?playlistId=cl-bgqaata6aw8x&integrationId=19syy24xjn1oqlpc',
+        'info_dict': {
+            'id': 'cl-bgqaata6aw8x',
+        },
+        'playlist_mincount': 2,
+        'params': {
+            'skip_download': True,
+        },
+    }]
+
+    @classmethod
+    def build_player_url(cls, video_id, integration, origin_url=None):
+        query_string = compat_urllib_parse_urlencode({
+            'playlistId': video_id,
+            'integrationId': integration,
+        })
+        player_url = '%s?%s' % (cls._BASE_PLAYER_URL, query_string)
+        if origin_url is not None:
+            player_url = cls._smuggle_origin_url(player_url, origin_url)
+        return player_url
+
+    @classmethod
+    def _extract_urls(cls, webpage, origin_url):
+        # make the scheme in _VALID_URL optional
+        _URL_RE = r'(?:https?:)?//' + cls._VALID_URL.split('://', 1)[1]
+        # simplify the query string part of _VALID_URL; after extracting iframe
+        # src, the URL will be matched again
+        _URL_RE = _URL_RE.split(r'\?', 1)[0] + r'\?(?:(?!(?P=_q1)).)+'
+        # https://docs.glomex.com/publisher/video-player-integration/javascript-api/
+        EMBED_RE = r'''(?x)
+        (?:
+            <iframe[^>]+?src=(?P<_q1>%(quot_re)s)
+                (?P<url>%(url_re)s)(?P=_q1)|
+            <(?P<html_tag>glomex-player|div)(?:
+                data-integration-id=(?P<_q2>%(quot_re)s)(?P<integration_html>(?:(?!(?P=_q2)).)+)(?P=_q2)|
+                data-playlist-id=(?P<_q3>%(quot_re)s)(?P<id_html>(?:(?!(?P=_q3)).)+)(?P=_q3)|
+                data-glomex-player=(?P<_q4>%(quot_re)s)(?P<glomex_player>true)(?P=_q4)|
+                [^>]*?
+            )+>|
+            # naive parsing of inline scripts for hard-coded integration parameters
+            <(?P<script_tag>script)[^<]*?>(?:
+                (?P<_stjs1>dataset\.)?integrationId\s*(?(_stjs1)=|:)\s*
+                    (?P<_q5>%(quot_re)s)(?P<integration_js>(?:(?!(?P=_q5)).)+)(?P=_q5)\s*(?(_stjs1);|,)?|
+                (?P<_stjs2>dataset\.)?playlistId\s*(?(_stjs2)=|:)\s*
+                    (?P<_q6>%(quot_re)s)(?P<id_js>(?:(?!(?P=_q6)).)+)(?P=_q6)\s*(?(_stjs2);|,)?|
+                (?:\s|.)*?
+            )+</script>
+        )
+        ''' % {'quot_re': r'["\']', 'url_re': _URL_RE}
+        for mobj in re.finditer(EMBED_RE, webpage):
+            url, html_tag, video_id_html, integration_html, glomex_player, \
+                script_tag, video_id_js, integration_js = \
+                mobj.group('url', 'html_tag', 'id_html',
+                           'integration_html', 'glomex_player', 'script_tag',
+                           'id_js', 'integration_js')
+            if url:
+                url = unescapeHTML(url)
+                if url.startswith('//'):
+                    scheme = compat_urllib_parse_urlparse(origin_url).scheme \
+                        if origin_url else 'https'
+                    url = '%s:%s' % (scheme, url)
+                if not cls.suitable(url):
+                    continue
+                yield cls._smuggle_origin_url(url, origin_url)
+            elif html_tag:
+                if html_tag == "div" and not glomex_player:
+                    continue
+                if not video_id_html or not integration_html:
+                    continue
+                yield cls.build_player_url(video_id_html, integration_html,
+                                           origin_url)
+            elif script_tag:
+                if not video_id_js or not integration_js:
+                    continue
+                yield cls.build_player_url(video_id_js, integration_js,
+                                           origin_url)
+
+    def _real_extract(self, url):
+        url, origin_url = self._unsmuggle_origin_url(url)
+        # must return a valid match since it was already tested when selecting the IE
+        try:
+            matches = self._VALID_URL_RE.match(url).groupdict()
+        except AttributeError:
+            matches = re.match(self._VALID_URL, url).groupdict()
+        # id is not enforced in the pattern, so do it now; ditto integration
+        video_id = matches['id']
+        integration = matches['integration']
+        return self._download_and_extract_api_data(video_id, integration,
+                                                   origin_url)

youtube_dl/extractor/youtube.py

@@ -91,12 +91,12 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
             'INNERTUBE_CONTEXT': {
                 'client': {
                     'clientName': 'IOS',
-                    'clientVersion': '19.45.4',
+                    'clientVersion': '20.10.4',
                     'deviceMake': 'Apple',
                     'deviceModel': 'iPhone16,2',
-                    'userAgent': 'com.google.ios.youtube/19.45.4 (iPhone16,2; U; CPU iOS 18_1_0 like Mac OS X;)',
+                    'userAgent': 'com.google.ios.youtube/20.10.4 (iPhone16,2; U; CPU iOS 18_3_2 like Mac OS X;)',
                     'osName': 'iPhone',
-                    'osVersion': '18.1.0.22B83',
+                    'osVersion': '18.3.2.22D82',
                 },
             },
             'INNERTUBE_CONTEXT_CLIENT_NAME': 5,
@@ -109,7 +109,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
             'INNERTUBE_CONTEXT': {
                 'client': {
                     'clientName': 'MWEB',
-                    'clientVersion': '2.20241202.07.00',
+                    'clientVersion': '2.20250311.03.00',
                     # mweb previously did not require PO Token with this UA
                     'userAgent': 'Mozilla/5.0 (iPad; CPU OS 16_7_10 like Mac OS X) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/16.6 Mobile/15E148 Safari/604.1,gzip(gfe)',
                 },
@@ -122,7 +122,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
             'INNERTUBE_CONTEXT': {
                 'client': {
                     'clientName': 'TVHTML5',
-                    'clientVersion': '7.20250120.19.00',
+                    'clientVersion': '7.20250312.16.00',
                     'userAgent': 'Mozilla/5.0 (ChromiumStylePlatform) Cobalt/Version',
                 },
             },
@@ -133,7 +133,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
             'INNERTUBE_CONTEXT': {
                 'client': {
                     'clientName': 'WEB',
-                    'clientVersion': '2.20241126.01.00',
+                    'clientVersion': '2.20250312.04.00',
                 },
             },
             'INNERTUBE_CONTEXT_CLIENT_NAME': 1,
@@ -692,7 +692,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
         'invidious': '|'.join(_INVIDIOUS_SITES),
     }
     _PLAYER_INFO_RE = (
-        r'/s/player/(?P<id>[a-zA-Z0-9_-]{8,})/player',
+        r'/s/player/(?P<id>[a-zA-Z0-9_-]{8,})//(?:tv-)?player',
         r'/(?P<id>[a-zA-Z0-9_-]{8,})/player(?:_ias\.vflset(?:/[a-zA-Z]{2,3}_[a-zA-Z]{2,3})?|-plasma-ias-(?:phone|tablet)-[a-z]{2}_[A-Z]{2}\.vflset)/base\.js$',
         r'\b(?P<id>vfl[a-zA-Z0-9_-]+)\b.*?\.js$',
     )
@@ -1857,7 +1857,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
     def _extract_n_function_code_jsi(self, video_id, jsi, player_id=None):
 
         var_ay = self._search_regex(
-            r'(?:[;\s]|^)\s*(var\s*[\w$]+\s*=\s*"[^"]+"\s*\.\s*split\("\{"\))(?=\s*[,;])',
+            r'(?:[;\s]|^)\s*(var\s*[\w$]+\s*=\s*"(?:\\"|[^"])+"\s*\.\s*split\("\W+"\))(?=\s*[,;])',
             jsi.code, 'useful values', default='')
 
         func_name = self._extract_n_function_name(jsi.code)