Merge efad43c197775eb3ef30b048c8d7b83c1ccbb57c 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/joj.py

@@ -2,6 +2,7 @@
 from __future__ import unicode_literals
 
 import re
+import json
 
 from .common import InfoExtractor
 from ..compat import compat_str
@@ -21,7 +22,8 @@ class JojIE(InfoExtractor):
                     (?P<id>[^/?#^]+)
                 '''
     _TESTS = [{
-        'url': 'https://media.joj.sk/embed/a388ec4c-6019-4a4a-9312-b1bee194e932',
+        'url':
+        'https://media.joj.sk/embed/a388ec4c-6019-4a4a-9312-b1bee194e932',
         'info_dict': {
             'id': 'a388ec4c-6019-4a4a-9312-b1bee194e932',
             'ext': 'mp4',
@@ -64,16 +66,20 @@ class JojIE(InfoExtractor):
                 r'(?s)(?:src|bitrates)\s*=\s*({.+?});', webpage, 'bitrates',
                 default='{}'),
             video_id, transform_source=js_to_json, fatal=False)
+        quality = self._search_regex(
+            r'var qualityLabels = ({.*?});', webpage, 'quality', default=None)
+        quality = json.loads(quality)
 
         formats = []
         for format_url in try_get(bitrates, lambda x: x['mp4'], list) or []:
             if isinstance(format_url, compat_str):
                 height = self._search_regex(
-                    r'(\d+)[pP]\.', format_url, 'height', default=None)
+                    r'-(\d+)p?\.', format_url, 'height', default=None)
                 formats.append({
                     'url': format_url,
                     'format_id': '%sp' % height if height else None,
                     'height': int(height),
+                    'resolution': quality.get(compat_str(height)+'p')
                 })
         if not formats:
             playlist = self._download_xml(
@@ -104,5 +110,5 @@ class JojIE(InfoExtractor):
             'title': title,
             'thumbnail': thumbnail,
             'duration': duration,
-            'formats': formats,
+            'formats': formats
         }

youtube_dl/extractor/markiza.py

@@ -1,21 +1,23 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+# update 15.01.2022 jastrab
 import re
+import json
 
 from .common import InfoExtractor
-from ..compat import compat_str
 from ..utils import (
     orderedSet,
-    parse_duration,
+    url_or_none,
-    try_get,
+    determine_ext
 )
 
 
 class MarkizaIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?videoarchiv\.markiza\.sk/(?:video/(?:[^/]+/)*|embed/)(?P<id>\d+)(?:[_/]|$)'
+    _VALID_URL = r'https:\/\/(?:www\.)?videoarchiv\.markiza\.sk\/(?:video\/(?:[^\/]+\/)*|embed\/)epizoda\/(?P<id>\d+)(?:[\_\/\-]|$)'
     _TESTS = [{
-        'url': 'http://videoarchiv.markiza.sk/video/oteckovia/84723_oteckovia-109',
+        'url': 'http://videoarchiv.markiza.sk/video/oteckovia/\
+            84723_oteckovia-109',
         'md5': 'ada4e9fad038abeed971843aa028c7b0',
         'info_dict': {
             'id': '139078',
@@ -26,54 +28,89 @@ class MarkizaIE(InfoExtractor):
             'duration': 2760,
         },
     }, {
-        'url': 'http://videoarchiv.markiza.sk/video/televizne-noviny/televizne-noviny/85430_televizne-noviny',
+        'url': ' https://videoarchiv.markiza.sk/video/laska-na-prenajom/epizoda/58779-seria-1-epizoda-14',
         'info_dict': {
             'id': '85430',
             'title': 'Televízne noviny',
         },
         'playlist_count': 23,
     }, {
-        'url': 'http://videoarchiv.markiza.sk/video/oteckovia/84723',
+        'url': 'https://videoarchiv.markiza.sk/video/oteckovia/84723',
         'only_matching': True,
     }, {
-        'url': 'http://videoarchiv.markiza.sk/video/84723',
+        'url': 'https://videoarchiv.markiza.sk/video/84723',
         'only_matching': True,
     }, {
-        'url': 'http://videoarchiv.markiza.sk/video/filmy/85190_kamenak',
+        'url': 'https://videoarchiv.markiza.sk/video/filmy/85190_kamenak',
         'only_matching': True,
     }, {
-        'url': 'http://videoarchiv.markiza.sk/video/reflex/zo-zakulisia/84651_pribeh-alzbetky',
+        'url': 'https://videoarchiv.markiza.sk/video/reflex/zo-zakulisia/84651_pribeh-alzbetky',
         'only_matching': True,
     }, {
-        'url': 'http://videoarchiv.markiza.sk/embed/85295',
+        'url': 'https://videoarchiv.markiza.sk/embed/85295',
         'only_matching': True,
     }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+        embed = self._search_regex(
+            r'<iframe src="(https:\/\/media.*?)" style="',
+            webpage, 'embed', fatal=False)
+        webpage = self._download_webpage(embed, video_id)
+        data = self._search_regex(
+            r'processAdTagModifier\((\{.*)\), {"video":',
+            webpage, 'embed', fatal=False)
+        data2 = self._search_regex(
+            r'processAdTagModifier\(\{.*\), ({"video":.*)\);',
+            webpage, 'embed', fatal=False)
+        data = re.sub('\\\\/', '/', data)
+        data2 = re.sub('\\\\/', '/', data2)
+        info = json.loads(data)
+        info2 = json.loads(data2)
 
-        data = self._download_json(
+        formats = []
-            'http://videoarchiv.markiza.sk/json/video_jwplayer7.json',
+        for format_id, format_list in info['tracks'].items():
-            video_id, query={'id': video_id})
+            if not isinstance(format_list, list):
-
+                format_list = [format_list]
-        info = self._parse_jwplayer_data(data, m3u8_id='hls', mpd_id='dash')
+            for format_dict in format_list:
-
+                if not isinstance(format_dict, dict):
-        if info.get('_type') == 'playlist':
+                    continue
-            info.update({
+                format_url = url_or_none(format_dict.get('src'))
-                'id': video_id,
+                format_type = format_dict.get('type')
-                'title': try_get(
+                ext = determine_ext(format_url)
-                    data, lambda x: x['details']['name'], compat_str),
+                if (format_type == 'application/x-mpegURL'
-            })
+                        or format_id == 'HLS' or ext == 'm3u8'):
-        else:
+                    formats.extend(self._extract_m3u8_formats(
-            info['duration'] = parse_duration(
+                        format_url, video_id, 'mp4',
-                try_get(data, lambda x: x['details']['duration'], compat_str))
+                        entry_protocol='m3u8_native', m3u8_id='hls',
-        return info
+                        fatal=False))
+                elif (format_type == 'application/dash+xml'
+                      or format_id == 'DASH' or ext == 'mpd'):
+                    formats.extend(self._extract_mpd_formats(
+                        format_url, video_id, mpd_id='dash', fatal=False))
+                else:
+                    formats.append({
+                        'url': format_url,
+                    })
+        thumbnail = info.get('plugins').get('thumbnails').get('url')
+        thumbnail = re.sub('$Num$', '001', thumbnail)
+        duration = info.get('duration')
+        title2 = info2.get('video').get('title')
+        title = info2.get('video').get('custom').get('show_title')
+        return {
+            'id': video_id,
+            'title': title + ' - ' + title2,
+            'thumbnail': thumbnail,
+            'duration': duration,
+            'formats': formats
+        }
 
 
 class MarkizaPageIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?(?:(?:[^/]+\.)?markiza|tvnoviny)\.sk/(?:[^/]+/)*(?P<id>\d+)_'
     _TESTS = [{
-        'url': 'http://www.markiza.sk/soubiz/zahranicny/1923705_oteckovia-maju-svoj-den-ti-slavni-nie-su-o-nic-menej-rozkosni',
+        'url': 'https://www.markiza.sk/soubiz/zahranicny/1923705_oteckovia-maju-svoj-den-ti-slavni-nie-su-o-nic-menej-rozkosni',
         'md5': 'ada4e9fad038abeed971843aa028c7b0',
         'info_dict': {
             'id': '139355',
@@ -87,29 +124,30 @@ class MarkizaPageIE(InfoExtractor):
             'skip_download': True,
         },
     }, {
-        'url': 'http://dajto.markiza.sk/filmy-a-serialy/1774695_frajeri-vo-vegas',
+        'url': 'https://dajto.markiza.sk/filmy-a-serialy/1774695_frajeri-vo-vegas',
         'only_matching': True,
     }, {
-        'url': 'http://superstar.markiza.sk/aktualne/1923870_to-je-ale-telo-spevacka-ukazala-sexy-postavicku-v-bikinach',
+        'url': 'https://superstar.markiza.sk/aktualne/\
+        1923870_to-je-ale-telo-spevacka-ukazala-sexy-postavicku-v-bikinach',
         'only_matching': True,
     }, {
-        'url': 'http://hybsa.markiza.sk/aktualne/1923790_uzasna-atmosfera-na-hybsa-v-poprade-superstaristi-si-prve-koncerty-pred-davom-ludi-poriadne-uzili',
+        'url': 'https://hybsa.markiza.sk/aktualne/1923790_uzasna-atmosfera-na-hybsa-v-poprade-superstaristi-si-prve-koncerty-pred-davom-ludi-poriadne-uzili',
         'only_matching': True,
     }, {
-        'url': 'http://doma.markiza.sk/filmy/1885250_moja-vysnivana-svadba',
+        'url': 'https://doma.markiza.sk/filmy/1885250_moja-vysnivana-svadba',
         'only_matching': True,
     }, {
-        'url': 'http://www.tvnoviny.sk/domace/1923887_po-smrti-manzela-ju-cakalo-poriadne-prekvapenie',
+        'url': 'https://www.tvnoviny.sk/domace/1923887_po-smrti-manzela-ju-cakalo-poriadne-prekvapenie',
         'only_matching': True,
     }]
 
     @classmethod
     def suitable(cls, url):
-        return False if MarkizaIE.suitable(url) else super(MarkizaPageIE, cls).suitable(url)
+        return False if MarkizaIE.suitable(url) else \
+            super(MarkizaPageIE, cls).suitable(url)
 
     def _real_extract(self, url):
         playlist_id = self._match_id(url)
-
         webpage = self._download_webpage(
             # Downloading for some hosts (e.g. dajto, doma) fails with 500
             # although everything seems to be OK, so considering 500
@@ -117,7 +155,8 @@ class MarkizaPageIE(InfoExtractor):
             url, playlist_id, expected_status=500)
 
         entries = [
-            self.url_result('http://videoarchiv.markiza.sk/video/%s' % video_id)
+            self.url_result(
+                'http://videoarchiv.markiza.sk/video/%s' % video_id)
             for video_id in orderedSet(re.findall(
                 r'(?:initPlayer_|data-entity=["\']|id=["\']player_)(\d+)',
                 webpage))]

youtube_dl/extractor/rtvs.py

@@ -1,7 +1,13 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
+from ..utils import (
+    url_or_none,
+    determine_ext
+)
 
 
 class RTVSIE(InfoExtractor):
@@ -26,7 +32,8 @@ class RTVSIE(InfoExtractor):
             'id': '63118',
             'ext': 'mp4',
             'title': 'Amaro Džives - Náš deň',
-            'description': 'Galavečer pri príležitosti Medzinárodného dňa Rómov.'
+            'description':
+            'Galavečer pri príležitosti Medzinárodného dňa Rómov.'
         },
         'params': {
             'skip_download': True,
@@ -36,12 +43,66 @@ class RTVSIE(InfoExtractor):
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
-        webpage = self._download_webpage(url, video_id)
+        if url.find('/radio/') != -1:
+            a2 = url.split('/')[-1]
+            a1 = url.split('/')[-2]
+            embed = self._download_webpage(
+                "https://www.rtvs.sk/embed/radio/archive/%s/%s" % (a1, a2),
+                video_id)
+            audio_id = re.search('audio5f.json?id=(?P<id>[^\"]+)', embed)
+            audio_id = audio_id.group('id')
+            info = self._download_json(
+                "https://www.rtvs.sk/json/audio5f.json?id=%s" % audio_id,
+                audio_id)
 
-        playlist_url = self._search_regex(
+            formats = []
-            r'playlist["\']?\s*:\s*(["\'])(?P<url>(?:(?!\1).)+)\1', webpage,
+            formats.append({
-            'playlist url', group='url')
+                'url': info['playlist'][0]['sources'][0]['src'],
+                'format_id':  None,
+                'height': 0
+                })
+            info = info['playlist'][0]
+            return {
+                'id': audio_id,
+                'title': info.get('title'),
+                'thumbnail': info.get('image'),
+                'formats': formats
+            }
+        else:
+            info = self._download_json(
+                "https://www.rtvs.sk/json/archive5f.json?id=%s" % video_id,
+                video_id)
+            info = info.get('clip')
 
-        data = self._download_json(
+            formats = []
-            playlist_url, video_id, 'Downloading playlist')[0]
+            for format_id, format_list in info.items():
-        return self._parse_jwplayer_data(data, video_id=video_id)
+                if not isinstance(format_list, list):
+                    format_list = [format_list]
+                for format_dict in format_list:
+                    if not isinstance(format_dict, dict):
+                        continue
+                    format_url = url_or_none(format_dict.get('src'))
+                    format_type = format_dict.get('type')
+                    ext = determine_ext(format_url)
+                    if (format_type == 'application/x-mpegURL'
+                            or format_id == 'HLS' or ext == 'm3u8'):
+                        formats.extend(self._extract_m3u8_formats(
+                            format_url, video_id, 'mp4',
+                            entry_protocol='m3u8_native', m3u8_id='hls',
+                            fatal=False))
+                    elif (format_type == 'application/dash+xml'
+                          or format_id == 'DASH' or ext == 'mpd'):
+                        pass
+                    else:
+                        formats.append({
+                            'url': format_url,
+                        })
+            formats = sorted(formats, key=lambda i: i['tbr'])
+            dt = info.get('datetime_create')
+            return {
+                'id': video_id,
+                'title': info.get('title') + '-' + dt[:10],
+                'thumbnail': info.get('image'),
+                'description': info.get('description'),
+                'formats': formats
+            }

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)