Merge 507b7c6d4a into f2a4983df7

Closes #11526
Authored by: jshumphrey

yt_dlp/extractor/_extractors.py

@@ -1085,10 +1085,6 @@ from .manoto import (
 )
 from .manyvids import ManyVidsIE
 from .maoritv import MaoriTVIE
-from .markiza import (
-    MarkizaIE,
-    MarkizaPageIE,
-)
 from .massengeschmacktv import MassengeschmackTVIE
 from .masters import MastersIE
 from .matchtv import MatchTVIE

yt_dlp/extractor/archiveorg.py

@@ -205,6 +205,26 @@ class ArchiveOrgIE(InfoExtractor):
                 },
             },
         ],
+    }, {
+        # The reviewbody is None for one of the reviews; just need to extract data without crashing
+        'url': 'https://archive.org/details/gd95-04-02.sbd.11622.sbeok.shnf/gd95-04-02d1t04.shn',
+        'info_dict': {
+            'id': 'gd95-04-02.sbd.11622.sbeok.shnf/gd95-04-02d1t04.shn',
+            'ext': 'mp3',
+            'title': 'Stuck Inside of Mobile with the Memphis Blues Again',
+            'creators': ['Grateful Dead'],
+            'duration': 338.31,
+            'track': 'Stuck Inside of Mobile with the Memphis Blues Again',
+            'description': 'md5:764348a470b986f1217ffd38d6ac7b72',
+            'display_id': 'gd95-04-02d1t04.shn',
+            'location': 'Pyramid Arena',
+            'uploader': 'jon@archive.org',
+            'album': '1995-04-02 - Pyramid Arena',
+            'upload_date': '20040519',
+            'track_number': 4,
+            'release_date': '19950402',
+            'timestamp': 1084927901,
+        },
     }]
 
     @staticmethod
@@ -335,7 +355,7 @@ class ArchiveOrgIE(InfoExtractor):
                 info['comments'].append({
                     'id': review.get('review_id'),
                     'author': review.get('reviewer'),
-                    'text': str_or_none(review.get('reviewtitle'), '') + '\n\n' + review.get('reviewbody'),
+                    'text': join_nonempty('reviewtitle', 'reviewbody', from_dict=review, delim='\n\n'),
                     'timestamp': unified_timestamp(review.get('createdate')),
                     'parent': 'root'})
 

yt_dlp/extractor/facebook.py

@@ -563,13 +563,13 @@ class FacebookIE(InfoExtractor):
                 return extract_video_data(try_get(
                     js_data, lambda x: x['jsmods']['instances'], list) or [])
 
-        def extract_dash_manifest(video, formats):
+        def extract_dash_manifest(vid_data, formats, mpd_url=None):
             dash_manifest = traverse_obj(
-                video, 'dash_manifest', 'playlist', 'dash_manifest_xml_string', expected_type=str)
+                vid_data, 'dash_manifest', 'playlist', 'dash_manifest_xml_string', 'manifest_xml', expected_type=str)
             if dash_manifest:
                 formats.extend(self._parse_mpd_formats(
                     compat_etree_fromstring(urllib.parse.unquote_plus(dash_manifest)),
-                    mpd_url=url_or_none(video.get('dash_manifest_url'))))
+                    mpd_url=url_or_none(video.get('dash_manifest_url')) or mpd_url))
 
         def process_formats(info):
             # Downloads with browser's User-Agent are rate limited. Working around
@@ -619,9 +619,12 @@ class FacebookIE(InfoExtractor):
                         video = video['creation_story']
                         video['owner'] = traverse_obj(video, ('short_form_video_context', 'video_owner'))
                         video.update(reel_info)
-                    fmt_data = traverse_obj(video, ('videoDeliveryLegacyFields', {dict})) or video
+
                     formats = []
                     q = qualities(['sd', 'hd'])
+
+                    # Legacy formats extraction
+                    fmt_data = traverse_obj(video, ('videoDeliveryLegacyFields', {dict})) or video
                     for key, format_id in (('playable_url', 'sd'), ('playable_url_quality_hd', 'hd'),
                                            ('playable_url_dash', ''), ('browser_native_hd_url', 'hd'),
                                            ('browser_native_sd_url', 'sd')):
@@ -629,7 +632,7 @@ class FacebookIE(InfoExtractor):
                         if not playable_url:
                             continue
                         if determine_ext(playable_url) == 'mpd':
-                            formats.extend(self._extract_mpd_formats(playable_url, video_id))
+                            formats.extend(self._extract_mpd_formats(playable_url, video_id, fatal=False))
                         else:
                             formats.append({
                                 'format_id': format_id,
@@ -638,6 +641,28 @@ class FacebookIE(InfoExtractor):
                                 'url': playable_url,
                             })
                     extract_dash_manifest(fmt_data, formats)
+
+                    # New videoDeliveryResponse formats extraction
+                    fmt_data = traverse_obj(video, ('videoDeliveryResponseFragment', 'videoDeliveryResponseResult'))
+                    mpd_urls = traverse_obj(fmt_data, ('dash_manifest_urls', ..., 'manifest_url', {url_or_none}))
+                    dash_manifests = traverse_obj(fmt_data, ('dash_manifests', lambda _, v: v['manifest_xml']))
+                    for idx, dash_manifest in enumerate(dash_manifests):
+                        extract_dash_manifest(dash_manifest, formats, mpd_url=traverse_obj(mpd_urls, idx))
+                    if not dash_manifests:
+                        # Only extract from MPD URLs if the manifests are not already provided
+                        for mpd_url in mpd_urls:
+                            formats.extend(self._extract_mpd_formats(mpd_url, video_id, fatal=False))
+                    for prog_fmt in traverse_obj(fmt_data, ('progressive_urls', lambda _, v: v['progressive_url'])):
+                        format_id = traverse_obj(prog_fmt, ('metadata', 'quality', {str.lower}))
+                        formats.append({
+                            'format_id': format_id,
+                            # sd, hd formats w/o resolution info should be deprioritized below DASH
+                            'quality': q(format_id) - 3,
+                            'url': prog_fmt['progressive_url'],
+                        })
+                    for m3u8_url in traverse_obj(fmt_data, ('hls_playlist_urls', ..., 'hls_playlist_url', {url_or_none})):
+                        formats.extend(self._extract_m3u8_formats(m3u8_url, video_id, 'mp4', fatal=False, m3u8_id='hls'))
+
                     if not formats:
                         # Do not append false positive entry w/o any formats
                         return

yt_dlp/extractor/markiza.py

@@ -1,123 +0,0 @@
-import re
-
-from .common import InfoExtractor
-from ..utils import (
-    orderedSet,
-    parse_duration,
-    try_get,
-)
-
-
-class MarkizaIE(InfoExtractor):
-    _WORKING = False
-    _VALID_URL = r'https?://(?:www\.)?videoarchiv\.markiza\.sk/(?:video/(?:[^/]+/)*|embed/)(?P<id>\d+)(?:[_/]|$)'
-    _TESTS = [{
-        'url': 'http://videoarchiv.markiza.sk/video/oteckovia/84723_oteckovia-109',
-        'md5': 'ada4e9fad038abeed971843aa028c7b0',
-        'info_dict': {
-            'id': '139078',
-            'ext': 'mp4',
-            'title': 'Oteckovia 109',
-            'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
-            'thumbnail': r're:^https?://.*\.jpg$',
-            'duration': 2760,
-        },
-    }, {
-        'url': 'http://videoarchiv.markiza.sk/video/televizne-noviny/televizne-noviny/85430_televizne-noviny',
-        'info_dict': {
-            'id': '85430',
-            'title': 'Televízne noviny',
-        },
-        'playlist_count': 23,
-    }, {
-        'url': 'http://videoarchiv.markiza.sk/video/oteckovia/84723',
-        'only_matching': True,
-    }, {
-        'url': 'http://videoarchiv.markiza.sk/video/84723',
-        'only_matching': True,
-    }, {
-        'url': 'http://videoarchiv.markiza.sk/video/filmy/85190_kamenak',
-        'only_matching': True,
-    }, {
-        'url': 'http://videoarchiv.markiza.sk/video/reflex/zo-zakulisia/84651_pribeh-alzbetky',
-        'only_matching': True,
-    }, {
-        'url': 'http://videoarchiv.markiza.sk/embed/85295',
-        'only_matching': True,
-    }]
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-
-        data = self._download_json(
-            'http://videoarchiv.markiza.sk/json/video_jwplayer7.json',
-            video_id, query={'id': video_id})
-
-        info = self._parse_jwplayer_data(data, m3u8_id='hls', mpd_id='dash')
-
-        if info.get('_type') == 'playlist':
-            info.update({
-                'id': video_id,
-                'title': try_get(
-                    data, lambda x: x['details']['name'], str),
-            })
-        else:
-            info['duration'] = parse_duration(
-                try_get(data, lambda x: x['details']['duration'], str))
-        return info
-
-
-class MarkizaPageIE(InfoExtractor):
-    _WORKING = False
-    _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',
-        'md5': 'ada4e9fad038abeed971843aa028c7b0',
-        'info_dict': {
-            'id': '139355',
-            'ext': 'mp4',
-            'title': 'Oteckovia 110',
-            'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
-            'thumbnail': r're:^https?://.*\.jpg$',
-            'duration': 2604,
-        },
-        'params': {
-            'skip_download': True,
-        },
-    }, {
-        'url': 'http://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',
-        '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',
-        'only_matching': True,
-    }, {
-        'url': 'http://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',
-        'only_matching': True,
-    }]
-
-    @classmethod
-    def suitable(cls, url):
-        return False if MarkizaIE.suitable(url) else super().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
-            # status code to be expected.
-            url, playlist_id, expected_status=500)
-
-        entries = [
-            self.url_result(f'http://videoarchiv.markiza.sk/video/{video_id}')
-            for video_id in orderedSet(re.findall(
-                r'(?:initPlayer_|data-entity=["\']|id=["\']player_)(\d+)',
-                webpage))]
-
-        return self.playlist_result(entries, playlist_id)

yt_dlp/extractor/nova.py

@@ -6,6 +6,7 @@ from ..utils import (
     determine_ext,
     int_or_none,
     js_to_json,
+    strip_or_none,
     traverse_obj,
     unified_strdate,
     url_or_none,
@@ -13,7 +14,12 @@ from ..utils import (
 
 
 class NovaEmbedIE(InfoExtractor):
-    _VALID_URL = r'https?://media(?:tn)?\.cms\.nova\.cz/embed/(?P<id>[^/?#&]+)'
+    _DOMAINS = [
+        r'media(?:tn)?\.cms\.nova\.cz',
+        r'media\.cms\.(?:markiza|tvnoviny)\.sk',
+    ]
+    _VALID_URL = [rf'https?://{domain}/embed/(?P<id>[^/?#&"\']+)' for domain in _DOMAINS]
+    _EMBED_REGEX = [rf'(?x)<iframe[^>]+\b(?:data-)?src=["\'](?P<url>{url})' for url in _VALID_URL]
     _TESTS = [{
         'url': 'https://media.cms.nova.cz/embed/8o0n0r?autoplay=1',
         'info_dict': {
@@ -43,19 +49,179 @@ class NovaEmbedIE(InfoExtractor):
             'id': 'EU5ELEsmOHt',
             'ext': 'mp4',
             'title': 'Haptické křeslo, bionická ruka nebo roboti. Reportérka se podívala na Týden inovací',
-            'thumbnail': r're:^https?://.*\.jpg',
+            'thumbnail': r're:^https?://cloudia\.cms\.nova\.cz/.+',
             'duration': 1780,
         },
         'params': {'skip_download': 'm3u8'},
     }]
+    _WEBPAGE_TESTS = [{
+        'url': 'http://www.markiza.sk/soubiz/zahranicny/1923705_oteckovia-maju-svoj-den-ti-slavni-nie-su-o-nic-menej-rozkosni',
+        'md5': 'a478390ea7f36aeb36004a107db8b031',
+        'info_dict': {
+            'id': '4q3zP2DsORO',
+            'ext': 'mp4',
+            'title': 'Oteckovia 110',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'duration': 2603,
+        },
+    }, {
+        'url': 'https://tvnoviny.sk/domace/clanok/141815-byvaly-sportovec-udajne-vyrabal-mast-z-marihuany-sud-mu-vymeral-20-rocny-trest-a-vzal-aj-rodinny-dom',
+        'md5': '51de0754352a36b4d623f98c9636a5e1',
+        'info_dict': {
+            'id': '2LcfYRqGuYP',
+            'ext': 'mp4',
+            'title': 'Marihuanový mastičkár si vypočul vysoký trest a prepad majetku',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'duration': 119,
+        },
+    }, {
+        'url': 'https://tvnoviny.sk/domace/clanok/144055-robert-z-kosic-dostal-najnizsi-mozny-trest-za-to-co-spravil-je-to-aj-tak-vela-tvrdia-blizki',
+        'md5': 'c9a8467b37951877336a9ae6309558b0',
+        'info_dict': {
+            'id': '82N7FrJK7cR',
+            'ext': 'mp4',
+            'title': 'Robovi z Košíc znížili trest za marihuanu, odsúdili ho na päť rokov',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'duration': 152,
+        },
+    }, {
+        'url': 'https://tvnoviny.sk/domace/clanok/338907-preco-sa-mnozia-utoky-tinedzerov-podla-psychologiciek-je-za-tym-rastuca-frustracia',
+        'md5': '869b589e99d7c19dd66f024a7d088502',
+        'info_dict': {
+            'id': 'DeiezcjCJmg',
+            'ext': 'mp4',
+            'title': '2022-11-03-TN-2-Nasilie-medzi-mladymi',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'duration': 142,
+        },
+    }, {
+        'url': 'http://tvnoviny.sk/domace/clanok/890183-vlada-chysta-postavit-novu-nemocnicu-v-presove-informoval-premier-robert-fico',
+        'md5': 'b9ef0b4917deee2c930f2248b568a90c',
+        'info_dict': {
+            'id': '7VCyuyfGsNZ',
+            'ext': 'mp4',
+            'title': '2024-04-15-PTN-1-Co-caka-zdravotnictvo',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'duration': 137,
+        },
+    }, {
+        'url': 'https://www.markiza.sk/live/1-markiza',
+        'info_dict': {
+            'id': 'markiza-live',
+            'ext': 'mp4',
+            'title': r're:^CRA Markiza SD \d{4}-\d{2}-\d{2} \d{2}:\d{2}$',
+            'thumbnail': r're:^https?://cloudia\.cms\.markiza\.sk/.+',
+            'live_status': 'is_live',
+        },
+    }, {
+        'url': 'http://www.tvnoviny.sk/domace/1923887_po-smrti-manzela-ju-cakalo-poriadne-prekvapenie',
+        'md5': 'e3e0f1e98172ea64147cada308276df8',
+        'info_dict': {
+            'id': 'JxqRvQkFwHK',
+            'ext': 'mp4',
+            'title': 'Po smrti manžela ju čakalo prekvapenie',
+            'thumbnail': r're:^https?://.*\.(?:jpg)',
+            'duration': 108,
+        },
+    }, {
+        'url': 'http://videoarchiv.markiza.sk/video/reflex/zo-zakulisia/84651_pribeh-alzbetky',
+        'md5': 'b40d04d5cb4cf529e2ff14d6726a3548',
+        'info_dict': {
+            'id': '9ZnlOQp2MRa',
+            'ext': 'mp4',
+            'title': 'Príbeh Alžbetky',
+            'thumbnail': r're:^https?://.*\.(?:jpg)',
+            'duration': 361,
+        },
+    }, {
+        'url': 'https://www.markiza.sk/relacie/superstar/clanok/549972-v-zakulisi-superstar-to-bolo-obcas-drsne-moderator-priznal-ze-musel-pouzit-aj-hrubu-silu',
+        'info_dict': {
+            'id': '549972-v-zakulisi-superstar-to-bolo-obcas-drsne-moderator-priznal-ze-musel-pouzit-aj-hrubu-silu',
+            'title': 'V zákulisí SuperStar to bolo občas drsné. Moderátor priznal, že musel použiť aj hrubú silu | TV Markíza',
+            'description': 'md5:02e240e302bddfd0cd352bc886d95161',
+            'thumbnail': r're:^https?://cmesk-ott-images-avod\.ssl\.cdn\.cra\.cz/.+',
+            'age_limit': 0,
+        },
+        'playlist_count': 2,
+    }, {
+        'url': 'https://voyo.markiza.sk/filmy/6702-vysnivana-svadba',
+        'info_dict': {
+            'id': '20kSOHBD8DQ',
+            'title': 'Vysnívaná svadba - 0000',
+            'thumbnail': r're:^https?://.*\.(?:jpg)',
+            'duration': 4924,
+        },
+        'params': {
+            'skip_download': True,
+            'ignore_no_formats_error': True,
+        },
+        'expected_warnings': [
+            'Requested format is not available',
+            'This video is DRM protected',
+        ],
+        'skip': 'premium member only',
+    }, {
+        # Another URLs:
+        #   http://videoarchiv.markiza.sk/video/84723
+        'url': 'http://videoarchiv.markiza.sk/video/oteckovia/84723_oteckovia-109',
+        'info_dict': {
+            'id': '2a5fQmhjvYm',
+            'title': 'Oteckovia 109',
+            'thumbnail': r're:^https?://.*\.(?:jpg)',
+            'duration': 2759,
+        },
+        'params': {
+            'skip_download': True,
+            'ignore_no_formats_error': True,
+        },
+        'expected_warnings': [
+            'Requested format is not available',
+            'This video is DRM protected',
+        ],
+        'skip': 'premium member only',
+    }, {
+        'url': 'https://voyo.markiza.sk/filmy/1377-frajeri-vo-vegas#player-fullscreen',
+        'info_dict': {
+            'id': '1377-frajeri-vo-vegas#player-fullscreen',
+            'title': 'Frajeri vo Vegas | Voyo',
+            'description': 'md5:7f16168f669f144986d862312949627c',
+            'thumbnail': r're:^https?://cmesk-ott-images-svod\.ssl\.cdn\.cra\.cz/.+',
+            'age_limit': 0,
+        },
+        'playlist': [{
+            'info_dict': {
+                'id': 'K8H4IvKNBbw',
+                'ext': 'mp4',
+                'title': 'frajeri-vo-vegas-hd-15_frajeri-trailer',
+                'duration': 90,
+                'thumbnail': r're:^https?://.*\.(?:jpg)',
+            },
+        },
+            # BUG: The 2nd item (CDjGcqcCYKy) is the movie itself and it's DRM-protected.
+            #      The "ext" field can neither be here nor omitted.
+        ],
+        'playlist_count': 2,
+        'params': {
+            'skip_download': True,
+            'ignore_no_formats_error': True,
+        },
+        'expected_warnings': [
+            'Requested format is not available',
+            'This video is DRM protected',
+        ],
+        'skip': 'premium member only',
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
         webpage = self._download_webpage(url, video_id)
+        if 'player_not_logged_in' in webpage:
+            self.raise_login_required()
 
         has_drm = False
         duration = None
+        is_live = False
         formats = []
 
         def process_format_list(format_list, format_id=''):
@@ -77,11 +243,11 @@ class NovaEmbedIE(InfoExtractor):
                     formats.extend(self._extract_m3u8_formats(
                         format_url, video_id, 'mp4',
                         entry_protocol='m3u8_native', m3u8_id='hls',
-                        fatal=False))
+                        fatal=False, headers={'Referer': url}))
                 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))
+                        format_url, video_id, mpd_id='dash', fatal=False, headers={'Referer': url}))
                 else:
                     formats.append({
                         'url': format_url,
@@ -93,6 +259,7 @@ class NovaEmbedIE(InfoExtractor):
             for src in traverse_obj(player, ('lib', 'source', 'sources', ...)):
                 process_format_list(src)
             duration = traverse_obj(player, ('sourceInfo', 'duration', {int_or_none}))
+            is_live = player.get('isLive', False)
         if not formats and not has_drm:
             # older code path, in use before August 2023
             player = self._parse_json(
@@ -108,11 +275,11 @@ class NovaEmbedIE(InfoExtractor):
         if not formats and has_drm:
             self.report_drm(video_id)
 
-        title = self._og_search_title(
+        title = strip_or_none(self._og_search_title(
             webpage, default=None) or self._search_regex(
             (r'<value>(?P<title>[^<]+)',
              r'videoTitle\s*:\s*(["\'])(?P<value>(?:(?!\1).)+)\1'), webpage,
-            'title', group='value')
+            'title', group='value'))
         thumbnail = self._og_search_thumbnail(
             webpage, default=None) or self._search_regex(
             r'poster\s*:\s*(["\'])(?P<value>(?:(?!\1).)+)\1', webpage,
@@ -127,6 +294,8 @@ class NovaEmbedIE(InfoExtractor):
             'thumbnail': thumbnail,
             'duration': duration,
             'formats': formats,
+            'http_headers': {'Referer': url},
+            'is_live': is_live,
         }