[ie] Add browser impersonation support to more extractors (#16029)

Closes #7001, Closes #7444, Closes #16004
Authored by: bashonly

yt_dlp/extractor/aenetworks.py

@@ -5,10 +5,12 @@ from ..utils import (
     ExtractorError,
     GeoRestrictedError,
     int_or_none,
+    make_archive_id,
     remove_start,
-    traverse_obj,
     update_url_query,
+    url_or_none,
 )
+from ..utils.traversal import traverse_obj
 
 
 class AENetworksBaseIE(ThePlatformIE):  # XXX: Do not subclass from concrete IE
@@ -29,6 +31,19 @@ class AENetworksBaseIE(ThePlatformIE):  # XXX: Do not subclass from concrete IE
         'historyvault.com': (None, 'historyvault', None),
         'biography.com': (None, 'biography', None),
     }
+    _GRAPHQL_QUERY = '''
+        query getUserVideo($videoId: ID!) {
+            video(id: $videoId) {
+                title
+                publicUrl
+                programId
+                tvSeasonNumber
+                tvSeasonEpisodeNumber
+                series {
+                    title
+                }
+            }
+        }'''
 
     def _extract_aen_smil(self, smil_url, video_id, auth=None):
         query = {
@@ -73,19 +88,39 @@ class AENetworksBaseIE(ThePlatformIE):  # XXX: Do not subclass from concrete IE
 
     def _extract_aetn_info(self, domain, filter_key, filter_value, url):
         requestor_id, brand, software_statement = self._DOMAIN_MAP[domain]
+        if filter_key == 'canonical':
+            webpage = self._download_webpage(url, filter_value)
+            graphql_video_id = self._search_regex(
+                r'<meta\b[^>]+\bcontent="[^"]*\btpid/(\d+)"', webpage,
+                'id') or self._html_search_meta('videoId', webpage, 'GraphQL video ID', fatal=True)
+        else:
+            graphql_video_id = filter_value
+
         result = self._download_json(
-            f'https://feeds.video.aetnd.com/api/v2/{brand}/videos',
-            filter_value, query={f'filter[{filter_key}]': filter_value})
-        result = traverse_obj(
-            result, ('results',
-                     lambda k, v: k == 0 and v[filter_key] == filter_value),
-            get_all=False)
-        if not result:
+            'https://yoga.appsvcs.aetnd.com/', graphql_video_id,
+            query={
+                'brand': brand,
+                'mode': 'live',
+                'platform': 'web',
+            },
+            data=json.dumps({
+                'operationName': 'getUserVideo',
+                'variables': {
+                    'videoId': graphql_video_id,
+                },
+                'query': self._GRAPHQL_QUERY,
+            }).encode(),
+            headers={
+                'Content-Type': 'application/json',
+            })
+
+        result = traverse_obj(result, ('data', 'video', {dict}))
+        media_url = traverse_obj(result, ('publicUrl', {url_or_none}))
+        if not media_url:
             raise ExtractorError('Show not found in A&E feed (too new?)', expected=True,
                                  video_id=remove_start(filter_value, '/'))
         title = result['title']
-        video_id = result['id']
-        media_url = result['publicUrl']
+        video_id = result['programId']
         theplatform_metadata = self._download_theplatform_metadata(self._search_regex(
             r'https?://link\.theplatform\.com/s/([^?]+)', media_url, 'theplatform_path'), video_id)
         info = self._parse_theplatform_metadata(theplatform_metadata)
@@ -100,9 +135,13 @@ class AENetworksBaseIE(ThePlatformIE):  # XXX: Do not subclass from concrete IE
         info.update(self._extract_aen_smil(media_url, video_id, auth))
         info.update({
             'title': title,
-            'series': result.get('seriesName'),
-            'season_number': int_or_none(result.get('tvSeasonNumber')),
-            'episode_number': int_or_none(result.get('tvSeasonEpisodeNumber')),
+            'display_id': graphql_video_id,
+            '_old_archive_ids': [make_archive_id(self, graphql_video_id)],
+            **traverse_obj(result, {
+                'series': ('series', 'title', {str}),
+                'season_number': ('tvSeasonNumber', {int_or_none}),
+                'episode_number': ('tvSeasonEpisodeNumber', {int_or_none}),
+            }),
         })
         return info
 
@@ -116,7 +155,7 @@ class AENetworksIE(AENetworksBaseIE):
         (?:shows/[^/?#]+/)?videos/[^/?#]+
     )'''
     _TESTS = [{
-        'url': 'http://www.history.com/shows/mountain-men/season-1/episode-1',
+        'url': 'https://www.history.com/shows/mountain-men/season-1/episode-1',
         'info_dict': {
             'id': '22253814',
             'ext': 'mp4',
@@ -139,11 +178,11 @@ class AENetworksIE(AENetworksBaseIE):
         },
         'params': {'skip_download': 'm3u8'},
         'add_ie': ['ThePlatform'],
-        'skip': 'Geo-restricted - This content is not available in your location.',
+        'skip': 'This content requires a valid, unexpired auth token',
     }, {
-        'url': 'http://www.aetv.com/shows/duck-dynasty/season-9/episode-1',
+        'url': 'https://www.aetv.com/shows/duck-dynasty/season-9/episode-1',
         'info_dict': {
-            'id': '600587331957',
+            'id': '147486',
             'ext': 'mp4',
             'title': 'Inlawful Entry',
             'description': 'md5:57c12115a2b384d883fe64ca50529e08',
@@ -160,6 +199,8 @@ class AENetworksIE(AENetworksBaseIE):
             'season_number': 9,
             'series': 'Duck Dynasty',
             'age_limit': 0,
+            'display_id': '600587331957',
+            '_old_archive_ids': ['aenetworks 600587331957'],
         },
         'params': {'skip_download': 'm3u8'},
         'add_ie': ['ThePlatform'],
@@ -186,6 +227,7 @@ class AENetworksIE(AENetworksBaseIE):
         },
         'params': {'skip_download': 'm3u8'},
         'add_ie': ['ThePlatform'],
+        'skip': '404 Not Found',
     }, {
         'url': 'https://www.aetv.com/specials/hunting-jonbenets-killer-the-untold-story',
         'info_dict': {
@@ -209,6 +251,7 @@ class AENetworksIE(AENetworksBaseIE):
         },
         'params': {'skip_download': 'm3u8'},
         'add_ie': ['ThePlatform'],
+        'skip': 'This content requires a valid, unexpired auth token',
     }, {
         'url': 'http://www.fyi.tv/shows/tiny-house-nation/season-1/episode-8',
         'only_matching': True,
@@ -259,7 +302,7 @@ class AENetworksListBaseIE(AENetworksBaseIE):
         domain, slug = self._match_valid_url(url).groups()
         _, brand, _ = self._DOMAIN_MAP[domain]
         playlist = self._call_api(self._RESOURCE, slug, brand, self._FIELDS)
-        base_url = f'http://watch.{domain}'
+        base_url = f'https://watch.{domain}'
 
         entries = []
         for item in (playlist.get(self._ITEMS_KEY) or []):

yt_dlp/extractor/learningonscreen.py

@@ -29,7 +29,7 @@ class LearningOnScreenIE(InfoExtractor):
     }]
 
     def _real_initialize(self):
-        if not self._get_cookies('https://learningonscreen.ac.uk/').get('PHPSESSID-BOB-LIVE'):
+        if not self._get_cookies('https://learningonscreen.ac.uk/').get('PHPSESSID-LOS-LIVE'):
             self.raise_login_required(method='session_cookies')
 
     def _real_extract(self, url):

yt_dlp/extractor/mixcloud.py

@@ -25,7 +25,7 @@ class MixcloudBaseIE(InfoExtractor):
     %s
   }
 }''' % (lookup_key, username, f', slug: "{slug}"' if slug else '', object_fields),  # noqa: UP031
-            })['data'][lookup_key]
+            }, impersonate=True)['data'][lookup_key]
 
 
 class MixcloudIE(MixcloudBaseIE):

yt_dlp/extractor/opencast.py

@@ -33,7 +33,8 @@ class OpencastBaseIE(InfoExtractor):
                             vid\.igb\.illinois\.edu|
                             cursosabertos\.c3sl\.ufpr\.br|
                             mcmedia\.missioncollege\.org|
-                            clases\.odon\.edu\.uy
+                            clases\.odon\.edu\.uy|
+                            oc-p\.uni-jena\.de
                         )'''
     _UUID_RE = r'[\da-fA-F]{8}-[\da-fA-F]{4}-[\da-fA-F]{4}-[\da-fA-F]{4}-[\da-fA-F]{12}'
 
@@ -106,7 +107,7 @@ class OpencastBaseIE(InfoExtractor):
 
 class OpencastIE(OpencastBaseIE):
     _VALID_URL = rf'''(?x)
-        https?://(?P<host>{OpencastBaseIE._INSTANCES_RE})/paella/ui/watch\.html\?
+        https?://(?P<host>{OpencastBaseIE._INSTANCES_RE})/paella[0-9]*/ui/watch\.html\?
         (?:[^#]+&)?id=(?P<id>{OpencastBaseIE._UUID_RE})'''
 
     _API_BASE = 'https://%s/search/episode.json?id=%s'
@@ -131,8 +132,12 @@ class OpencastIE(OpencastBaseIE):
 
     def _real_extract(self, url):
         host, video_id = self._match_valid_url(url).group('host', 'id')
-        return self._parse_mediapackage(
-            self._call_api(host, video_id)['search-results']['result']['mediapackage'])
+        response = self._call_api(host, video_id)
+        package = traverse_obj(response, (
+            ('search-results', 'result'),
+            ('result', ...),  # Path needed for oc-p.uni-jena.de
+            'mediapackage', {dict}, any)) or {}
+        return self._parse_mediapackage(package)
 
 
 class OpencastPlaylistIE(OpencastBaseIE):

yt_dlp/extractor/soundcloud.py

@@ -6,6 +6,7 @@ import re
 from .common import InfoExtractor, SearchInfoExtractor
 from ..networking import HEADRequest
 from ..networking.exceptions import HTTPError
+from ..networking.impersonate import ImpersonateTarget
 from ..utils import (
     ExtractorError,
     float_or_none,
@@ -833,6 +834,30 @@ class SoundcloudPagedPlaylistBaseIE(SoundcloudBaseIE):
             'entries': self._entries(base_url, playlist_id),
         }
 
+    @functools.cached_property
+    def _browser_impersonate_target(self):
+        available_targets = self._downloader._get_available_impersonate_targets()
+        if not available_targets:
+            # impersonate=True gives a generic warning when no impersonation targets are available
+            return True
+
+        # Any browser target older than chrome-116 is 403'd by Datadome
+        MIN_SUPPORTED_TARGET = ImpersonateTarget('chrome', '116', 'windows', '10')
+        version_as_float = lambda x: float(x.version) if x.version else 0
+
+        # Always try to use the newest Chrome target available
+        filtered = sorted([
+            target[0] for target in available_targets
+            if target[0].client == 'chrome' and target[0].os in ('windows', 'macos')
+        ], key=version_as_float)
+
+        if not filtered or version_as_float(filtered[-1]) < version_as_float(MIN_SUPPORTED_TARGET):
+            # All available targets are inadequate or newest available Chrome target is too old, so
+            # warn the user to upgrade their dependency to a version with the minimum supported target
+            return MIN_SUPPORTED_TARGET
+
+        return filtered[-1]
+
     def _entries(self, url, playlist_id):
         # Per the SoundCloud documentation, the maximum limit for a linked partitioning query is 200.
         # https://developers.soundcloud.com/blog/offset-pagination-deprecated
@@ -847,7 +872,9 @@ class SoundcloudPagedPlaylistBaseIE(SoundcloudBaseIE):
                 try:
                     response = self._call_api(
                         url, playlist_id, query=query, headers=self._HEADERS,
-                        note=f'Downloading track page {i + 1}')
+                        note=f'Downloading track page {i + 1}',
+                        # See: https://github.com/yt-dlp/yt-dlp/issues/15660
+                        impersonate=self._browser_impersonate_target)
                     break
                 except ExtractorError as e:
                     # Downloading page may result in intermittent 502 HTTP error

yt_dlp/extractor/spankbang.py

@@ -3,6 +3,7 @@ import re
 from .common import InfoExtractor
 from ..utils import (
     ExtractorError,
+    clean_html,
     determine_ext,
     merge_dicts,
     parse_duration,
@@ -12,6 +13,7 @@ from ..utils import (
     urlencode_postdata,
     urljoin,
 )
+from ..utils.traversal import find_element, traverse_obj, trim_str
 
 
 class SpankBangIE(InfoExtractor):
@@ -122,7 +124,7 @@ class SpankBangIE(InfoExtractor):
                 }), headers={
                     'Referer': url,
                     'X-Requested-With': 'XMLHttpRequest',
-                })
+                }, impersonate=True)
 
             for format_id, format_url in stream.items():
                 if format_url and isinstance(format_url, list):
@@ -178,9 +180,9 @@ class SpankBangPlaylistIE(InfoExtractor):
     def _real_extract(self, url):
         mobj = self._match_valid_url(url)
         playlist_id = mobj.group('id')
-
-        webpage = self._download_webpage(
-            url, playlist_id, headers={'Cookie': 'country=US; mobile=on'})
+        country = self.get_param('geo_bypass_country') or 'US'
+        self._set_cookie('.spankbang.com', 'country', country.upper())
+        webpage = self._download_webpage(url, playlist_id, impersonate=True)
 
         entries = [self.url_result(
             urljoin(url, mobj.group('path')),
@@ -189,8 +191,8 @@ class SpankBangPlaylistIE(InfoExtractor):
                 r'<a[^>]+\bhref=(["\'])(?P<path>/?[\da-z]+-(?P<id>[\da-z]+)/playlist/[^"\'](?:(?!\1).)*)\1',
                 webpage)]
 
-        title = self._html_search_regex(
-            r'<em>([^<]+)</em>\s+playlist\s*<', webpage, 'playlist title',
-            fatal=False)
+        title = traverse_obj(webpage, (
+            {find_element(tag='h1', attr='data-testid', value='playlist-title')},
+            {clean_html}, {trim_str(end=' Playlist')}))
 
         return self.playlist_result(entries, playlist_id, title)

yt_dlp/extractor/truth.py

@@ -51,7 +51,8 @@ class TruthIE(InfoExtractor):
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        status = self._download_json(f'https://truthsocial.com/api/v1/statuses/{video_id}', video_id)
+        status = self._download_json(
+            f'https://truthsocial.com/api/v1/statuses/{video_id}', video_id, impersonate=True)
         uploader_id = strip_or_none(traverse_obj(status, ('account', 'username')))
         return {
             'id': video_id,

yt_dlp/extractor/xhamster.py

@@ -268,7 +268,7 @@ class XHamsterIE(InfoExtractor):
         display_id = mobj.group('display_id') or mobj.group('display_id_2')
 
         desktop_url = re.sub(r'^(https?://(?:.+?\.)?)m\.', r'\1', url)
-        webpage, urlh = self._download_webpage_handle(desktop_url, video_id)
+        webpage, urlh = self._download_webpage_handle(desktop_url, video_id, impersonate=True)
 
         error = self._html_search_regex(
             r'<div[^>]+id=["\']videoClosed["\'][^>]*>(.+?)</div>',