Merge a060e2681e into f2a4983df7

Closes #11526
Authored by: jshumphrey

yt_dlp/extractor/_extractors.py

@@ -1982,6 +1982,10 @@ from .streetvoice import StreetVoiceIE
 from .stretchinternet import StretchInternetIE
 from .stripchat import StripchatIE
 from .stv import STVPlayerIE
+from .subsplash import (
+    SubsplashPlaylistIE,
+    SubsplashVideoIE,
+)
 from .substack import SubstackIE
 from .sunporno import SunPornoIE
 from .sverigesradio import (

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/chaturbate.py

@@ -9,7 +9,7 @@ from ..utils import (
 
 
 class ChaturbateIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:[^/]+\.)?chaturbate\.com/(?:fullvideo/?\?.*?\bb=)?(?P<id>[^/?&#]+)'
+    _VALID_URL = r'https?://(?:[^/]+\.)?chaturbate\.(?P<tld>com|eu|global)/(?:fullvideo/?\?.*?\bb=)?(?P<id>[^/?&#]+)'
     _TESTS = [{
         'url': 'https://www.chaturbate.com/siswet19/',
         'info_dict': {
@@ -29,15 +29,24 @@ class ChaturbateIE(InfoExtractor):
     }, {
         'url': 'https://en.chaturbate.com/siswet19/',
         'only_matching': True,
+    }, {
+        'url': 'https://chaturbate.eu/siswet19/',
+        'only_matching': True,
+    }, {
+        'url': 'https://chaturbate.eu/fullvideo/?b=caylin',
+        'only_matching': True,
+    }, {
+        'url': 'https://chaturbate.global/siswet19/',
+        'only_matching': True,
     }]
 
     _ROOM_OFFLINE = 'Room is currently offline'
 
     def _real_extract(self, url):
-        video_id = self._match_id(url)
+        video_id, tld = self._match_valid_url(url).group('id', 'tld')
 
         webpage = self._download_webpage(
-            f'https://chaturbate.com/{video_id}/', video_id,
+            f'https://chaturbate.{tld}/{video_id}/', video_id,
             headers=self.geo_verification_headers())
 
         found_m3u8_urls = []

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/subsplash.py

@@ -0,0 +1,112 @@
+import functools
+import math
+
+from .common import InfoExtractor
+from ..utils import (
+    InAdvancePagedList,
+    int_or_none,
+    str_or_none,
+    traverse_obj,
+    unified_strdate,
+    url_or_none,
+)
+
+
+class SubsplashVideoIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?subsplash\.com/u/[^/]+/media/d/(?P<id>\w+)'
+    _TESTS = [{
+        'url': 'https://subsplash.com/u/skywatchtv/media/d/5whnx5s-the-grand-delusion-taking-place-right-now',
+        'md5': '2d67c50deac3c6c49c6e25c4a5b25afe',
+        'info_dict': {
+            'id': '33f8d305-68ab-414c-acf2-f2317a0abe21',
+            'ext': 'mp4',
+            'title': 'THE GRAND DELUSION TAKING PLACE RIGHT NOW!',
+            'description': 'md5:220a630865c3697b0ec9dcb3a70cbc33',
+            'upload_date': '20240901',
+            'duration': 1710,
+            'thumbnail': r're:https?://.*\.(?:jpg|png)$',
+        },
+    }, {
+        'url': 'https://subsplash.com/u/prophecywatchers/media/d/n4dr8b2-the-transhumanist-plan-for-humanity-billy-crone',
+        'md5': 'f7b4109ba7f012dff953391d6b400730',
+        'info_dict': {
+            'id': 'e16348f1-040e-4596-b922-77b45fa8d253',
+            'ext': 'mp4',
+            'title': 'The Transhumanist Plan for Humanity | Billy Crone',
+            'description': None,
+            'upload_date': '20240903',
+            'duration': 1709,
+            'thumbnail': r're:https?://.*\.(?:jpg|png)$',
+        },
+    }]
+
+    def _fetch_json(self, url, display_id, token):
+        return self._download_json(url, display_id, headers={'Authorization': f'Bearer {token}'})
+
+    def _extract_metadata(self, data, display_id):
+        return traverse_obj(data, {
+            'id': ('id', {str_or_none}),
+            'title': ('title', {str_or_none}),
+            'description': ('summary_text', {str_or_none}),
+            'thumbnail': ('_embedded', 'images', 0, '_links', 'related', 'href', {url_or_none}),
+            'duration': ('_embedded', 'video', 'duration', {lambda x: int_or_none(x, 1000)}),
+            'upload_date': ('published_at', {unified_strdate}),
+            'formats': ('_embedded', 'video', '_embedded', 'playlists', 0, '_links', 'related', 'href',
+                        {lambda url: self._extract_m3u8_formats(url, display_id)}),
+        })
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+        webpage, urlh = self._download_webpage_handle(url, display_id)
+        token = urlh.get_header('set-cookie').split(';')[0].split('=')[1].strip()
+        metadata_url = f'https://core.subsplash.com/media/v1/media-items?filter[short_code]={display_id}&include=images,audio.audio-outputs,audio.video,video.video-outputs,video.playlists,document,broadcast'
+        metadata = self._fetch_json(metadata_url, display_id, token)
+        return self._extract_metadata(traverse_obj(metadata, ('_embedded', 'media-items', 0, {dict})), display_id)
+
+
+class SubsplashPlaylistIE(SubsplashVideoIE):
+    IE_NAME = 'subsplash:playlist'
+    _VALID_URL = r'https?://(?:www\.)?subsplash\.com/[^/]+/(?:our-videos|media)/ms/\+(?P<id>\w+)'
+    _PAGE_SIZE = 15
+    _TESTS = [{
+        'url': 'https://subsplash.com/skywatchtv/our-videos/ms/+dbyjzp8',
+        'info_dict': {
+            'id': 'dbyjzp8',
+            'title': 'Five in Ten',
+        },
+        'playlist_mincount': 24,
+    }, {
+        'url': 'https://subsplash.com/prophecywatchers/media/ms/+n42mr48',
+        'info_dict': {
+            'id': 'n42mr48',
+            'title': 'Road to Zion Series',
+        },
+        'playlist_mincount': 13,
+    }, {
+        'url': 'https://subsplash.com/prophecywatchers/media/ms/+918b9f6',
+        'only_matching': True,
+    }]
+
+    def _get_entries(self, token, series_id, page):
+        url = f'https://core.subsplash.com/media/v1/media-items?filter[broadcast.status|broadcast.status]=null|on-demand&filter[media_series]={series_id}&filter[status]=published&include=images,audio.audio-outputs,audio.video,video.video-outputs,video.playlists,document&page[number]={page + 1}&page[size]={self._PAGE_SIZE}&sort=-position'
+        data = self._fetch_json(url, f'{series_id}_{page}', token)
+        entries = traverse_obj(data, ('_embedded', 'media-items', {list}))
+        for entry in entries:
+            yield self._extract_metadata(entry, series_id)
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+        webpage, urlh = self._download_webpage_handle(url, display_id)
+        token = urlh.get_header('x-api-token')
+
+        series_url = f'https://core.subsplash.com/media/v1/media-series?filter[short_code]={display_id}'
+        json_data = self._fetch_json(series_url, display_id, token)
+        series_data = traverse_obj(json_data, ('_embedded', 'media-series', 0, {
+            'id': ('id', {str}),
+            'title': ('title', {str}),
+            'count': ('media_items_count', {int}),
+        }))
+        total_pages = math.ceil(series_data['count'] / self._PAGE_SIZE)
+
+        entries = InAdvancePagedList(functools.partial(self._get_entries, token, series_data['id']), total_pages, self._PAGE_SIZE)
+        return self.playlist_result(entries, display_id, series_data['title'])