Merge a060e2681e into 6b43a8d84b

yt_dlp/extractor/archiveorg.py

@@ -205,26 +205,6 @@ 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
@@ -355,7 +335,7 @@ class ArchiveOrgIE(InfoExtractor):
                 info['comments'].append({
                     'id': review.get('review_id'),
                     'author': review.get('reviewer'),
-                    'text': join_nonempty('reviewtitle', 'reviewbody', from_dict=review, delim='\n\n'),
+                    'text': str_or_none(review.get('reviewtitle'), '') + '\n\n' + review.get('reviewbody'),
                     '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\.(?P<tld>com|eu|global)/(?:fullvideo/?\?.*?\bb=)?(?P<id>[^/?&#]+)'
+    _VALID_URL = r'https?://(?:[^/]+\.)?chaturbate\.com/(?:fullvideo/?\?.*?\bb=)?(?P<id>[^/?&#]+)'
     _TESTS = [{
         'url': 'https://www.chaturbate.com/siswet19/',
         'info_dict': {
@@ -29,24 +29,15 @@ 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, tld = self._match_valid_url(url).group('id', 'tld')
+        video_id = self._match_id(url)
 
         webpage = self._download_webpage(
-            f'https://chaturbate.{tld}/{video_id}/', video_id,
+            f'https://chaturbate.com/{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(vid_data, formats, mpd_url=None):
+        def extract_dash_manifest(video, formats):
             dash_manifest = traverse_obj(
-                vid_data, 'dash_manifest', 'playlist', 'dash_manifest_xml_string', 'manifest_xml', expected_type=str)
+                video, 'dash_manifest', 'playlist', 'dash_manifest_xml_string', 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')) or mpd_url))
+                    mpd_url=url_or_none(video.get('dash_manifest_url'))))
 
         def process_formats(info):
             # Downloads with browser's User-Agent are rate limited. Working around
@@ -619,12 +619,9 @@ 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')):
@@ -632,7 +629,7 @@ class FacebookIE(InfoExtractor):
                         if not playable_url:
                             continue
                         if determine_ext(playable_url) == 'mpd':
-                            formats.extend(self._extract_mpd_formats(playable_url, video_id, fatal=False))
+                            formats.extend(self._extract_mpd_formats(playable_url, video_id))
                         else:
                             formats.append({
                                 'format_id': format_id,
@@ -641,28 +638,6 @@ 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