Merge 50b4b87a3d into f2a4983df7

Closes #11526
Authored by: jshumphrey

yt_dlp/extractor/_extractors.py

@@ -1649,6 +1649,7 @@ from .radiokapital import (
     RadioKapitalIE,
     RadioKapitalShowIE,
 )
+from .radioradicale import RadioRadicaleIE
 from .radiozet import RadioZetPodcastIE
 from .radlive import (
     RadLiveChannelIE,

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/radioradicale.py

@@ -0,0 +1,105 @@
+from .common import InfoExtractor
+from ..utils import url_or_none
+from ..utils.traversal import traverse_obj
+
+
+class RadioRadicaleIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?radioradicale\.it/scheda/(?P<id>[0-9]+)'
+    _TESTS = [{
+        'url': 'https://www.radioradicale.it/scheda/471591',
+        'md5': 'eb0fbe43a601f1a361cbd00f3c45af4a',
+        'info_dict': {
+            'id': '471591',
+            'ext': 'mp4',
+            'title': 'md5:e8fbb8de57011a3255db0beca69af73d',
+            'description': 'md5:5e15a789a2fe4d67da8d1366996e89ef',
+            'location': 'Napoli',
+            'duration': 2852.0,
+            'timestamp': 1459987200,
+            'upload_date': '20160407',
+            'thumbnail': 'https://www.radioradicale.it/photo400/0/0/9/0/1/00901768.jpg',
+        },
+    }, {
+        'url': 'https://www.radioradicale.it/scheda/742783/parlamento-riunito-in-seduta-comune-11a-della-xix-legislatura',
+        'info_dict': {
+            'id': '742783',
+            'title': 'Parlamento riunito in seduta comune (11ª della XIX legislatura)',
+            'description': '-) Votazione per l\'elezione di un giudice della Corte Costituzionale (nono scrutinio)',
+            'location': 'CAMERA',
+            'duration': 5868.0,
+            'timestamp': 1730246400,
+            'upload_date': '20241030',
+        },
+        'playlist': [{
+            'md5': 'aa48de55dcc45478e4cd200f299aab7d',
+            'info_dict': {
+                'id': '742783-0',
+                'ext': 'mp4',
+                'title': 'Parlamento riunito in seduta comune (11ª della XIX legislatura)',
+            },
+        }, {
+            'md5': 'be915c189c70ad2920e5810f32260ff5',
+            'info_dict': {
+                'id': '742783-1',
+                'ext': 'mp4',
+                'title': 'Parlamento riunito in seduta comune (11ª della XIX legislatura)',
+            },
+        }, {
+            'md5': 'f0ee4047342baf8ed3128a8417ac5e0a',
+            'info_dict': {
+                'id': '742783-2',
+                'ext': 'mp4',
+                'title': 'Parlamento riunito in seduta comune (11ª della XIX legislatura)',
+            },
+        }],
+    }]
+
+    def _entries(self, videos_info, page_id):
+        for idx, video in enumerate(traverse_obj(
+                videos_info, ('playlist', lambda _, v: v['sources']))):
+            video_id = f'{page_id}-{idx}'
+            formats = []
+            subtitles = {}
+
+            for m3u8_url in traverse_obj(video, ('sources', ..., 'src', {url_or_none})):
+                fmts, subs = self._extract_m3u8_formats_and_subtitles(m3u8_url, video_id)
+                formats.extend(fmts)
+                self._merge_subtitles(subs, target=subtitles)
+            for sub in traverse_obj(video, ('subtitles', ..., lambda _, v: url_or_none(v['src']))):
+                self._merge_subtitles({sub.get('srclang') or 'und': [{
+                    'url': sub['src'],
+                    'name': sub.get('label'),
+                }]}, target=subtitles)
+
+            yield {
+                'id': video_id,
+                'title': video.get('title'),
+                'formats': formats,
+                'subtitles': subtitles,
+            }
+
+    def _real_extract(self, url):
+        page_id = self._match_id(url)
+        webpage = self._download_webpage(url, page_id)
+
+        videos_info = self._search_json(
+            r'jQuery\.extend\(Drupal\.settings\s*,',
+            webpage, 'videos_info', page_id)['RRscheda']
+
+        entries = list(self._entries(videos_info, page_id))
+
+        common_info = {
+            'id': page_id,
+            'title': self._og_search_title(webpage),
+            'description': self._og_search_description(webpage),
+            'location': videos_info.get('luogo'),
+            **self._search_json_ld(webpage, page_id),
+        }
+
+        if len(entries) == 1:
+            return {
+                **entries[0],
+                **common_info,
+            }
+
+        return self.playlist_result(entries, multi_video=True, **common_info)