Merge 7642837a49 into f2a4983df7

Closes #11526
Authored by: jshumphrey

yt_dlp/extractor/_extractors.py

@@ -1549,6 +1549,7 @@ from .pluralsight import (
     PluralsightIE,
 )
 from .plutotv import PlutoTVIE
+from .plvideo import PlVideoVideoIE
 from .podbayfm import (
     PodbayFMChannelIE,
     PodbayFMIE,

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/plvideo.py

@@ -0,0 +1,91 @@
+from yt_dlp.utils._utils import qualities
+
+from .common import InfoExtractor
+
+
+class PlVideoVideoIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?plvideo\.ru/watch\?v=(?P<id>\w+)&?(.+)?'  # type: ignore
+    _TESTS = [
+        {
+            'url': 'https://plvideo.ru/watch?v=lYmu2gcUKOa9',
+            'md5': 'eb3e7830abb375a782d943f593d2646b',
+            'info_dict': {
+                'id': 'lYmu2gcUKOa9',
+                'ext': 'mp4',
+                'title': 'Запретная страсть. Премьера 2024. 18+Мелодрама. Триллер. 18+',
+                'uploader_id': 'y__S081jJiUt',
+                'uploader': 'Tvoja Mediateka',
+                'duration': 6238333,
+                'like_count': int,
+                'description': str,
+                'comment_count': int,
+                'thumbnail': r're:^https?://.*\.jpg',
+                'type': 'video',
+                'view_count': int,
+                'dislike_count': int,
+            },
+        },
+    ]
+
+    def _quality_to_dimensions(self, quality):
+        mapped = {
+            '240p': (426, 240),
+            '360p': (640, 360),
+            '468p': (720, 468),
+            '480p': (720, 480),
+            '720p': (1280, 720),
+            '1080p': (1920, 1080),
+        }
+        return mapped.get(quality)
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        api_url = f'https://api.g1.plvideo.ru/v1/videos/{video_id}?Aud=18'
+
+        result = self._download_json(api_url, video_id, 'Downloading video JSON')
+        assert result.get('code') == 200, 'Failed to download video JSON'
+
+        item = result.get('item')
+        assert item is not None, 'Bad API response'
+
+        thumbnail = item.get('cover').get('paths').get('original').get('src')
+
+        formats = []
+        preference = qualities(['240p', '360p', '468p', '480p', '720p', '1080p'])
+
+        for key, value in item.get('profiles').items():
+            hlsurl = value.get('hls')
+            dimensions = self._quality_to_dimensions(key)
+            fmt = {
+                'url': hlsurl,
+                'ext': 'mp4',
+                'quality': preference(key),
+                'width': dimensions[0],
+                'height': dimensions[1],
+                'format_id': key,
+                'protocol': 'm3u8_native',
+                'aspect_ratio': float(value.get('aspectRatio')),
+            }
+
+            formats.append(fmt)
+
+        result = {
+            'id': video_id,
+            'title': item.get('title'),
+            'formats': formats,
+            'thumbnails': [{'url': thumbnail}],
+            'uploader': item.get('channel').get('name'),
+            'duration': item.get('uploadFile').get('videoDuration'),
+            'uploader_id': item.get('channel').get('id'),
+            'view_count': item.get('stats').get('viewTotalCount'),
+            'like_count': item.get('stats').get('likeCount'),
+            'comment_count': item.get('stats').get('commentCount'),
+            'dislike_count': item.get('stats').get('dislikeCount'),
+            'type': item.get('type'),
+        }
+
+        description = item.get('description')
+        if description:
+            result['description'] = description
+
+        return result