[ie/appleconnect] Rework extractor

yt_dlp/extractor/appleconnect.py

@@ -1,47 +1,92 @@
 from .common import InfoExtractor
-from ..utils import ExtractorError, str_to_int
+from ..utils import (
+    ExtractorError,
+    float_or_none,
+    parse_resolution,
+    qualities,
+    url_or_none,
+)
+from ..utils.traversal import traverse_obj
 
 
 class AppleConnectIE(InfoExtractor):
-    _VALID_URL = r'https?://itunes\.apple\.com/\w{0,2}/?post/(?:id)?sa\.(?P<id>[\w-]+)'
+    IE_NAME = 'apple:music:connect'
+    IE_DESC = 'Apple Music Connect'
+
+    _HEADERS = {
+        'Authorization': 'Bearer eyJhbGciOiJFUzI1NiIsInR5cCI6IkpXVCIsImtpZCI6IldlYlBsYXlLaWQifQ.eyJpc3MiOiJBTVBXZWJQbGF5IiwiaWF0IjoxNzQ2NjM3MTY2LCJleHAiOjE3NTM4OTQ3NjYsInJvb3RfaHR0cHNfb3JpZ2luIjpbImFwcGxlLmNvbSJdfQ.ONPUnh6UMOJ1VWujIxxWuTdi2ueBAM01B8xMg4NkNy9mdE_C1Y15-xKGoZ6Qg6mgC-ZMdfFHt5Xf4hL4X4-lMw',
+        'Origin': 'https://music.apple.com',
+    }
+    _QUALITIES = {
+        'provisionalUploadVideo': (None, None),
+        'sdVideo': (640, 480),
+        'sdVideoWithPlusAudio': (640, 480),
+        'sd480pVideo': (720, 480),
+        '720pHdVideo': (1280, 720),
+        '1080pHdVideo': (1440, 1080),
+    }
+    _VALID_URL = r'https?://music\.apple\.com/\w{0,2}/post/(?P<id>\d+)'
     _TESTS = [{
-        'url': 'https://itunes.apple.com/us/post/idsa.4ab17a39-2720-11e5-96c5-a5b38f6c42d3',
+        'url': 'https://music.apple.com/us/post/1018290019',
-        'md5': 'c1d41f72c8bcaf222e089434619316e4',
         'info_dict': {
-            'id': '4ab17a39-2720-11e5-96c5-a5b38f6c42d3',
+            'id': '1018290019',
             'ext': 'm4v',
             'title': 'Energy',
-            'uploader': 'Drake',
+            'duration': 177.911,
-            'thumbnail': r're:^https?://.*\.jpg$',
+            'thumbnail': r're:https?://.+\.png',
             'upload_date': '20150710',
-            'timestamp': 1436545535,
+            'uploader': 'Drake',
         },
     }, {
-        'url': 'https://itunes.apple.com/us/post/sa.0fe0229f-2457-11e5-9f40-1bb645f2d5d9',
+        'url': 'https://music.apple.com/us/post/1016746627',
-        'only_matching': True,
+        'info_dict': {
+            'id': '1016746627',
+            'ext': 'm4v',
+            'title': 'Body Shop (Madonna) - Chellous Lima (Acoustic Cover)',
+            'duration': 210.278,
+            'thumbnail': r're:https?://.+\.png',
+            'upload_date': '20150706',
+            'uploader': 'Chellous Lima',
+        },
     }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
 
-        try:
+        if not (videos := traverse_obj(self._download_json(
-            video_json = self._html_search_regex(
+            'https://amp-api.music.apple.com/v1/catalog/us/uploaded-videos',
-                r'class="auc-video-data">(\{.*?\})', webpage, 'json')
+            video_id, headers=self._HEADERS, query={'ids': video_id, 'l': 'en-US'},
-        except ExtractorError:
+        ), ('data', ..., 'attributes', any), default={})):
-            raise ExtractorError('This post doesn\'t contain a video', expected=True)
+            raise ExtractorError('Failed to fetch video information')
 
-        video_data = self._parse_json(video_json, video_id)
+        formats = []
-        timestamp = str_to_int(self._html_search_regex(r'data-timestamp="(\d+)"', webpage, 'timestamp'))
+        quality = qualities(list(self._QUALITIES.keys()))
-        like_count = str_to_int(self._html_search_regex(r'(\d+) Loves', webpage, 'like count', default=None))
+        for format_id, src_url in traverse_obj(videos, (
+            'assetTokens', {dict.items}, lambda _, v: url_or_none(v[1]),
+        )):
+            formats.append({
+                'ext': 'm4v',
+                'format_id': format_id,
+                'quality': quality(format_id),
+                'url': src_url,
+                **parse_resolution(src_url),
+                **traverse_obj(self._QUALITIES, (format_id, {
+                    'height': 1,
+                    'width': 0,
+                })),
+            })
 
         return {
             'id': video_id,
-            'url': video_data['sslSrc'],
+            'formats': formats,
-            'title': video_data['title'],
+            'thumbnail': self._html_search_meta(
-            'description': video_data['description'],
+                ('og:image', 'og:image:secure_url', 'twitter:image'), webpage),
-            'uploader': video_data['artistName'],
+            **traverse_obj(videos, {
-            'thumbnail': video_data['artworkUrl'],
+                'title': ('name', {str}),
-            'timestamp': timestamp,
+                'duration': ('durationInMilliseconds', {float_or_none(scale=1000)}),
-            'like_count': like_count,
+                'upload_date': ('uploadDate', {str}, {lambda x: x.replace('-', '')}),
+                'uploader': (('artistName', 'uploadingArtistName'), {str}, any),
+                'webpage_url': ('postUrl', {url_or_none}),
+            }),
         }