yt_dlp/extractor/mx3.py

@@ -1,15 +1,14 @@
 import re
 
 from .common import InfoExtractor
-from ..networking import HEADRequest
-from ..utils import (
-    get_element_by_class,
-    int_or_none,
-    try_call,
-    url_or_none,
-    urlhandle_detect_ext,
-)
 from ..utils.traversal import traverse_obj
+from ..utils import (
+    urlhandle_detect_ext,
+    url_or_none,
+    int_or_none,
+    unified_timestamp,
+)
+from ..networking import HEADRequest
 
 
 class Mx3BaseIE(InfoExtractor):
@@ -18,18 +17,30 @@ class Mx3BaseIE(InfoExtractor):
     def _real_extract(self, url):
         track_id = self._match_id(url)
         webpage = self._download_webpage(url, track_id)
-        data = self._download_json(
-            f'https://{self._MX3_DOMAIN}/t/{track_id}.json', track_id, fatal=False)
+        json = self._download_json(f'https://{self._MX3_DOMAIN}/t/{track_id}.json', track_id)
+
+        artists = []
+        if json.get('artist'):
+            artists.append(json['artist'])
+        performer = json.get('performer_name')
+        if performer and performer not in artists:
+            artists.append(performer)
+
+        genre = self._html_search_regex(r'<div\b[^>]+class="single-band-genre"[^>]*>([^<]+)</div>',
+                                        webpage, 'genre', fatal=False, flags=re.DOTALL)
 
         formats = []
 
-        def add_format(fmt):
-            urlh = self._request_webpage(
-                HEADRequest(fmt['url']), track_id, fatal=False, expected_status=404,
-                note=f'Checking for format {fmt["format_id"]}')
+        def add_format(fmt, fatal):
+            if fatal:
+                urlh = self._request_webpage(HEADRequest(fmt['url']), track_id, note='Fetching default media headers')
+            else:
+                urlh = self._request_webpage(HEADRequest(fmt['url']), track_id, fatal=False, expected_status=404,
+                                             note=f'Trying media headers for optional format {fmt["format_id"]}')
             if urlh and urlh.status == 200:
                 fmt['ext'] = urlhandle_detect_ext(urlh)
                 fmt['filesize'] = int_or_none(urlh.headers.get('Content-Length'))
+                fmt['timestamp'] = unified_timestamp(urlh.headers.get('Last-Modified'))
                 formats.append(fmt)
 
         track_url = f'https://{self._MX3_DOMAIN}/tracks/{track_id}'
@@ -37,42 +48,26 @@ class Mx3BaseIE(InfoExtractor):
             'url': f'{track_url}/player_asset',
             'format_id': 'default',
             'quality': 1,
-        })
+        }, fatal=True)
+        # the formats below don't always exist
         add_format({
             'url': f'{track_url}/player_asset?quality=hd',
             'format_id': 'hd',
             'quality': 10,
-        })
+        }, fatal=False)
         add_format({
             'url': f'{track_url}/download',
             'format_id': 'download',
             'quality': 11,
-        })
-        add_format({
-            'url': f'{track_url}/player_asset?quality=source',
-            'format_id': 'source',
-            'quality': 11,
-        })
-
-        more_info = get_element_by_class('single-more-info', webpage)
-
-        def get_info_field(name):
-            return self._html_search_regex(
-                rf'<dt[^>]*>\s*{name}\s*</dt>\s*<dd[^>]*>(.*?)</dd>',
-                more_info, name, default=None, flags=re.DOTALL)
+        }, fatal=False)
 
         return {
             'id': track_id,
             'formats': formats,
-            'genre': self._html_search_regex(
-                r'<div\b[^>]+class="single-band-genre"[^>]*>([^<]+)</div>', webpage, 'genre', fatal=False),
-            'release_year': int_or_none(get_info_field('Year of creation')),
-            'description': get_info_field('Description'),
-            'tags': try_call(lambda: get_info_field('Tag').split(', '), list),
-            **traverse_obj(data, {
+            'artist': ', '.join(artists),
+            'genre': genre,
+            **traverse_obj(json, {
                 'title': ('title', {str}),
-                'artist': (('performer_name', 'artist'), {str}),
-                'album_artist': ('artist', {str}),
                 'composer': ('composer_name', {str}),
                 'thumbnail': (('picture_url_xlarge', 'picture_url'), {url_or_none}),
             }, get_all=False),
@@ -84,49 +79,45 @@ class Mx3IE(Mx3BaseIE):
     _VALID_URL = r'https?://(?:www\.)?mx3\.ch/t/(?P<id>[0-9A-Za-z]+)'
     _TESTS = [{
         'url': 'https://mx3.ch/t/1Cru',
-        'md5': '7ba09e9826b4447d4e1ce9d69e0e295f',
+        'md5': '82510bf4c21f17da41bff7e1ffd84e78',
         'info_dict': {
             'id': '1Cru',
-            'ext': 'wav',
-            'artist': 'Godina',
-            'album_artist': 'Tortue Tortue',
+            # This one is audio-only. It's a mp3, but we have to make a HEAD request to find out.
+            'ext': 'mp3',
+            'artist': 'Tortue Tortue, Godina',
             'composer': 'Olivier Godinat',
             'genre': 'Rock',
             'thumbnail': 'https://mx3.ch/pictures/mx3/file/0101/4643/square_xlarge/1-s-envoler-1.jpg?1630272813',
-            'title': "S'envoler",
-            'release_year': 2021,
-            'tags': [],
+            'title': 'S\'envoler',
+            'timestamp': 1630272831,
         }
     }, {
         'url': 'https://mx3.ch/t/1LIY',
-        'md5': '48293cb908342547827f963a5a2e9118',
+        'md5': '4117489dff8c763ecfbb0b95a67d6c8e',
         'info_dict': {
             'id': '1LIY',
-            'ext': 'mov',
-            'artist': 'Tania Kimfumu',
-            'album_artist': 'The Broots',
+            # This is a music video. 'file' says: ISO Media, MP4 Base Media v1 [ISO 14496-12:2003]
+            'ext': 'mp4',
+            'artist': 'The Broots, Tania Kimfumu',
             'composer': 'Emmanuel Diserens',
             'genre': 'Electro',
             'thumbnail': 'https://mx3.ch/pictures/mx3/file/0110/0003/video_xlarge/frame_0000.png?1686963670',
             'title': 'The Broots-Larytta remix "Begging For Help"',
-            'release_year': 2023,
-            'tags': ['the broots', 'cassata records', 'larytta'],
-            'description': '"Begging for Help" Larytta Remix Official Video\nRealized By Kali Donkilie in 2023',
+            'timestamp': 1686963636,
         }
     }, {
         'url': 'https://mx3.ch/t/1C6E',
         'md5': '1afcd578493ddb8e5008e94bb6d97e25',
         'info_dict': {
             'id': '1C6E',
+            # This one has a download button, yielding a WAV.
             'ext': 'wav',
             'artist': 'Alien Bubblegum',
-            'album_artist': 'Alien Bubblegum',
             'composer': 'Alien Bubblegum',
             'genre': 'Punk',
             'thumbnail': 'https://mx3.ch/pictures/mx3/file/0101/1551/square_xlarge/pandora-s-box-cover-with-title.png?1627054733',
             'title': 'Wide Awake',
-            'release_year': 2021,
-            'tags': ['alien bubblegum', 'bubblegum', 'alien', 'pop punk', 'poppunk'],
+            'timestamp': 1627054732,
         }
     }]
 
@@ -136,18 +127,16 @@ class Mx3NeoIE(Mx3BaseIE):
     _VALID_URL = r'https?://(?:www\.)?neo.mx3\.ch/t/(?P<id>[0-9A-Za-z]+)'
     _TESTS = [{
         'url': 'https://neo.mx3.ch/t/1hpd',
-        'md5': '6d9986bbae5cac3296ec8813bf965eb2',
+        'md5': 'ff0b2b91ce0b8931c0a358715758dc78',
         'info_dict': {
             'id': '1hpd',
-            'ext': 'wav',
-            'artist': 'Baptiste Lopez',
-            'album_artist': 'Kammerorchester Basel',
+            'ext': 'mp3',
+            'artist': 'Kammerorchester Basel, Baptiste Lopez',
             'composer': 'Jannik Giger',
             'genre': 'Composition, Orchestra',
             'title': 'Troisième œil. Für Kammerorchester (2023)',
             'thumbnail': 'https://neo.mx3.ch/pictures/neo/file/0000/0241/square_xlarge/kammerorchester-basel-group-photo-2_c_-lukasz-rajchert.jpg?1560341252',
-            'release_year': 2023,
-            'tags': [],
+            'timestamp': 1705055012,
         }
     }]
 
@@ -162,12 +151,10 @@ class Mx3VolksmusikIE(Mx3BaseIE):
             'id': 'Zx',
             'ext': 'mp3',
             'artist': 'Ländlerkapelle GrischArt',
-            'album_artist': 'Ländlerkapelle GrischArt',
             'composer': 'Urs Glauser',
             'genre': 'Instrumental, Graubünden',
             'title': 'Chämilouf',
             'thumbnail': 'https://volksmusik.mx3.ch/pictures/vxm/file/0000/3815/square_xlarge/grischart1.jpg?1450530120',
-            'release_year': 2012,
-            'tags': [],
+            'timestamp': 1450532809,
         }
     }]