Merge 2be0c24897 into c699bafc50

Closes #11537

Authored by: bashonly

test/test_traversal.py

@@ -481,7 +481,7 @@ class TestTraversalHelpers:
             'id': 'name',
             'data': 'content',
             'url': 'url',
-        }, all, {subs_list_to_dict}]) == {
+        }, all, {subs_list_to_dict(lang=None)}]) == {
             'de': [{'url': 'https://example.com/subs/de.ass'}],
             'en': [{'data': 'content'}],
         }, 'subs with mandatory items missing should be filtered'
@@ -507,6 +507,54 @@ class TestTraversalHelpers:
             {'url': 'https://example.com/subs/en1', 'ext': 'ext'},
             {'url': 'https://example.com/subs/en2', 'ext': 'ext'},
         ]}, '`quality` key should sort subtitle list accordingly'
+        assert traverse_obj([
+            {'name': 'de', 'url': 'https://example.com/subs/de.ass'},
+            {'name': 'de'},
+            {'name': 'en', 'content': 'content'},
+            {'url': 'https://example.com/subs/en'},
+        ], [..., {
+            'id': 'name',
+            'url': 'url',
+            'data': 'content',
+        }, all, {subs_list_to_dict(lang='en')}]) == {
+            'de': [{'url': 'https://example.com/subs/de.ass'}],
+            'en': [
+                {'data': 'content'},
+                {'url': 'https://example.com/subs/en'},
+            ],
+        }, 'optionally provided lang should be used if no id available'
+        assert traverse_obj([
+            {'name': 1, 'url': 'https://example.com/subs/de1'},
+            {'name': {}, 'url': 'https://example.com/subs/de2'},
+            {'name': 'de', 'ext': 1, 'url': 'https://example.com/subs/de3'},
+            {'name': 'de', 'ext': {}, 'url': 'https://example.com/subs/de4'},
+        ], [..., {
+            'id': 'name',
+            'url': 'url',
+            'ext': 'ext',
+        }, all, {subs_list_to_dict(lang=None)}]) == {
+            'de': [
+                {'url': 'https://example.com/subs/de3'},
+                {'url': 'https://example.com/subs/de4'},
+            ],
+        }, 'non str types should be ignored for id and ext'
+        assert traverse_obj([
+            {'name': 1, 'url': 'https://example.com/subs/de1'},
+            {'name': {}, 'url': 'https://example.com/subs/de2'},
+            {'name': 'de', 'ext': 1, 'url': 'https://example.com/subs/de3'},
+            {'name': 'de', 'ext': {}, 'url': 'https://example.com/subs/de4'},
+        ], [..., {
+            'id': 'name',
+            'url': 'url',
+            'ext': 'ext',
+        }, all, {subs_list_to_dict(lang='de')}]) == {
+            'de': [
+                {'url': 'https://example.com/subs/de1'},
+                {'url': 'https://example.com/subs/de2'},
+                {'url': 'https://example.com/subs/de3'},
+                {'url': 'https://example.com/subs/de4'},
+            ],
+        }, 'non str types should be replaced by default id'
 
     def test_trim_str(self):
         with pytest.raises(TypeError):
@@ -525,7 +573,7 @@ class TestTraversalHelpers:
     def test_unpack(self):
         assert unpack(lambda *x: ''.join(map(str, x)))([1, 2, 3]) == '123'
         assert unpack(join_nonempty)([1, 2, 3]) == '1-2-3'
-        assert unpack(join_nonempty(delim=' '))([1, 2, 3]) == '1 2 3'
+        assert unpack(join_nonempty, delim=' ')([1, 2, 3]) == '1 2 3'
         with pytest.raises(TypeError):
             unpack(join_nonempty)()
         with pytest.raises(TypeError):

test/test_utils.py

@@ -72,7 +72,6 @@ from yt_dlp.utils import (
     intlist_to_bytes,
     iri_to_uri,
     is_html,
-    join_nonempty,
     js_to_json,
     limit_length,
     locked_file,
@@ -2158,10 +2157,6 @@ Line 1
         assert int_or_none(v=10) == 10, 'keyword passed positional should call function'
         assert int_or_none(scale=0.1)(10) == 100, 'call after partial application should call the function'
 
-        assert callable(join_nonempty(delim=', ')), 'varargs positional should apply partially'
-        assert callable(join_nonempty()), 'varargs positional should apply partially'
-        assert join_nonempty(None, delim=', ') == '', 'passed varargs should call the function'
-
 
 if __name__ == '__main__':
     unittest.main()

yt_dlp/YoutubeDL.py

@@ -4381,7 +4381,9 @@ class YoutubeDL:
             return None
 
         for idx, t in list(enumerate(thumbnails))[::-1]:
-            thumb_ext = (f'{t["id"]}.' if multiple else '') + determine_ext(t['url'], 'jpg')
+            thumb_ext = t.get('ext') or determine_ext(t['url'], 'jpg')
+            if multiple:
+                thumb_ext = f'{t["id"]}.{thumb_ext}'
             thumb_display_id = f'{label} thumbnail {t["id"]}'
             thumb_filename = replace_extension(filename, thumb_ext, info_dict.get('ext'))
             thumb_filename_final = replace_extension(thumb_filename_base, thumb_ext, info_dict.get('ext'))

yt_dlp/extractor/_extractors.py

@@ -1983,6 +1983,7 @@ from .stretchinternet import StretchInternetIE
 from .stripchat import StripchatIE
 from .stv import STVPlayerIE
 from .substack import SubstackIE
+from .suno import SunoIE, SunoPlaylistIE
 from .sunporno import SunPornoIE
 from .sverigesradio import (
     SverigesRadioEpisodeIE,

yt_dlp/extractor/afreecatv.py

@@ -66,6 +66,14 @@ class AfreecaTVBaseIE(InfoExtractor):
             extensions={'legacy_ssl': True}), display_id,
             'Downloading API JSON', 'Unable to download API JSON')
 
+    @staticmethod
+    def _fixup_thumb(thumb_url):
+        if not url_or_none(thumb_url):
+            return None
+        # Core would determine_ext as 'php' from the url, so we need to provide the real ext
+        # See: https://github.com/yt-dlp/yt-dlp/issues/11537
+        return [{'url': thumb_url, 'ext': 'jpg'}]
+
 
 class AfreecaTVIE(AfreecaTVBaseIE):
     IE_NAME = 'soop'
@@ -155,7 +163,7 @@ class AfreecaTVIE(AfreecaTVBaseIE):
             'uploader': ('writer_nick', {str}),
             'uploader_id': ('bj_id', {str}),
             'duration': ('total_file_duration', {int_or_none(scale=1000)}),
-            'thumbnail': ('thumb', {url_or_none}),
+            'thumbnails': ('thumb', {self._fixup_thumb}),
         })
 
         entries = []
@@ -226,8 +234,7 @@ class AfreecaTVCatchStoryIE(AfreecaTVBaseIE):
 
         return self.playlist_result(self._entries(data), video_id)
 
-    @staticmethod
+    def _entries(self, data):
-    def _entries(data):
         # 'files' is always a list with 1 element
         yield from traverse_obj(data, (
             'data', lambda _, v: v['story_type'] == 'catch',
@@ -238,7 +245,7 @@ class AfreecaTVCatchStoryIE(AfreecaTVBaseIE):
                 'title': ('title', {str}),
                 'uploader': ('writer_nick', {str}),
                 'uploader_id': ('writer_id', {str}),
-                'thumbnail': ('thumb', {url_or_none}),
+                'thumbnails': ('thumb', {self._fixup_thumb}),
                 'timestamp': ('write_timestamp', {int_or_none}),
             }))
 

yt_dlp/extractor/common.py

@@ -279,6 +279,7 @@ class InfoExtractor:
     thumbnails:     A list of dictionaries, with the following entries:
                         * "id" (optional, string) - Thumbnail format ID
                         * "url"
+                        * "ext" (optional, string) - actual image extension if not given in URL
                         * "preference" (optional, int) - quality of the image
                         * "width" (optional, int)
                         * "height" (optional, int)

yt_dlp/extractor/suno.py

@@ -0,0 +1,152 @@
+import re
+
+from .common import InfoExtractor
+from ..utils import ExtractorError, unescapeHTML
+
+
+class SunoBaseIE(InfoExtractor):
+    def _get_title(self, webpage):
+        return self._html_search_meta(
+            ['og:title', 'twitter:title'], webpage, 'title',
+            default=None) or self._html_extract_title(webpage)
+
+    def _get_description(self, webpage):
+        return self._html_search_meta(
+            ['og:description', 'description', 'twitter:description'],
+            webpage, 'description', default=None)
+
+    def _get_thumbnail(self, webpage):
+        return self._html_search_meta(
+            ['og:image', 'twitter:image'], webpage, 'thumbnail', default=None)
+
+
+class SunoIE(SunoBaseIE):
+    _VALID_URL = r'https?://(?:www\.)?suno\.com/song/(?P<id>[-a-f0-9]+)'
+    _TESTS = [
+        {
+            'url': 'https://suno.com/song/ab39a04d-b2e6-463b-9b8e-ddea725422f5',
+            'md5': 'ef850763b175d8a3c7fba5e2dbdc6bc5',
+            'info_dict': {
+                'id': 'ab39a04d-b2e6-463b-9b8e-ddea725422f5',
+                'title': 'Life\'s a Soundtrack · AI Funk Factory @ YT by @funk | Suno',
+                'description': 'groovy funk, melodic song. Listen and make your own with Suno.',
+                'thumbnail': r're:https?://.*903f2bd7-ccc0-4029-a76a-887f07ebc2df.*\.jpeg$',
+                'ext': 'mp3',
+            },
+        },
+        {
+            'url': 'https://suno.com/song/9cbcb5f4-f367-4f1c-8a32-23ec62bdc47e',
+            'md5': '2f038badef88d189891d5f8cd8d8804d',
+            'info_dict': {
+                'id': '9cbcb5f4-f367-4f1c-8a32-23ec62bdc47e',
+                'title': 'Pequenos Prazeres da Vida by @groovebot | Suno',
+                'description': 'pop bossa nova song. Listen and make your own with Suno.',
+                'thumbnail': r're:https?://.*9cbcb5f4-f367-4f1c-8a32-23ec62bdc47e.*\.jpeg$',
+                'ext': 'mp3',
+            },
+        },
+    ]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+
+        url = self._html_search_meta(
+            ['og:audio', 'twitter:player:stream'], webpage, 'url', default=None)
+
+        return {
+            'id': video_id,
+            'title': self._get_title(webpage),
+            'description': self._get_description(webpage),
+            'thumbnail': self._get_thumbnail(webpage),
+            'url': url,
+        }
+
+
+class SunoPlaylistIE(SunoBaseIE):
+    _VALID_URL = r'https?://(?:www\.)?suno\.com/playlist/(?P<id>[-a-f0-9]+)'
+    _TESTS = [
+        {
+            'url': 'https://suno.com/playlist/01f2ac32-c32e-4d26-b10c-221107c02946',
+            'info_dict': {
+                'id': '01f2ac32-c32e-4d26-b10c-221107c02946',
+                'title': 'Main 0 by @contemplativetranspositions367 | Suno',
+                'description': 'Hopefully the test case passed',
+                'thumbnail': r're:https?://.*19d6d518-1b87-43b3-90b9-2a476ca5824a.*\.jpeg$',
+            },
+            'playlist': [{
+                'info_dict': {
+                    'id': '19d6d518-1b87-43b3-90b9-2a476ca5824a',
+                    'title': 'Ceaseless <Echoes>',
+                    'ext': 'mp3',
+                },
+            }],
+            'playlist_count': 1,
+        },
+        {
+            'url': 'https://www.suno.com/playlist/568eeaab-dfbf-4da6-aa0a-0fb1a32330de',
+            'info_dict': {
+                'id': '568eeaab-dfbf-4da6-aa0a-0fb1a32330de',
+                'title': 'Piano by @kunal | Suno',
+                'description': 'Here are some good piano',
+                'thumbnail': r're:https?://.*0ecc0956-3b17-4d4b-8504-55849dd75e22.*\.jpeg$',
+            },
+            'playlist': [
+                {
+                    'info_dict': {
+                        'id': '0ecc0956-3b17-4d4b-8504-55849dd75e22',
+                        'title': 'ST',
+                        'ext': 'mp3',
+                    },
+                },
+                {
+                    'info_dict': {
+                        'id': '3fef7d44-c5a3-4181-9de3-d81542af23ef',
+                        'title': 'ST',
+                        'ext': 'mp3',
+                    },
+                },
+                {
+                    'info_dict': {
+                        'id': '15e797fa-06c0-4e11-8cc0-3b2580476039',
+                        'title': 'ST - 2',
+                        'ext': 'mp3',
+                    },
+                },
+            ],
+            'playlist_count': 3,
+        },
+    ]
+
+    def _real_extract(self, url):
+        playlist_id = self._match_id(url)
+        webpage = self._download_webpage(url, playlist_id)
+
+        # There are <a>s whose href is a song/ID path. The <span>s directly
+        # within them have the respective song title as their innerHTML.
+        # Alternatively, this info can be extracted through parsing an escaped
+        # JSON object inside a <script> array, though that seems even less stable
+        # than this HTML.
+        songs_regex = r'/song/(?P<id>[-a-f0-9]+)["\'][^>]*>\s*<span[^>]*>\s*(?P<title>[^<]+)</span>'
+        songs = re.findall(songs_regex, webpage)
+
+        og_audio_regex = self._og_regexes('audio')[0]
+        audio_urls = [matches[0] for matches in re.findall(og_audio_regex, webpage)]
+
+        if len(songs) != len(audio_urls):
+            raise ExtractorError('Unexpected mismatch between song HTML list and og audio URLs')
+
+        return {
+            '_type': 'playlist',
+            'id': playlist_id,
+            'title': self._get_title(webpage),
+            'description': self._get_description(webpage),
+            'thumbnail': self._get_thumbnail(webpage),
+
+            'entries': [{
+                'id': song_tuple[0],
+                'title': unescapeHTML(song_tuple[1]),
+                'url': audio_urls[i],
+
+            } for i, song_tuple in enumerate(songs)],
+        }

yt_dlp/utils/_utils.py

@@ -216,7 +216,7 @@ def partial_application(func):
     sig = inspect.signature(func)
     required_args = [
         param.name for param in sig.parameters.values()
-        if param.kind in (inspect.Parameter.POSITIONAL_ONLY, inspect.Parameter.POSITIONAL_OR_KEYWORD, inspect.Parameter.VAR_POSITIONAL)
+        if param.kind in (inspect.Parameter.POSITIONAL_ONLY, inspect.Parameter.POSITIONAL_OR_KEYWORD)
         if param.default is inspect.Parameter.empty
     ]
 
@@ -4837,7 +4837,6 @@ def number_of_digits(number):
     return len('%d' % number)
 
 
-@partial_application
 def join_nonempty(*values, delim='-', from_dict=None):
     if from_dict is not None:
         values = (traversal.traverse_obj(from_dict, variadic(v)) for v in values)

yt_dlp/utils/traversal.py

@@ -332,14 +332,14 @@ class _RequiredError(ExtractorError):
 
 
 @typing.overload
-def subs_list_to_dict(*, ext: str | None = None) -> collections.abc.Callable[[list[dict]], dict[str, list[dict]]]: ...
+def subs_list_to_dict(*, lang: str | None = 'und', ext: str | None = None) -> collections.abc.Callable[[list[dict]], dict[str, list[dict]]]: ...
 
 
 @typing.overload
-def subs_list_to_dict(subs: list[dict] | None, /, *, ext: str | None = None) -> dict[str, list[dict]]: ...
+def subs_list_to_dict(subs: list[dict] | None, /, *, lang: str | None = 'und', ext: str | None = None) -> dict[str, list[dict]]: ...
 
 
-def subs_list_to_dict(subs: list[dict] | None = None, /, *, ext=None):
+def subs_list_to_dict(subs: list[dict] | None = None, /, *, lang='und', ext=None):
     """
     Convert subtitles from a traversal into a subtitle dict.
     The path should have an `all` immediately before this function.
@@ -352,7 +352,7 @@ def subs_list_to_dict(subs: list[dict] | None = None, /, *, ext=None):
     `quality`  The sort order for each subtitle
     """
     if subs is None:
-        return functools.partial(subs_list_to_dict, ext=ext)
+        return functools.partial(subs_list_to_dict, lang=lang, ext=ext)
 
     result = collections.defaultdict(list)
 
@@ -360,10 +360,16 @@ def subs_list_to_dict(subs: list[dict] | None = None, /, *, ext=None):
         if not url_or_none(sub.get('url')) and not sub.get('data'):
             continue
         sub_id = sub.pop('id', None)
-        if sub_id is None:
+        if not isinstance(sub_id, str):
-            continue
+            if not lang:
-        if ext is not None and not sub.get('ext'):
+                continue
-            sub['ext'] = ext
+            sub_id = lang
+        sub_ext = sub.get('ext')
+        if not isinstance(sub_ext, str):
+            if not ext:
+                sub.pop('ext', None)
+            else:
+                sub['ext'] = ext
         result[sub_id].append(sub)
     result = dict(result)
 
@@ -452,9 +458,9 @@ def trim_str(*, start=None, end=None):
     return trim
 
 
-def unpack(func):
+def unpack(func, **kwargs):
     @functools.wraps(func)
-    def inner(items, **kwargs):
+    def inner(items):
         return func(*items, **kwargs)
 
     return inner