Merge eec17abd84 into c699bafc50

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/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(data):
+    def _entries(self, 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/youtube.py

@@ -50,6 +50,7 @@ from ..utils import (
     parse_iso8601,
     parse_qs,
     qualities,
+    remove_end,
     remove_start,
     smuggle_url,
     str_or_none,
@@ -515,6 +516,8 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
     _YT_HANDLE_RE = r'@[\w.-]{3,30}'  # https://support.google.com/youtube/answer/11585688?hl=en
     _YT_CHANNEL_UCID_RE = r'UC[\w-]{22}'
 
+    _NETRC_MACHINE = 'youtube'
+
     def ucid_or_none(self, ucid):
         return self._search_regex(rf'^({self._YT_CHANNEL_UCID_RE})$', ucid, 'UC-id', default=None)
 
@@ -573,10 +576,19 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
         self._initialize_consent()
         self._check_login_required()
 
+    def _perform_login(self, username, password):
+        if username.startswith('oauth'):
+            raise ExtractorError(
+                f'Login with OAuth is no longer supported. {self._youtube_login_hint}', expected=True)
+
+        self.report_warning(
+            f'Login with password is not supported for YouTube. {self._youtube_login_hint}')
+
     @property
     def _youtube_login_hint(self):
-        return (f'{self._login_hint(method="cookies")}. '
-                'See  https://github.com/yt-dlp/yt-dlp/wiki/Extractors#exporting-youtube-cookies  for help with cookies')
+        return (f'{self._login_hint(method="cookies")}. Also see  '
+                'https://github.com/yt-dlp/yt-dlp/wiki/Extractors#exporting-youtube-cookies  '
+                'for tips on effectively exporting YouTube cookies')
 
     def _check_login_required(self):
         if self._LOGIN_REQUIRED and not self.is_authenticated:
@@ -851,11 +863,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
         for alert_type, alert_message in (warnings + errors[:-1]):
             self.report_warning(f'YouTube said: {alert_type} - {alert_message}', only_once=only_once)
         if errors:
-            msg = errors[-1][1]
-            if msg and 'sign in' in msg.lower():
-                expected = True
-                msg += '\n' + self._youtube_login_hint
-            raise ExtractorError(f'YouTube said: {msg}', expected=expected)
+            raise ExtractorError(f'YouTube said: {errors[-1][1]}', expected=expected)
 
     def _extract_and_report_alerts(self, data, *args, **kwargs):
         return self._report_alerts(self._extract_alerts(data), *args, **kwargs)
@@ -4454,6 +4462,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                     self.raise_geo_restricted(subreason, countries, metadata_available=True)
                 reason += f'. {subreason}'
             if reason:
+                if 'sign in' in reason.lower():
+                    reason = remove_end(reason, 'This helps protect our community. Learn more')
+                    reason = f'{remove_end(reason.strip(), ".")}. {self._youtube_login_hint}'
                 self.raise_no_formats(reason, expected=True)
 
         keywords = get_first(video_details, 'keywords', expected_type=list) or []
@@ -6973,7 +6984,7 @@ class YoutubeTabIE(YoutubeTabBaseInfoExtractor):
         raise ExtractorError('Unable to recognize tab page')
 
 
-class YoutubePlaylistIE(InfoExtractor):
+class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
     IE_DESC = 'YouTube playlists'
     _VALID_URL = r'''(?x)(?:
                         (?:https?://)?
@@ -7087,7 +7098,7 @@ class YoutubePlaylistIE(InfoExtractor):
         return self.url_result(url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
 
 
-class YoutubeYtBeIE(InfoExtractor):
+class YoutubeYtBeIE(YoutubeBaseInfoExtractor):
     IE_DESC = 'youtu.be'
     _VALID_URL = rf'https?://youtu\.be/(?P<id>[0-9A-Za-z_-]{{11}})/*?.*?\blist=(?P<playlist_id>{YoutubeBaseInfoExtractor._PLAYLIST_ID_RE})'
     _TESTS = [{
@@ -7138,7 +7149,7 @@ class YoutubeYtBeIE(InfoExtractor):
             }), ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
 
 
-class YoutubeLivestreamEmbedIE(InfoExtractor):
+class YoutubeLivestreamEmbedIE(YoutubeBaseInfoExtractor):
     IE_DESC = 'YouTube livestream embeds'
     _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/embed/live_stream/?\?(?:[^#]+&)?channel=(?P<id>[^&#]+)'
     _TESTS = [{
@@ -7153,7 +7164,7 @@ class YoutubeLivestreamEmbedIE(InfoExtractor):
             ie=YoutubeTabIE.ie_key(), video_id=channel_id)
 
 
-class YoutubeYtUserIE(InfoExtractor):
+class YoutubeYtUserIE(YoutubeBaseInfoExtractor):
     IE_DESC = 'YouTube user videos; "ytuser:" prefix'
     IE_NAME = 'youtube:user'
     _VALID_URL = r'ytuser:(?P<id>.+)'
@@ -7440,7 +7451,7 @@ class YoutubeMusicSearchURLIE(YoutubeTabBaseInfoExtractor):
         return self.playlist_result(self._search_results(query, params, default_client='web_music'), title, title)
 
 
-class YoutubeFeedsInfoExtractor(InfoExtractor):
+class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
     """
     Base class for feed extractors
     Subclasses must re-define the _FEED_NAME property.
@@ -7457,7 +7468,7 @@ class YoutubeFeedsInfoExtractor(InfoExtractor):
             f'https://www.youtube.com/feed/{self._FEED_NAME}', ie=YoutubeTabIE.ie_key())
 
 
-class YoutubeWatchLaterIE(InfoExtractor):
+class YoutubeWatchLaterIE(YoutubeBaseInfoExtractor):
     IE_NAME = 'youtube:watchlater'
     IE_DESC = 'Youtube watch later list; ":ytwatchlater" keyword (requires cookies)'
     _VALID_URL = r':ytwatchlater'
@@ -7511,7 +7522,7 @@ class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
     }]
 
 
-class YoutubeShortsAudioPivotIE(InfoExtractor):
+class YoutubeShortsAudioPivotIE(YoutubeBaseInfoExtractor):
     IE_DESC = 'YouTube Shorts audio pivot (Shorts using audio of a given video)'
     IE_NAME = 'youtube:shorts:pivot:audio'
     _VALID_URL = r'https?://(?:www\.)?youtube\.com/source/(?P<id>[\w-]{11})/shorts'
@@ -7535,7 +7546,7 @@ class YoutubeShortsAudioPivotIE(InfoExtractor):
             ie=YoutubeTabIE)
 
 
-class YoutubeTruncatedURLIE(InfoExtractor):
+class YoutubeTruncatedURLIE(YoutubeBaseInfoExtractor):
     IE_NAME = 'youtube:truncated_url'
     IE_DESC = False  # Do not list
     _VALID_URL = r'''(?x)
@@ -7694,7 +7705,7 @@ class YoutubeConsentRedirectIE(YoutubeBaseInfoExtractor):
         return self.url_result(redirect_url)
 
 
-class YoutubeTruncatedIDIE(InfoExtractor):
+class YoutubeTruncatedIDIE(YoutubeBaseInfoExtractor):
     IE_NAME = 'youtube:truncated_id'
     IE_DESC = False  # Do not list
     _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'

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)
 
@@ -360,9 +360,15 @@ 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):
+            if not lang:
                 continue
-        if ext is not None and not sub.get('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