Merge 19fa262ecb 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/_extractors.py

@@ -2083,6 +2083,7 @@ from .theplatform import (
 from .thestar import TheStarIE
 from .thesun import TheSunIE
 from .theweatherchannel import TheWeatherChannelIE
+from .thirtydaysinger import ThirtyDaySingerIE, ThirtyDaySingerPlaylistIE
 from .thisamericanlife import ThisAmericanLifeIE
 from .thisoldhouse import ThisOldHouseIE
 from .thisvid import (

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/thirtydaysinger.py

@@ -0,0 +1,104 @@
+import re
+
+from .wistia import WistiaBaseIE
+from ..utils import clean_html, get_elements_html_by_class
+
+
+class ThirtyDaySingerBase(WistiaBaseIE):
+    _INDEX_EXTRACTION_RE = r'/tutorial/[\w-]+/(?P<index>[\w-]+)'
+
+    def _extract_for_url(self, url):
+        lesson_index = re.search(self._INDEX_EXTRACTION_RE, url).group('index')
+        webpage = self._download_webpage(url, lesson_index)
+        match = next(self._extract_wistia_async_embed(webpage))
+        embed_config = self._download_embed_config('medias', match.group('id'), url)
+
+        embed_infojson = self._extract_media(embed_config)
+        webpage_infojson = self._extract_webpage_data(webpage)
+
+        return {**embed_infojson, **webpage_infojson}
+
+    def _extract_webpage_data(self, webpage):
+        title = self._html_search_regex(r'<h1>([^<]+)</h1>', webpage, 'title')
+        fallback_title = self._html_extract_title(webpage)
+        description = self._html_search_meta('description', webpage, fatal=False)
+
+        return {
+            'title': title or fallback_title,
+            'description': clean_html(self._format_html_list(description)),
+        }
+
+    # The site makes extensive use of HTML lists for formatting and `clean_html`
+    # doesn't handle them well. This is needed to keep lists readable.
+    def _format_html_list(self, html):
+        replacements = {
+            '<ul>': '<br>',
+            '</ul': '<br>',
+            '<li>': '<br>- ',
+            '</li>': '',
+        }
+
+        for k, v in replacements.items():
+            html = html.replace(k, v)
+
+        return html
+
+
+class ThirtyDaySingerIE(ThirtyDaySingerBase):
+    _VALID_URL = r'https?://www\.30daysinger\.com/tutorial/[\w-]+/[\w-]+'
+
+    _TESTS = [{
+        'url': 'https://www.30daysinger.com/tutorial/30-day-beginner-course-with-jonathan-estabrooks/1',
+        'md5': '56bb11529b9777899b27b599d4b16cf6',
+        'info_dict': {
+            'id': 'tegd38l3d5',
+            'ext': 'mp4',
+            'thumbnail': 'http://embed.wistia.com/deliveries/c26a85cb98e32efa8a5e12a0576e63355af66230.jpg',
+            'upload_date': '20190608',
+            'description': 'md5:d3291de8988be57b1d3e411126ba4d33',
+            'duration': 344.22,
+            'timestamp': 1559952526,
+            'title': 'Welcome to 30 Day Singer',
+        },
+    }]
+
+    def _real_extract(self, url):
+        return self._extract_for_url(url)
+
+
+class ThirtyDaySingerPlaylistIE(ThirtyDaySingerBase):
+    _URI_BASE = 'https://www.30daysinger.com'
+    _VALID_URL = r'https?://www\.30daysinger\.com/tutorial/(?P<playlist_id>[\w-]+)/?(?:$|[#?])'
+
+    _TESTS = [{
+        'url': 'https://www.30daysinger.com/tutorial/30-day-beginner-course-with-jonathan-estabrooks',
+        'info_dict': {
+            'id': '30-day-beginner-course-with-jonathan-estabrooks',
+            'description': 'md5:8cf6d6c7c377895653c9cde9dfc4104f',
+            'title': '30 Day Beginner Course with Jonathan Estabrooks',
+        },
+        'playlist_count': 1,
+        'expected_warnings': ['This video is for premium members only'],
+    }]
+
+    def _real_extract(self, url):
+        playlist_id = self._match_valid_url(url).group('playlist_id')
+        webpage = self._download_webpage(url, playlist_id)
+        playlist_attrs = self._extract_webpage_data(webpage)
+
+        entries = []
+        for html_element in get_elements_html_by_class('playlist-item-link', webpage):
+            href = self._search_regex(r'href="([^"]+)"', html_element, 'href')
+
+            if not href:
+                continue
+            # Often _some_ content is free so we should still download that but warn the user
+            # when we encounter premium content.
+            # NOTE: this only applies to the playlist extractor, not the single video extractor
+            if 'upgrade' in href:
+                self.report_warning('This video is for premium members only')
+                continue
+
+            entries.append(self._extract_for_url(self._URI_BASE + href))
+
+        return self.playlist_result(entries, playlist_id, playlist_attrs['title'], playlist_attrs['description'])

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:
-            continue
-        if ext is not None and not sub.get('ext'):
-            sub['ext'] = ext
+        if not isinstance(sub_id, str):
+            if not lang:
+                continue
+            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