Merge 049862b803 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/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/rtp.py

@@ -4,40 +4,97 @@ import re
 import urllib.parse
 
 from .common import InfoExtractor
-from ..utils import js_to_json
+from ..utils import (
+    ExtractorError,
+    determine_ext,
+    join_nonempty,
+    js_to_json,
+)
 
 
 class RTPIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?rtp\.pt/play/(?:(?:estudoemcasa|palco|zigzag)/)?p(?P<program_id>[0-9]+)/(?P<id>[^/?#]+)'
+    _VALID_URL = r'https?://(?:(?:www\.)?rtp\.pt/play/(?P<subarea>.*/)?p(?P<program_id>\d+)/|arquivos\.rtp\.pt/conteudos/)(?P<id>[^/?#]+)'
     _TESTS = [{
-        'url': 'http://www.rtp.pt/play/p405/e174042/paixoes-cruzadas',
+        'url': 'https://www.rtp.pt/play/p9165/e562949/por-do-sol',
-        'md5': 'e736ce0c665e459ddb818546220b4ef8',
         'info_dict': {
-            'id': 'e174042',
+            'id': 'e562949',
-            'ext': 'mp3',
-            'title': 'Paixões Cruzadas',
-            'description': 'As paixões musicais de António Cartaxo e António Macedo',
-            'thumbnail': r're:^https?://.*\.jpg',
-        },
-    }, {
-        'url': 'https://www.rtp.pt/play/zigzag/p13166/e757904/25-curiosidades-25-de-abril',
-        'md5': '9a81ed53f2b2197cfa7ed455b12f8ade',
-        'info_dict': {
-            'id': 'e757904',
             'ext': 'mp4',
-            'title': '25 Curiosidades, 25 de Abril',
+            'title': 'Pôr do Sol Episódio 1',
-            'description': 'Estudar ou não estudar - Em cada um dos episódios descobrimos uma curiosidade acerca de como era viver em Portugal antes da revolução do 25 de abr',
+            'description': 'Madalena Bourbon de Linhaça vive atormentada pelo segredo que esconde desde 1990. Matilde Bourbon de Linhaça sonha fugir com o seu amor proibido. O',
-            'thumbnail': r're:^https?://.*\.jpg',
+            'thumbnail': r're:https?://.*\.(?:jpg|png)',
         },
     }, {
-        'url': 'http://www.rtp.pt/play/p831/a-quimica-das-coisas',
+        'url': 'https://www.rtp.pt/play/p12646/e738493/telejornal',
-        'only_matching': True,
+        'info_dict': {
+            'id': 'e738493',
+            'ext': 'mp4',
+            'title': 'Telejornal de 01 jan 2024 PARTE 1',
+            'description': 'A mais rigorosa seleção de notícias, todos os dias às 20h00. De segunda a domingo, João Adelino Faria, José Rodrigues dos Santos e Ana Lourenço',
+            'thumbnail': r're:https?://.*\.(?:jpg|png)',
+        },
     }, {
-        'url': 'https://www.rtp.pt/play/estudoemcasa/p7776/portugues-1-ano',
+        'url': 'https://www.rtp.pt/play/p6646/e457262/grande-entrevista',
-        'only_matching': True,
+        'info_dict': {
+            'id': 'e457262',
+            'ext': 'mp4',
+            'title': 'Grande Entrevista Episódio 7 - de 19 fev 2020',
+            'description': 'Bruno Nogueira - É um dos mais originais humoristas portugueses e de maior êxito! Bruno Nogueira na Grande Entrevista com Vítor Gonçalves.',
+            'thumbnail': r're:https?://.*\.(?:jpg|png)',
+        },
     }, {
-        'url': 'https://www.rtp.pt/play/palco/p13785/l7nnon',
+        'url': 'https://www.rtp.pt/play/p1525/e738522/a-mosca',
-        'only_matching': True,
+        'info_dict': {
+            'id': 'e738522',
+            'ext': 'mp4',
+            'title': 'A Mosca de 02 jan 2024',
+            'description': 'Ano novo, vida nova - Ano novo, vida nova',
+            'thumbnail': r're:https?://.*\.(?:jpg|png)',
+        },
+    }, {
+        'url': 'https://www.rtp.pt/play/estudoemcasa/p7776/e539826/portugues-1-ano',
+        'info_dict': {
+            'id': 'e539826',
+            'ext': 'mp4',
+            'title': 'Português - 1.º ano , aula 45 - 27 abr 2021',
+            'description': 'A História do Pedrito Coelho, de Beatrix Potter. O dígrafo \'lh\' - A História do Pedrito Coelho, de Beatrix Potter. O dígrafo \'lh\'.',
+            'thumbnail': r're:https?://.*\.(?:jpg|png)',
+        },
+    }, {
+        'url': 'https://www.rtp.pt/play/zigzag/p13857/e794575/zig-zag-zzz-e-amigos',
+        'info_dict': {
+            'id': 'e794575',
+            'ext': 'mp4',
+            'title': 'Zig, Zag, Zzz e Amigos Episódio 1 - de 16 set 2024',
+            'description': 'O Brinquedo Perdido - Zig, Zag e Zzz são três amigos inseparáveis que partilham aventuras emocionantes e cheias de imaginação. Exploram o mundo <20>',
+            'thumbnail': r're:https?://.*\.(?:jpg|png)',
+        },
+    }, {
+        'url': 'https://www.rtp.pt/play/palco/p13151/premio-miguel-rovisco-2023-requiem-por-isabel',
+        'info_dict': {
+            'id': 'premio-miguel-rovisco-2023-requiem-por-isabel',
+            'ext': 'mp4',
+            'title': 'Prémio Miguel Rovisco 23: Requiem Por Isabel de 30 mar 2024',
+            'description': 'Lucrécia foi a atriz mais famosa e requisitada do seu tempo. Este já não é o seu tempo. A debater-se com a decrepitude física e financeira, foi o',
+            'thumbnail': r're:https?://.*\.(?:jpg|png)',
+        },
+    }, {
+        'url': 'https://arquivos.rtp.pt/conteudos/liga-dos-ultimos-152/',
+        'info_dict': {
+            'id': 'liga-dos-ultimos-152',
+            'ext': 'mp4',
+            'title': 'Liga dos Últimos – RTP Arquivos',
+            'description': 'Magazine desportivo, com apresentação de Álvaro Costa e comentários em estúdio do professor Hernâni Gonçalves e do sociólogo João Nuno Coelho. Destaque para os jogos de futebol das equipas dos escalões secundários de Portugal, com momentos dos jogos: Agrário de Lamas vs Pampilhoense e Apúlia vs Fragoso.',
+            'thumbnail': r're:https?://.*\.(?:jpg|png)',
+        },
+    }, {
+        'url': 'https://www.rtp.pt/play/p510/e786608/aleixo-fm',
+        'info_dict': {
+            'id': 'e786608',
+            'ext': 'mp3',
+            'title': 'Aleixo FM de 31 jul 2024',
+            'description': 'Melhor dia pra casar - Já o diz Joaquim de Magalhães Fernandes Barreiros, comummente conhecido como Quim Barreiros. Mas será mesmo este o melhor di',
+            'thumbnail': r're:https?://.*\.(?:jpg|png)',
+        },
     }]
 
     _RX_OBFUSCATION = re.compile(r'''(?xs)
@@ -60,42 +117,77 @@ class RTPIE(InfoExtractor):
         video_id = self._match_id(url)
 
         webpage = self._download_webpage(url, video_id)
-        title = self._html_search_meta(
-            'twitter:title', webpage, display_name='title', fatal=True)
 
+        # Title tag includes relevant data
+        title = self._html_search_regex(r'<title>(.+?)</title>', webpage, 'title', default='')
+
+        # Raise error if episode is unavailable
+        if 'Este episódio não se encontra disponível' in title:
+            raise ExtractorError('Episode unavailable', expected=True)
+
+        # Replace irrelevant text in title
+        title = re.sub(r' -  ?(RTP Play|Estudo Em Casa|Zig Zag Play|RTP Palco)( - RTP)?', '', title)
+
+        # Check if it's a episode split in parts
+        part = self._html_search_regex(r'section\-parts.*<span.*>(.+?)</span>.*</ul>', webpage, 'part', default=None)
+
+        # Add episode part identification to title if it exists
+        title = join_nonempty(title, part, delim=' ')
+
+        # Extract f and config from page
         f, config = self._search_regex(
             r'''(?sx)
                 (?:var\s+f\s*=\s*(?P<f>".*?"|{[^;]+?});\s*)?
-                var\s+player1\s+=\s+new\s+RTPPlayer\s*\((?P<config>{(?:(?!\*/).)+?})\);(?!\s*\*/)
+                var\s+player1?\s+=\s+new\s+RTPPlayer\s*\((?P<config>{(?:(?!\*/).)+?})\);(?!\s*\*/)
             ''', webpage,
             'player config', group=('f', 'config'))
 
         config = self._parse_json(
             config, video_id,
             lambda data: self.__unobfuscate(data, video_id=video_id))
+
+        # Estudo em Casa / Zig Zag / Palco / RTP Arquivos subareas don't include f
         f = config['file'] if not f else self._parse_json(
             f, video_id,
             lambda data: self.__unobfuscate(data, video_id=video_id))
 
         formats = []
         if isinstance(f, dict):
-            f_hls = f.get('hls')
+            file_hls = f.get('hls')
-            if f_hls is not None:
+            file_fps = f.get('fps')
-                formats.extend(self._extract_m3u8_formats(
-                    f_hls, video_id, 'mp4', 'm3u8_native', m3u8_id='hls'))
 
-            f_dash = f.get('dash')
+            if file_fps is not None:
-            if f_dash is not None:
+                # RTP Arquivos specific use case
-                formats.extend(self._extract_mpd_formats(f_dash, video_id, mpd_id='dash'))
+                if '/arquivo/' in file_fps:
+                    file_key = config['fileKey']
+                    split_file_key = file_key.split('/')
+                    filename = split_file_key[-1]
+                    del split_file_key[-1]
+                    split_file_key.extend([f'index.m3u8?tlm=hls&streams={filename}.m3u8'])
+
+                    path = '/'.join(split_file_key)
+
+                    file_hls = f'https://streaming-arquivo-ondemand.rtp.pt/nas2.share{path}'
+                elif file_hls is None:
+                    file_hls = file_fps.replace('drm-fps', 'hls')
+
+            formats.extend(self._extract_m3u8_formats(
+                file_hls, video_id, 'mp4', 'm3u8_native', m3u8_id='hls'))
         else:
-            formats.append({
+            ext = determine_ext(f)
-                'format_id': 'f',
+
-                'url': f,
+            if ext == 'm3u8':
-                'vcodec': 'none' if config.get('mediaType') == 'audio' else None,
+                formats.extend(self._extract_m3u8_formats(
-            })
+                    f, video_id, 'mp4', 'm3u8_native', m3u8_id='hls'))
+
+            else:
+                formats.append({
+                    'format_id': 'f',
+                    'url': f,
+                    'vcodec': 'none' if config.get('mediaType') == 'audio' else None,
+                })
 
         subtitles = {}
-
         vtt = config.get('vtt')
         if vtt is not None:
             for lcode, lname, url in vtt:
@@ -108,7 +200,7 @@ class RTPIE(InfoExtractor):
             'id': video_id,
             'title': title,
             'formats': formats,
-            'description': self._html_search_meta(['description', 'twitter:description'], webpage),
+            'description': self._html_search_meta(['og:description', 'description'], webpage),
             'thumbnail': config.get('poster') or self._og_search_thumbnail(webpage),
             'subtitles': subtitles,
         }

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