Merge 8f5a765e25 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

@@ -1518,8 +1518,12 @@ from .pgatour import PGATourIE
 from .philharmoniedeparis import PhilharmonieDeParisIE
 from .phoenix import PhoenixIE
 from .photobucket import PhotobucketIE
+from .pialive import PiaLiveIE
 from .piapro import PiaproIE
-from .piaulizaportal import PIAULIZAPortalIE
+from .piaulizaportal import (
+    PIAULIZAPortalAPIIE,
+    PIAULIZAPortalIE,
+)
 from .picarto import (
     PicartoIE,
     PicartoVodIE,

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/pialive.py

@@ -0,0 +1,126 @@
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    clean_html,
+    extract_attributes,
+    get_element_by_class,
+    get_element_html_by_class,
+    multipart_encode,
+    unified_timestamp,
+    url_or_none,
+)
+from ..utils.traversal import traverse_obj
+
+
+class PiaLiveIE(InfoExtractor):
+    PLAYER_ROOT_URL = 'https://player.pia-live.jp/'
+    PIA_LIVE_API_URL = 'https://api.pia-live.jp'
+    API_KEY = 'kfds)FKFps-dms9e'
+    _VALID_URL = r'https?://player\.pia-live\.jp/stream/(?P<id>[\w-]+)'
+
+    _TESTS = [
+        {
+            'url': 'https://player.pia-live.jp/stream/4JagFBEIM14s_hK9aXHKf3k3F3bY5eoHFQxu68TC6krUDqGOwN4d61dCWQYOd6CTxl4hjya9dsfEZGsM4uGOUdax60lEI4twsXGXf7crmz8Gk__GhupTrWxA7RFRVt76',
+            'info_dict': {
+                'id': '88f3109a-f503-4d0f-a9f7-9f39ac745d84',
+                'display_id': '2431867_001',
+                'title': 'こながめでたい日２０２４の視聴ページ | PIA LIVE STREAM(ぴあライブストリーム)',
+                'live_status': 'was_live',
+                'comment_count': int,
+            },
+            'params': {
+                'getcomments': True,
+                'skip_download': True,
+                'ignore_no_formats_error': True,
+            },
+            'skip': 'The video is no longer available',
+        },
+        {
+            'url': 'https://player.pia-live.jp/stream/4JagFBEIM14s_hK9aXHKf3k3F3bY5eoHFQxu68TC6krJdu0GVBVbVy01IwpJ6J3qBEm3d9TCTt1d0eWpsZGj7DrOjVOmS7GAWGwyscMgiThopJvzgWC4H5b-7XQjAfRZ',
+            'info_dict': {
+                'id': '9ce8b8ba-f6d1-4d1f-83a0-18c3148ded93',
+                'display_id': '2431867_002',
+                'title': 'こながめでたい日２０２４の視聴ページ | PIA LIVE STREAM(ぴあライブストリーム)',
+                'live_status': 'was_live',
+                'comment_count': int,
+            },
+            'params': {
+                'getcomments': True,
+                'skip_download': True,
+                'ignore_no_formats_error': True,
+            },
+            'skip': 'The video is no longer available',
+        },
+    ]
+
+    def _extract_vars(self, variable, html):
+        return self._search_regex(
+            rf'(?:var|const|let)\s+{variable}\s*=\s*(["\'])(?P<value>(?:(?!\1).)+)\1',
+            html, f'variable {variable}', group='value')
+
+    def _real_extract(self, url):
+        video_key = self._match_id(url)
+        webpage = self._download_webpage(url, video_key)
+
+        program_code = self._extract_vars('programCode', webpage)
+        article_code = self._extract_vars('articleCode', webpage)
+        title = self._html_extract_title(webpage)
+
+        if get_element_html_by_class('play-end', webpage):
+            raise ExtractorError('The video is no longer available', expected=True, video_id=program_code)
+
+        if start_info := clean_html(get_element_by_class('play-waiting__date', webpage)):
+            date, time = self._search_regex(
+                r'(?P<date>\d{4}/\d{1,2}/\d{1,2})\([月火水木金土日]\)(?P<time>\d{2}:\d{2})',
+                start_info, 'start_info', fatal=False, group=('date', 'time'))
+            if date and time:
+                release_timestamp_str = f'{date} {time} +09:00'
+                release_timestamp = unified_timestamp(release_timestamp_str)
+                self.raise_no_formats(f'The video will be available after {release_timestamp_str}', expected=True)
+                return {
+                    'id': program_code,
+                    'title': title,
+                    'live_status': 'is_upcoming',
+                    'release_timestamp': release_timestamp,
+                }
+
+        payload, content_type = multipart_encode({
+            'play_url': video_key,
+            'api_key': self.API_KEY,
+        })
+        api_kwargs = {
+            'video_id': program_code,
+            'data': payload,
+            'headers': {'Content-Type': content_type, 'Referer': self.PLAYER_ROOT_URL},
+        }
+
+        player_tag_list = self._download_json(
+            f'{self.PIA_LIVE_API_URL}/perf/player-tag-list/{program_code}', **api_kwargs,
+            note='Fetching player tag list', errnote='Unable to fetch player tag list')
+        chat_room_url = None
+        if self.get_param('getcomments'):
+            chat_room_url = traverse_obj(self._download_json(
+                f'{self.PIA_LIVE_API_URL}/perf/chat-tag-list/{program_code}/{article_code}', **api_kwargs,
+                note='Fetching chat info', errnote='Unable to fetch chat info', fatal=False),
+                ('data', 'chat_one_tag', {extract_attributes}, 'src', {url_or_none}))
+
+        return self.url_result(
+            extract_attributes(player_tag_list['data']['movie_one_tag'])['src'], url_transparent=True,
+            video_title=title, display_id=program_code, __post_extractor=self.extract_comments(
+                program_code, chat_room_url))
+
+    def _get_comments(self, video_id, chat_room_url):
+        if not chat_room_url:
+            return
+        if comment_page := self._download_webpage(
+                chat_room_url, video_id, headers={'Referer': self.PLAYER_ROOT_URL},
+                note='Fetching comment page', errnote='Unable to fetch comment page', fatal=False):
+            yield from traverse_obj(self._search_json(
+                r'var\s+_history\s*=', comment_page, 'comment list',
+                video_id, contains_pattern=r'\[(?s:.+)\]', fatal=False), (..., {
+                    'timestamp': 0,
+                    'author_is_uploader': (1, {lambda x: x == 2}),
+                    'author': 2,
+                    'text': 3,
+                    'id': 4,
+                }))

yt_dlp/extractor/piaulizaportal.py

@@ -1,11 +1,62 @@
 from .common import InfoExtractor
-from ..utils import (
-    ExtractorError,
-    int_or_none,
-    parse_qs,
-    time_seconds,
-    traverse_obj,
-)
+from ..utils import ExtractorError, int_or_none, parse_qs, time_seconds
+from ..utils.traversal import traverse_obj
+
+
+class PIAULIZAPortalAPIIE(InfoExtractor):
+    _VALID_URL = r'https://player-api\.p\.uliza\.jp/v1/players/[^?#]+\?(?:[^#]*&)?name=(?P<id>[^#&]+)'
+    _TESTS = [
+        {
+            'url': 'https://player-api.p.uliza.jp/v1/players/timeshift-disabled/pia/admin?type=normal&playerobjectname=ulizaPlayer&name=livestream01_dvr&repeatable=true',
+            'info_dict': {
+                'id': '88f3109a-f503-4d0f-a9f7-9f39ac745d84',
+                'ext': 'mp4',
+                'title': '88f3109a-f503-4d0f-a9f7-9f39ac745d84',
+                'live_status': 'was_live',
+            },
+        },
+        {
+            'url': 'https://player-api.p.uliza.jp/v1/players/uliza_jp_gallery_normal/promotion/admin?type=presentation&name=cookings&targetid=player1',
+            'info_dict': {
+                'id': 'ae350126-5e22-4a7f-a8ac-8d0fd448b800',
+                'ext': 'mp4',
+                'title': 'ae350126-5e22-4a7f-a8ac-8d0fd448b800',
+                'live_status': 'not_live',
+            },
+        },
+        {
+            'url': 'https://player-api.p.uliza.jp/v1/players/default-player/pia/admin?type=normal&name=pia_movie_uliza_fix&targetid=ulizahtml5&repeatable=true',
+            'info_dict': {
+                'id': '0644ecc8-e354-41b4-b957-3b08a2d63df1',
+                'ext': 'mp4',
+                'title': '0644ecc8-e354-41b4-b957-3b08a2d63df1',
+                'live_status': 'not_live',
+            },
+        },
+    ]
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+        player_data = self._download_webpage(
+            url, display_id, headers={'Referer': 'https://player-api.p.uliza.jp/'},
+            note='Fetching player data', errnote='Unable to fetch player data')
+
+        m3u8_url = self._search_regex(
+            r'["\'](https://vms-api\.p\.uliza\.jp/v1/prog-index\.m3u8[^"\']+)', player_data, 'm3u8 url')
+        video_id = parse_qs(m3u8_url).get('ss', [display_id])[0]
+
+        formats = self._extract_m3u8_formats(m3u8_url, video_id)
+        m3u8_type = self._search_regex(
+            r'/hls/(dvr|video)/', traverse_obj(formats, (0, 'url')), 'm3u8 type', default=None)
+        return {
+            'id': video_id,
+            'title': video_id,
+            'formats': formats,
+            'live_status': {
+                'video': 'is_live',
+                'dvr': 'was_live',  # short-term archives
+            }.get(m3u8_type, 'not_live'),  # VOD or long-term archives
+        }
 
 
 class PIAULIZAPortalIE(InfoExtractor):
@@ -14,7 +65,8 @@ class PIAULIZAPortalIE(InfoExtractor):
     _TESTS = [{
         'url': 'https://ulizaportal.jp/pages/005f18b7-e810-5618-cb82-0987c5755d44',
         'info_dict': {
-            'id': '005f18b7-e810-5618-cb82-0987c5755d44',
+            'id': 'ae350126-5e22-4a7f-a8ac-8d0fd448b800',
+            'display_id': '005f18b7-e810-5618-cb82-0987c5755d44',
             'title': 'プレゼンテーションプレイヤーのサンプル',
             'live_status': 'not_live',
         },
@@ -25,7 +77,8 @@ class PIAULIZAPortalIE(InfoExtractor):
     }, {
         'url': 'https://ulizaportal.jp/pages/005e1b23-fe93-5780-19a0-98e917cc4b7d?expires=4102412400&signature=f422a993b683e1068f946caf406d211c17d1ef17da8bef3df4a519502155aa91&version=1',
         'info_dict': {
-            'id': '005e1b23-fe93-5780-19a0-98e917cc4b7d',
+            'id': '0644ecc8-e354-41b4-b957-3b08a2d63df1',
+            'display_id': '005e1b23-fe93-5780-19a0-98e917cc4b7d',
             'title': '【確認用】視聴サンプルページ（ULIZA）',
             'live_status': 'not_live',
         },
@@ -44,27 +97,9 @@ class PIAULIZAPortalIE(InfoExtractor):
 
         webpage = self._download_webpage(url, video_id)
 
-        player_data = self._download_webpage(
-            self._search_regex(
-                r'<script [^>]*\bsrc="(https://player-api\.p\.uliza\.jp/v1/players/[^"]+)"',
-                webpage, 'player data url'),
-            video_id, headers={'Referer': 'https://ulizaportal.jp/'},
-            note='Fetching player data', errnote='Unable to fetch player data')
-
-        formats = self._extract_m3u8_formats(
-            self._search_regex(
-                r'["\'](https://vms-api\.p\.uliza\.jp/v1/prog-index\.m3u8[^"\']+)', player_data,
-                'm3u8 url', default=None),
-            video_id, fatal=False)
-        m3u8_type = self._search_regex(
-            r'/hls/(dvr|video)/', traverse_obj(formats, (0, 'url')), 'm3u8 type', default=None)
-
-        return {
-            'id': video_id,
-            'title': self._html_extract_title(webpage),
-            'formats': formats,
-            'live_status': {
-                'video': 'is_live',
-                'dvr': 'was_live',  # short-term archives
-            }.get(m3u8_type, 'not_live'),  # VOD or long-term archives
-        }
+        player_data_url = self._search_regex(
+            r'<script [^>]*\bsrc="(https://player-api\.p\.uliza\.jp/v1/players/[^"]+)"',
+            webpage, 'player data url')
+        return self.url_result(
+            player_data_url, PIAULIZAPortalAPIIE, url_transparent=True,
+            display_id=video_id, video_title=self._html_extract_title(webpage))

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