[extractor/newspicks] Add extractor (#4725)

Authored by: Lesmiscore

diff --git a/yt_dlp/extractor/_extractors.py b/yt_dlp/extractor/_extractors.py
index 60e1b716f1c982bf96b8b2623dba3f629ecaafa4..1cded3ddf84e842254d3e16a90452831f03b2988 100644 (file)
--- a/yt_dlp/extractor/_extractors.py
+++ b/yt_dlp/extractor/_extractors.py
@@ -1083,6 +1083,7 @@
     NewgroundsPlaylistIE,
     NewgroundsUserIE,
 )
+from .newspicks import NewsPicksIE
 from .newstube import NewstubeIE
 from .newsy import NewsyIE
 from .nextmedia import (

diff --git a/yt_dlp/extractor/common.py b/yt_dlp/extractor/common.py
index f950d28ed3f4d256fcb6bf0923392d9ac8945034..b792219553706a5a9aead2846ee4199b8240d09b 100644 (file)
--- a/yt_dlp/extractor/common.py
+++ b/yt_dlp/extractor/common.py
@@ -3260,7 +3260,7 @@ def _media_formats(src, cur_media_type, type_info=None):
                 'subtitles': {},
             }
             media_attributes = extract_attributes(media_tag)
-            src = strip_or_none(media_attributes.get('src'))
+            src = strip_or_none(dict_get(media_attributes, ('src', 'data-video-src', 'data-src', 'data-source')))
             if src:
                 f = parse_content_type(media_attributes.get('type'))
                 _, formats = _media_formats(src, media_type, f)
@@ -3271,7 +3271,7 @@ def _media_formats(src, cur_media_type, type_info=None):
                     s_attr = extract_attributes(source_tag)
                     # data-video-src and data-src are non standard but seen
                     # several times in the wild
-                    src = strip_or_none(dict_get(s_attr, ('src', 'data-video-src', 'data-src')))
+                    src = strip_or_none(dict_get(s_attr, ('src', 'data-video-src', 'data-src', 'data-source')))
                     if not src:
                         continue
                     f = parse_content_type(s_attr.get('type'))

diff --git a/yt_dlp/extractor/newspicks.py b/yt_dlp/extractor/newspicks.py
new file mode 100644 (file)
index 0000000..0232d53
--- /dev/null
+++ b/yt_dlp/extractor/newspicks.py
@@ -0,0 +1,54 @@
+import re
+
+from .common import InfoExtractor
+from ..utils import ExtractorError
+
+
+class NewsPicksIE(InfoExtractor):
+    _VALID_URL = r'https://newspicks.com/movie-series/(?P<channel_id>\d+)\?movieId=(?P<id>\d+)'
+
+    _TESTS = [{
+        'url': 'https://newspicks.com/movie-series/11?movieId=1813',
+        'info_dict': {
+            'id': '1813',
+            'title': '日本の課題を破壊せよ【ゲスト：成田悠輔】',
+            'description': 'md5:09397aad46d6ded6487ff13f138acadf',
+            'channel': 'HORIE ONE',
+            'channel_id': '11',
+            'release_date': '20220117',
+            'thumbnail': r're:https://.+jpg',
+            'ext': 'mp4',
+        },
+    }]
+
+    def _real_extract(self, url):
+        video_id, channel_id = self._match_valid_url(url).group('id', 'channel_id')
+        webpage = self._download_webpage(url, video_id)
+        entries = self._parse_html5_media_entries(
+            url, webpage.replace('movie-for-pc', 'movie'), video_id, 'hls')
+        if not entries:
+            raise ExtractorError('No HTML5 media elements found')
+        info = entries[0]
+        self._sort_formats(info['formats'])
+
+        title = self._html_search_meta('og:title', webpage, fatal=False)
+        description = self._html_search_meta(
+            ('og:description', 'twitter:title'), webpage, fatal=False)
+        channel = self._html_search_regex(
+            r'value="11".+?<div\s+class="title">(.+?)</div', webpage, 'channel name', fatal=False)
+        if not title or not channel:
+            title, channel = re.split(r'\s*|\s*', self._html_extract_title(webpage))
+
+        release_date = self._search_regex(
+            r'<span\s+class="on-air-date">\s*(\d+)年(\d+)月(\d+)日\s*</span>',
+            webpage, 'release date', fatal=False, group=(1, 2, 3))
+
+        info.update({
+            'id': video_id,
+            'title': title,
+            'description': description,
+            'channel': channel,
+            'channel_id': channel_id,
+            'release_date': ('%04d%02d%02d' % tuple(map(int, release_date))) if release_date else None,
+        })
+        return info