[extractor/kompas] Add extractor (#4562)

Authored by: HobbyistDev

diff --git a/yt_dlp/extractor/_extractors.py b/yt_dlp/extractor/_extractors.py
index c3d9474838b7b1422149ff80edcc773321a58951..3abae19b01b6f3d084b234be2fc74d49e2dd98a1 100644 (file)
--- a/yt_dlp/extractor/_extractors.py
+++ b/yt_dlp/extractor/_extractors.py
@@ -765,6 +765,7 @@
 from .kickstarter import KickStarterIE
 from .kinja import KinjaEmbedIE
 from .kinopoisk import KinoPoiskIE
+from .kompas import KompasVideoIE
 from .konserthusetplay import KonserthusetPlayIE
 from .koo import KooIE
 from .kth import KTHIE

diff --git a/yt_dlp/extractor/kompas.py b/yt_dlp/extractor/kompas.py
new file mode 100644 (file)
index 0000000..d400c42
--- /dev/null
+++ b/yt_dlp/extractor/kompas.py
@@ -0,0 +1,68 @@
+from .common import InfoExtractor
+from ..utils import (
+    clean_html,
+    float_or_none,
+    traverse_obj,
+    try_call,
+)
+
+# Video from www.kompas.tv and video.kompas.com seems use jixie player
+# see [1] https://jixie.atlassian.net/servicedesk/customer/portal/2/article/1339654214?src=-1456335525,
+# [2] https://scripts.jixie.media/jxvideo.3.1.min.js for more info
+
+
+class KompasVideoIE(InfoExtractor):
+    _VALID_URL = r'https?://video\.kompas\.com/\w+/(?P<id>\d+)/(?P<slug>[\w-]+)'
+    _TESTS = [{
+        'url': 'https://video.kompas.com/watch/164474/kim-jong-un-siap-kirim-nuklir-lawan-as-dan-korsel',
+        'info_dict': {
+            'id': '164474',
+            'ext': 'mp4',
+            'title': 'Kim Jong Un Siap Kirim Nuklir Lawan AS dan Korsel',
+            'description': 'md5:262530c4fb7462398235f9a5dba92456',
+            'uploader_id': '9262bf2590d558736cac4fff7978fcb1',
+            'display_id': 'kim-jong-un-siap-kirim-nuklir-lawan-as-dan-korsel',
+            'duration': 85.066667,
+            'categories': ['news'],
+            'thumbnail': 'https://video.jixie.media/1001/164474/164474_1280x720.jpg',
+            'tags': 'count:9',
+        }
+    }]
+
+    def _real_extract(self, url):
+        video_id, display_id = self._match_valid_url(url).group('id', 'slug')
+        webpage = self._download_webpage(url, display_id)
+
+        json_data = self._download_json(
+            'https://apidam.jixie.io/api/public/stream', display_id,
+            query={'metadata': 'full', 'video_id': video_id})['data']
+
+        formats, subtitles = [], {}
+        for stream in json_data['streams']:
+            if stream.get('type') == 'HLS':
+                fmt, sub = self._extract_m3u8_formats_and_subtitles(stream.get('url'), display_id, ext='mp4')
+                formats.extend(fmt)
+                self._merge_subtitles(sub, target=subtitles)
+            else:
+                formats.append({
+                    'url': stream.get('url'),
+                    'width': stream.get('width'),
+                    'height': stream.get('height'),
+                    'ext': 'mp4',
+                })
+
+        self._sort_formats(formats)
+        return {
+            'id': video_id,
+            'display_id': display_id,
+            'formats': formats,
+            'subtitles': subtitles,
+            'title': json_data.get('title') or self._html_search_meta(['og:title', 'twitter:title'], webpage),
+            'description': (clean_html(traverse_obj(json_data, ('metadata', 'description')))
+                            or self._html_search_meta(['description', 'og:description', 'twitter:description'], webpage)),
+            'thumbnails': traverse_obj(json_data, ('metadata', 'thumbnails')),
+            'duration': float_or_none(traverse_obj(json_data, ('metadata', 'duration'))),
+            'tags': try_call(lambda: json_data['metadata']['keywords'].split(',')),
+            'categories': try_call(lambda: json_data['metadata']['categories'].split(',')),
+            'uploader_id': json_data.get('owner_id'),
+        }