1 files changed, 96 insertions, 0 deletions
diff --git a/yt_dlp/extractor/airtv.py b/yt_dlp/extractor/airtv.py
new file mode 100644
index 0000000..0b73a96
--- /dev/null
+++ b/yt_dlp/extractor/airtv.py
@@ -0,0 +1,96 @@
+from .common import InfoExtractor
+from .youtube import YoutubeIE
+from ..utils import (
+    determine_ext,
+    int_or_none,
+    mimetype2ext,
+    parse_iso8601,
+    traverse_obj
+)
+
+
+class AirTVIE(InfoExtractor):
+    _VALID_URL = r'https?://www\.air\.tv/watch\?v=(?P<id>\w+)'
+    _TESTS = [{
+        # without youtube_id
+        'url': 'https://www.air.tv/watch?v=W87jcWleSn2hXZN47zJZsQ',
+        'info_dict': {
+            'id': 'W87jcWleSn2hXZN47zJZsQ',
+            'ext': 'mp4',
+            'release_date': '20221003',
+            'release_timestamp': 1664792603,
+            'channel_id': 'vgfManQlRQKgoFQ8i8peFQ',
+            'title': 'md5:c12d49ed367c3dadaa67659aff43494c',
+            'upload_date': '20221003',
+            'duration': 151,
+            'view_count': int,
+            'thumbnail': 'https://cdn-sp-gcs.air.tv/videos/W/8/W87jcWleSn2hXZN47zJZsQ/b13fc56464f47d9d62a36d110b9b5a72-4096x2160_9.jpg',
+            'timestamp': 1664792603,
+        }
+    }, {
+        # with youtube_id
+        'url': 'https://www.air.tv/watch?v=sv57EC8tRXG6h8dNXFUU1Q',
+        'info_dict': {
+            'id': '2ZTqmpee-bQ',
+            'ext': 'mp4',
+            'comment_count': int,
+            'tags': 'count:11',
+            'channel_follower_count': int,
+            'like_count': int,
+            'uploader': 'Newsflare',
+            'thumbnail': 'https://i.ytimg.com/vi_webp/2ZTqmpee-bQ/maxresdefault.webp',
+            'availability': 'public',
+            'title': 'Geese Chase Alligator Across Golf Course',
+            'uploader_id': 'NewsflareBreaking',
+            'channel_url': 'https://www.youtube.com/channel/UCzSSoloGEz10HALUAbYhngQ',
+            'description': 'md5:99b21d9cea59330149efbd9706e208f5',
+            'age_limit': 0,
+            'channel_id': 'UCzSSoloGEz10HALUAbYhngQ',
+            'uploader_url': 'http://www.youtube.com/user/NewsflareBreaking',
+            'view_count': int,
+            'categories': ['News & Politics'],
+            'live_status': 'not_live',
+            'playable_in_embed': True,
+            'channel': 'Newsflare',
+            'duration': 37,
+            'upload_date': '20180511',
+        }
+    }]
+
+    def _get_formats_and_subtitle(self, json_data, video_id):
+        formats, subtitles = [], {}
+        for source in traverse_obj(json_data, 'sources', 'sources_desktop', ...):
+            ext = determine_ext(source.get('src'), mimetype2ext(source.get('type')))
+            if ext == 'm3u8':
+                fmts, subs = self._extract_m3u8_formats_and_subtitles(source.get('src'), video_id)
+                formats.extend(fmts)
+                self._merge_subtitles(subs, target=subtitles)
+            else:
+                formats.append({'url': source.get('src'), 'ext': ext})
+        return formats, subtitles
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id)
+
+        nextjs_json = self._search_nextjs_data(webpage, display_id)['props']['pageProps']['initialState']['videos'][display_id]
+        if nextjs_json.get('youtube_id'):
+            return self.url_result(
+                f'https://www.youtube.com/watch?v={nextjs_json.get("youtube_id")}', YoutubeIE)
+
+        formats, subtitles = self._get_formats_and_subtitle(nextjs_json, display_id)
+        return {
+            'id': display_id,
+            'title': nextjs_json.get('title') or self._html_search_meta('og:title', webpage),
+            'formats': formats,
+            'subtitles': subtitles,
+            'description': nextjs_json.get('description') or None,
+            'duration': int_or_none(nextjs_json.get('duration')),
+            'thumbnails': [
+                {'url': thumbnail}
+                for thumbnail in traverse_obj(nextjs_json, ('default_thumbnails', ...))],
+            'channel_id': traverse_obj(nextjs_json, 'channel', 'channel_slug'),
+            'timestamp': parse_iso8601(nextjs_json.get('created')),
+            'release_timestamp': parse_iso8601(nextjs_json.get('published')),
+            'view_count': int_or_none(nextjs_json.get('views')),
+        }