Merge c1589ea3fe into f3411af12e

Authored by: megumintyan

yt_dlp/extractor/facebook.py

@@ -476,6 +476,29 @@ def _extract_from_url(self, url, video_id):
         webpage = self._download_webpage(
             url.replace('://m.facebook.com/', '://www.facebook.com/'), video_id)
 
+        def extract_follower_count(webpage):
+            post_data = [self._parse_json(j, video_id, fatal=False) for j in re.findall(
+                r'data-sjs>({.*?ScheduledServerJS.*?})</script>', webpage)]
+            post = traverse_obj(post_data, (
+                ..., 'require', ..., ..., ..., '__bbox', 'require', ..., ..., ..., '__bbox', 'result', 'data'), expected_type=dict) or []
+
+            followers = get_first(post, ('user', 'profile_header_renderer', 'user', 'profile_social_context', 'content', ..., 'text',
+                                         lambda k, v: k == 'text' and isinstance(v, str) and v.endswith('followers'))) or None
+            if not isinstance(followers, str):
+                return None
+
+            matches = re.search(r"(\d+)([K|M])?", followers)
+            if matches is None:
+                return None
+
+            count = int(matches[1])
+            unit = matches[2]
+            if unit == "K":
+                count *= 1_000
+            elif unit == "M":
+                count *= 1_000_000
+            return count
+
         def extract_metadata(webpage):
             post_data = [self._parse_json(j, video_id, fatal=False) for j in re.findall(
                 r'data-sjs>({.*?ScheduledServerJS.*?})</script>', webpage)]
@@ -513,6 +536,8 @@ def extract_metadata(webpage):
             # in https://www.facebook.com/yaroslav.korpan/videos/1417995061575415/
             if thumbnail and not re.search(r'\.(?:jpg|png)', thumbnail):
                 thumbnail = None
+            profile_url = get_first(post, ('attachments', ..., 'media', 'creation_story', 'comet_sections', 'actor_photo', 'story', 'actors', ..., 'profile_url'))
+            follower_count = extract_follower_count(self._download_webpage(profile_url, None))
             info_dict = {
                 'description': description,
                 'uploader': uploader,
@@ -524,6 +549,7 @@ def extract_metadata(webpage):
                     webpage, 'view count', default=None)),
                 'concurrent_view_count': get_first(post, (
                     ('video', (..., ..., 'attachments', ..., 'media')), 'liveViewerCount', {int_or_none})),
+                'channel_follower_count': follower_count,
             }
 
             info_json_ld = self._search_json_ld(webpage, video_id, default={})

yt_dlp/extractor/matchtv.py

@@ -1,51 +1,35 @@
-import random
-
 from .common import InfoExtractor
-from ..utils import xpath_text
 
 
 class MatchTVIE(InfoExtractor):
-    _VALID_URL = r'https?://matchtv\.ru(?:/on-air|/?#live-player)'
+    _VALID_URL = [
+        r'https?://matchtv\.ru/on-air/?(?:$|[?#])',
+        r'https?://video\.matchtv\.ru/iframe/channel/106/?(?:$|[?#])',
+    ]
     _TESTS = [{
-        'url': 'http://matchtv.ru/#live-player',
+        'url': 'http://matchtv.ru/on-air/',
         'info_dict': {
             'id': 'matchtv-live',
-            'ext': 'flv',
+            'ext': 'mp4',
             'title': r're:^Матч ТВ - Прямой эфир \d{4}-\d{2}-\d{2} \d{2}:\d{2}$',
-            'is_live': True,
+            'live_status': 'is_live',
         },
         'params': {
             'skip_download': True,
         },
     }, {
-        'url': 'http://matchtv.ru/on-air/',
+        'url': 'https://video.matchtv.ru/iframe/channel/106',
         'only_matching': True,
     }]
 
     def _real_extract(self, url):
         video_id = 'matchtv-live'
-        video_url = self._download_json(
-            'http://player.matchtv.ntvplus.tv/player/smil', video_id,
-            query={
-                'ts': '',
-                'quality': 'SD',
-                'contentId': '561d2c0df7159b37178b4567',
-                'sign': '',
-                'includeHighlights': '0',
-                'userId': '',
-                'sessionId': random.randint(1, 1000000000),
-                'contentType': 'channel',
-                'timeShift': '0',
-                'platform': 'portal',
-            },
-            headers={
-                'Referer': 'http://player.matchtv.ntvplus.tv/embed-player/NTVEmbedPlayer.swf',
-            })['data']['videoUrl']
-        f4m_url = xpath_text(self._download_xml(video_url, video_id), './to')
-        formats = self._extract_f4m_formats(f4m_url, video_id)
+        webpage = self._download_webpage('https://video.matchtv.ru/iframe/channel/106', video_id)
+        video_url = self._html_search_regex(
+            r'data-config="config=(https?://[^?"]+)[?"]', webpage, 'video URL').replace('/feed/', '/media/') + '.m3u8'
         return {
             'id': video_id,
             'title': 'Матч ТВ - Прямой эфир',
             'is_live': True,
-            'formats': formats,
+            'formats': self._extract_m3u8_formats(video_url, video_id, 'mp4', live=True),
         }