[ie/huya:live] Fix extractor (#13520)

Authored by: doe1080

yt_dlp/extractor/huya.py

@@ -7,12 +7,13 @@
 from .common import InfoExtractor
 from ..utils import (
     ExtractorError,
+    clean_html,
     int_or_none,
     parse_duration,
     str_or_none,
     try_get,
     unescapeHTML,
-    unified_strdate,
+    update_url,
     update_url_query,
     url_or_none,
 )
@@ -22,8 +23,8 @@
 class HuyaLiveIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.|m\.)?huya\.com/(?!(?:video/play/))(?P<id>[^/#?&]+)(?:\D|$)'
     IE_NAME = 'huya:live'
-    IE_DESC = 'huya.com'
+    IE_DESC = '虎牙直播'
-    TESTS = [{
+    _TESTS = [{
         'url': 'https://www.huya.com/572329',
         'info_dict': {
             'id': '572329',
@@ -149,63 +150,94 @@ class HuyaVideoIE(InfoExtractor):
             'id': '1002412640',
             'ext': 'mp4',
             'title': '8月3日',
-            'thumbnail': r're:https?://.*\.jpg',
+            'categories': ['主机游戏'],
-            'duration': 14,
+            'duration': 14.0,
             'uploader': '虎牙-ATS欧卡车队青木',
             'uploader_id': '1564376151',
             'upload_date': '20240803',
             'view_count': int,
             'comment_count': int,
             'like_count': int,
+            'thumbnail': r're:https?://.+\.jpg',
+            'timestamp': 1722675433,
         },
-    },
+    }, {
-        {
         'url': 'https://www.huya.com/video/play/556054543.html',
         'info_dict': {
             'id': '556054543',
             'ext': 'mp4',
             'title': '我不挑事 也不怕事',
-            'thumbnail': r're:https?://.*\.jpg',
+            'categories': ['英雄联盟'],
-            'duration': 1864,
+            'description': 'md5:58184869687d18ce62dc7b4b2ad21201',
+            'duration': 1864.0,
             'uploader': '卡尔',
             'uploader_id': '367138632',
             'upload_date': '20210811',
             'view_count': int,
             'comment_count': int,
             'like_count': int,
+            'tags': 'count:4',
+            'thumbnail': r're:https?://.+\.jpg',
+            'timestamp': 1628675950,
+        },
+    }, {
+        # Only m3u8 available
+        'url': 'https://www.huya.com/video/play/1063345618.html',
+        'info_dict': {
+            'id': '1063345618',
+            'ext': 'mp4',
+            'title': '峡谷第一中！黑铁上钻石顶级教学对抗elo',
+            'categories': ['英雄联盟'],
+            'comment_count': int,
+            'duration': 21603.0,
+            'like_count': int,
+            'thumbnail': r're:https?://.+\.jpg',
+            'timestamp': 1749668803,
+            'upload_date': '20250611',
+            'uploader': '北枫CC',
+            'uploader_id': '2183525275',
+            'view_count': int,
         },
     }]
 
     def _real_extract(self, url: str):
         video_id = self._match_id(url)
-        video_data = self._download_json(
+        moment = self._download_json(
-            'https://liveapi.huya.com/moment/getMomentContent', video_id,
+            'https://liveapi.huya.com/moment/getMomentContent',
-            query={'videoId': video_id})['data']['moment']['videoInfo']
+            video_id, query={'videoId': video_id})['data']['moment']
 
         formats = []
-        for definition in traverse_obj(video_data, ('definitions', lambda _, v: url_or_none(v['url']))):
+        for definition in traverse_obj(moment, (
-            formats.append({
+            'videoInfo', 'definitions', lambda _, v: url_or_none(v['m3u8']),
-                'url': definition['url'],
+        )):
-                **traverse_obj(definition, {
+            fmts = self._extract_m3u8_formats(definition['m3u8'], video_id, 'mp4', fatal=False)
-                    'format_id': ('defName', {str}),
+            for fmt in fmts:
-                    'width': ('width', {int_or_none}),
+                fmt.update(**traverse_obj(definition, {
-                    'height': ('height', {int_or_none}),
                     'filesize': ('size', {int_or_none}),
-                }),
+                    'format_id': ('defName', {str}),
-            })
+                    'height': ('height', {int_or_none}),
+                    'quality': ('definition', {int_or_none}),
+                    'width': ('width', {int_or_none}),
+                }))
+            formats.extend(fmts)
 
         return {
             'id': video_id,
             'formats': formats,
-            **traverse_obj(video_data, {
+            **traverse_obj(moment, {
+                'comment_count': ('commentCount', {int_or_none}),
+                'description': ('content', {clean_html}, filter),
+                'like_count': ('favorCount', {int_or_none}),
+                'timestamp': ('cTime', {int_or_none}),
+            }),
+            **traverse_obj(moment, ('videoInfo', {
                 'title': ('videoTitle', {str}),
-                'thumbnail': ('videoCover', {url_or_none}),
+                'categories': ('category', {str}, filter, all, filter),
                 'duration': ('videoDuration', {parse_duration}),
+                'tags': ('tags', ..., {str}, filter, all, filter),
+                'thumbnail': (('videoBigCover', 'videoCover'), {url_or_none}, {update_url(query=None)}, any),
                 'uploader': ('nickName', {str}),
                 'uploader_id': ('uid', {str_or_none}),
-                'upload_date': ('videoUploadTime', {unified_strdate}),
                 'view_count': ('videoPlayNum', {int_or_none}),
-                'comment_count': ('videoCommentNum', {int_or_none}),
+            })),
-                'like_count': ('favorCount', {int_or_none}),
-            }),
         }