更新部分爬虫以兼容本地运行及数据库存储

MindSpider/DeepSentimentCrawling/MediaCrawler/store/douyin/_store_impl.py

@@ -88,6 +88,30 @@ class DouyinDbStoreImplement(AbstractStore):
             content_item: content item dict
         """
         aweme_id = content_item.get("aweme_id")
+        if not aweme_id:
+            return
+        
+        # 关键词过滤：仅在落库时进行，仅对主贴/视频过滤，不过滤评论
+        # 支持精确匹配和模糊匹配两种模式
+        try:
+            import sys
+            from pathlib import Path
+            project_root = Path(__file__).resolve().parents[4]
+            if str(project_root) not in sys.path:
+                sys.path.insert(0, str(project_root))
+            from config import settings
+            
+            desc = content_item.get("desc", "")
+            strict_keywords = getattr(settings, 'STRICT_KEYWORDS', None)
+            fuzzy_keywords = getattr(settings, 'FUZZY_KEYWORDS', None)
+            
+            if strict_keywords or fuzzy_keywords:
+                if not utils.check_keyword_match_with_modes(desc, strict_keywords, fuzzy_keywords):
+                    utils.logger.warning(f"[DouyinDbStoreImplement.store_content] ❌ Filtered aweme {aweme_id} - content does not match any keyword")
+                    return
+        except Exception as e:
+            utils.logger.debug(f"[DouyinDbStoreImplement.store_content] Failed to load keyword config: {e}")
+        
         async with get_session() as session:
             result = await session.execute(select(DouyinAweme).where(DouyinAweme.aweme_id == aweme_id))
             aweme_detail = result.scalar_one_or_none()