更新部分爬虫以兼容本地运行及数据库存储

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

@@ -95,6 +95,32 @@ class TieBaDbStoreImplement(AbstractStore):
             content_item: content item dict
         """
         note_id = content_item.get("note_id")
+        if not note_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
+            
+            title = content_item.get("title", "")
+            text = content_item.get("text", "")
+            content_text = title + " " + text
+            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(content_text, strict_keywords, fuzzy_keywords):
+                    utils.logger.warning(f"[TiebaDbStoreImplement.store_content] ❌ Filtered note {note_id} - content does not match any keyword")
+                    return
+        except Exception as e:
+            utils.logger.debug(f"[TiebaDbStoreImplement.store_content] Failed to load keyword config: {e}")
+        
         async with get_session() as session:
             stmt = select(TiebaNote).where(TiebaNote.note_id == note_id)
             res = await session.execute(stmt)