Method _parse_file

agentclaw/node/document.py:207–240

解析单个文件

(self, file_path: str)

            return f"[解析失败: {e}]"
    
    def _parse_file(self, file_path: str) -> str:
        """解析单个文件"""
        path = Path(file_path)
        
        if not path.exists():
            logger.warning(f"DocumentNode: 文件不存在: {file_path}")
            return ""
        
        try:
            md = _get_markitdown()
            result = md.convert(str(path))
            
            content = result.text_content if hasattr(result, 'text_content') else str(result)
            
            # 添加元数据
            if self.include_metadata:
                metadata = [
                    f"**文件名**: {path.name}",
                    f"**大小**: {path.stat().st_size} bytes",
                ]
                if hasattr(result, 'title') and result.title:
                    metadata.append(f"**标题**: {result.title}")
                content = "\n".join(metadata) + "\n\n" + content
            
            # 限制长度
            if self.max_length > 0 and len(content) > self.max_length:
                content = content[:self.max_length] + "\n\n... [内容已截断]"
            
            logger.info(f"DocumentNode: 解析完成 {path.name} ({len(content)} 字符)")
            return content
            
        except Exception as e:
            logger.error(f"DocumentNode: 解析失败 {file_path}: {e}")
            return f"[解析失败: {e}]"


class DocumentExtractNode(DocumentNode):