batch_resize
Resize multiple images to specified dimensions while optionally maintaining aspect ratio and converting formats in batch processing workflows.
Instructions
批量调整图片大小
Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| image_sources | Yes | 图片源列表，每个元素可以是文件路径或base64编码的图片数据 | |
| target_width | Yes | 目标宽度（像素） | |
| target_height | Yes | 目标高度（像素） | |
| keep_aspect_ratio | No | 是否保持宽高比 | |
| output_format | No | 输出格式：PNG、JPEG、WEBP 等 | PNG |
Output Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| result | Yes |
Implementation Reference
- tools/advanced.py:326-439 (handler)Core implementation of the batch_resize tool handler. Processes a list of image sources in parallel, resizes each image using PIL (with optional aspect ratio preservation), encodes outputs to base64, and returns JSON results including success/failure stats.
async def batch_resize(arguments: Dict[str, Any]) -> List[TextContent]: """ 批量调整多张图片的大小 Args: arguments: 包含图片源列表和调整参数的字典 Returns: List[TextContent]: 处理结果 """ try: # 参数验证 image_sources = arguments.get("image_sources", []) if not image_sources: raise ValidationError("image_sources不能为空") width = arguments.get("width") height = arguments.get("height") maintain_aspect_ratio = arguments.get("maintain_aspect_ratio", True) resample_method = arguments.get("resample_method", "LANCZOS") output_format = arguments.get("output_format", DEFAULT_IMAGE_FORMAT) # 验证参数 validate_numeric_range(width, MIN_IMAGE_SIZE, MAX_IMAGE_SIZE, "width") validate_numeric_range(height, MIN_IMAGE_SIZE, MAX_IMAGE_SIZE, "height") # 获取重采样方法 resample_map = { "LANCZOS": Image.Resampling.LANCZOS, "BILINEAR": Image.Resampling.BILINEAR, "BICUBIC": Image.Resampling.BICUBIC, "NEAREST": Image.Resampling.NEAREST } resample = resample_map.get(resample_method, Image.Resampling.LANCZOS) processor = ImageProcessor() results = [] failed_count = 0 # 使用线程池进行并行处理 def resize_single_image(image_source): try: ensure_valid_image_source(image_source) image = processor.load_image(image_source) if maintain_aspect_ratio: # 保持宽高比 image.thumbnail((width, height), resample) resized_image = image else: # 强制调整到指定尺寸 resized_image = image.resize((width, height), resample) # 输出图片 output_info = processor.output_image(resized_image, "batch_resize", output_format) return output_info except Exception as e: return f"ERROR: {str(e)}" # 并行处理 with ThreadPoolExecutor(max_workers=min(len(image_sources), MAX_CONCURRENT_TASKS)) as executor: futures = [executor.submit(resize_single_image, source) for source in image_sources] for i, future in enumerate(futures): result = future.result() if result.startswith("ERROR:"): failed_count += 1 results.append({ "index": i, "success": False, "error": result[7:] # 移除"ERROR: "前缀 }) else: results.append({ "index": i, "success": True, "result": result }) return [TextContent( type="text", text=json.dumps({ "success": True, "message": f"批量调整完成，成功: {len(results) - failed_count}, 失败: {failed_count}", "results": results, "metadata": { "total_images": len(image_sources), "successful": len(results) - failed_count, "failed": failed_count, "target_size": f"{width}x{height}", "maintain_aspect_ratio": maintain_aspect_ratio, "resample_method": resample_method, "format": output_format } }, ensure_ascii=False) )] except ValidationError as e: return [TextContent( type="text", text=json.dumps({ "success": False, "error": f"参数验证失败: {str(e)}" }, ensure_ascii=False) )] except Exception as e: return [TextContent( type="text", text=json.dumps({ "success": False, "error": f"批量调整失败: {str(e)}" }, ensure_ascii=False) )] - main.py:649-673 (registration)MCP tool registration for batch_resize using FastMCP @mcp.tool() decorator. Defines input schema via Annotated parameters, constructs arguments dict, and delegates execution to the advanced async handler.
@mcp.tool() def batch_resize( image_sources: Annotated[list, Field(description="图片源列表，每个元素可以是文件路径或base64编码的图片数据")], target_width: Annotated[int, Field(description="目标宽度（像素）", ge=1)], target_height: Annotated[int, Field(description="目标高度（像素）", ge=1)], keep_aspect_ratio: Annotated[bool, Field(description="是否保持宽高比", default=True)], output_format: Annotated[str, Field(description="输出格式：PNG、JPEG、WEBP 等", default="PNG")] ) -> str: """批量调整图片大小""" try: arguments = { "image_sources": image_sources, "target_width": target_width, "target_height": target_height, "keep_aspect_ratio": keep_aspect_ratio, "output_format": output_format } result = safe_run_async(advanced_batch_resize(arguments)) return result[0].text except Exception as e: return json.dumps({ "success": False, "error": f"批量调整大小失败: {str(e)}" }, ensure_ascii=False, indent=2) - tools/advanced.py:35-80 (schema)Input schema definition for batch_resize tool within get_advanced_tools(), specifying JSON schema for parameters including image_sources array, dimensions, aspect ratio, resample method, and output format.
Tool( name="batch_resize", description="批量调整多张图片的大小", inputSchema={ "type": "object", "properties": { "image_sources": { "type": "array", "description": "图片源列表（文件路径或base64编码）", "items": {"type": "string"}, "minItems": 1, "maxItems": 10 }, "width": { "type": "integer", "description": "目标宽度", "minimum": MIN_IMAGE_SIZE, "maximum": MAX_IMAGE_SIZE }, "height": { "type": "integer", "description": "目标高度", "minimum": MIN_IMAGE_SIZE, "maximum": MAX_IMAGE_SIZE }, "maintain_aspect_ratio": { "type": "boolean", "description": "是否保持宽高比", "default": True }, "resample_method": { "type": "string", "description": "重采样方法", "enum": ["LANCZOS", "BILINEAR", "BICUBIC", "NEAREST"], "default": "LANCZOS" }, "output_format": { "type": "string", "description": "输出格式", "enum": ["PNG", "JPEG", "WEBP"], "default": "PNG" } }, "required": ["image_sources", "width", "height"] } ),