CameraMachine/python/service/customer_template_service.py

from email.policy import default
from settings import *
from middleware import UnicornException
import copy
import requests
from PIL import Image
from io import BytesIO
import base64,shutil
from logger import logger
'''前端生图接口'''
generate_templace = "/generate"
class CustomerTemplateService:
    def __init__(self):
        pass
    def generateTemplate(self,config_data,template_json,template_name,save_path):
        '''
        参数：
        config_data: 配置数据
        template_json: 模板数据
        template_name: 模板名称
        save_path: 保存路径
        '''
        print("开始生成模板")
        # print("config_data",config_data)
        handler_config_data,model_image,scene_image = self.__handler_config_data(config_data)
        goods_no = list(handler_config_data.keys())[0]
        headers = {"Content-Type": "application/json"}
        json_data = {"goodsList":[handler_config_data],"canvasList":template_json}
        json_data = json.dumps(json_data,ensure_ascii=False)
        # print("json_data",json_data)
        template_result = requests.post(CUSTOMER_TEMPLATE_URL+generate_templace,data=json_data,headers=headers)
        resultJson = template_result.json()
        code = resultJson.get("code")
        msg = resultJson.get("msg")
        images = resultJson.get("images",[])
        if code != 0:
            raise UnicornException(f"详情页生成失败,请检查模板数据是否正确:{msg}")
        concat_images_array = []
        for image in images:
            canvasIndex = image.get("canvasIndex")
            dataUrl = image.get("dataUrl")
            save_name = f"{save_path}/切片图-{template_name}/{goods_no}（{int(canvasIndex)+1}）.png"
            match dataUrl:
                case "model":
                    # 复制模特图进行拼接
                    if model_image:
                        model_copy_res = self.copyImage(model_image,save_name)
                        if model_copy_res:
                            model_image_pil = Image.open(model_image)
                            concat_images_array.append(model_image_pil)
                case "scene":
                    # 复制场景图进行拼接
                    if scene_image:
                        scene_copy_res = self.copyImage(scene_image,save_name)
                        if scene_copy_res:
                            scene_image_pil = Image.open(scene_image)
                            concat_images_array.append(scene_image_pil)
                case _:
                    pillowImage = self.save_base64_image(dataUrl,save_name)
                    concat_images_array.append(pillowImage)
        print("模板生成成功")
    
    def concat_images_vertically(image_array, custom_width=None):
        """
        按照顺序将图片数组拼接成长图，并统一图片宽度

        参数：
        - image_array: list，存放 Pillow 图片对象的数组
        - custom_width: int，可选参数，指定统一的图片宽度（默认为第一张图的宽度）

        返回：
        - concatenated_image: PIL.Image，拼接后的长图对象
        """
        if not image_array:
            raise ValueError("图片数组为空，无法拼接")

        # 1. 确定统一宽度
        base_width = custom_width or image_array[0].width

        # 2. 计算总高度和调整图片尺寸
        total_height = 0
        resized_images = []
        for img in image_array:
            # 调整图片宽度并保持宽高比
            width_ratio = base_width / img.width
            new_height = int(img.height * width_ratio)
            resized_img = img.resize((base_width, new_height), Image.ANTIALIAS)
            resized_images.append(resized_img)
            total_height += new_height

        # 3. 创建空白画布
        concatenated_image = Image.new("RGB", (base_width, total_height))

        # 4. 按顺序拼接图片
        y_offset = 0
        for resized_img in resized_images:
            concatenated_image.paste(resized_img, (0, y_offset))
            y_offset += resized_img.height

        return concatenated_image
    def __handler_config_data(self,config_data):
        '''
        处理配置数据，返回一个新的数据对象
        '''
        # 深拷贝原始数据，确保不改变原数据对象
        new_config_data = copy.deepcopy(config_data)
        model_image = None
        scene_image = None
        # 如果输入是字典，则将其转换为目标结构
        if isinstance(new_config_data, dict):
            # result = []
            for key, item in new_config_data.items():
                # 提取需要添加的数据
                additional_data = {k: v for k, v in item.items() if k not in ["款号", "货号资料"]}
                
                # 遍历货号资料，将额外数据添加到每个货号对象中
                for product in item.get("货号资料", []):
                    product.update(additional_data)
                    # 处理 pics 字段中的 xx-抠图 转换为 Base64 并新增字段
                    pics = product.get("pics", {})
                    if not model_image:
                        model_image = product.get("模特图", None)
                    if not scene_image:
                        scene_image = product.get("场景图", None)
                    new_pics = {}
                    for pic_key, pic_path in pics.items():
                        if "-抠图" in pic_key:
                            # 读取图片并转换为 Base64
                            try:
                                # base64_data = self.crop_image_and_convert_to_base64(pic_path)
                                # 新增字段（去除 -抠图）
                                new_key = pic_key.replace("-抠图", "")
                                new_pics[new_key] = pic_path
                            except Exception as e:
                                print(f"读取图片失败: {pic_path}, 错误: {e}")
                        else:
                            # 非 -抠图 字段保持不变
                            new_pics[pic_key] = pic_path
                    
                    # 更新 pics 字段
                    product["pics"] = new_pics
                
                # 构建目标结构
            #     result.append({key: item})
            # return result
        
        return new_config_data,model_image,scene_image
    def save_base64_image(self,base64_data, output_path):
        """
        将 Base64 编码的图像保存到本地文件

        参数：
        - base64_data: str，Base64 编码的图像数据（不包含前缀如 "data:image/png;base64,"）
        - output_path: str，保存图像的本地路径
        """
        if "data:image/jpeg;base64," in base64_data:
            base64_data = base64_data.split(",")[1]
        try:
            # 1. 解码 Base64 数据
            image_data = base64.b64decode(base64_data)
            
            # 2. 加载图像数据
            image = Image.open(BytesIO(image_data))
            # 3. 检查路径是否存在，如果不存在则创建
            directory = os.path.dirname(output_path)
            if directory and not os.path.exists(directory):
                os.makedirs(directory)
                print(f"目录已创建: {directory}")
            # 4. 保存图像到本地
            image.save(output_path)
            print(f"图像已成功保存到 {output_path}")
            return image
        except Exception as e:
            print(f"保存图像失败: {e}")
            print(f"Base64 数据前 100 字符: {base64_data[:100]}")
            return None
            
    def crop_image_and_convert_to_base64(self,pic_path):
        """
        使用 Pillow 裁剪图片并生成带有前缀的 Base64 图片

        参数：
        - pic_path: str，图片文件路径

        返回：
        - base64_data_with_prefix: str，带有前缀的 Base64 编码图片数据
        """
        try:
            # 1. 加载图片
            with Image.open(pic_path) as image:
                # 2. 获取图片的非透明区域边界框 (bounding box)
                bbox = image.getbbox()
                if not bbox:
                    raise ValueError("图片可能是完全透明的，无法获取边界框")
                
                # 3. 裁剪图片
                cropped_image = image.crop(bbox)
                
                # 4. 将裁剪后的图片转换为 Base64
                buffered = BytesIO()
                cropped_image.save(buffered, format="PNG")  # 保存为 PNG 格式
                base64_data = base64.b64encode(buffered.getvalue()).decode("utf-8")
                
                # 5. 添加 Base64 前缀
                base64_data_with_prefix = f"data:image/png;base64,{base64_data}"
                return base64_data_with_prefix

        except Exception as e:
            print(f"处理图片失败: {pic_path}, 错误: {e}")
            return None
    def copyImage(self,src_path,limit_path):
        try:
          shutil.copy(src_path, limit_path)
          return True
        except Exception as e:
            logger.info(f"copyImage 复制模特图/场景图出错:{str(e)}",src_path,limit_path)
            return False