Pydantic v2 教程：优雅地解析逗号分隔的浮点数字符串

from pydantic import BaseModel, model_validator, ValidationError
from typing import Dict, Any

class User(BaseModel):
    name: str
    balance: float
    weight: float
    # ... 其他可能需要修正的浮点数字段

    @model_validator(mode='before')
    @classmethod
    def fix_float_comma_strings(cls, data: Dict[str, Any]) -> Dict[str, Any]:
        """
        在 Pydantic 模型解析之前，检查并修正字典中所有声明为 float 类型
        且当前值为字符串类型，并包含逗号作为小数分隔符的字段。
        将逗号替换为句点，以确保 Pydantic 能够正确解析为浮点数。
        """
        # 创建数据的副本以进行修改，避免直接修改原始输入
        processed_data = data.copy() 
        for k, v in processed_data.items():
            # 1. 检查字段 k 是否在模型的类型注解中声明
            # 2. 检查声明的类型是否为 float (或其子类)
            # 3. 检查当前值 v 是否为字符串类型
            if k in cls.__annotations__ and \
               issubclass(float, cls.__annotations__[k]) and \
               isinstance(v, str):
                # 如果字符串包含逗号，则替换为句点
                if ',' in v:
                    processed_data[k] = v.replace(',', '.')
        return processed_data

# --- 示例使用 ---
if __name__ == "__main__":
    # 包含逗号的浮点数字符串数据
    user_data_with_comma = {"name": "Bob", "balance": "13,7", "weight": "75,25"}
    # 正常浮点数字符串数据
    user_data_normal_str = {"name": "Alice", "balance": "123.45", "weight": "60.5"}
    # 混合数据
    user_data_mixed = {"name": "Charlie", "balance": 99.9, "weight": "88,88"}
    # 包含非浮点数字符串的错误数据
    user_data_invalid_str = {"name": "David", "balance": "abc", "weight": "70.0"}

    print("--- 解析含逗号的浮点数字符串 ---")
    try:
        user1 = User(**user_data_with_comma)
        print(f"User 1: {user1.model_dump()}")
        print(f"Type of balance: {type(user1.balance)}, Type of weight: {type(user1.weight)}")
    except ValidationError as e:
        print(f"Error parsing user_data_with_comma: {e}")

    print("\n--- 解析正常浮点数字符串 ---")
    try:
        user2 = User(**user_data_normal_str)
        print(f"User 2: {user2.model_dump()}")
        print(f"Type of balance: {type(user2.balance)}, Type of weight: {type(user2.weight)}")
    except ValidationError as e:
        print(f"Error parsing user_data_normal_str: {e}")

    print("\n--- 解析混合数据（部分已为浮点数，部分为逗号字符串） ---")
    try:
        user3 = User(**user_data_mixed)
        print(f"User 3: {user3.model_dump()}")
        print(f"Type of balance: {type(user3.balance)}, Type of weight: {type(user3.weight)}")
    except ValidationError as e:
        print(f"Error parsing user_data_mixed: {e}")

    print("\n--- 尝试解析无效数据（非浮点数字符串） ---")
    try:
        user4 = User(**user_data_invalid_str)
        print(f"User 4: {user4.model_dump()}")
    except ValidationError as e:
        print(f"Error parsing user_data_invalid_str: {e}")
        # 预期会在这里捕获错误，因为 'abc' 无法转换为 float，且 validator 仅替换逗号