FastAPI从入门到实战（2）——Pydantic模型

绝地灬酷狼 2024-03-31 12:25 120阅读 0赞

> 前面了解了一下python的类型提示，这里就接着记录一下Pydantic这个用来执行数据校验的库。而且FastAPI就是基于python的类型提示和Padantic实现的数据验证。

### 简介 ###

*  官网：[https://pydantic-docs.helpmanual.io/][https_pydantic-docs.helpmanual.io]
 *  Pydantic就是一个基于Python类型提示来定义数据验证、序列化和文档(使用JSON模式)的库；
 *  使用Python的类型提示来进行数据校验和settings管理；
 *  可以在代码运行的时候提供类型提示，数据校验失败的时候提供友好的错误提示；
 *  定义数据应该如何在纯规范的Python代码中保存，并用Pydantic验证；

### 基本用法 ###

*  数据规范的情况
    
    > 这里的`**`符号是为了分配参数用的，可以分配字典

# -*- coding: utf-8 -*-
    # @Time: 2022/11/23 15:42
    # @Author: MinChess
    # @File: pydantic_tutorial.py
    # @Software: PyCharm
    
    from pydantic import BaseModel
    from datetime import datetime
    from typing import List,Optional
    
    class User(BaseModel):
        id: int #必填字段（无默认值的时候，其为必填字段）
        name: str = "MinChess" #有默认值，选填字段
        signup_ts: Optional[datetime] = None
        friends: List[int] = [] # 列表中的元素需要是int类型或者能转换为int类型的str
    
    external_data = {
        
        "id":"123",
        "signuo_ts": "2011-2-12 12:23:20",
        "friends": [1,2,'3']
    }
    
    user = User(**external_data)
    print(user.id,user.friends)
    print(repr(user.signup_ts))
    print(user.dict())

运行结果：

123 [1, 2, 3]
    None
    {
        'id': 123, 'name': 'MinChess', 'signup_ts': None, 'friends': [1, 2, 3]}

> 字符串类型的数据也转为了int型

*  数据不规范：id不为int类型或可转为int的str类型

external_data = {
        
        "id":"123h",
        "signuo_ts": "2011-2-12 12:23:20",
        "friends": [1,2,'3']
    }

输出：

Traceback (most recent call last):
      File "E:\project\PythonProject\FastApiProject\pydantic_tutorial.py", line 23, in <module>
        user = User(**external_data)
      File "pydantic\main.py", line 342, in pydantic.main.BaseModel.__init__
    pydantic.error_wrappers.ValidationError: 1 validation error for User
    id
      value is not a valid integer (type=type_error.integer)

> 报错说id是int型，这里就可以看出，提示和规范还是非常友好的了

### 校验失败处理 ###

*  给属性赋错误类型

try:
        User(id=1,signup_ts=datetime.now(),friends=[1,5,'dsd']) # 直接给属性赋值
    except ValidationError as e:
        print(e.json()) # 错误json格式化

*  输出结果：

[
      {
        
        "loc": [
          "friends",
          2
        ],
        "msg": "value is not a valid integer",
        "type": "type_error.integer"
      }
    ]

### 模型类的属性和方法 ###

*  解析和转换

print(user.dict()) # 转换为字典
    print(user.json()) # 转换为json
    print(user.copy()) # 浅copy
    
    print(User.parse_obj(obj=external_data)) # 直接解析字典数据
    print(User.parse_raw('{"id": 123, "name": "MinChess", "signup_ts": null, "friends": [1, 2, 3]}')) # 解析标准格式的数据，里面是双引号
    
    path = Path('pydantic_tutorial.json')
    path.write_text('{"id": 123, "name": "MinChess", "signup_ts": null, "friends": [1, 2, 3]}')
    print(User.parse_file(path)) # 解析文件
    
    print(user.schema()) #此方法用来告诉我们实例的数据格式的方案
    print(user.schema_json())
    
    user_data = {
        "id": 123, "name": "MinChess", "signup_ts": "2022-12-20 12:12:30", "friends": [1, 2, 3]}
    print(User.construct(**user_data)) # 此方法不校验数据，直接创建模型类
    
    print(User.__fields__.keys()) # （这里查看所有字段）定义模型类的时候，所有字段都注明类型，字段顺序就不会乱

*  输出：

{
        'id': 123, 'name': 'MinChess', 'signup_ts': None, 'friends': [1, 2, 3]}
    {
        "id": 123, "name": "MinChess", "signup_ts": null, "friends": [1, 2, 3]}
    id=123 name='MinChess' signup_ts=None friends=[1, 2, 3]
    id=123 name='MinChess' signup_ts=None friends=[1, 2, 3]
    id=123 name='MinChess' signup_ts=None friends=[1, 2, 3]
    id=123 name='MinChess' signup_ts=None friends=[1, 2, 3]
    {
        'title': 'User', 'type': 'object', 'properties': {
        'id': {
        'title': 'Id', 'type': 'integer'}, 'name': {
        'title': 'Name', 'default': 'MinChess', 'type': 'string'}, 'signup_ts': {
        'title': 'Signup Ts', 'type': 'string', 'format': 'date-time'}, 'friends': {
        'title': 'Friends', 'default': [], 'type': 'array', 'items': {
        'type': 'integer'}}}, 'required': ['id']}
    {
        "title": "User", "type": "object", "properties": {
        "id": {
        "title": "Id", "type": "integer"}, "name": {
        "title": "Name", "default": "MinChess", "type": "string"}, "signup_ts": {
        "title": "Signup Ts", "type": "string", "format": "date-time"}, "friends": {
        "title": "Friends", "default": [], "type": "array", "items": {
        "type": "integer"}}}, "required": ["id"]}
    id=123 name='MinChess' signup_ts='2022-12-20 12:12:30' friends=[1, 2, 3]
    dict_keys(['id', 'name', 'signup_ts', 'friends'])

### 递归模型 ###

*  创建递归

class Sound(BaseModel):
        sound:str
    
    class Dog(BaseModel):
        birthday: date
        weight: float = Optional[None]
        sound: List[Sound]
    
    dogs = Dog(birthday=date.today(),weight=6.25,sound=[{
        "sound":"wang wang"},{
        "sound":"ying ying"}])
    print(dogs.json())

*  输出

{
        "birthday": "2022-11-23", "sound": [{
        "sound": "wang wang"}, {
        "sound": "ying ying"}]}

### ORM模型 ###

> ORM 全称 Object Relational Mapping, 叫对象关系映射。简单的说，ORM 将数据库中的表与面向对象语言中的类建立了一种对应关系。

*  利用`sqlalchemy`创建模型

Base = declarative_base()
    
    class CompanyOrm(Base):
        __tablename__ = 'companies'
        id = Column(Integer,primary_key = True,nullable = False)
        public_key = Column(String(20),index = True,nullable = False ,unique = True)
        name = Column(String(63),unique=True)
        domains = Column(ARRAY(String(255)))
    
    class CompanyModel(BaseModel):
        id:int
        public_key:constr(max_length=20)
        name:constr(max_length=63)
        domains:List[constr(max_length=255)]
    
        class Config:
            orm_mode = True
    
    co_orm = CompanyOrm(
        id = 12,
        public_key= 'dddddd',
        name= 'MinChess',
        domains=['eeeeeee','fffffff']
    )
    
    print(CompanyModel.from_orm(co_orm))

*  输出

id=12 public_key='dddddd' name='MinChess' domains=['eeeeeee', 'fffffff']

### Pydantic支持的所有字段类型 ###

> Pydantic支持很多类型的数据，除了常用的那些基本类型外，还有一些不常用的类型，具体参看官网：
> 
> [https://pydantic-docs.helpmanual.io/usage/types/][https_pydantic-docs.helpmanual.io_usage_types]

![image-20221123170940674][]

### 源码 ###

# -*- coding: utf-8 -*-
    # @Time: 2022/11/23 15:42
    # @Author: MinChess
    # @File: pydantic_tutorial.py
    # @Software: PyCharm
    
    from pydantic import BaseModel,ValidationError,constr
    from datetime import datetime,date
    from pathlib import Path
    from typing import List,Optional
    from sqlalchemy import Column,Integer,String
    from sqlalchemy.dialects.postgresql import ARRAY
    from sqlalchemy.ext.declarative import declarative_base
    
    class User(BaseModel):
        id: int #必填字段（无默认值的时候，其为必填字段）
        name: str = "MinChess" #有默认值，选填字段
        signup_ts: Optional[datetime] = None
        friends: List[int] = [] # 列表中的元素需要是int类型或者能转换为int类型的str
    
    external_data = {
        
        "id":"123",
        "signuo_ts": "2011-2-12 12:23:20",
        "friends": [1,2,'3']
    }
    
    user = User(**external_data)
    print("====="*6,'基本使用','====='*6)
    print(user.id,user.friends)
    print(repr(user.signup_ts))
    print(user.dict())
    
    print("====="*6,'校验失败处理','====='*6)
    try:
        User(id=1,signup_ts=datetime.now(),friends=[1,5,'dsd']) # 直接给属性赋值
    except ValidationError as e:
        print(e.json()) # 错误json格式化
    
    print("====="*6,'模型类的属性和方法','====='*6)
    print(user.dict()) # 转换为字典
    print(user.json()) # 转换为json
    print(user.copy()) # 浅copy
    
    print(User.parse_obj(obj=external_data)) # 直接解析字典数据
    print(User.parse_raw('{"id": 123, "name": "MinChess", "signup_ts": null, "friends": [1, 2, 3]}')) # 解析标准格式的数据，里面是双引号
    
    path = Path('pydantic_tutorial.json')
    path.write_text('{"id": 123, "name": "MinChess", "signup_ts": null, "friends": [1, 2, 3]}')
    print(User.parse_file(path)) # 解析文件
    
    print(user.schema()) #此方法用来告诉我们实例的数据格式的方案
    print(user.schema_json())
    
    user_data = {
        "id": 123, "name": "MinChess", "signup_ts": "2022-12-20 12:12:30", "friends": [1, 2, 3]}
    print(User.construct(**user_data)) # 此方法不校验数据，直接创建模型类
    
    print(User.__fields__.keys()) # （这里查看所有字段）定义模型类的时候，所有字段都注明类型，字段顺序就不会乱
    
    print("====="*6,'递归模型','====='*6)
    
    class Sound(BaseModel):
        sound:str
    
    class Dog(BaseModel):
        birthday: date
        weight: float = Optional[None]
        sound: List[Sound]
    
    dogs = Dog(birthday=date.today(),weight=6.25,sound=[{
        "sound":"wang wang"},{
        "sound":"ying ying"}])
    print(dogs.json())
    
    print("====="*6,'ORM模型','====='*6)
    
    Base = declarative_base()
    
    class CompanyOrm(Base):
        __tablename__ = 'companies'
        id = Column(Integer,primary_key = True,nullable = False)
        public_key = Column(String(20),index = True,nullable = False ,unique = True)
        name = Column(String(63),unique=True)
        domains = Column(ARRAY(String(255)))
    
    class CompanyModel(BaseModel):
        id:int
        public_key:constr(max_length=20)
        name:constr(max_length=63)
        domains:List[constr(max_length=255)]
    
        class Config:
            orm_mode = True
    
    co_orm = CompanyOrm(
        id = 12,
        public_key= 'dddddd',
        name= 'MinChess',
        domains=['eeeeeee','fffffff']
    )
    
    print(CompanyModel.from_orm(co_orm))

--------------------

感谢阅读  
博客链接：[https://blog.jiumoz.com/archives/fastapi-cong-ru-men-dao-shi-zhan-2pydantic-mo-xing][https_blog.jiumoz.com_archives_fastapi-cong-ru-men-dao-shi-zhan-2pydantic-mo-xing]

![30b7b3ba49344b6691969078c7df196e.png][]欢 迎 关 注 博 主 个 人 小 程 序！

[https_pydantic-docs.helpmanual.io]: https://pydantic-docs.helpmanual.io/
[https_pydantic-docs.helpmanual.io_usage_types]: https://pydantic-docs.helpmanual.io/usage/types/
[image-20221123170940674]: https://image.dandelioncloud.cn/pgy_files/images/2024/03/31/0163409c3dea40f6873d0b0c9c1d174d.png
[https_blog.jiumoz.com_archives_fastapi-cong-ru-men-dao-shi-zhan-2pydantic-mo-xing]: https://blog.jiumoz.com/archives/fastapi-cong-ru-men-dao-shi-zhan-2pydantic-mo-xing
[30b7b3ba49344b6691969078c7df196e.png]: https://image.dandelioncloud.cn/pgy_files/images/2024/03/31/54db4fd581404572b98f59c27d64639e.png