Pydantic Schema生成指南：自定义JSON Schema

当前位置:

首页 > Python基础教程 >

Pydantic Schema生成指南：自定义JSON Schema

第一章：Schema生成基础
1.1 默认Schema生成机制
PYTHON

from pydantic import BaseModel


class User(BaseModel):
    id: int
    name: str = Field(..., max_length=50)


print(User.schema_json(indent=2))

输出特征：

JSON

{
  "title": "User",
  "type": "object",
  "properties": {
    "id": {
      "title": "Id",
      "type": "integer"
    },
    "name": {
      "title": "Name",
      "type": "string",
      "maxLength": 50
    }
  },
  "required": [
    "id",
    "name"
  ]
}

1.2 Schema生成流程
graph TD
A[字段定义] --> B[元数据收集]
B --> C[类型映射]
C --> D[约束转换]
D --> E[Schema组装]
第二章：核心定制技术
2.1 字段级元数据注入
PYTHON

from pydantic import BaseModel, Field


class Product(BaseModel):
    sku: str = Field(
        ...,
        json_schema_extra={
            "x-frontend": {"widget": "search-input"},
            "x-docs": {"example": "ABC-123"}
        }
    )


print(Product.schema()["properties"]["sku"])

输出：

JSON

{
  "title": "Sku",
  "type": "string",
  "x-frontend": {
    "widget": "search-input"
  },
  "x-docs": {
    "example": "ABC-123"
  }
}

2.2 类型映射重载
PYTHON

from pydantic import BaseModel
from pydantic.json_schema import GenerateJsonSchema


class CustomSchemaGenerator(GenerateJsonSchema):
    def generate(self, schema):
        if schema["type"] == "string":
            schema["format"] = "custom-string"
        return schema


class DataModel(BaseModel):
    content: str


print(DataModel.schema(schema_generator=CustomSchemaGenerator))

第三章：动态Schema生成
3.1 运行时Schema构建
PYTHON

from pydantic import create_model
from pydantic.fields import FieldInfo


def dynamic_model(field_defs: dict):
    fields = {}
    for name, config in field_defs.items():
        fields[name] = (
            config["type"],
            FieldInfo(**config["field_params"])
        )
    return create_model('DynamicModel', **fields)


model = dynamic_model({
    "timestamp": {
        "type": int,
        "field_params": {"ge": 0, "json_schema_extra": {"unit": "ms"}}
    }
})

3.2 环境感知Schema
PYTHON

from pydantic import BaseModel, ConfigDict


class EnvAwareSchema(BaseModel):
    model_config = ConfigDict(json_schema_mode="dynamic")

    @classmethod
    def __get_pydantic_json_schema__(cls, core_schema, handler):
        schema = handler(core_schema)
        if os.getenv("ENV") == "prod":
            schema["required"].append("audit_info")
        return schema

第四章：企业级应用模式
4.1 OpenAPI增强方案
PYTHON

from pydantic import BaseModel


class OpenAPICompatible(BaseModel):
    model_config = dict(
        json_schema_extra={
            "components": {
                "schemas": {
                    "ErrorResponse": {
                        "type": "object",
                        "properties": {
                            "code": {"type": "integer"},
                            "message": {"type": "string"}
                        }
                    }
                }
            }
        }
    )

4.2 版本化Schema管理
PYTHON

from pydantic import BaseModel, field_validator


class VersionedModel(BaseModel):
    model_config = ConfigDict(extra="allow")

    @classmethod
    def __get_pydantic_json_schema__(cls, core_schema, handler):
        schema = handler(core_schema)
        schema["x-api-version"] = "2.3"
        return schema


class V1Model(VersionedModel):
    @classmethod
    def __get_pydantic_json_schema__(cls, *args):
        schema = super().__get_pydantic_json_schema__(*args)
        schema["x-api-version"] = "1.2"
        return schema

第五章：错误处理与优化
5.1 Schema验证错误
PYTHON

try:
    class InvalidSchemaModel(BaseModel):
        data: dict = Field(format="invalid-format")
except ValueError as e:
    print(f"Schema配置错误: {e}")

5.2 性能优化策略
PYTHON

from functools import lru_cache


class CachedSchemaModel(BaseModel):
    @classmethod
    @lru_cache(maxsize=128)
    def schema(cls, **kwargs):
        return super().schema(**kwargs)

课后Quiz
Q1：如何添加自定义Schema属性？
A) 使用json_schema_extra
B) 修改全局配置
C) 继承GenerateJsonSchema

Q2：处理版本兼容的正确方式？

动态注入版本号
创建子类覆盖Schema
维护多个模型
Q3：优化Schema生成性能应使用？

LRU缓存
增加校验步骤
禁用所有缓存
错误解决方案速查表
错误信息原因分析解决方案
ValueError: 无效的format类型不支持的Schema格式检查字段类型与格式的兼容性
KeyError: 缺失必需字段动态Schema未正确注入验证__get_pydantic_json_schema__实现
SchemaGenerationError 自定义生成器逻辑错误检查类型映射逻辑
MemoryError 大规模模型未缓存启用模型Schema缓存
架构箴言：Schema设计应遵循”契约优先”原则，建议使用Git版本控制管理Schema变更，通过CI/CD流水线实现Schema的自动化测试与文档生成，建立Schema变更通知机制保障多团队协作。

来源：https://blog.cmdragon.cn/posts/3bd5ffd5fdcb/

栏目列表