Pydantic Schema生成指南：自定义JSON Schema

扫描二维码关注或者微信搜一搜：编程智域前端至全栈交流与成长

第一章：Schema生成基础

1.1 默认Schema生成机制

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：

{
  "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

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：

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

2.2 类型映射重载

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

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

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

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

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

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

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。

余下文章内容请点击跳转至个人博客页面或者扫码关注或者微信搜一搜：编程智域前端至全栈交流与成长，阅读完整的文章：