如何在Python中通过Prisma将字符串插入PostgreSQL枚举字段

问题 #

尝试在Python中通过Prisma将字符串类型数据插入PostgreSQL数据库，数据库字段在Prisma Schema中定义为枚举类型，通过字符串映射插入时失败，提示不支持字符串类型。

代码示例 #

from enum import Enum
from prisma import Prisma
import os
import logging

db = Prisma()
os.environ['DATABASE_URL'] = ''
db.connect()

class DifficultyLevel(Enum):
    EASY = 'EASY'
    MEDIUM = 'MEDIUM'
    HARD = 'HARD'
    LUCK = 'LUCK'

def map_question_difficulty(generated_difficulty):
    difficulty_mapping = {
        'Easy': DifficultyLevel.EASY,
        'Medium': DifficultyLevel.MEDIUM,
        'Hard': DifficultyLevel.HARD,
        'LUCK': DifficultyLevel.LUCK
    }
    return difficulty_mapping.get(generated_difficulty, DifficultyLevel.EASY).value  # Default to EASY if difficulty not found

async def insert_questions_to_db(questions):
    try:
        for question in questions:
            db.elementgenerated.create(
                data={
                    "difficulty": map_question_difficulty(question['difficulty'])
                }
            )
        logging.info(f"Successfully inserted {len(questions)} questions into the database.")
    except Exception as e:
        logging.error(f"Error inserting questions into the database: {e}")
        raise

Prisma Schema枚举定义 #

enum DifficultyLevel {
  EASY
  MEDIUM
  HARD
}

错误信息 #

prisma.errors.DataError: Error converting field "difficulty" of expected non-nullable type "String", found incompatible value of "EASY".

其中question['difficulty']是'Easy'、'Medium'或'Hard'这类字符串。

解决方法 #

1. 核心修复：使用Prisma自动生成的枚举类 #

Prisma会根据Schema中的枚举定义自动生成对应的Python枚举类，无需手动定义。直接导入使用即可保证类型完全匹配。

修改后的完整代码：

from prisma import Prisma
import os
import logging
from prisma.enums import DifficultyLevel  # 导入Prisma自动生成的枚举

db = Prisma()
os.environ['DATABASE_URL'] = ''
db.connect()

def map_question_difficulty(generated_difficulty):
    difficulty_mapping = {
        'Easy': DifficultyLevel.EASY,
        'Medium': DifficultyLevel.MEDIUM,
        'Hard': DifficultyLevel.HARD,
        # 移除Schema中不存在的LUCK条目，避免插入报错
    }
    return difficulty_mapping.get(generated_difficulty, DifficultyLevel.EASY)  # 直接返回枚举实例，不要取.value

async def insert_questions_to_db(questions):
    try:
        for question in questions:
            await db.elementgenerated.create(  # 异步方法必须加await
                data={
                    "difficulty": map_question_difficulty(question['difficulty'])
                }
            )
        logging.info(f"Successfully inserted {len(questions)} questions into the database.")
    except Exception as e:
        logging.error(f"Error inserting questions into the database: {e}")
        raise

2. 关键修复点说明 #

• 删除自定义枚举：手动定义的枚举与Prisma生成的类型不兼容，直接用prisma.enums下的类即可。
• 不要取.value：Prisma期望接收枚举实例而非字符串值，直接传入枚举成员（如DifficultyLevel.EASY）即可。
• 添加await：create()是异步方法，必须用await触发执行，否则不会实际操作数据库。
• 同步枚举值：Schema中没有LUCK枚举，需从映射字典中移除该条目，避免插入时出现枚举值不存在的错误。

3. 可选方案：直接映射字符串（无需枚举类） #

如果不想导入枚举类，也可以直接传入与Schema枚举值完全匹配的字符串（注意大小写一致）：

def map_question_difficulty(generated_difficulty):
    difficulty_mapping = {
        'Easy': 'EASY',
        'Medium': 'MEDIUM',
        'Hard': 'HARD',
    }
    return difficulty_mapping.get(generated_difficulty, 'EASY')

这种方式需严格保证字符串与Schema枚举值的大小写、拼写完全一致。

内容的提问来源于stack exchange，提问作者Jerry Chen