使用Mongoose通过_id(ObjectId)查询返回空数组的问题排查

我编写了如下findCustomer函数：

export const findCustomer = async (query?: FindCustomerParams) => {
  try {
    await database.connect();
  } catch (err: any) {
    throw new Error("Could not connect to the database. Root error: " + err);
  }

  if (!query) {
    return await customer.find();
  }

  findCustomerParams.parse(query);

  let returnCustomers;
  try {
    const fetchedCustomers: (Omit<Customer, "address"> & { address?: FindOneCustomerParams["address"] })[] | null =
      await customer.find(query ? removeFalsyValues({ ...query, _id: new mongoose.Types.ObjectId(query.id) }) : {});

    if (!fetchedCustomers) {
      throw new Error();
    }

    if (query.address) {
      const fetchedAddresses = await addressMethods.find(query.address);

      for (let i = 0; i < fetchedAddresses.length; i++) {
        if (!fetchedAddresses || !fetchedAddresses[i].customers.includes(fetchedCustomers[i]._id)) {
          throw new Error(`Could find address on customer ${fetchedCustomers[i]._id}`);
        }

        fetchedCustomers[i].address = fetchedAddresses[i];
      }
    }

    returnCustomers = fetchedCustomers;
  } catch (err: any) {
    throw new Error("Failed to find a customer. Root error: " + err);
  }

  await database.close();

  return returnCustomers;
};

该函数除通过_id查询外，其他查询均正常工作。当通过_id查询时，会返回空数组，而实际应返回单个客户数据。

可正常返回结果的查询示例：

await findCustomer({firstName: "Ethan"})

无法返回结果的查询示例：

await findCustomer({_id: id})

我已尝试以下操作：

• 验证ID正确性
• 确认已连接Atlas集群
• 尝试将_id从string类型转为ObjectId类型
• 分别使用id和_id查询均无效
• 控制台无报错信息

解决方法 #

问题根源 #

核心bug出在查询构造逻辑：当传入{_id: id}时，代码里写的是new mongoose.Types.ObjectId(query.id)，但你传入的参数里只有_id字段，query.id是undefined，转成ObjectId后会生成一个无效的空ID，自然查不到数据。

另外还有两个潜在问题：

1. 用customer.find()查单个ID会返回数组，更适合用customer.findOne()，性能更好且符合单文档查询的预期。
2. 每次查询都手动执行connect()和close()，Mongoose默认会维护连接池，频繁开闭连接会导致性能损耗甚至连接异常。

修复后的代码 #

export const findCustomer = async (query?: FindCustomerParams) => {
  // 移除手动连接/关闭逻辑：Mongoose会自动维护连接池，在应用启动时初始化一次即可
  if (!query) {
    return await customer.find();
  }

  findCustomerParams.parse(query);

  let returnCustomers;
  try {
    let queryObj = { ...query };

    // 正确处理ID查询：判断传入的是_id还是id字段，转换为ObjectId
    if (query._id) {
      queryObj._id = new mongoose.Types.ObjectId(query._id);
      delete queryObj.id; // 清理冗余字段
    } else if (query.id) {
      queryObj._id = new mongoose.Types.ObjectId(query.id);
      delete queryObj.id;
    }

    // 清理空值
    queryObj = removeFalsyValues(queryObj);

    // 单个ID查询用findOne，其他条件用find
    if (queryObj._id) {
      const fetchedCustomer = await customer.findOne(queryObj);
      if (!fetchedCustomer) {
        throw new Error("Customer not found");
      }

      // 关联查询地址（按客户ID匹配，而非原逻辑的索引匹配）
      if (query.address) {
        const fetchedAddress = await addressMethods.find({
          ...query.address,
          customers: fetchedCustomer._id
        });
        if (!fetchedAddress?.length) {
          throw new Error(`Could not find address on customer ${fetchedCustomer._id}`);
        }
        fetchedCustomer.address = fetchedAddress[0];
      }
      returnCustomers = [fetchedCustomer]; // 保持返回数组格式兼容原有逻辑
    } else {
      const fetchedCustomers = await customer.find(queryObj);
      if (!fetchedCustomers) {
        throw new Error();
      }

      if (query.address) {
        const fetchedAddresses = await addressMethods.find(query.address);
        // 按客户ID匹配地址，避免原逻辑索引错位的问题
        for (let customer of fetchedCustomers) {
          const matchedAddress = fetchedAddresses.find(addr => addr.customers.includes(customer._id));
          if (!matchedAddress) {
            throw new Error(`Could not find address on customer ${customer._id}`);
          }
          customer.address = matchedAddress;
        }
      }
      returnCustomers = fetchedCustomers;
    }
  } catch (err: any) {
    throw new Error("Failed to find a customer. Root error: " + err);
  }

  return returnCustomers;
};

关键修复点 #

• 修正_id转换逻辑：根据传入的字段（_id或id）正确获取ID值并转换为ObjectId
• 针对单个ID查询使用findOne，优化性能和结果准确性
• 移除冗余的连接/关闭逻辑，改用Mongoose连接池管理
• 优化地址关联查询的匹配逻辑，避免原代码中按索引匹配的潜在错误

内容的提问来源于stack exchange，提问作者Ethan