如何在SYMBIAN60中编写DLL,比较经典!

Prerequisites

  • Knowledge of C++ and COM.
  • Knowledge of Symbian Programming idioms

Introduction

Writing a polymorphic DLL involves the following steps:

  1. Design and write interfaces.
  2. Implement the interface.

Lets explore these in more detail:

Design and write interface consisting of method declarations.

An interface is essentially an abstract class. An example would clarify it:

class ICalculator {

public:
virtual TInt Sum(TInt, TInt) = 0;

};

Above snippet declares a class ICalculator with one method Sum which would take 2 TInt parameters and would return a TInt value. Take a look at the name given to the class. This is done to follow a convention i.e. an interface name should start with an 'I'. Now if one wishes to implement interface ICalculator, he/she has to implement the method Sum.

COM programmers would be wondering that how it can be an interface without a UUID or something like that. Yes! in Symbian too, an interface requires a UUID. While writing an interface this is how UUID is given to it.

static const TUid KCalculator_DLLUID = {0x100039CE};
static const TUid KCalculator_DLLUID_Ver1 = {0x0352D96B};
class ICalculator {

public:
virtual TInt Sum(TInt, TInt) = 0;

};

Your polymorphic interface is ready. You have to write it in a .h file so that it can be implemented in other .cpp files.

Implement the interface.

Writing a polymorphic DLL is half done when interface declaration is done. Now one would require to implement it. Implementing an interface is as simple as inheriting abstract classes. You have to inherit from an interface and implement pure virtual methods of the interface.

Again an example would help us understand this:

class CImpCalculator: public ICalculator {

public:

EXPORT_C CImpCalculator* NewImpClass();

//function would be exposed through Function table
TInt Sum(TInt num1,TInt num2);
};


GLDEF_C TInt E32Dll(TDllReason) {

return KErrNone;
}

EXPORT_C CImpCalculator* CImpCalculator::NewImpClass() {

return new (ELeave) CImpCalculator;
}

TInt CImpCalculator::Sum(TInt num1,TInt num2) {

return num1 + num2;

}

The example above does something more than just implementing the interface. The class CImpCalculator has a method of its own, NewImpClass. This method has a qualifier EXPORT_C. Clearly this is the method which is exposed from the implementation class. In Symbian it is not required to export or expose all the methods from an interface. Only method which needs to be exposed is which returns the object of the interface (or should I say the implementor class?). Rest of the methods can be accessed from function table. The NewImpClass method returns a pointer to the newly created object.

E32Dll method has to be written in every Polymorohic DLL as it is required by the Symbian run time. You would get linking errors in case you forget to write it. This method is responsible to allocate thread local store for the DLL instance.

Now you are ready to compile and build your own DLL. Build it and start using it.

Lastly these DLLs are called polymorphic because one can write many implementations of the same interface. Further all the Symbian GUI applications are Polymorphic DLLs. Check the GUI SDK examples, you will find E32Dll method implemented.

That's all you would require to write a DLL on Symbian!! Note that I have tested the code on NOKIA Series 60 emulator.

 
  • 0
    点赞
  • 0
    收藏
    觉得还不错? 一键收藏
  • 0
    评论
在风能领域,准确预测风速对于风电场的运行与管理至关重要。Matlab作为一个强大的数学计算和数据分析平台,被广泛应用于风速预测模型的构建。本文将深入探讨基于四种风速——随机风、基本风、阵风和渐变风的组合风速预测技术。 我们来理解这四种风速类型: 1. **随机风**:随机风是指风速呈现出随机性的变化,通常由大气湍流引起。在建模,通常通过统计方法如高斯分布或Weibull分布来模拟这种不确定性。 2. **基本风**:基本风速是指在无特定扰动条件下的平均风速,它是长期观测结果的平均值,通常用于结构设计和风能评估。 3. **阵风**:阵风是短时间内风速显著增强的现象,对建筑物和风力发电机造成的主要威胁之一。阵风的预测涉及到风的脉动特性分析。 4. **渐变风**:渐变风是指风速随时间和空间逐渐变化的过程,常见于风向转变或地形影响下的风场变化。 在Matlab,利用这四种风速类型进行组合预测,可以提高预测的准确性。预测模型可能包括以下几个步骤: 1. **数据收集与预处理**:收集历史风速数据,包括随机风、基本风、阵风和渐变风的数据,进行异常值检测、缺失值填充以及数据标准化。 2. **特征工程**:提取风速变化的相关特征,如平均值、标准差、极值、频率分布等,这些特征可能对预测有重要影响。 3. **模型选择**:可以选择多种预测模型,如时间序列分析(ARIMA、状态空间模型等)、机器学习算法(线性回归、决策树、支持向量机、神经网络等)或深度学习模型(LSTM、GRU等)。 4. **模型训练**:利用历史数据训练选定的模型,调整模型参数以优化性能,例如通过交叉验证来避免过拟合。 5. **模型验证与评估**:使用独立的测试集验证模型预测效果,常见的评估指标有均方误差(MSE)、平均绝对误差(MAE)和决定系数(R²)。 6. **组合预测**:结合四种风速的不同模型预测结果,可以采用加权平均、集成学习(如bagging、boosting)等方式,以提升整体预测精度。 7. **实时更新与动态调整**:实际应用,模型需要不断接收新的风速数据并进行在线更新,以适应风场环境的变化。 通过以上步骤,可以构建一个综合考虑各种风速特性的预测系统,这对于风电场的功率输出预测、风电设备的维护计划以及电网调度都具有重要价值。然而,需要注意的是,每个风场的地理环境、气候条件和设备状况都有所不同,因此模型的建立应根据实际情况进行定制和优
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值