C++ cli pure virtual function overloading (or disabling?) - c++-cli

I don't know if it is possible, but there is simple example what I should achieve:
I have to use the first pure virtual function, because there are a lot of derived classes that needs it. But a few derived classes needs the second parameter too.
There is the base class Car:
public ref class Car abstract
{
.....
public:
virtual void move(Road ^ road) = 0; //I am not allowed to delete this line
//virtual void move(Road ^ road, Parameter2 ^ parameter2) = 0; //overload
}
Is it possible to overload the pure virtual function? Or in the worst case to disable that function in that few classes which need two parameters?
I am just learning, sorry for stupid questions..

It is certaily possible to overload a virtual function. Which one is called is based on the number (and type) of the function parameters. Note that pure virtual functions ('=0') impose the requirement on all derived classes to implement all pure virtual functions of the base class (or be abstract themselves as well, passing down the requirement to further derived classes).
If you can modify the signature/body of the first declaration, you can get away with only one function, though:
virtual void move(Road ^ road, Parameter2 ^ parameter2 = 0) = 0;
which assigns a default parameter to parameter2, when none is supplied explicitly at the time of calling. Then have the logic be identical to your first case (no parameter2), when the value of parameter2 is 0.

Related

Passing all arguments to class constructor vs passing arguments to member functions

This question is about OOP design.
What are the advantages/disadvantages of passing all arguments to class constructor vs passing arguments to member functions ?
In my case I know all arguments in the beginning of the program and I don't need to change them until the program is over.
In C++ the situation would be something like that (although in my code I need to parse more arguments and the member functions are more complex):
// All arguments in class constructor
Rectangle::Rectangle(float base, float height, string rectColor){
this->area = 0;
this->base = base;
this->height = height;
this->rectColor = rectColor;
}
void Rectangle::calcArea(){
area = base * height;
}
void Rectangle::paintRectangle(){
// use area
// whatever
}
vs
// Arguments in member functions
Rectangle::Rectancle(){
this->area = 0;
}
void Rectangle::calcArea(float base, float height){
area = base * height;
}
void Rectangle::paintRectangle(string rectColor){
// use area
// whatever
}
One strategy that I'm using is: If I need the variable in multiple member functions I make it a class variable. Is that good or the best approach ?
A bad thing about passing everything into constructor is that it would have lots of arguments. And also I wouldn't need to call the class member functions in my main.
Please explain the main principles I should follow.
Benefits for putting arguments in the constructor:
The instance is more completely initialized (no problems with the order of functions to be called to get a 'complete'useful instance.
Benefits for putting arguments to specific functions:
More flexibility, since the functions use variables instead of 'constants' passed to the constructor
The arguments passed to functions tend to belong better to the functions (e.g. for paintRectangle it is logical to pass the color, but if it never change, why making the flexibility to change the color afterwards?)
To prevent too many arguments in the cnstructor
Create a structure to pass the variables
Create a sub class ... if you need more than 5-7 parameters, possibly the responsibility of the class is too big.
Use named arguments (more clearer, but still the same amount of arguments)
In general, make classes as limited as possible, do not make them more flexible than needed UNLESS you know beforehand the functionality is needed at a later stage.
The one of the most important thing about constructors is that they make your OOP code consistent. For any object, it is good approach, if you already know, that someone created with properties need to have to exist.
i.e. Rectangle cannot exist without "height".
Therefore the "minimum parameters constructor" is great advantage. (the minimum required parameters need to have for object, to be usable and for not able to crash, when computing i.e. area)
If you have more parameters, that they are not neccesary, it is good to create more constructors based on what is probably "often use" of your object.

Static methods in OOP

I've always known what static methods are by definition, but I've always avoided using them at school because I was afraid of what I didn't know.
I already understand that you can use it as a counter throughout your entire project.
Now that I am interning I want to know when exactly static methods are used. From my observation so far, static classes/methods are used when it contains a lot of functions that will be used in many different classes and itself doesn't contain too many critical local variables within the class where it is not necessary to create an instant of it.
So as an example, you can have a static class called Zip that zips and unzips files and provide it to many different classes for them to do whatever with it.
Am I right? Do I have the right idea? I'm pretty sure there are many ways to use it.
Static functions are helpful as they do not rely on an instantiated member of whatever class they are attached to.
Static functions can provide functionality related to an a particular class without requiring the programmer to first create an instance of that class.
See this comparison:
class Numbers
{
public int Add(int x, int y)
{
return x + y;
}
public static int AddNumbers(int x, int y)
{
return x + y;
}
}
class Main
{
//in this first case, we use the non-static version of the Add function
int z1 = (new Numbers()).Add(2, 4);
//in the second case, we use the static one
int z2 = Numbers.AddNumbers(3, 5);
}
Technically, answers above are correct.
But the examples are not correct from the OOP point of view.
For example you have a class like this:
class Zip
{
public static function zipFile($fileName)
{
//
}
public static function unzipFile($fileName)
{
//
}
}
The truth is that there is nothing object-oriented here. You just defined two functions which you need to call using the fancy syntax like Zip::zipFile($myFile) instead of just zipFile($myFile).
You don't create any objects here and the Zip class is only used as a namespace.
So in this case it is better to just define these functions outside of class, as regular functions. There are namespaces in php since version 5.3, you can use them if you want to group your functions.
With the OOP approach, your class would look like this:
class ZipArchive
{
private $_archiveFileName;
private $_files;
public function __construct($archiveFileName) {
$this->_archiveFileName = $archiveFileName;
$this->_files = [];
}
public function add($fileName)
{
$this->_files[] = $fileName;
return $this; // allows to chain calls
}
public function zip()
{
// zip the files into archive specified
// by $_archiveFileName
}
}
And then you can use it like this:
$archive = new ZipArchive('path/to/archive.zip');
$archive->add('file1')->add('file2')->zip();
What is more important, you can now use the zip functionality in an OOP way.
For example, you can have a base class Archive and sub-classes like ZipArchive, TarGzArchive, etc.
Now, you can create an instance of the specific sub-class and pass it to other code which will not even know if files are going to be zip-ped or tag.gz-ipped. For example:
if ($config['archive_type'] === 'targz') {
// use tar.gz if specified
$archive = new TarGzArchive($path);
} else {
// use zip by default
$archive = new ZipArchive($path);
}
$backup = new Backup($archive /*, other params*/);
$backup->run();
Now the $backup object will use the specified archive type. Internally it doesn't know and doesn't care how exactly files will be archived.
You can even have a CopyArchive class which will simply copy files to another location.
It is easy to do it this way because your archive support is written in OOP way. You have small object responsible for specific things, you create and combine them and get the result you want.
And if you just have a bunch of static methods instead of real class, you will be forced to write the procedural-style code.
So I would not recommend to use static methods to implement actual features of your application.
Static methods may be helpful to support logging, debugging, testing and similar things. Like if you want to count number of objects created, you can use class-level static counter, increment it in the constructor and you can have a static method which reads the counter and prints it or writes to the log file.
Yes, static classes are used for problems that require stateless computation. Such as adding two numbers. Zipping a file. Etc.
If your class requires state, where you need to store connections or other longer living entities, then you wouldn't use static.
AFAIK. Static methods does not depends on a class instance. Just that.
As an example:
If you have an single thread program that will have only ONE database connection and will do several queries against the database it will be better to implement it as a static class (note that I specified that you will not connect, ever to several databases or have several threads).
So you will not need to create several connection objects, because you already know that you will only use one. And you will not need to create several objects. Singletons in this scenario are, also, an option.
There are other examples.
If you create an class to convert values.
class Convert{
static std::string fromIntToString(int value);
}
This way you will not need to create the class convert every time you need to convert from integer to an string.
std::string value = Convert::fromIntToString(10).
If you haven't done that you would need to instantiate this class several times through your program.
I know that you can find several other examples. It is up to you and your scenario to decide when you are going to do that.

Possibility of having "dynamically-binded" and "implicit" interface?

Is there any construct that allows all classes which implemented a set of functions to be considered as a certain interface, even when the classes themselves do not explicitly implement the interface?
To make the question clearer, I'll make an example. Suppose we want to implement LinearSearch, which look through the whole array and search for certain key, and return the index of the key upon discovery. Essentially, the psudeocode might look something like this:
LinearSearch(A, key)
for (k = 0; k < A.length(); k++)
if (A.get(k) == key)
return k
return NULL
In that case, any classes which implemented length and get will be able to search through the structure. We could implement this on DynamicArray, which acts the same as ArrayList in Java. We could implement this on a LinkedList, ignoring the fact the get takes linear time per query. Similarly for other structures that implement these 2 functions. However, such classes might not have explicitly implemented a common interface, even though it is favorable to have them being in one.
While writing this question, I feel a sense of insecurity tinkering within me about such a construct, but I cannot put it into words. So, is there any reason you think that this might not be a good construct in actual languages?
It's called "duck typing". Message-based object models like Smalltalk allow sending any message to an object as long as its name and parameters match.
In languages like C++, you can emulate this using "signals" and "slots", which, at their most primitive, can be implemented by writing a little template adapter class like
class CallGetLengthAdapterBase
{
public:
int length() = 0;
key_type key() = 0;
};
template<class N>
class CallGetLengthAdapter : public CallGetLengthAdapterBase
{
public:
CallGetLengthAdapter( N* obj ) { mObject = obj; };
int length() { return mObject->length(); };
key_type key() { return mObject->key(); };
protected:
N* mObject;
};
So the LinearSearch would just know about CallGetLengthAdapterBase, and would take a pointer to an object of this type. Whoever owns and connects both of these objects would call them like:
LinearSearch( CallGetLengthAdapter<A_type>(&A), key );
That's all.
From Wikipedia:
Go has "interface" types that are compatible with any type that supports a given set of methods (the type does not need to explicitly implement the interface). The empty interface, interface{}, is compatible with all types.
It sounds like this is what you mean, so it is another sense of interface than we might be used to from Java or such. This is a structural typing kind of interface, where the structure of methods involved are the important part, not a name given to the interface.
More formally, it seems that this is called a type class.

Why is overriding of static methods left out of most OOP languages?

It is certainly not for good OOP design - as the need for common behavior of all instances of a derived class is quite valid conceptually. Moreover, it would make for so much cleaner code if one could just say Data.parse(file), have the common parse() code in the base class and let overriding do its magic than having to implement mostly similar code in all data subtypes and be careful to call DataSybtype.parse(file) - ugly ugly ugly
So there must be a reason - like Performance ?
As a bonus - are there OOP languages that do allow this ?
Java-specific arguments are welcome as that's what I am used to - but I believe the answer is language agnostic.
EDIT : one could ideally :
<T> void method(Iface<? extends T> ifaceImpl){
T.staticMeth(); // here the right override would be called
}
This will also fail due to erasure (in java at least) - if erasure is at work one needs (would need) to actually pass the class :
<T, K extends T> void method(Iface<K> ifaceImpl, Class<K> cls){
cls.staticMeth(); // compile error
}
Does it make sense ? Are there languages doing this already ? Is there a workaround apart from reflection ?
Speaking to C++
class Foo {
public:
static void staticFn(int i);
virtual void virtFn(int i);
};
The virtual function is a member function - that is, it is called with a this pointer from which to look up the vtable and find the correct function to call.
The static function, explicitly, does not operate on a member, so there is no this object from which to look up the vtable.
When you invoke a static member function as above, you are explicitly providing a fixed, static, function pointer.
foo->virtFn(1);
expands out to something vaguely like
foo->_vtable[0](foo, 1);
while
foo->staticFn(1);
expands to a simple function call
Foo##staticFn(1);
The whole point of "static" is that it is object-independent. Thus it would be impossible to virtualize.

Injection of class with multiple constructors

Resolving a class that has multiple constructors with NInject doesn't seem to work.
public class Class1 : IClass
{
public Class1(int param) {...}
public Class1(int param2, string param3) { .. }
}
the following doesn’t seem to work:
IClass1 instance =
IocContainer.Get<IClass>(With.Parameters.ConstructorArgument(“param”, 1));
The hook in the module is simple, and worked before I added the extra constructor:
Bind().To();
The reason that it doesn't work is that manually supplied .ctor arguments are not considered in the .ctor selection process. The .ctors are scored according to how many parameters they have of which there is a binding on the parameter type. During activation, the manually supplied .ctor arguments are applied. Since you don't have bindings on int or string, they are not scored. You can force a scoring by adding the [Inject] attribute to the .ctor you wish to use.
The problem you're having is that Ninject selects .ctors based on the number of bound parameters available to it. That means that Ninject fundamentally doesn't understand overloading.
You can work around this problem by using the .ToConstructor() function in your bindings and combining it with the .Named() function. That lets you create multiple bindings for the same class to different constructors with different names. It's a little kludgy, but it works.
I maintain my own software development blog so this ended up being a post on it. If you want some example code and a little more explanation you should check it out.
http://www.nephandus.com/2013/05/10/overloading-ninject/