This question already has answers here:
Closed 12 years ago.
Possible Duplicates:
Public Data members vs Getters, Setters
Purpose of private members in a class
What is the use of getters and setters when you can just make your variables public and avoid the hassle of such lines as A.setVariableX(A.getVariableY())?
To hide implementation details. If you choose to make your properties public instead, then later decide to change the way they work, you have to change all of the associated code.
With getters and setters, you can often change the internal implementation details without changing the visible API (and thus avoid having to change any of the code that uses the API of the class in question).
Getters and setters can do validation, and lazy instantiation, whereas public members cannot.
As an aside, this isn't language agnostic, as most of the languages that implement properties abstract away the implementation so they look like public members in code.
Because sometimes there are some constraints as to what your variable might be and having to check that when you set it make sense.
for instance, if your Age is not allowed to be -1, you would check that in your setter.
if(value >= 0)
{
_age = value;
}
else
{
throw new InvalidAgeException("Age may not be less than 0");
}
Setters and getters apply to properties, whatever "properties" are in your code.
For instance, you can have an angle variable, which stores an angle value expressed in radians. you can then define setters and getters for this variable, with any angle unit that you think relevant.
It's part of the encapsulation principle. You use setters and getters as an interface to interact with your object so that you can - for example - later add any functionality (like validating input data) without touching the code that uses the class(es).
Related
I'm new to learning Kotlin, and have just covered the fact that Getters and Setters for every object property are automatically generated behind the scenes, like Lombok is built directly into the language (great as far as I'm concerned!), but now I'm left with the question of why does it even bother?
I know you can override the defaults to change how they act, but why is this not just that case of them being created when you need to do something more complex that just getting or setting the value, but just accessing the property normally otherwise?
Having used Java a lot previously, I understand there are arguments for and against the use of Getters and Setters. Many say they're actually pointless and a bad approach to OOP, others would say they're best practice and should be used everywhere. The way many modern frameworks and libraries are written makes them necessary in lots of cases. This argument spans many different languages, but this question isn't really about the theoretical reasons for them in OOP as a whole, I'm looking for insight specifically into why they're used in Kotlin by default, when to my (perhaps naive) eyes it looks like they needn't be.
What would be the practical difference if Kotlin didn't generate standard Getters and Setters and dot notation simply accessed the property directly? Syntactically, as far as I can see, it produces the same result, just with extra steps.
What would be the practical difference if Kotlin didn't generate standard Getters and Setters and dot notation simply accessed the property directly?
So what you are suggesting is, for properties without custom getters/setters,
// Accessing x would access the field backing the property directly
// no getter or setter
var x: Int = 1
And for properties with custom getters/setters,
// both a field and a setter is generated for y
var y: Int = 1
set(newValue) {
field = newValue.also { println("y is being set!") }
}
// only a getter is generated for z, no field
val z get() = 1
The appropriated getter or setter or both will be generated depending on which ones you wrote.
Well, this would be problematic in a few ways. Here's what I could think of.
Firstly, all of this "when (not) to generate a getter/setter" logic is going to make the compiler more complicated. Why do that, when you can just simply generate a getter and setter, unconditionally, for every property?
Secondly, for Java code interacting with the Kotlin code, they would only have a field to work with, in the case of x, and public fields in Java doesn't look idiomatic at all... Let's say at some point you add a getter to x so that it's computed from something else, rather than backed by a field, then all the Java code would break, because there is no longer a field. I'm not sure whether this is also true on Kotlin/JS.
Thirdly, accessing the values of properties through reflection would be more annoying. You would need to check if it has a getter. If it does, call it. Otherwise, access the backing field instead. They could provide you with a helper property on KProperty that does on this for you, but still, that's a lot more complicated than simply calling the getter and be done with it.
Do note that just for JVM, there is the annotation #JvmField that you can use to make a property not have getters and setters, as far as the JVM is concerned. Kotlin's reflection API will still find a getter and setter though.
The two main arguments against getters and setters:
Verbose. But they aren't verbose in Kotlin.
Performance overhead. But it's not a concern with modern JVM's.
It's much safer for future-proofing to use getters and setters from the original design of a class, even if they are just passing through reads and writes to a field. It leaves open the possibility of adding side effects if you need to later. Forcing the use of properties prevents the possibility of you designing yourself into a corner.
Also, a language is easier to read and write when there are fewer ways of doing the same task. You never have to remember if the property you want to access on a class uses a getter function syntax versus standard property syntax.
What is the purpose of Kotlin's implicit calling of getter/setter functions for properties when you try to access them? Isn't the point of getter and setters already that you can easily call them if you intend to use them? Kotlin's version basically just introduces an additional complexity with the 'field' identifier and introduces weirdness like the following, where an object may not behave like its interface intended:
interface Counter {
var count: Int
fun increment() {
count = count + 1
}
}
class WeirdCounter: Counter {
override var count: Int = 0
get() = field
set(value) {println("ignore the value")}
}
Just trying to understand the intent behind this.
The way Kotlin implements properties using getters and setters is basically what's common practice — and best practice — in many other languages.
‘Bare’ fields, as in Java, are simple, clear, and easy to use; but bare fields have problems:
They expose an implementation detail (the field, and especially its type), preventing it from being changed in future.
They don't allow the class to control its own state.
Of course, these aren't a problem for simple value classes. But for more complex classes, they can be a real problem.
For example, you may want to change the way a class stores its state (e.g. replacing a long with a BigDecimal), but if that class is part of a popular library's public interface then thousands of users would get pretty annoyed.
Or suppose it would be really convenient if you could ensure that a String property was always stored in lower-case without leading or trailing whitespace. But with a ‘bare’ property there's no way to enforce that.
So the usual pattern is to have a field that's private, and only accessible from within the class itself (which you control); and provide accessor methods.
That gives you full control. You can change the internal representation, as long as you update the accessor methods to convert to/from the new form as needed. And your setter can do any normalisation, formatting, or whatever to enforce any restrictions on the state.
However, in languages like Java, that's more awkward and long-winded than a simple field: accessor methods turn a one-line field into seven lines (excluding blank lines, and excluding doc comments, so that's probably more like turning 3 lines into 21). And while calling a getter method is only a few characters longer (with get and ()) than referencing a field, calling a setter is a lot less intuitive than a simple assignment.
The result is that either developers do the right thing and fill their classes with boilerplate (with all the implications for maintainability and risk of error), or they don't bother and risk the problems above.
Kotlin, though, gives the best of both worlds: a simple property looks just like a field, both when defining and when accessing it. So you get the lean, concise, clear code. But it's implemented with a private backing field (if needed) and accessor method(s)s, so you get all the advantages of those too. And if you ever need to add validation or change the representation or log all access or whatever, you have the option of replacing the default accessors with your own implementations.
Your WeirdCounter example is odd, but not as scary (or as likely) as you might think. In an object-oriented language, a class is master of its own state, and other classes generally don't and shouldn't know about its internals. (That way, they're insulated from changes to those internals.) If a class needs to do something counter-intuitive in a setter, that's only a concern if it breaks the class's contract — but that would be a bug, and should become obvious in tests, if not elsewhere.
In practice, the ability for classes to control access to their state is more important than the risk of a class using that to do something stupid or malicious (that would be fairly easy to spot).
I am from a C# background and have been doing programming for quite some time now. But only recently i started giving some thoughts on how i program. Apparently, my OOP is very bad.
I have a few questions maybe someone can help me out. They are basic but i want to confirm.
1- In C#, we can declare class properties like
private int _test;
and there setter getters like
public int Test {get; set;}
Now, lets say i have to use this property inside the class. Which one will i use ? the private one or the public one ? or they both are the same ?
2- Lets say that i have to implement a class that does XML Parsing. There can be different things that we can use as input for the class like "FILE PATH". Should i make this a class PROPERTY or should i just pass it as an argument to a public function in the class ? Which approach is better. Check the following
I can create a class property and use like this
public string FilePath {get; set;}
public int Parse()
{
var document = XDocument.Load(this.FilePath);
.........//Remaining code
}
Or
I can pass the filepath as a parameter
public int Parse(string filePath)
On what basis should i make a decision that i should make a property or i should pass something as argument ?
I know the solutions of these questions but i want to know the correct approach. If you can recommend some video lectures or books that will be nice also.
Fields vs Properties
Seems like you've got a few terms confused.
private int _test;
This is an instance field (also called member).
This field will allow direct access to the value from inside the class.
Note that I said "inside the class". Because it is private, it is not accessible from outside the class. This is important to preserve encapsulation, a cornerstone of OOP. Encapsulation basically tells us that instance members can't be accessed directly outside the class.
For this reason we make the member private and provide methods that "set" and "get" the variable (at least: in Java this is the way). These methods are exposed to the outside world and force whoever is using your class to go trough your methods instead of accessing your variable directly.
It should be noted that you also want to use your methods/properties when you're inside the current class. Each time you don't, you risk bypassing validation rules. Play it safe and always use the methods instead of the backing field.
The netto result from this is that you can force your logic to be applied to changes (set) or retrieval (get). The best example is validation: by forcing people to use your method, your validation logic will be applied before (possibly) setting a field to a new value.
public int Test {get; set;}
This is an automatically implemented property. A property is crudely spoken an easier way of using get/set methods.
Behind the scenes, your code translates to
private int _somevariableyoudontknow;
public void setTest(int t){
this._somevariableyoudontknow = t;
}
public int getTest(){
return this._somevariableyoudontknow;
}
So it is really very much alike to getters and setters. What's so nice about properties is that you can define on one line the things you'd do in 7 lines, while still maintaining all the possibilities from explicit getters and setters.
Where is my validation logic, you ask?
In order to add validation logic, you have to create a custom implemented property.
The syntax looks like this:
private int _iChoseThisName;
public int Test {
get {
return _iChoseThisName;
}
set {
if(value > 5) { return _iChoseThisName; }
throw new ArgumentException("Value must be over 5!");
}
}
Basically all we did was provide an implementation for your get and set. Notice the value keyword!
Properties can be used as such:
var result = SomeClass.Test; // returns the value from the 'Test' property
SomeClass.Test = 10; // sets the value of the 'Test' property
Last small note: just because you have a property named Test, does not mean the backing variable is named test or _test. The compiler will generate a variablename for you that serves as the backing field in a manner that you will never have duplication.
XML Parsing
If you want your second answer answered, you're going to have to show how your current architecture looks.
It shouldn't be necessary though: it makes most sense to pass it as a parameter with your constructor. You should just create a new XmlParser (random name) object for each file you want to parse. Once you're parsing, you don't want to change the file location.
If you do want this: create a method that does the parsing and let it take the filename as a parameter, that way you still keep it in one call.
You don't want to create a property for the simple reason that you might forget to both set the property and call the parse method.
There are really two questions wrapped in your first question.
1) Should I use getters and setters (Accessors and Mutators) to access a member variable.
The answer depends on whether the implementation of the variable is likely to change. In some cases, the interface type (the type returned by the getter, and set by the setter) needs to be kept consistent but the underlying mechanism for storing the data may change. For instance, the type of the property may be a String but in fact the data is stored in a portion of a much larger String and the getter extracts that portion of the String and returns it to the user.
2) What visibility should I give a property?
Visibility is entirely dependent on use. If the property needs to be accessible to other classes or to classes that inherit from the base class then the property needs to be public or protected.
I never expose implementation to external concerns. Which is to say I always put a getter and setter on public and protected data because it helps me ensure that I will keep the interface the same even if the underlying implementation changes. Another common issue with external changes is that I want a chance to intercept an outside user's attempt to modify a property, maybe to prevent it, but more likely to keep the objects state in a good or safe state. This is especially important for cached values that may be exposed as properties. Think of a property that sums the contents of an array of values. You don't want to recalculate the value every time it is referenced so you need to be certain that the setter for the elements in the array tells the object that the sum needs to be recalculated. This way you keep the calculation to a minimum.
I think the second question is: When do I make a value that I could pass in to a constructor public?
It depends on what the value is used for. I generally think that there are two distinct types of variables passed in to constructors. Those that assist in the creation of the object (your XML file path is a good example of this) and those that are passed in because the object is going to be responsible for their management. An example of this is in collections which you can often initialize the collection with an array.
I follow these guidelines.
If the value passed in can be changed without damaging the state of the object then it can be made into a property and publicly visible.
If changing the value passed in will damage the state of the object or redefine its identity then it should be left to the constructor to initialize the state and not be accesible again through property methods.
A lot of these terms are confusing because of the many different paradigms and languages in OO Design. The best place to learn about good practices in OO Design is to start with a good book on Patterns. While the so-called Gang of Four Book http://en.wikipedia.org/wiki/Design_Patterns was the standard for many years, there have since been many better books written.
Here are a couple resources on Design Patterns:
http://sourcemaking.com/design_patterns
http://www.oodesign.com/
And a couple on C# specific.
http://msdn.microsoft.com/en-us/magazine/cc301852.aspx
http://www.codeproject.com/Articles/572738/Building-an-application-using-design-patterns-and
I can possibly answer your first question. You asked "I have to use this property inside the class." That sounds to me like you need to use your private variable. The public method which you provided I believe will only do two things: Allow a client to set one of your private variables, or to allow a client to "see" (get) the private variable. But if you want to "use this property inside the class", the private variable is the one that should be your focus while working with the data within the class. Happy holidays :)
The following is my personal opinion based on my personal experience in various programming languages. I do not think that best practices are necessarily static for all projects.
When to use getters, when to use private instance variables directly
it depends.
You probably know that, but let's talk about why we usually want getters and setters instead of public instance variables: it allows us to aquire the full power of OOP.
While an instance variable is just some dump piece of memory (the amount of dumbness surely depends on the language you're working in), a getter is not bound to a specific memory location. The getter allows childs in the OOP hirarchy to override the behaviour of the "instance variable" without being bound to it. Thus, if you have an interface with various implementations, some may use ab instance variable, while others may use IO to fetch data from the network, calculate it from other values, etc.
Thus, getters do not necessarily return the instance variable (in some languages this is more complicated, such as c++ with the virtual keyword, but I'll try to be language-independent here).
Why is that related to the inner class behaviour? If you have a class with a non-final getter, the getter and the inner variable may return different values. Thus, if you need to be sure it is the inner value, use it directly. If you, however, rely on the "real" value, always use the getter.
If the getter is final or the language enforces the getter to be equal (and this case is way more common than the first case), I personally prefer accessing the private field directly; this makes code easy to read (imho) and does not yield any performance penalty (does not apply to all languages).
When to use parameters, when to use instance variables/properties
use parameters whereever possible.
Never use instance variables or properties as parameters. A method should be as self-contained as possible. In the example you stated, the parameterized version is way better imo.
Intance variables (with getters or not) are properties of the instance. As they are part of the instance, they should be logically bound to it.
Have a look at your example. If you hear the word XMLParser, what do you think about it? Do you think that a parser can only parse a single file it is bound to? Or do you think that a parser can parse any files? I tend to the last one (additionally, using an instance variable would additionally kill thread-safety).
Another example: You wish to create an XMLArchiver, taking multiple xml documents into a single archive. When implementing, you'd have the filename as a parameter of the constructor maybe opening an outputstream towards the file and storing a reference to it as an instance variable. Then, you'd call archiver.add(stuff-to-add) multiple times. As you see, the file (thus, the filename) is naturally bound to the XMLArchiver instance, not to the method adding files to it.
What are setters and getters? Why do I need them? What is a good example of them in use in an effective way? What is the point of a setter and getter?
Update:
Can I get some coding examples please?
A getter is a method that gets the value of a property. A setter is a method that sets the value of a property. There is some contention about their efficacy, but the points are generally:
for completeness of encapsulation
to maintain a consistent interface in case internal details change
More useful is when you need to add some logic around getting or setting, like validating a value before you write it.
A getter/setter is used to hide a private field from the publicity (you can avoid direct access to a field).
The getter allows you to check a provided value before you use it in your internal field. The setter allows you for instance to apply a different format or just to restrict write access (e.g. to derived classes).
A useful application of a getter can be some kind of lazy loading: The backing field (the private field that is hidden by the getter) is initialized to null. When you ask the getter to return the value, it will check for null and load the value with a more time consuming method. This will happen only the first call, later the getter will provide the already loaded value all the time.
Getters & setters separate interface (getter/setter functions) from implementation (how the data is actually stored).
Getters and Setters allow you to control how data members of an object can be accessed or changed.
In contrast, if you expose your data members directly to the user of the object, the user can change them at will, and the object wouldn't even know that they had been changed.
Don't want people to read a data member? Make the data member private, and don't write a getter that gives the value back. Don't want people to modify a data member? Make the data member private, and don't write a setter for it. Want to control the range of allowed values? Put that in the setter.
One question which might pop out of this is if using a method instead of a direct field access might decrease performance.
Answer is not really as compilers optimize code so that if your method is only doing return field;, where field is the field in your class that you hide with the setter/getter, it will actually access the field directly. Thus you get in most cases the same performance, at the same time keeping the option of later on change what set/get methods do.
Effective Java Programming of Joshua Block is a great book with tips on how to write good code, and explains why as well. Why using setter/getter is one of the hints.
Note: You might notice that in some books/documentation fields that present a setter/getter instead of being directly accessible are called 'properties' instead of fields. E.g. in C#, you can even specify that a field is a property and you don't need to define set/get anymore (nice feature I think).
public accessors(getter and setter) make sometimes sense.
(I'm annoyed that I have not only to document the member variable of a class but also the 2 mostly meaningless accessor methods. )
It usually doesn't help with encapsulation except in cases mentioned by Jason S.
An java example for some char loaded from a database but should be represented as a boolean value
char boolFromDb;
public boolean getBoolFromDb() {
return boolFromDb == 'T';
}
public void setBoolFromDb(boolean newValue) {
boolFromDb = newValue ? 'T' : 'F';
}
Should you ever use protected member variables? What are the the advantages and what issues can this cause?
Should you ever use protected member variables?
Depends on how picky you are about hiding state.
If you don't want any leaking of internal state, then declaring all your member variables private is the way to go.
If you don't really care that subclasses can access internal state, then protected is good enough.
If a developer comes along and subclasses your class they may mess it up because they don't understand it fully. With private members, other than the public interface, they can't see the implementation specific details of how things are being done which gives you the flexibility of changing it later.
Generally, if something is not deliberately conceived as public, I make it private.
If a situation arises where I need access to that private variable or method from a derived class, I change it from private to protected.
This hardly ever happens - I'm really not a fan at all of inheritance, as it isn't a particularly good way to model most situations. At any rate, carry on, no worries.
I'd say this is fine (and probably the best way to go about it) for the majority of developers.
The simple fact of the matter is, if some other developer comes along a year later and decides they need access to your private member variable, they are simply going to edit the code, change it to protected, and carry on with their business.
The only real exceptions to this are if you're in the business of shipping binary dll's in black-box form to third parties. This consists basically of Microsoft, those 'Custom DataGrid Control' vendors, and maybe a few other large apps that ship with extensibility libraries. Unless you're in that category, it's not worth expending the time/effort to worry about this kind of thing.
The general feeling nowadays is that they cause undue coupling between derived classes and their bases.
They have no particular advantage over protected methods/properties (once upon a time they might have a slight performance advantage), and they were also used more in an era when very deep inheritance was in fashion, which it isn't at the moment.
The key issue for me is that once you make a variable protected, you then cannot allow any method in your class to rely on its value being within a range, because a subclass can always place it out of range.
For example, if I have a class that defines width and height of a renderable object, and I make those variables protected, I then can make no assumptions over (for example), aspect ratio.
Critically, I can never make those assumptions at any point from the moment that code's released as a library, since even if I update my setters to maintain aspect ratio, I have no guarantee that the variables are being set via the setters or accessed via the getters in existing code.
Nor can any subclass of my class choose to make that guarantee, as they can't enforce the variables values either, even if that's the entire point of their subclass.
As an example:
I have a rectangle class with width and height being stored as protected variables.
An obvious sub-class (within my context) is a "DisplayedRectangle" class, where the only difference is that I restrict the widths and heights to valid values for a graphical display.
But that's impossible now, since my DisplayedRectangle class cannot truly constrain those values, as any subclass of it could override the values directly, while still being treated as a DisplayedRectangle.
By constraining the variables to be private, I can then enforce the behaviour I want through setters or getters.
In general, I would keep your protected member variables to the rare case where you have total control over the code that uses them as well. If you are creating a public API, I'd say never. Below, we'll refer to the member variable as a "property" of the object.
Here's what your superclass cannot do after making a member variable protected rather than private-with-accessors:
lazily create a value on the fly when the property is being read. If you add a protected getter method, you can lazily create the value and pass it back.
know when the property been modified or deleted. This can introduce bugs when the superclass makes assumptions about the state of that variable. Making a protected setter method for the variable keeps that control.
Set a breakpoint or add debug output when the variable is read or written to.
Rename that member variable without searching through all the code that might use it.
In general, I think it'd be the rare case that I'd recommend making a protected member variable. You are better off spending a few minutes exposing the property through getters/setters than hours later tracking down a bug in some other code that modified the protected variable. Not only that, but you are insured against adding future functionality (such as lazy loading) without breaking dependent code. It's harder to do it later than to do it now.
At the design level it might be appropriate to use a protected property, but for implementation I see no advantage in mapping this to a protected member variable rather than accessor/mutator methods.
Protected member variables have significant disadvantages because they effectively allow client code (the sub-class) access to the internal state of the base class class. This prevents the base class from effectively maintaining its invariants.
For the same reason, protected member variables also make writing safe multi-threaded code significantly more difficult unless guaranteed constant or confined to a single thread.
Accessor/mutator methods offer considerably more API stability and implementation flexibility under maintenance.
Also, if you're an OO purist, objects collaborate/communicate by sending messages, not reading/setting state.
In return they offer very few advantages. I wouldn't necessarily remove them from somebody else's code, but I don't use them myself.
Just for the record, under Item 24 of "Exceptional C++", in one of the footnotes, Sutter goes
"you would never write a class that has a public or protected member variable. right? (Regardless of the poor example set by some libraries.)"
Most of the time, it is dangerous to use protected because you break somewhat the encapsulation of your class, which could well be broken down by a poorly designed derived class.
But I have one good example: Let's say you can some kind of generic container. It has an internal implementation, and internal accessors. But you need to offer at least 3 public access to its data: map, hash_map, vector-like. Then you have something like:
template <typename T, typename TContainer>
class Base
{
// etc.
protected
TContainer container ;
}
template <typename Key, typename T>
class DerivedMap : public Base<T, std::map<Key, T> > { /* etc. */ }
template <typename Key, typename T>
class DerivedHashMap : public Base<T, std::hash_map<Key, T> > { /* etc. */ }
template <typename T>
class DerivedVector : public Base<T, std::vector<T> > { /* etc. */ }
I used this kind of code less than a month ago (so the code is from memory). After some thinking, I believe that while the generic Base container should be an abstract class, even if it can live quite well, because using directly Base would be such a pain it should be forbidden.
Summary Thus, you have protected data used by the derived class. Still, we must take int o account the fact the Base class should be abstract.
In short, yes.
Protected member variables allow access to the variable from any sub-classes as well as any classes in the same package. This can be highly useful, especially for read-only data. I don't believe that they are ever necessary however, because any use of a protected member variable can be replicated using a private member variable and a couple of getters and setters.
For detailed info on .Net access modifiers go here
There are no real advantages or disadvantages to protected member variables, it's a question of what you need in your specific situation. In general it is accepted practice to declare member variables as private and enable outside access through properties. Also, some tools (e.g. some O/R mappers) expect object data to be represented by properties and do not recognize public or protected member variables. But if you know that you want your subclasses (and ONLY your subclasses) to access a certain variable there is no reason not to declare it as protected.