Related
I've read a lot of articles about "public vs getter/setter", but I still wonder if there is any good part about public variable.
Or the question is:
If you're going to make a new awesome programming languange, are you still going to support public variable and why??
I agree with almost everything that's been said by everyone else, but wanted to add this:
Public isn't automatically bad. Public is bad if you're writing an Object Class. Data Classes are just fine. There's nothing wrong with this class:
public class CommentRecord
{
public int id;
public string comment;
}
... why? Because the class isn't using the variables for anything. It's just a data object - it's meant to be just a simple data repository.
But there's absolutely something wrong with this class:
public class CommentRecord
{
public int id;
public string comment;
public void UpdateInSQL()
{
// code to update the SQL table for the row with commentID = this.id
// and set its UserComment column to this.comment
}
}
... why is this bad? Because it's not a data class. It's a class that actually does stuff with its variables - and because of that, making them public forces the person using the class to know the internals of the class. The person using it needs to know "If I want to update the comment, I have to change the public variable, but not change the id, then call the UpdateInSQL() method." Worse, if they screw up, they use the class in a way it wasn't intended and in a way that'll cause unforseen consequences down the line!
If you want to get some more info on this, take a look at Clean Code by Robert Martin, Chapter 6, on "Data/Object Anti-Symmetry"
A public variable essentially means you have a global accessible/changeable variable within the scope of an object. Is there really a use case for this?
Take this example: you have a class DatabaseQueryHandler which has a variable databaseAccessor. Under what circumstances would you want this variable to be:
Publicly accessible (i.e. gettable)
Publicly settable
Option #1 I can think of a few - you may want to get the last insert ID after an insert operation, you may want to check any errors the last query generated, commit or rollback transactions, etc., and it might make more logical sense to have these methods written in the class DatabaseAccessor than DatabaseQueryHandler.
Option #2 is less desirable, especially if you are doing OOP and abiding by SOLID principles, in particular regards to the ISP and DIP principles. In that case, when would you want to set the variable databaseAccessor in DatabaseQueryHandler? Probably on construction only, and never at any time after that. You probably also want it type-hinted at the interface level as well, so that you can code to interfaces. Also, why would you need an arbitrary object to be able to alter the database accessor? What happens if Foo changes the variable DatabaseQueryHandler->databaseAccessor to be NULL and then Bar tries to call DatabaseQueryHandler->databaseAccessor->beginTransaction()?
I'm just giving one example here, and it is by no means bullet proof. I program in PHP (dodges the hurled rotten fruit) and take OOP and SOLID very seriously given the looseness of the language. I'm sure there will be arguments on both sides of the fence, but I would say that if you're considering using a public class variable, instead consider what actually needs to access it, and how that variable is to be used. In most cases the functionality can be exposed via public methods without allowing unexpected alteration of the variable type.
Simple answer is: yes, they are bad. There are many reasons to that like coupling and unmaintanable code. In practice you should not use them. In OOP the public variable alternative is Singleton, which is considered a bad pracitce. Check out here.
It has a lot to do with encapsulation. You don't want your variable to be accessed anyhow. Other languages like iOS (objective-c) use properties:
#property (nonatomic, strong) NSArray* array;
then the compiler will generate the instance variable with it's getter and setter implicitly. In this case there is no need to use a variable (though other developers still prefer to use variables). You can then make this property public by declaring it in the .h file or private by declaring it in the .m file.
I am designing a process that should be run daily at work, and I've written a class to do the work. It looks something like this:
class LeadReport {
public $posts = array();
public $fields = array();
protected _getPost() {
// get posts of a certain type
// set them to the property $this->posts
}
protected _getFields() {
// use $this->posts to get fields
// set $this->fields
}
protected _writeCsv() {
// use the properties to write a csv
}
protected _sendMail() {
// attach a csv and send it
}
public function main() {
$this->out('Lead Report');
$this->out("Getting Yesterday's Posts...");
$this->_getPosts();
$this->out("Done.");
$this->out("Next, Parsing the Fields...");
$this->_getFields();
$this->out("Done.");
$this->out("Next, Writing the CSVs...");
$this->_writeCsv();
$this->out("Done.");
$this->out("Finally, Sending Mail");
$this->_sendMail();
$this->out('Bye!');
}
}
After showing this code to one of my colleagues, he commented that the _get() methods should have return values, and that the _write() and _sendMail() methods should use those values as parameters.
So, two questions:
1) Which is "correct" in this case (properties or return values)?
2) Is there a general rule or principle that governs when to use properties over when to use return values in object oriented programming?
I think maybe the source of your question comes from the fact that you are not entirely convinced that using properties is better than having public fields. For example here, common practice says that should not have posts and fields as public. You should use the getField method or a Field protected property to regulate access to those fields. Having a protected getField and a public fields doesn't really make sense.
In this case your colleague may be pointing at two things. The fact that you need to use Properties and not public fields and also the fact that it is probably better to pass the post into the method and not have the method access a property if you can. That way you don't have to set a property before calling the method. Think of it as a way of documenting what the method needs for it to operate. In this way another developer doesn't need to know which properties need to be set for the method to work. Everything the method needs should be passed in.
Regarding why we need properties in the first place? why shouldn't you use public fields. Isn't it more convenient? It sure is. The reason we use properties and not public fields is that just like most other concepts in OOP, you want your object to hide its details from the outside world and just project well defined interfaces of its state. Why? Ultimately to hide implementation details and keep internal change to ripple out(Encapsulation). Also, accessing properties has the added benefit of debugging. You can simply set a breakpoint in a property to see when a variable is changed or simply do a check if the variable is of certain value. Instead of littering your code with said check all over the place. There are many more goodies that come with this, returning readonly values, access control, etc.
To sum up, fields are though of as internal state. Properties(actual get/set methods) are though of as methods that interact with internal state. Having an outside object interact with interfaces is smiley face. Having outside class interact with internal state directly is frowny face.
This question already has answers here:
Are static local variables bad practice?
(2 answers)
Closed 8 years ago.
Consider the following functionally two code snippets in a single-threaded environment. Assuming there are no other methods in Foo I believe these are functionally identical.
Class Foo
Private _Bar As Bar
Public ReadOnly Property GetBar As Bar
Get
If IsNothing(_Bar) Then
_Bar = New Bar
End If
Return _Bar
End Get
End Property
End Class
And
Class Foo
Public ReadOnly Property GetBar2 As Bar
Get
Static _Bar As New Bar
Return _Bar
End Get
End Property
End Class
Today I was challenged on code following the 2nd method because "the New will be called each time". I already know that is false, but the primary objection was with regards to the use of Static. I found several references to Static variables indicating that they may be dangerous, but they were all talking about Java. However, I was not able to find any good explanations as to why.
How are these two methods different? Is the 2nd method dangerous? If so, why?
Static in VB.Net is not that same as static in Java, C#, C, or C++. VB.Net's analog to that construct is Shared. The documentation on the Static keyword is here:
http://msdn.microsoft.com/en-us/library/z2cty7t8.aspx
In particular, I'd like to point out this snippet:
Behavior
When you declare a static variable in a Shared procedure, only one copy of the static variable is available for the whole application. You call a Shared procedure by using the class name, not a variable that points to an instance of the class.
When you declare a static variable in a procedure that isn't Shared, only one copy of the variable is available for each instance of the class. You call a non-shared procedure by using a variable that points to a specific instance of the class.
It's likely the objection comes from believing that Static always behaves like the first paragraph, even in instance methods, when we can see here that it's clearly documented that this is not the case.
Instead, Static allows you to declare a variable whose lifetime-scope is that of the class instance, but whose access-scope is limited to a single method. It's a way to narrow the potential scope of a variable, and therefore is a good thing. Additionally, variables declared as Static are rewritten by the compiler to be protected via the Monitor class (at least for the Shared version), giving them a measure of thread-safety. In other words, a variable declared as Static is more likely to have any needed locking done verses a similar class-scoped variable.
In this particular case, though, I fail to see the point. You don't really gain anything beyond an auto-implemented property like this:
Public ReadOnly Property GetBar2 As New Bar()
This probably is confusing the VB.net concepts of Static and Shared because some languages use the keyword Static to mean what VB uses Shared for: a variable/field/property/method that is shared or common to all instances of a class.
But Static doesn't mean that in VB. Instead it means a routine-local variable that persists beyond the invocation of the routine (i.e., its lifetime is object-scoped rather than routine invocation-scoped).
REF: http://msdn.microsoft.com/en-us/library/z2cty7t8.aspx
So in VB, Static means "routine-scoped visibility, object-scoped lifetime".
Whereas Shared means "class-scoped visibilty, class/program-scoped lifetime".
I would avoid the second approach if for no other reason than the fact that C and C# have a static keyword whose meaning is totally different from that of the VB.NET Static keyword. I generally dislike language features which look like features of other languages but aren't. If it's necessary to use a language feature despite its unfortunate resemblance to the other language's feature, I'll use it, but the VB.NET static keyword doesn't really add much here. Effectively, it asks the compiler to make the variable Private field, give it an arbitrary name which differs from that of any other field, and replace all references to the variable's given name within the method with references to the invented name.
Conceptually, use of such "localized" fields may be regarded as dubious because while one may expect that a field will only need to be used within one method, that may turn out not to be true. I wouldn't worry too much about that issue in vb.net, however, because a Static variable may easily be turned into an ordinary private field if the need arises. If when that need does arise a field exists with the same name, one may easily rename the Static variable before moving it.
I am starting a new project from the ground up and want it to be clean / have good coding standards. In what order do the seasoned developers on here like to lay things out within a class?
A : 1) public methods 2) private methods 3) public vars 4) private vars
B : 1) public vars 2) private vars 3) public methods 4) private methods
C : 1) public vars 2) public methods 3) private methods 4)private vars
I generally like to put public static vars at the top, but then would a public static method be listed ahead of your constructor, or should the constructor always be listed first? That sort of thing...
I know it's finnicky but I just wondered: what are best practices for this?
PS: no I don't use Cc#. I know. I'm a luddite.
In Clean Code, Robert C. Martin advises coders to always put member variables at the top of the class (constants first, then private members) and methods should be ordered in such a way so that they read like a story that doesn't cause the reader to need to jump around the code too much. This is a more sensible way to organize code rather than by access modifier.
The best practice is to be consistent.
Personally, I prefer putting public methods first, followed by protected methods, following by private methods. Member data should in general always be private or protected, unless you have a good reason for it not to be so.
My rationale for putting public methods at the top is that it defines the interface for your class, so anyone perusing your header file should be able to see this information immediately.
In general, private and protected members are less important to most people looking at the header file, unless they are considering modifying the internals of the class. Keeping them "out of the way" ensures this information is maintained only on a need to know basis, one of the more important aspects of encapsulation.
Personally I like to have public at top, protected and then private. The reason for this is that when somebody cracks open the header he/she sees what he/she can access first, then more details as he/she scrolls down.
One should not have to look at the implementation details of a class in order to use it, then the class design is not done well.
I think I have a different philosophy on this than most. I prefer to group related items together. I can't stand having to jump around to work with a class. The code should flow and using a rather artificial ordering based on accessibility (public, private, protected etc. ) or instance versus static or member versus property versus function doesn't help keep a nice flow. So if I nave a public method Method that is implemented by private helper methods HelperMethodA, HelperMethodB etc. then rather than have these method far apart from each other in the file, I will keep them close to each other. Similarly, if i have an instance method that is implemented by a static method, I will group these together too.
So my classes often look like this:
class MyClass {
public string Method(int a) {
return HelperMethodA(a) + HelperMethodB(this.SomeStringMember);
}
string HelperMethodA(int a) { // returns some string }
string HelperMethodB(string s) { // returns some string }
public bool Equals(MyClass other) { return MyClass.Equals(this, other); }
public static bool Equals(MyClass left, MyClass right) { // return some bool }
public double SomeCalculation(double x, double y) {
if(x < 0) throw new ArgumentOutOfRangeException("x");
return DoSomeCalculation(x, y);
}
const double aConstant;
const double anotherConstant;
double DoSomeCalculation(double x, double y) {
return Math.Pow(aConstant, x) * Math.Sin(y)
+ this.SomeDoubleMember * anotherConstant;
}
}
This would be my ordering
Static Variables
Static Methods
Public Variables
Protected Variables
Private Variables
Constructors
Public Methods
Protected Methods
Private Methods
I use the following rules:
static before anything
variables before constructors before methods (i consider
constructors to be in the category of
methods)
public before protected before private
The idea is that you define the object (the data), before the behaviours (methods). Statics need to be separated because they aren't really part of the object, nor it's behaviour.
I used to care a lot. Over the last several years using modern IDEs pretty much everything is only 1 or 2 keystrokes away, I've let my standards relax substantially. Now, I start with statics, member variables, then constructors after that I don't worry about it much.
In C# I do let Resharper organize things automatically.
I generally agree with the public, protected, private order as well as the static data, member data, member functions order.
Though I sometimes group like members (getters & setters) I generally prefer listing members within a group ALPHABETICALLY so that they can be located more easily.
I also like lining up the data/functions vertically. I tab/space over to the right enough so that all names are aligned in the same column.
To each his own, and as Elzo says, modern IDEs have made it easier to find members and their modifiers in an easy way with colored icons in drop-down menus and such.
My take is that it is more important for the programmer to know what the class was designed for, and how it can be expected to behave.
So, if it is a Singleton, I put the semantics (static getInstance() class) first.
If it is a concrete factory, I put the getNew() function and the register / initialize functions first.
... and so on. When I say first, I mean soon after the c'tors and d'tor - since they are the default way of instantiating any class.
The functions that follow are then in:
logical call-order (e.g. initialize(), preProcess(), process(), postProcess() ), or
related functions together (like accessors, utilities, manipulators etc),
depending on if the class was meant primarily to be a data-store with some functions, or function provider with a few data members.
Some editors, like Eclipse and its offspring, allow you to reorder in the outline view the the vars and the methods, alphabetically or as in page.
The sequence of public followed by protected and private is more readable to me, It's better to describe the class logic in comments at top of the header file simply and function call orders to understand what a class dose and algorithms used inside.
I am using Qt c++ for a while and see some new sort of keywords like signal and slot I prefer to keep ordering like above and share my idea with you here.
#ifndef TEMPLATE_H
#define TEMPLATE_H
class ClassName
{
Q_OBJECT
Q_PROPERTY(qreal startValue READ startValue WRITE setStartValue)
Q_ENUMS(MyEnum)
public:
enum MyEnum {
Hello = 0x0,
World = 0x1
};
// constructors
explicit ClassName(QObject *parent = Q_NULLPTR);
~ClassName();
// getter and setters of member variables
// public functions (normal & virtual) -> orderby logic
public slots:
signals:
protected:
// protected functions it's rule followed like public functions
private slots:
private:
// methods
// members
};
#endif // TEMPLATE_H
I am working on a little pinball-game project for a hobby and am looking for a pattern to encapsulate constant variables.
I have a model, within which there are values which will be constant over the life of that model e.g. maximum speed/maximum gravity etc. Throughout the GUI and other areas these values are required in order to correctly validate input. Currently they are included either as references to a public static final, or just plain hard-coded. I'd like to encapsulate these "constant variables" in an object which can be injected into the model, and retrieved by the view/controller.
To clarify, the value of the "constant variables" may not necessarily be defined at compile-time, they could come from reading in a file; user input etc. What is known at compile time is which ones are needed. A way which may be easier to explain it is that whatever this encapsulation is, the values it provides are immutable.
I'm looking for a way to achieve this which:
has compile time type-safety (i.e. not mapping a string to variable at runtime)
avoids anything static (including enums, which can't be extended)
I know I could define an interface which has the methods such as:
public int getMaximumSpeed();
public int getMaximumGravity();
... and inject an instance of that into the model, and make it accessible in some way. However, this results in a lot of boilerplate code, which is pretty tedious to write/test etc (I am doing this for funsies :-)).
I am looking for a better way to do this, preferably something which has the benefits of being part of a shared vocabulary, as with design patterns.
Is there a better way to do this?
P.S. I've thought some more about this, and the best trade-off I could find would be to have something like:
public class Variables {
enum Variable {
MaxSpeed(100),
MaxGravity(10)
Variable(Object variableValue) {
// assign value to field, provide getter etc.
}
}
public Object getVariable(Variable v) { // look up enum and get member }
} // end of MyVariables
I could then do something like:
Model m = new Model(new Variables());
Advantages: the lookup of a variable is protected by having to be a member of the enum in order to compile, variables can be added with little extra code
Disadvantages: enums cannot be extended, brittleness (a recompile is needed to add a variable), variable values would have to be cast from Object (to Integer in this example), which again isn't type safe, though generics may be an option for that... somehow
Are you looking for the Singleton or, a variant, the Monostate? If not, how does that pattern fail your needs?
Of course, here's the mandatory disclaimer that Anything Global Is Evil.
UPDATE: I did some looking, because I've been having similar debates/issues. I stumbled across a list of "alternatives" to classic global/scope solutions. Thought I'd share.
Thanks for all the time spent by you guys trying to decipher what is a pretty weird question.
I think, in terms of design patterns, the closest that comes to what I'm describing is the factory pattern, where I have a factory of pseudo-constants. Technically it's not creating an instance each call, but rather always providing the same instance (in the sense of a Guice provider). But I can create several factories, which each can provide different psuedo-constants, and inject each into a different model, so the model's UI can validate input a lot more flexibly.
If anyone's interested I've came to the conclusion that an interface providing a method for each psuedo-constant is the way to go:
public interface IVariableProvider {
public int maxGravity();
public int maxSpeed();
// and everything else...
}
public class VariableProvider {
private final int maxGravity, maxSpeed...;
public VariableProvider(int maxGravity, int maxSpeed) {
// assign final fields
}
}
Then I can do:
Model firstModel = new Model(new VariableProvider(2, 10));
Model secondModel = new Model(new VariableProvider(10, 100));
I think as long as the interface doesn't provide a prohibitively large number of variable getters, it wins over some parameterised lookup (which will either be vulnerable at run-time, or will prohibit extension/polymorphism).
P.S. I realise some have been questioning what my problem is with static final values. I made the statement (with tongue in cheek) to a colleague that anything static is an inherently not object-oriented. So in my hobby I used that as the basis for a thought exercise where I try to remove anything static from the project (next I'll be trying to remove all 'if' statements ;-D). If I was on a deadline and I was satisfied public static final values wouldn't hamstring testing, I would have used them pretty quickly.
If you're just using java/IOC, why not just dependency-inject the values?
e.g. Spring inject the values via a map, specify the object as a singleton -
<property name="values">
<map>
<entry> <key><value>a1</value></key><value>b1</value></entry>
<entry> <key><value>a2</value></key><value>b3</value></entry>
</map>
</property>
your class is a singleton that holds an immutable copy of the map set in spring -
private Map<String, String> m;
public String getValue(String s)
{
return m.containsKey(s)?m.get(s):null;
}
public void setValues(Map m)
{
this.m=Collections.unmodifiableMap(m):
}
From what I can tell, you probably don't need to implement a pattern here -- you just need access to a set of constants, and it seems to me that's handled pretty well through the use of a publicly accessible static interface to them. Unless I'm missing something. :)
If you simply want to "objectify" the constants though, for some reason, than the Singleton pattern would probably be called for, if any; I know you mentioned in a comment that you don't mind creating multiple instances of this wrapper object, but in response I'd ask, then why even introduce the sort of confusion that could arise from having multiple instances at all? What practical benefit are you looking for that'd be satisfied with having the data in object form?
Now, if the values aren't constants, then that's different -- in that case, you probably do want a Singleton or Monostate. But if they really are constants, just wrap a set of enums or static constants in a class and be done! Keep-it-simple is as good a "pattern" as any.