Class Versus Struct

Question

In C++ and other influenced languages there is a construct called Structure (struct), and another called the class. Both are capable of holding functions and variables. Some differences are:

1. Class is given memory in the heap and struct is given memory in the stack (remark: this is wrong for C++, but maybe correct in what the OP called "influenced languages")
2. Class variable are private by default and in struct they are public

My question is: was the struct somehow abandoned for Class? If so, why? Other than the differences above, a struct can do all the same things that a class does. So why abandon it?

• prometheuspk: by abandon I meant why one is used over the other.
• Umair A.: your question answers this itself I think.
• Job: The difference between classes and structs is language-dependent. Some lessons from C++ do not really apply to C#.
• jojo: Uhhhh. In C++ you can allocate objects on the stack or heap. Wherever you want.
• Aaronaught: Wrong, wrong, wrong. Stack vs. heap has nothing to do with the difference.

Answer 1

It is not abandoned at all. In fact, even modern languages like C# which make heavy use of class still offer you struct. As for when it's useful to choose one over the other, I refer you to this article:

Choosing Between Classes and Structures

Quoted from the MSDN article:

Consider defining a structure instead of a class if instances of the type are small and commonly short-lived or are commonly embedded in other objects.

Do not define a structure unless the type has all of the following characteristics:

• It logically represents a single value, similar to primitive types (integer, double, and so on).
• It has an instance size smaller than 16 bytes.
• It is immutable.
• It will not have to be boxed frequently.

• Klaim: This answer is really misguided as .Net/C# struct/class meaning is different than C++ one: they use the same name but have different semantic depending on the language!!! In C++ there is almost no difference between struct and class and the question is totally wrong on the first point, which is right in c# and D but not in c++. So the link to this article, which is about .Net, is really really wrong. It's not C++!
• jk.: @Klaim technically its not right for C# either (structs can be on the heap) The important distinction in C# is between value semantics and reference semantics, a distinction that C++ doesn't need as it can have values or references for anything

Answer 2

You are mistaken about C++: the only significant difference between class and struct is the default access specifier difference. Struct and class are for all intents and purposes synonyms, I believe struct is kept around for backwards compatibility to C.

• quant_dev: Not only. Struct is a good choice when you just want to dump a bunch of data in one object and don't add any behaviour to it.
• Aaronaught: @quant: That's a terrible reason to use a struct. I really hope that comment was tongue-in-cheek.
• quant_dev: @Aaronaught Why? My view on the usage of struct is fairly standard, see this answer for example: stackoverflow.com/questions/54585/…
• quant_dev: @Aaronaught How do you return multiple values then? Say, a procing model returns the clean price, dirty price, and sum of accrued payments.
• Joris Timmermans: @quant_dev - you can decide to attach some meta-meaning to a struct for your purposes (and for former C programmers one that actually does make a lot of sense), but that does not change the fact that the only difference between the struct and class keywords in C++ is the default access.

Answer 3

the language D has created a greater distinction between class and struct

a struct in d is nothing but a stack allocated data record with some functions you can call on it (there is no option for inheritance unless you use a enum + union setup i.e. implement the polymorfism yourself) that is passed by value

a class is like we are used to: virtual functions, heap allocation, (single) inheritance, passed by ref