Why some object names starts with single or multiple underscores in c++?

Question

I've seen a lot of times in c++ libraries tones of objects whose names start with underscores. I know some cases when this is useful for example:

    class Person {
    private: 
       unsigned int age;
       //...
    public:
       Person(unsigned int _age): age(_age) {}
    };

here it's useful to distinguish logically same objects _age, age.

But there are also cases when global functions names (not from a class) also have single or multiple underscores. So I would like to understand this in general and more ideologically.

So:

What are the benefits for this kind of naming ?

When it's need to use single and when multiple underscores at the start ?

Double underscores are usually reserved for compiler's internal use as a convention. Single underscores are reserved for library functions to prevent collisions. — Parag Agarwal, Jun 10 '16 at 12:07
In this specific case, `Person(unsigned int age):age(age){}` would be legal as well (and produce the same code). — IInspectable, Jun 10 '16 at 12:11
@ParagAgarwal - double underscores are **always** reserved, not sometimes. — Pete Becker, Jun 10 '16 at 12:24

score 0 · Answer 1 · answered Jun 10 '16 at 12:05

0

There's never a requirement to use an initial single or a double-underline prefix to variable names.

This is simply someone's naming convention, or a habit.

answered Jun 10 '16 at 12:05

Sam Varshavchik

114,536
5
94
148

5

But there is a requirement not to use them per the standard. – NathanOliver Jun 10 '16 at 12:06
Only certain types; namely an underscore followed by an uppercase letter. – Sam Varshavchik Jun 10 '16 at 12:06
1

@SamVarshavchik: Also double underscores. – Kerrek SB Jun 10 '16 at 12:06
And all double underscores and underscores in the global namespace. – NathanOliver Jun 10 '16 at 12:07

Why some object names starts with single or multiple underscores in c++?

1 Answers1