When to make constructor explicit in C++

After reading the following blog :

http://xania.org/200711/ambiguous-overloading

I started asking myself "should I not always explicit define my constructors?"

So I started reading more than found out this article :

http://www.sjbrown.co.uk/2004/05/01/always-use-explicit/

Which shows another example, and also explains his thoughts behind it. But of course this is one blogger's thoughts.

I would be happy to hear from some of you,what your thought on the manner, what is your experience with the subject and a few example for either way would be nice.

The traditional wisdom is that constructors taking one parameter (explicitly or effectively through the use of default parameters) should be marked explicit, unless they do define a conversion (std::string being convertible from const char* being one example of the latter). You've figured out the reasons yourself, in that implicit conversions can indeed make life harder than it has to be.

A perhaps obvious exception to that would be the copy constructor. Or perhaps another way is to consider that most types are convertible from and to themselves, and that as such the copy constructor is not marked explicit most of the time.

While it may appear that marking all other kinds of constructors explicit does not hurt, I'd argue against it. Because while explicit has no effect on a constructor taking multiple arguments in C++03, it does have an effect in C++11. To put it into code:

struct foo {
    explicit foo(int i);
    foo(int i, int j);
    explicit foo(int i, int j, int k);
};

foo make_foo()
{
    /* Not C++11-specific: */
    // Error: no conversion from int to foo
    return 42;

    // Okay: construction, not conversion
    return foo(42);

    // Okay: constructions
    return foo(42, 42);
    return foo(42, 42, 42);

    /* C++11 specific: */
    // Error: no conversion from int to foo
    return { 42 };

    // Not an error, not a conversion
    return { 42, 42 };

    // Error! Constructor is explicit
    return { 42, 42, 42 };
    // Not an error, direct-initialization syntax
    return foo { 42, 42, 42 };
}

I personally find it needlessly verbose that in a function that returns foo I have to explicitly return foo { 42, 42, 42 }. I don't see what explicit is protecting me from. I really want the { initializers... } syntax to mean 'construct object from given initializers', and explicit gets into the way of that while saving me from nothing. (Since { i } does boil down to i in the context of copy-initialization -- most of the time -- I'll gladly give that one up.)

So I'd say get into the habit of using explicit for unary constructors, and those only.