Hacker News Reader

Comment by rossmohax | Hacker News Reader

If not exception, then how to fail constructor?

Don't write constructors that can fail unless it is failure that would be appropriate to crash for. That works out a lot better than it might naively sound. It is hard for people to reason about the possibility of constructors/destructors failing, so actually rather nice to just forbid it.

And for those super-rare cases where it's appropriate to crash / fail hard in a constructor but you still have requirements about reporting or even recovery... setjmp/longjmp still exist.

You have always had the choice available to do something overwhelmingly worse than throwing. That is not a reason to do it.

Signal failure via an out parameter reference. For destructors store the reference in the object and call an error method on errors.

I think that would be problematical in those cases where the special constructors are called, and complicate using constructors in expressions and argument lists.

That is the worst of all possible choices.

Just throw.

The “Herbceptions” proposal addressed this. http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2018/p070...

you can still set a static flag or something in ctor then check it, but you can't do anything about dtor indeed if you need throw there.

Except you are not supposed to throw in a destructor.

https://www.cs.technion.ac.il/users/yechiel/c++-faq/dtors-sh... but what if dtor fails(e.g. can't release a resource), is the by-default action terminate or ignore?

Well, the "environment" may decide for you in some cases, but there is no default action as such. You log an error and either terminate or ignore - depending on which causes less harm.

It has been many years since I used C++

But returning null is an obvious choice.

Combined types which return the structure or an error are another. Has the advantage of returning error information, why failure.

There are many ways

You can’t return null from a constructor (constructors have no return)

What dumb design.

I thought the constructor returned a pointer to the memory created.

No return? Stupid!

FWIW, constructors do not "create" (I assume you mean allocate) memory. That's the job of operator new. A constructor, given a block of untyped memory, will construct an object in it.

Right.

Been a long time.

I would still expect a constructor to return. How can you tell if it failed?

> But returning null is an obvious choice.

It's not because C++ is not an everything-is-a-reference language.

> Combined types which return the structure or an error are another. Has the advantage of returning error information, why failure.

Sum types are great, but changing constructors to return sum types would break all existing code. It would also lead to a whole slew of questions about how the construction of arrays of values would work. (Does an array of an objects now turn into an array of wrapper types? Byebye SIMD optimization!) It would also break the consistency of constructing POD vs non POD types, which would make writing templates that need to generalize over both a huge PITA.

Reply on news.ycombinator.com