[Bug c++/103668] New: constexpr std::vector not working as expected

[Bug c++/103668] New: constexpr std::vector not working as expected

[andrei.popa105 at yahoo dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=103668

            Bug ID: 103668
           Summary: constexpr std::vector not working as expected
           Product: gcc
           Version: 12.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c++
          Assignee: unassigned at gcc dot gnu.org
          Reporter: andrei.popa105 at yahoo dot com
  Target Milestone: ---

Hi,

Currently gcc supports constexpr std::vector and std::string. Suppose that you
want to write code like this:

#include <vector>

constexpr auto get_vector() {
        std::vector<int> vec{ 1, 2, 3, 4, 5, 6 };
        return vec;
}

constexpr auto get_vector_size() {
        constexpr auto vec = get_vector();
        return vec.size();
}

int main() {
        constexpr auto vec_size = get_vector_size();
        return 0;
}

Unfortunately, this code does not compile and the error is the following:

error: ‘get_vector()()’ is not a constant expression because it refers to a
result of ‘operator new’

But std::vector is not used in runtime world, the function get_vector is called
in a constexpr context and then returns the size of the vector. Am I missing
something or it is a bug?

I compiled this code with g++ file.cpp -std=c++23

[Bug c++/103668] constexpr std::vector not working as expected

[redi at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=103668

--- Comment #1 from Jonathan Wakely <redi at gcc dot gnu.org> ---
I don't think this is valid C++.

A vector created in a constexpr function must be destroyed there as well, it
cannot escape from the function.

[Bug c++/103668] constexpr std::vector not working as expected

[andrei.popa105 at yahoo dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=103668

--- Comment #2 from Andrei-Edward Popa <andrei.popa105 at yahoo dot com> ---
This code is valid in MSVC compiler, that's why I'm wondering about it.

[Bug c++/103668] constexpr std::vector not working as expected

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=103668

Jakub Jelinek <jakub at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |jakub at gcc dot gnu.org

--- Comment #3 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
(In reply to Jonathan Wakely from comment #1)
> I don't think this is valid C++.
> 
> A vector created in a constexpr function must be destroyed there as well, it
> cannot escape from the function.

Not a C++ lawyer, but I think returning it from constexpr function is just
fine,
what is wrong is if it isn't destructed in the same constant evaluation as the
construction.  So the bug in the testcase IMHO is the constexpr keyword before
auto vec = get_vector();, that means one constant evaluation is the computation
of the initializer of vec and that doesn't destruct what it constructed.
Now, if you remove that bogus constexpr keyword or replace it with const
keyword,
it works just fine, vec is constructed, vec.size() is computed and then vec is
destructed.

[Bug c++/103668] constexpr std::vector not working as expected

[andrei.popa105 at yahoo dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=103668

--- Comment #4 from Andrei-Edward Popa <andrei.popa105 at yahoo dot com> ---
Yes, I really missed this constexpr instead of const, this is clear for me.
Thank you! 

I think this thread can be closed now.

[Bug c++/103668] constexpr std::vector not working as expected

[redi at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=103668

Jonathan Wakely <redi at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |RESOLVED
         Resolution|---                         |INVALID

--- Comment #5 from Jonathan Wakely <redi at gcc dot gnu.org> ---
Ah yes, Jakub is right.