Re: [boost] [review][variant2] Variant2 Review Starts April 1

Subject: Re: [boost] [review][variant2] Variant2 Review Starts April 1
From: degski (degski_at_[hidden])
Date: 2019-03-26 11:38:29

On Tue, 26 Mar 2019 at 12:18, dariomt--- via Boost <boost_at_[hidden]>
wrote:

> Well, Boost.Variant has been around so much more time than std::variant, so
> this seems obvious to me.

This is bound to be the case as Boost pretended to be the breeding ground
for the STL. This seems to have changed slightly nowadays. Things are
included in the STL, by-passing Boost and conversely Boost is now on its
way to re-do functionality that's already standardized.

Also, I'd say the cost of migrating from Boost.Variant to std::variant is
> non-zero, and the benefits are not clear (to me), so I'd expect not many
> people to undergo this migration.
>

One advantage is that every-one using a C++17 compiler has a STL that
includes an implementation (no downloading, building, maintaining etc.).

Well, in my (limited) experience Boost is ubiquitous, so being in std is
> not that much of a head start.
>

Boost itself is built upon the (a) STL, so surely the STL is by definition
more ubiquitous than Boost. C++17-STL probably isn't, I'd agree with that,
but this is just a temporary situation (and is changing rapidly in my
opinion).

I wonder if code dealing explicitly with valueless_by_exception() will
> break if you make this change.
>

Picking a whatever::variant will/has become rocket-science [on top of
everything else that modern C++ throws at you].

degski
-- 
*Microsoft, please kill Paint3D*