pwned VS cppfront

modern C++ is fine, as long as you don't try to get too clever with return types, move semantics, constexpr and template parameter expansions.

They all have their places, but if your code is littered with them, it feels like every line is a puzzle.

Here's an attempt to wrap sqlite in a typesafe manner using template expansion. All very simple C++11. https://github.com/nurettin/pwned/blob/master/sqlsafe/sqlsaf...

(some types had to be spelled out because the compilers weren't ready)

Here's some more C++11 without using any of the mentioned complexities https://github.com/nurettin/pwned/blob/master/server/server....

This shows that simply spelling out your types and not getting crazy with polymorphic metaprogramming makes code much more readable. Ideally, code should just be a bunch of ifs and loops.

CPP2/cppfront:

https://github.com/hsutter/cppfront

I hope we see this in C++26 as optional mode i.e. #safe and #unsafe and same for #impdef or so.

C++23 is done. But C++ is not! In this talk, the author shares his personal perspectives on an ongoing and very active evolution of C++, updates on his cppfront experimental compiler, and why compatibility is essential to the further success of the C++ development.

https://github.com/hsutter/cppfront

But his side project at Microsoft didn't gain traction with gcc, clang, etc and everybody else in the industry. So at this point, the C++ committee will be perceived as "so far behind" ... because there's nothing for them to vote on.

C++ has major flaws that cannot be rectified without serious breaking changes. With that said, Herb has been experimenting with a new cpp frontend with sane defaults [1].

In my opinion, the world is on standby until Anders Hejlsberg feels like tackling a modern, next generation systems language.

[1] https://github.com/hsutter/cppfront