Re: [boost] [outcome] Change semantics on UB from peer review agreed semantics?

Subject: Re: [boost] [outcome] Change semantics on UB from peer review agreed semantics?
From: Gavin Lambert (boost_at_[hidden])
Date: 2018-09-13 00:11:54

Mere moments ago, quoth I:
> Perhaps boost::throw_error, which is expected to internally construct an 
> appropriate exception based on its argument type and then call 
> boost::throw_exception?

To clarify, I was wondering if we should:

  1. Implement an error-to-exception conversion trait as a generic 
template which does not compile by default.
  2. Provide a specialisation for boost::system::error_code which 
returns boost::system::system_error.
  3. (Where supported by the STL) Provide a specialisation for 
std::error_code which returns std::system_error.
  4. Allow users to provide additional specialisations for other error 
types.
  5. Provide generic boost::throw_error which calls 
boost::throw_exception on the result of the conversion trait.
  6. Make Outcome's default policy call boost::throw_error instead of 
hard UB.

The people who want hard UB can still use a policy that requests it.

Everyone else will get a compile error until they either implement the 
conversion trait, use an error_type which is already implemented, or 
switch to a UB policy.

Bonus: the conversion trait provides an easy way to ask metaprogramming 
questions such as "what is the exception type for this error code" or 
"is this a throwable error code type".