Re: [boost] [config] Rethinking feature macros?

Subject: Re: [boost] [config] Rethinking feature macros?
From: Andrey Semashev (andrey.semashev_at_[hidden])
Date: 2017-11-06 18:22:02

On 11/06/17 20:59, Peter Dimov via Boost wrote:
> Andrey Semashev wrote:
>> > 3. Even if a compiler is out of wide use, we still maintain Config 
>> for > it.
>> >
>> > 
>> https://github.com/boostorg/config/blob/develop/include/boost/config/compiler/digitalmars.hpp 
>>
>>
>> Sure, as long as we choose to. And we would still have to maintain and 
>> test those files with positive macros.
> 
> I feel like we're talking past of each other here. What is there to 
> maintain about digitalmars.hpp? It stays the same for eternity because 
> the compiler is no longer developed so it's impossible for it to acquire 
> a new feature.

I'm not aware of the Digital Mars compiler activity, but if it's long 
abandoned then we might drop that file entirely. Although I can see some 
C++14 and 17 additions in it.

Anyway, if we keep supporting it, adding a #define for each new 
Boost.Config macro in that file doesn't seem that much of a problem. 
Naming confusion is worse for users and mass switching to positive 
naming is more expensive.

Dead compilers are not the problem. The most time consuming part is 
related to the compilers that are still alive. It is those compilers, 
especially the ones that are hard to come by, require research about 
whether they support a feature and since which version. For those 
compilers you have to do the same amount of work to define the macro 
either way.