Re: msvc-inval: more options

[Jim Meyering]

Bruno Haible wrote:
> Jim Meyering wrote:
>> > -#if MSVC_INVALID_PARAMETER_HANDLING == DEFAULT_HANDLING
>> > +#if HAVE_MSVC_INVALID_PARAMETER_HANDLER \
>> > +    && MSVC_INVALID_PARAMETER_HANDLING == DEFAULT_HANDLING
>> >    gl_msvc_inval_ensure_handler ();
>> >  #endif
>>
>> That is a detail specific to MSVC.  Why impose the readability and
>> maintenance hassle of requiring such an expression on all callers?
>
> It's not for all callers. It's only for specific tests. I could introduce
> a macro
>   
> MSVC_INIT_INVALID_PARAMETER_HANDLER_FOR_TESTS_WITH_FUNCTIONS_WITHOUT_GNULIB_WORKAROUND()
> but that would be so special-purpose that I really wonder whether it's worth
> adding to msvc-inval.h.

I see now.
If there will only ever be these four guards, then I agree
that it's not worth the trouble.

Thanks for explaining.