Re: [PATCH 1/5] add SWAP macro

["René Scharfe" <l.s.r@web.de>]

Am 01.02.2017 um 19:33 schrieb Junio C Hamano:
> René Scharfe <l.s.r@web.de> writes:
> 
>> Size checks could be added to SIMPLE_SWAP as well.
> 
> Between SWAP() and SIMPLE_SWAP() I am undecided.
> 
> If the compiler can infer the type and the size of the two
> "locations" given to the macro, there is no technical reason to
> require the caller to specify the type as an extra argument, so
> SIMPLE_SWAP() may not necessarily an improvement over SWAP() from
> that point of view.  If the redundancy is used as a sanity check,
> I'd be in favor of SIMPLE_SWAP(), though.
> 
> If the goal of SIMPLE_SWAP(), on the other hand, were to support the
> "only swap char with int for small value" example earlier in the
> thread, it means you cannot sanity check the type of things being
> swapped in the macro, and the readers of the code need to know about
> the details of what variables are being swapped.  It looks to me
> that it defeats the main benefit of using a macro.

Full type inference could be done with C11's _Generic for basic types,
while typeof would be needed for complex ones, I guess.  Checking that
sizes match is better than nothing and portable to ancient platforms,
though.  Having an explicit type given is portable and easy to use for
checks, of course, e.g. like this:

#define SIMPLE_SWAP(T, a, b) do { \
	T swap_tmp_ = a + BUILD_ASSERT_OR_ZERO(sizeof(T) == sizeof(a)); \
	a = b + BUILD_ASSERT_OR_ZERO(sizeof(a) == sizeof(b)); \
	b = swap_tmp_; \
} while(0)

It doesn't support expressions with side effects, but that's probably
not much of a concern.

Swapping between different types would then still have to be done
manually, but I wonder how common that is -- couldn't find such a case
in our tree.

René