On Mon, Aug 17, 2015 at 12:36 PM, Andres Freund <andres@anarazel.de> wrote:
> On 2015-08-17 12:30:56 -0400, Robert Haas wrote:
>> - The possibility that may repeatedly break #define FRONTEND
>> compilation when we add static inline functions, where instead adding
>> macros would not have caused breakage, thus resulting in continual
>> tinkering with the header files.
>
> Again, that's really independent. Inlines have that problem, even with
> STATIC_IF_INLINE. C.f. MemoryContextSwitch() and a9baeb361d.
Inlines, yes, but macros don't.
I'm not saying we shouldn't do this, but I *am* saying that we need to
be prepared to treat breaking FRONTEND compilation as a problem, not
just today and tomorrow, but way off into the future. It's not at all
a stretch to think that we could still be hitting fallout from these
changes in 2-3 years time.
--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company