On 2024/09/21 12:09, Tom Lane wrote:
> Fujii Masao <masao.fujii@oss.nttdata.com> writes:
>> I don’t have any objections to this commit, but I’d like to confirm
>> whether we really want to proactively reorder #include directives,
>> even for standard C library headers.
>
> I'm hesitant to do that. We can afford to insist that our own header
> files be inclusion-order-independent, because we have the ability to
> fix any problems that might arise. We have no ability to do something
> about it if the system headers on $random_platform have inclusion
> order dependencies. (In fact, I'm fairly sure there are already
> places in plperl and plpython where we know we have to be careful
> about inclusion order around those languages' headers.)
>
> So I would tread pretty carefully around making changes of this
> type, especially in long-established code. I have no reason to
> think that the committed patch will cause any problems, but
> I do think it's mostly asking for trouble.
Sounds reasonable to me.
Regards,
--
Fujii Masao
Advanced Computing Technology Center
Research and Development Headquarters
NTT DATA CORPORATION
