Re: make installcheck is broken in HEAD on mingw

Itagaki Takahiro <itagaki(dot)takahiro(at)oss(dot)ntt(dot)co(dot)jp> writes:
> WARNING: getopt(p) = ostgres
> FATAL: attempted change of parameter "port" ignored

> The first argument 'postgres' was interpreted as "-p ostgres" by getopt().

Ugh.

> We might need to re-initilaize variables for getopt() because we call
> process_postgres_switches twice(). So getopt() is also called twice.

Yeah. process_postgres_switches attempts to do that, using code that
is the same as in postmaster.c. However I suppose that in mingw the
intervening fork/exec might manage to reset getopt() in some other way.

> (Are there any known bugs in mingw's getopt?)

No idea, but maybe you are running a different mingw version than the
buildfarm machines are using?

> BTW, the following code seems to be a bit storange.
> The part of { argv++; argc--; } removes the first arugment,
> but '--single' argument (argv[1]) is not removed, no?

This is just the same as what was there for the last year or so.
I only moved it to a different place ...

regards, tom lane