I wrote:
> Albe Laurenz <laurenz.albe@wien.gv.at> writes:
> Anyway, I have prepared a patch along the lines you suggest.
> Pushed, we'll see if the buildfarm likes this iteration any better.
And the answer is "not very much". The Windows builds aren't actually
failing, but they are producing lots of warnings:
lquery_op.obj : warning LNK4197: export '_ltq_regex' specified multiple times; using first specification
lquery_op.obj : warning LNK4197: export '_ltq_rregex' specified multiple times; using first specification
lquery_op.obj : warning LNK4197: export '_lt_q_regex' specified multiple times; using first specification
lquery_op.obj : warning LNK4197: export '_lt_q_rregex' specified multiple times; using first specification
ltree_gist.obj : warning LNK4197: export '_ltree_compress' specified multiple times; using first specification
ltree_gist.obj : warning LNK4197: export '_ltree_same' specified multiple times; using first specification
ltree_gist.obj : warning LNK4197: export '_ltree_union' specified multiple times; using first specification
ltree_gist.obj : warning LNK4197: export '_ltree_penalty' specified multiple times; using first specification
ltree_gist.obj : warning LNK4197: export '_ltree_picksplit' specified multiple times; using first specification
ltree_gist.obj : warning LNK4197: export '_ltree_consistent' specified multiple times; using first specification
ltree_op.obj : warning LNK4197: export '_ltree_isparent' specified multiple times; using first specification
ltree_op.obj : warning LNK4197: export '_ltree_risparent' specified multiple times; using first specification
ltree_op.obj : warning LNK4197: export '_lca' specified multiple times; using first specification
ltxtquery_op.obj : warning LNK4197: export '_ltxtq_exec' specified multiple times; using first specification
ltxtquery_op.obj : warning LNK4197: export '_ltxtq_rexec' specified multiple times; using first specification
This is evidently from the places where there are two "extern"
declarations for a function, one in a header and one in
PG_FUNCTION_INFO_V1. The externs are identical now, but nonetheless
MSVC insists on whining about it.
I'm inclined to give this up as a bad job and go back to the
previous state. We have a solution that works and doesn't
produce warnings; third-party authors who don't want to use it
are on their own.
regards, tom lane