Re: Should contrib modules install .h files?

>>>>> "Tom" == Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> writes:

Tom> Something that copes with different modules installing headers
Tom> with the same base name. Allowing for that was the driving force
Tom> for going with subdirectory-per-extension, but if we really want
Tom> that to work, there seems to be no alternative but for extensions
Tom> to write qualified header names (#include "hstore/hstore.h" not
Tom> #include "hstore.h"). Andres, for one, seemed to think that
Tom> wouldn't play nicely with PGXS,

I think that was me, not Andres?

But I think I was partially wrong and that it's possible that this can
be made to work at least in most cases, as long as we can rely on the
same-directory rule for #include "foo.h". (i.e. the first place to look
is always the same directory as the file containing the #include
statement).

I'm going to test this now, trying to do an out-of-both-trees build
of a transform function for an out-of-tree PL that uses multiple .h
files.

--
Andrew (irc:RhodiumToad)