Re: BUG #12869: PostGIS 2.2 can't compile against 9.5 dev branch

> According to the C and POSIX standards, DBL_DIG is defined by <float.h>.
> It sounds like you were indirectly depending on some Postgres header to
> #include that, and said header no longer does.  I'm not sure what bit
> of refactoring might've had such an effect, but it doesn't much matter
> --- you really ought to #include <float.h> for yourself.

Tom,

Thanks for the response.  We were able to fix that issue by adding a
<float.h> that was missing in one of our files.
That unfortunately unearthed another failure when it got to regression
testing which I have logged here:  http://trac.osgeo.org/postgis/ticket/3080

It seems that under some conditions functions are being called multiple
times causing additional NOTICES to be emitted.  This is causing a failure
in one of our raster tests.

In  prior versions they were emitted only once (I assume they were only
called once).  I've come up with a simple example that doesn't involve
PostGIS to replicate the issue.


CREATE OR REPLACE FUNCTION dummy_notice(variadic integer[]) RETURNS
integer[] AS
$$
BEGIN
    RAISE NOTICE 'This is a dummy notice';
    RETURN $1;
END;
$$
language plpgsql IMMUTABLE STRICT;


SELECT  v[1] As v1, v[2] As v2, v[3] As v3
FROM
    (SELECT dummy_notice(1,2,3) As v) As t;


In prior versions of PostgreSQL, this simple test would just emit:

NOTICE:  This is a dummy notice


In PostgreSQL 9.5 as of when I had reported the original issue, it emits for
every single column in the outter output.
NOTICE:  This is a dummy notice
NOTICE:  This is a dummy notice
NOTICE:  This is a dummy notice

 I presume that means the dummy_notice function is now being called 3 times
for the above.

This behavior started when I reported the float issue and  is still doing
that with latest PostgreSQL 9.5 snapshot.

I'm not sure if this is an intended change or a side-effect of something
else that needs checking.


Thanks,
Regina