Thread: pgsql: Make float4byval and float8byval configurable on MSVC.

pgsql: Make float4byval and float8byval configurable on MSVC.

From
mha@postgresql.org (Magnus Hagander)
Date:
Log Message:
-----------
Make float4byval and float8byval configurable on MSVC.

Modified Files:
--------------
    pgsql/src/tools/msvc:
        Solution.pm (r1.38 -> r1.39)
        (http://anoncvs.postgresql.org/cvsweb.cgi/pgsql/src/tools/msvc/Solution.pm?r1=1.38&r2=1.39)
        config.pl (r1.11 -> r1.12)
        (http://anoncvs.postgresql.org/cvsweb.cgi/pgsql/src/tools/msvc/config.pl?r1=1.11&r2=1.12)

Re: pgsql: Make float4byval and float8byval configurable on MSVC.

From
Andrew Dunstan
Date:

Magnus Hagander wrote:
> Log Message:
> -----------
> Make float4byval and float8byval configurable on MSVC.
>
> Modified Files:
> --------------
>     pgsql/src/tools/msvc:
>         Solution.pm (r1.38 -> r1.39)
>         (http://anoncvs.postgresql.org/cvsweb.cgi/pgsql/src/tools/msvc/Solution.pm?r1=1.38&r2=1.39)
>         config.pl (r1.11 -> r1.12)
>         (http://anoncvs.postgresql.org/cvsweb.cgi/pgsql/src/tools/msvc/config.pl?r1=1.11&r2=1.12)
>

Should we now remove the entries for these 2 defines in pg_config.h.win32 ?

cheers

andrew

Re: pgsql: Make float4byval and float8byval configurable on MSVC.

From
Tom Lane
Date:
Andrew Dunstan <andrew@dunslane.net> writes:
> Magnus Hagander wrote:
>> Make float4byval and float8byval configurable on MSVC.

> Should we now remove the entries for these 2 defines in pg_config.h.win32 ?

It looked to me like the patch-as-applied relied on their being there.
I thought it was a bit ugly though.  +1 for reverting my change to
pg_config.h.win32 and making the script insert the correct #defines in
all four cases.

            regards, tom lane

Re: pgsql: Make float4byval and float8byval configurable on MSVC.

From
Magnus Hagander
Date:
Tom Lane wrote:
> Andrew Dunstan <andrew@dunslane.net> writes:
> > Magnus Hagander wrote:
> >> Make float4byval and float8byval configurable on MSVC.
>
> > Should we now remove the entries for these 2 defines in
> > pg_config.h.win32 ?
>
> It looked to me like the patch-as-applied relied on their being there.
> I thought it was a bit ugly though.  +1 for reverting my change to
> pg_config.h.win32 and making the script insert the correct #defines in
> all four cases.

It does rely on it, because I thought it was needed by the mingw port.
And yes, I realise that makes no sense at all, now that I think about
it :-) Will fix.

//Magnus