From: Andrew Dunstan <andrew.dunstan@2ndquadrant.com>
Subject: Re: pgsql: Prevent infinity and NaN in jsonb/plperl transform
To: Tom Lane <tgl@sss.pgh.pa.us>,
 Peter Eisentraut <peter.eisentraut@2ndquadrant.com>
Cc: Andrew Dunstan <andrew@dunslane.net>,
 pgsql-committers@lists.postgresql.org
References: <E1fDCX9-0005yu-7q@gemulon.postgresql.org>
 <21194.1525110966@sss.pgh.pa.us>
 <59e5787f-314e-6e5d-db5e-bcb6b9dbc25c@2ndquadrant.com>
 <23691.1525113718@sss.pgh.pa.us>
Message-ID: <213242a0-0ed8-ade6-822c-0270f7064799@2ndQuadrant.com>
Date: Mon, 30 Apr 2018 15:13:05 -0400
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101
 Thunderbird/52.4.0
MIME-Version: 1.0
In-Reply-To: <23691.1525113718@sss.pgh.pa.us>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
Content-Language: en-MW
Precedence: bulk


On 04/30/2018 02:41 PM, Tom Lane wrote:
> Peter Eisentraut <peter.eisentraut@2ndquadrant.com> writes:
>> On 4/30/18 13:56, Tom Lane wrote:
>>> So this isn't working on Windows: ...
>>> The reason seems to be this kluge in plperl.h:
>>>
>>> /* stop perl headers from hijacking stdio and other stuff on Windows */
>>> #ifdef WIN32
>>> #define WIN32IO_IS_STDIO
>>> /*
>>> * isnan is defined in both the perl and mingw headers. We don't use it,
>>> * so this just clears up the compile warning.
>>> */
>>> #ifdef isnan
>>> #undef isnan
>>> #endif
>>> #endif							/* WIN32 */
>>>
>>> Looks like the half-life of that hack just expired.  What shall
>>> we do about it?
>> I have removed this for now.  If it's really just about a compiler
>> warning, then we can find a different workaround later.
> I was just about to propose exactly that fix.  This undef'ing of isnan
> traces back to Andrew's 2006 commit ea73a78b0.  I think perhaps that's not
> necessary with any still-supported version of Perl; I see no evidence on
> my machines that any Perl header #defines isnan.  There is this bit:
>
> #ifdef UNDER_CE
> int isnan(double d);
> #endif
>
> which would cause trouble if isnan is a macro, but we can hope that
> that doesn't apply to us.
>
> 			


Fingers crossed we'll be OK. Looking at the perl headers on a Windows
machine I'm doing some other stuff on, the #define is protected by an
#ifndef. I'm always happy when my hacks can be removed :-)

cheers

andrew

-- 
Andrew Dunstan                https://www.2ndQuadrant.com
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services