Re: Needless additional partition check in INSERT?

On Fri, May 11, 2018 at 06:12:38PM +1200, David Rowley wrote:
> On 11 May 2018 at 17:48, Amit Langote <Langote_Amit_f8@lab.ntt.co.jp> wrote:
>> By the way,
>>
>> +            !resultRelInfo->ri_PartitionRoot)
>>
>> This should be resultRelInfo->ri_PartitionRoot == NULL, because the above
>> gives an impression that ri_PartitionRoot is a Boolean.
> 
> If this is some new coding rule, then that's the first I've heard of it.
> 
> Scanning over the result of git grep -E "if \(!\w{1,}\)" it looks like
> we have a bit of cleanup work to do before we can comply.
> 
> FWIW, I've previously been told off for the opposite.

NULL maps to 0 so that's not really worth worrying.  A lot of code paths
use one way or the other for pointers.  That's really up to the patch
author at the end (I prefer matching with NULL, but usually it is better
to comply with the surroundings for consistency).
--
Michael