On Sat, Jan 25, 2014 at 04:56:37PM -0500, Tom Lane wrote:
> Bruce Momjian <bruce@momjian.us> writes:
> > On Sat, Jan 25, 2014 at 10:40:28PM +0100, Andres Freund wrote:
> >> I think this style of pinhole copy editing is pretty pointless. There's
> >> dozen checks just like this around. If somebody wants to change the rules
> >> or improve comment it takes more than picking a random one.
>
> > OK, change made.
>
> FWIW, I don't find that an improvement either. As Andres says, this
> is just applying the same rule that's used in many other places, ie
> return null if the requested attnum is off the end of the tuple.
OK, I can revert it, but I don't see any other cases of the string
'return NULL if' in the executor code. What the code really is doing is
"Assume NULL so return true if". The code was never returning NULL, it
was assuming the attribute was NULL and returning true. Am I missing
something?
-- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB
http://enterprisedb.com
+ Everyone has their own god. +
