On 2017/07/04 18:15, Amit Langote wrote:
> On 2017/07/04 17:55, Etsuro Fujita wrote:
>> This comment in an error handling in ExecPartitionCheck():
>>
>> if (!ExecCheck(resultRelInfo->ri_PartitionCheckExpr, econtext))
>> {
>> char *val_desc;
>> Relation orig_rel = rel;
>>
>> /* See the comment above. */
>> if (resultRelInfo->ri_PartitionRoot)
>>
>> should be updated because we don't have any comment on that above in the
>> code. Since we have a comment on that in ExecConstraints() defined just
>> below that function, I think the comment should be something like this:
>> "See the comment in ExecConstraints().". Attached is a patch for that.
>
> Thanks for fixing that. I forgot to do the same in the patch that got
> committed as 15ce775faa428 [1], which moved that code block from
> ExecConstraints() to ExecPartitionCheck().
Thanks for the explanation!
In relation to this, I found odd behavior in the error handling. Since
that is another topic, I'll start a new thread.
Best regards,
Etsuro Fujita