Re: Report error position in partition bound check

On Fri, 10 Jul 2020 at 23:31, Alexandra Wang <alexandra(dot)wanglei(at)gmail(dot)com>
wrote:

>
>
> Thank you Daniel. Here's the rebased patch. I also squashed the two
> patches into one so it's easier to review.
>
> Thanks for rebasing patch. It applies cleanly still. Here are some comments
@@ -3320,7 +3338,9 @@ make_one_partition_rbound(PartitionKey key, int
index, List *datums, bool lower)
* partition_rbound_cmp
*
* Return for two range bounds whether the 1st one (specified in datums1,

I think it's better to reword it as. "For two range bounds decide whether
...

- * kind1, and lower1) is <, =, or > the bound specified in *b2.
+ * kind1, and lower1) is <, =, or > the bound specified in *b2. 0 is
returned if
+ * equal and the 1-based index of the first mismatching bound if unequal;
+ * multiplied by -1 if the 1st bound is smaller.

This sentence makes sense after the above correction. I liked this change,
requires very small changes in other parts.

/*
@@ -3495,7 +3518,7 @@ static int
partition_range_bsearch(int partnatts, FmgrInfo *partsupfunc,
Oid *partcollation,
PartitionBoundInfo boundinfo,
- PartitionRangeBound *probe, bool *is_equal)
+ PartitionRangeBound *probe, bool *is_equal, int32
*cmpval)

Please update the prologue explaining the new argument.

After this change, the patch will be ready for a committer.
--
Best Wishes,
Ashutosh