Re: HashAgg's batching counter starts at 0, but Hash's starts at 1. (now: incremental sort)

On Wed, Jul 29, 2020 at 09:18:44PM -0700, Peter Geoghegan wrote:
> On Wed, Jul 29, 2020 at 9:05 PM Justin Pryzby <pryzby(at)telsasoft(dot)com> wrote:
> > So my 2ndary suggestion is to conditionalize based on the method rather than
> > value of space used.
>
> What's the advantage of doing it that way?

Because filtering out zero values is exactly what's intended to be avoided for
nontext output.

I think checking whether the method was used should result in the same output,
without the literal check for zero value (which itself sets a bad example).

--- a/src/backend/commands/explain.c
+++ b/src/backend/commands/explain.c
@@ -2824,13 +2824,13 @@ show_incremental_sort_group_info(IncrementalSortGroupInfo *groupInfo,
appendStringInfo(&groupName, "%s Groups", groupLabel);
ExplainOpenGroup("Incremental Sort Groups", groupName.data, true, es);
ExplainPropertyInteger("Group Count", NULL, groupInfo->groupCount, es);

ExplainPropertyList("Sort Methods Used", methodNames, es);

- if (groupInfo->maxMemorySpaceUsed > 0)
+ if (groupInfo->sortMethods & SORT_TYPE_QUICKSORT)
{
long avgSpace = groupInfo->totalMemorySpaceUsed / groupInfo->groupCount;
const char *spaceTypeName;
StringInfoData memoryName;

spaceTypeName = tuplesort_space_type_name(SORT_SPACE_TYPE_MEMORY);
@@ -2841,13 +2841,13 @@ show_incremental_sort_group_info(IncrementalSortGroupInfo *groupInfo,
ExplainPropertyInteger("Average Sort Space Used", "kB", avgSpace, es);
ExplainPropertyInteger("Peak Sort Space Used", "kB",
groupInfo->maxMemorySpaceUsed, es);

ExplainCloseGroup("Sort Spaces", memoryName.data, true, es);
}
- if (groupInfo->maxDiskSpaceUsed > 0)
+ if (groupInfo->sortMethods & SORT_TYPE_EXTERNAL_SORT)
{
long avgSpace = groupInfo->totalDiskSpaceUsed / groupInfo->groupCount;
const char *spaceTypeName;
StringInfoData diskName;

spaceTypeName = tuplesort_space_type_name(SORT_SPACE_TYPE_DISK);