Re: Parallel Inserts in CREATE TABLE AS - Mailing list pgsql-hackers

On Wed, Jan 6, 2021 at 8:19 AM Zhihong Yu <zyu@yugabyte.com> wrote:
> For v20-0001-Parallel-Inserts-in-CREATE-TABLE-AS.patch :
>
> ParallelInsCmdEstimate :
>
> +   Assert(pcxt && ins_info &&
> +          (ins_cmd == PARALLEL_INSERT_CMD_CREATE_TABLE_AS));
> +
> +   if (ins_cmd == PARALLEL_INSERT_CMD_CREATE_TABLE_AS)
>
> Sinc the if condition is covered by the assertion, I wonder why the if check is still needed.
>
> Similar comment for SaveParallelInsCmdFixedInfo and SaveParallelInsCmdInfo

Thanks.

The idea is to have assertion with all the expected ins_cmd types, and
then later to have selective handling for different ins_cmds. For
example, if we add (in future) parallel insertion in Refresh
Materialized View, then the code in those functions will be something
like:

+static void
+ParallelInsCmdEstimate(ParallelContext *pcxt, ParallelInsertCmdKind ins_cmd,
+                       void *ins_info)
+{
+    Assert(pcxt && ins_info &&
+           (ins_cmd == PARALLEL_INSERT_CMD_CREATE_TABLE_AS ||
+            (ins_cmd == PARALLEL_INSERT_CMD_REFRESH_MAT_VIEW));
+
+    if (ins_cmd == PARALLEL_INSERT_CMD_CREATE_TABLE_AS)
+    {
+
+    }
+   else if (ns_cmd == PARALLEL_INSERT_CMD_REFRESH_MAT_VIEW)
+   {
+
+   }

Similarly for other functions as well.

With Regards,
Bharath Rupireddy.
EnterpriseDB: http://www.enterprisedb.com