Re: Set query_id for query contained in utility statement

On Tue, Oct 15, 2024 at 3:23 PM jian he <jian(dot)universality(at)gmail(dot)com> wrote:
> explain(verbose) SELECT 1, 2, 3\; explain SELECT 1, 2, 3, 4;
> will transformed to
> explain(verbose) SELECT 1, 2, 3; explain SELECT 1, 2, 3, 4;
>
> it seems to me your patch care about following position.
> explain(verbose) SELECT 1, 2, 3; explain SELECT 1, 2, 3, 4;
> ^
> but this patch [1] at another thread will get the top level statement
> (passed the raw parse, no syntax error) beginning more effortless.
> explain(verbose) SELECT 1, 2, 3; explain SELECT 1, 2, 3, 4;
> ^ ^
>
> can you try to looking at [1]. it may help to resolve this patch problem.
>
> [1] https://www.postgresql.org/message-id/2245576.1728418678%40sss.pgh.pa.us

The top level statement beginning doesn't have an impact on this
patch. The patch's focus is on the nested statement and RawStmt's
location and length are only used to get the nested statement length.
I will need the nested statement's location and length no matter what,
to handle cases like "COPY (UPDATE ...) to stdout;" and only report
the statement within parentheses.

The linked patch will allow a simplification: the "if (@$ < 0) @$ =
@2;" I've added won't be needed anymore. But I think that's the only
possible simplification? I've run the tests on top of the linked patch
and there was no change in the regression output.