| Lists: | pgsql-hackers |
|---|
| From: | Zhihong Yu <zyu(at)yugabyte(dot)com> |
|---|---|
| To: | "Andrey V(dot) Lepikhov" <a(dot)lepikhov(at)postgrespro(dot)ru>, PostgreSQL Developers <pgsql-hackers(at)lists(dot)postgresql(dot)org> |
| Subject: | Re: [POC] Allow flattening of subquery with a link to upper query |
| Date: | 2022-10-04 21:45:11 |
| Message-ID: | CALNJ-vTa5VgvV1NPRHnypdnbx-fhDu7vWp73EkMUbZRpNHTYQQ@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Lists: | pgsql-hackers |
Hi,
For contain_placeholders():
+ if (IsA(node, Query))
+ return query_tree_walker((Query *) node, contain_placeholders,
context, 0);
+ else if (IsA(node, PlaceHolderVar))
The `else` is not needed.
For correlated_t struct, it would be better if the fields have comments.
+ * (for grouping, as an example). So, revert its status
to
+ * a full valued entry.
full valued -> fully valued
Cheers
| From: | Andrey Lepikhov <a(dot)lepikhov(at)postgrespro(dot)ru> |
|---|---|
| To: | Zhihong Yu <zyu(at)yugabyte(dot)com>, PostgreSQL Developers <pgsql-hackers(at)lists(dot)postgresql(dot)org> |
| Subject: | Re: [POC] Allow flattening of subquery with a link to upper query |
| Date: | 2022-10-05 11:38:10 |
| Message-ID: | 35c8a3e8-d080-dfa8-2be3-cf5fe702010a@postgrespro.ru |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Lists: | pgsql-hackers |
On 5/10/2022 02:45, Zhihong Yu wrote:
> Hi,
> For contain_placeholders():
>
> + if (IsA(node, Query))
> + return query_tree_walker((Query *) node, contain_placeholders,
> context, 0);
> + else if (IsA(node, PlaceHolderVar))
Fixed
>
> The `else` is not needed.
>
> For correlated_t struct, it would be better if the fields have comments.
Ok, I've added some comments.
>
> + * (for grouping, as an example). So, revert its
> status to
> + * a full valued entry.
>
> full valued -> fully valued
Fixed
--
regards,
Andrey Lepikhov
Postgres Professional
| Attachment | Content-Type | Size |
|---|---|---|
| v3-0001-Transform-correlated-subquery-of-type-N-J-1-into-ord.patch | text/plain | 51.6 KB |
| From: | Zhihong Yu <zyu(at)yugabyte(dot)com> |
|---|---|
| To: | Andrey Lepikhov <a(dot)lepikhov(at)postgrespro(dot)ru> |
| Cc: | PostgreSQL Developers <pgsql-hackers(at)lists(dot)postgresql(dot)org> |
| Subject: | Re: [POC] Allow flattening of subquery with a link to upper query |
| Date: | 2022-10-05 14:11:54 |
| Message-ID: | CALNJ-vQ_1mB4BrObnP7V4VQxa48hPs9qw36A4bteve3VaH04Vw@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Lists: | pgsql-hackers |
On Wed, Oct 5, 2022 at 4:38 AM Andrey Lepikhov <a(dot)lepikhov(at)postgrespro(dot)ru>
wrote:
> On 5/10/2022 02:45, Zhihong Yu wrote:
> > Hi,
> > For contain_placeholders():
> >
> > + if (IsA(node, Query))
> > + return query_tree_walker((Query *) node, contain_placeholders,
> > context, 0);
> > + else if (IsA(node, PlaceHolderVar))
> Fixed
> >
> > The `else` is not needed.
> >
> > For correlated_t struct, it would be better if the fields have comments.
> Ok, I've added some comments.
> >
> > + * (for grouping, as an example). So, revert its
> > status to
> > + * a full valued entry.
> >
> > full valued -> fully valued
> Fixed
>
> --
> regards,
> Andrey Lepikhov
> Postgres Professional
>
Hi,
+ List *pulling_quals; /* List of expressions contained pulled
expressions */
contained -> containing
+ /* Does the var already exists in the target list? */
exists -> exist
+ {"optimize_correlated_subqueries", PGC_USERSET, QUERY_TUNING_METHOD,
Is it possible that in the future there would be other optimization for
correlated subqueries ?
If so, either rename the guc or, make the guc a string which represents an
enum.
Cheers