| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
|---|---|
| To: | Bruce Momjian <bruce(at)momjian(dot)us> |
| Cc: | Michael Paquier <michael(at)paquier(dot)xyz>, jian he <jian(dot)universality(at)gmail(dot)com>, Alvaro Herrera <alvherre(at)alvh(dot)no-ip(dot)org>, Kirill Reshke <reshkekirill(at)gmail(dot)com>, PostgreSQL-development <pgsql-hackers(at)postgresql(dot)org> |
| Subject: | Re: attndims, typndims still not enforced, but make the value within a sane threshold |
| Date: | 2025-01-21 21:30:17 |
| Message-ID: | 3485415.1737495017@sss.pgh.pa.us |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
I wrote:
> I wonder if we should try to fix the GIN AM to avoid that.
> The column being indexed is of an array type in these cases, but the
> index entries aren't. It seems inconsistent that it sets up the index
> column's attndims and atttypid this way.
Ah, I see the problem: it's not GIN's fault, it's that index.c's
ConstructTupleDescriptor is very sloppy about setting attndims
in some places and not others. Fortunately it's quite cheap to
fix that, since the places where this is missed already have their
hands on the pg_type entry for the index column's type. See v2
attached (now with a regression test).
regards, tom lane
| Attachment | Content-Type | Size |
|---|---|---|
| v2-preserve-attndims-better.patch | text/x-diff | 11.1 KB |
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Robert Treat | 2025-01-21 22:45:02 | Re: XMLDocument (SQL/XML X030) |
| Previous Message | Jeff Davis | 2025-01-21 21:26:52 | Re: Proposal: "query_work_mem" GUC, to distribute working memory to the query's individual operators |