Re: SQLFunctionCache and generic plans
От | Alexander Lakhin |
---|---|
Тема | Re: SQLFunctionCache and generic plans |
Дата | |
Msg-id | 0fde377a-3870-4d18-946a-ce008ee5bb88@gmail.com обсуждение исходный текст |
Ответ на | Re: SQLFunctionCache and generic plans (Tom Lane <tgl@sss.pgh.pa.us>) |
Ответы |
Re: SQLFunctionCache and generic plans
|
Список | pgsql-hackers |
Hello Tom,
02.04.2025 21:09, Tom Lane wrote:
02.04.2025 21:09, Tom Lane wrote:
Since feature freeze is fast approaching, I did a tiny bit more cosmetic work on this patchset and then pushed it. (There's still plenty of time for adjustments if you have further comments.)
I've discovered that starting from 0dca5d68d, the following query:
CREATE FUNCTION f(x anyelement) RETURNS anyarray AS '' LANGUAGE SQL;
SELECT f(0);
triggers:
TRAP: failed Assert("fcache->func->rettype == VOIDOID"), File: "functions.c", Line: 1737, PID: 3784779
On 0dca5d68d~1, it raises:
ERROR: return type mismatch in function declared to return integer[]
DETAIL: Function's final statement must be SELECT or INSERT/UPDATE/DELETE/MERGE RETURNING.
CONTEXT: SQL function "f" during startup
Best regards,
Alexander Lakhin
Neon (https://neon.tech)
В списке pgsql-hackers по дате отправления: