Re: Forbid to DROP temp tables of other sessions
От | Steven Niu |
---|---|
Тема | Re: Forbid to DROP temp tables of other sessions |
Дата | |
Msg-id | a9f0bf88-a726-4bfd-a602-ee55b2f724e4@gmail.com обсуждение исходный текст |
Ответ на | Re: Forbid to DROP temp tables of other sessions (Daniil Davydov <3danissimo@gmail.com>) |
Ответы |
Re: Forbid to DROP temp tables of other sessions
|
Список | pgsql-hackers |
在 2025/3/17 18:13, Daniil Davydov 写道: > Hi, > > On Mon, Mar 17, 2025 at 4:48 PM Steven Niu <niushiji@gmail.com> wrote: >> >> 1. namespace.c, if relation->schemaname is pg_temp but myTempNamespace >> isn't set, the error information might be misleading. Consider checking >> OidIsValid(myTempNamespace) first. > Could you please clarify exactly which place in the code we are > talking about? I think we handle this case in the > LookupExplicitNamespace call (with all appropriate error information). > I mean RangeVarGetRelidExtended(), you deleted the following code: if (!OidIsValid(myTempNamespace)) relId = InvalidOid; /* this probably can't happen? */ >> >> 2."you have not any temporary relations" --> "you have no any temporary >> relations" > I am not an English speaker, but it seems that "have not" would be > more correct. Someone has to judge us :) > >> >> 3. Regarding to the code "strncmp(nspname, "pg_temp", 7)", is it ok when >> the nspname contains something like "pg_temp_1234"? I think we should >> use strcmp instead of strncmp for exact matching. > Great catch! I'll fix it. Please, see v3 patch. > > -- > Best regards, > Daniil Davydov
В списке pgsql-hackers по дате отправления: