Re: Reject invalid databases in pg_get_database_ddl() - Mailing list pgsql-hackers

From Lakshmi N
Subject Re: Reject invalid databases in pg_get_database_ddl()
Date
Msg-id CA+3i_M_JzLyF14Uf8pbev8pZB6wOWjWnp1QerUz2REDr0PJrxQ@mail.gmail.com
Whole thread
In response to Re: Reject invalid databases in pg_get_database_ddl()  (Amit Langote <amitlangote09@gmail.com>)
Responses Re: Reject invalid databases in pg_get_database_ddl()
List pgsql-hackers
Hi,

On Thu, Apr 16, 2026 at 8:31 PM Amit Langote <amitlangote09@gmail.com> wrote:
Hi,

On Fri, Apr 17, 2026 at 11:49 AM Hu Xunqi <huxunqi.08@gmail.com> wrote:
> On Fri, Apr 17, 2026 at 10:16 AM Amit Langote <amitlangote09@gmail.com> wrote:
> +       /*
> +        * Reject invalid databases: datconnlimit = -2 would be emitted as
> +        * CONNECTION LIMIT = -2, which cannot be executed.
> +        */
>
> This comment looks a bit too centered on datconnlimit=-2, but the real issue is that an invalid pg_database row should not be deparsed into DDL. So, maybe rephrase like:
>
> /*
> * Reject invalid databases. Deparsing a pg_database row in invalid state
> * can produce SQL that is not executable, such as CONNECTION LIMIT = -2.
> */

I was trying to be precise about datconnlimit = -2 being the thing
that produces invalid SQL. But your version covers that with the "such
as CONNECTION LIMIT = -2" example, and it's closer to the original,
which was on the right track, just needed to be more precise. Let's go
with it.

This looks good to me. Thank you for reviewing and making the changes!
 
Regards,
Lakshmi

pgsql-hackers by date:

Previous
From: jian he
Date:
Subject: Re: COPY FROM ON_ERROR SET_NULL bypasses domain NOT NULL with partial column list
Next
From: Amit Langote
Date:
Subject: Re: Reject invalid databases in pg_get_database_ddl()