Re: [PATCH] Add pg_get_database_ddl() function to reconstruct CREATE DATABASE statement - Mailing list pgsql-hackers

From Japin Li
Subject Re: [PATCH] Add pg_get_database_ddl() function to reconstruct CREATE DATABASE statement
Date
Msg-id SY7PR01MB10921B6F9D478C3F1B2639901B67CA@SY7PR01MB10921.ausprd01.prod.outlook.com
Whole thread Raw
In response to [PATCH] Add pg_get_database_ddl() function to reconstruct CREATE DATABASE statement  (Akshay Joshi <akshay.joshi@enterprisedb.com>)
Responses Re: [PATCH] Add pg_get_database_ddl() function to reconstruct CREATE DATABASE statement
List pgsql-hackers
On Wed, 04 Mar 2026 at 18:29, Akshay Joshi <akshay.joshi@enterprisedb.com> wrote:
> Thanks for the review, Japin. I’ve addressed all of your comments. I also added a check to throw an error if an
option
> appears more than once.
>
> Attached is the v10 patch, now ready for further review.
>

Thanks for updating the patch.  Here are some comments on v10.

1.
+ * db_oid - OID/Name of the database for which to generate the DDL.

Should the comment be updated? The code only accepts an OID for `db_oid`,
database names are not supported.

2.
+       /* Set the OWNER in the DDL if owner is not omitted */
+       if (OidIsValid(dbform->datdba) && !(ddl_flags & PG_DDL_NO_OWNER))
+       {
+               get_formatted_string(&buf, pretty_flags, 8, "OWNER = %s",
+                                                        quote_identifier(dbowner));
+       }

`dbowner` is only needed inside this `if` — how about declaring it there to
reduce its scope?

3.
+               /* If is_with_defaults is true, then we skip default encoding check */
+               if (is_with_defaults ||
+                       (pg_strcasecmp(pg_encoding_to_char(dbform->encoding),
+                                                  DDL_DEFAULTS.DATABASE.ENCODING) != 0))
+               {
+                       get_formatted_string(&buf, pretty_flags, 8, "ENCODING = %s",
+                                                                quote_literal_cstr(
+
pg_encoding_to_char(dbform->encoding)));
+               }

How about cache the result of `pg_encoding_to_char()` in a local variable to
avoid calling it twice?

--
Regards,
Japin Li
ChengDu WenWu Information Technology Co., Ltd.



pgsql-hackers by date:

Previous
From: John Naylor
Date:
Subject: Re: Reduce timing overhead of EXPLAIN ANALYZE using rdtsc?
Next
From: Álvaro Herrera
Date:
Subject: Re: UBSAN crash in EventTriggerCollectAlterTSConfig (memcpy with NULL src)