From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
---|---|
To: | Stephen Frost <sfrost(at)snowman(dot)net> |
Cc: | Vik Fearing <vik(dot)fearing(at)2ndquadrant(dot)com>, "Johann 'Myrkraverk' Oskarsson" <johann(at)myrkraverk(dot)com>, pgsql-general(at)lists(dot)postgresql(dot)org |
Subject: | Re: Retroactively adding send and recv functions to a type? |
Date: | 2019-08-20 15:03:13 |
Message-ID: | 16030.1566313393@sss.pgh.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
Stephen Frost <sfrost(at)snowman(dot)net> writes:
> * Vik Fearing (vik(dot)fearing(at)2ndquadrant(dot)com) wrote:
>> We generally discourage updating the catalogs directly. This was why I
>> wrote the CREATE/ALTER DATABASE patch in 2014 that you helped me with
>> (fbb1d7d73f8).
>> I'm willing to work on a patch for ALTER TYPE if it has a chance of
>> being accepted.
> Seems pretty clear that it'd be a useful thing to have, so +1 from me,
> at least.
We'd have to be extremely circumspect about what aspects of a base
type could be altered after-the-fact. Adding binary I/O functions
seems unproblematic, but I'm not very sure what else is.
regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | Adrian Klaver | 2019-08-20 15:11:42 | Re: pg_dump problems: [archiver (db)] query failed: ERROR: relation "pg_opfamily" does not exist |
Previous Message | Adrian Klaver | 2019-08-20 15:02:07 | Re: pg_dump problems: [archiver (db)] query failed: ERROR: relation "pg_opfamily" does not exist |