Horton

Re: Typed table DDL loose ends

Tom Lane <tgl@sss.pgh.pa.us>

View thread
From: Tom Lane <tgl@sss.pgh.pa.us>
To: Robert Haas <robertmhaas@gmail.com>
Cc: Noah Misch <noah@leadboat.com>, pgsql-hackers@postgresql.org, Peter Eisentraut <peter_e@gmx.net>
Date: 2011-04-18T15:33:47Z
Lists: pgsql-hackers

Commits

Same data as JSON: GET /api/v1/messages/:b64id/commits the thread's linked commits as JSON, with link sources. API reference →

  1. Prevent a rowtype from being included in itself.

Robert Haas <robertmhaas@gmail.com> writes:
> On Mon, Apr 18, 2011 at 10:48 AM, Noah Misch <noah@leadboat.com> wrote:
>> FWIW, the term "stand-alone composite type" appears twice in our documentation.

> Hmm, OK.  Anyone else have an opinion on the relative merits of:

> ERROR: type stuff is not a composite type
> vs.
> ERROR: type stuff is not a stand-alone composite type

> The intent of adding "stand-alone" was, I believe, to clarify that it
> has to be a CREATE TYPE stuff AS ... type, not just a row type (that
> is, naturally, composite, in some less-pure sense).  I'm not sure
> whether the extra word actually makes it more clear, though.

In 99.9% of the code and docs, a table rowtype is a perfectly good
composite type.  I agree with Noah that just saying "composite type"
is inadequate here; but I'm not sure that "stand-alone" is a helpful
adjective either.  What about inverting the message phrasing, ie

ERROR: type stuff must not be a table's row type

You might need some extra logic to keep on giving "is not a composite
type" in cases where it's not composite at all.  But this is enough of a
departure from our usual behavior that I think the error message had
better be pretty darn clear.

			regards, tom lane