| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
|---|---|
| To: | Peter Eisentraut <peter_e(at)gmx(dot)net> |
| Cc: | Peter Geoghegan <peter(dot)geoghegan86(at)gmail(dot)com>, David Fetter <david(at)fetter(dot)org>, Marko Tiikkaja <marko(dot)tiikkaja(at)cs(dot)helsinki(dot)fi>, pgsql-hackers(at)postgresql(dot)org |
| Subject: | Re: "writable CTEs" |
| Date: | 2010-12-28 14:53:59 |
| Message-ID: | 18981.1293548039@sss.pgh.pa.us |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
Peter Eisentraut <peter_e(at)gmx(dot)net> writes:
> On tis, 2010-12-28 at 00:19 +0000, Peter Geoghegan wrote:
>> It's worth noting that officially (i.e. in the docs), we don't even
>> call CTEs CTEs at any point. We call them WITH queries. I think that
>> that's a mistake because we call them CTEs everywhere else.
> I think "WITH query" or "WITH clause" is more understandable than CTE,
> which to me is a term that has no relationship with anything else.
I'm with Peter on that. CTE is a completely meaningless term to most
users.
As for the problem at hand, couldn't we use "WITH ... RETURNING", or
some other phrase based on what users actually see/write? DML has
the same problem as CTE, namely it's just another damn TLA. It may
be one that more people have heard of, but that doesn't make it
particularly attractive.
regards, tom lane
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Tom Lane | 2010-12-28 14:58:28 | Re: pg_primary_conninfo |
| Previous Message | Kevin Grittner | 2010-12-28 14:53:06 | Re: estimating # of distinct values |