Re: WITH RECURSIVE ... CYCLE in vanilla SQL: issues with arrays of rows - Mailing list pgsql-hackers

From Tom Lane
Subject Re: WITH RECURSIVE ... CYCLE in vanilla SQL: issues with arrays of rows
Date
Msg-id 28968.1223996368@sss.pgh.pa.us
Whole thread Raw
In response to Re: WITH RECURSIVE ... CYCLE in vanilla SQL: issues with arrays of rows  ("Merlin Moncure" <mmoncure@gmail.com>)
Responses Re: WITH RECURSIVE ... CYCLE in vanilla SQL: issues with arrays of rows
List pgsql-hackers
"Merlin Moncure" <mmoncure@gmail.com> writes:
> Here is another use-case solved by the patch.  Previously, there was
> no easy way to index on a composite function result.  The following
> works in HEAD:

> create function func(f foo, a out int, b out int) returns record ...

> create index foo_idx on foo(func(foo));

Urk ... "works" for small values of "work", perhaps.  Did you try using
the index from a fresh session?

We could support this for named composite types but not for anonymous
record types.  I'm not quite sure how to enforce that distinction
considering that the opclass is defined to take "record".  Maybe we
should apply CheckAttributeType() to index column types?
        regards, tom lane


pgsql-hackers by date:

Previous
From: "Merlin Moncure"
Date:
Subject: Re: WITH RECURSIVE ... CYCLE in vanilla SQL: issues with arrays of rows
Next
From: Magnus Hagander
Date:
Subject: Re: 8.3 .4 + Vista + MingW + initdb = ACCESS_DENIED