From: | Royce Ausburn <royce(dot)ml(at)inomial(dot)com> |
---|---|
To: | Yeb Havinga <yebhavinga(at)gmail(dot)com> |
Cc: | Pg Hackers <pgsql-hackers(at)postgresql(dot)org> |
Subject: | Re: [REVIEW] Patch for cursor calling with named parameters |
Date: | 2011-10-11 11:55:09 |
Message-ID: | 29A8127B-BB37-40DC-8FB6-833710DAB22C@inomial.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
On 08/10/2011, at 1:56 AM, Yeb Havinga wrote:
> Attach is v2 of the patch.
>
> Mixed notation now raises an error.
>
> In contrast with what I said above, named parameter related errors are thrown before any syntax errors. I tested with raising syntax errors first but the resulting code was a bit more ugly and the sql checking under a error condition (i.e. double named parameter error means there is one parameter in short) was causing serious errors.
>
> Documentation was also added, regression tests updated.
I've tested this patch out and can confirm mixed notation produces an error:
psql:plsqltest.sql:27: ERROR: mixing positional and named parameter assignment not allowed in cursor "cur1"
LINE 10: open cur1( param2 := 4, 2, 5);
Just one small thing: it'd be nice to have an example for cursor declaration with named parameters. Your patch adds one for opening a cursor, but not for declaring one.
Other than that, I think the patch is good. Everything works as advertised =)
--Royce
From | Date | Subject | |
---|---|---|---|
Next Message | Alexander Korotkov | 2011-10-11 12:16:27 | Re: index-only scans |
Previous Message | Dave Page | 2011-10-11 11:18:33 | Buildfarm git failures |