"Jonathan S. Katz" writes:
> On 2/17/23 1:18 PM, Tom Lane wrote:
>> It can be reproduced with INSERT too, on the same principle as the others:
>> put the DML command inside a WITH, and give it an alias conflicting with
>> the outer query.
> Ah, I see based on your example below. I did not alias t
On 2/17/23 1:18 PM, Tom Lane wrote:
It can be reproduced with INSERT too, on the same principle as the others:
put the DML command inside a WITH, and give it an alias conflicting with
the outer query.
Ah, I see based on your example below. I did not alias the INSERT
statement in the way (and
"Jonathan S. Katz" writes:
> I spoke too soon -- I was looking at the wrong logs. I did reproduce it
> with UPDATE, but not INSERT.
It can be reproduced with INSERT too, on the same principle as the others:
put the DML command inside a WITH, and give it an alias conflicting with
the outer query.
On 2/17/23 11:19 AM, Jonathan S. Katz wrote:
On 2/17/23 10:09 AM, Tom Lane wrote:
Agreed on that -- while it's harder to set up, I do prefer the original
example[1] to demonstrate this, as it shows the issue given it does not
have those multiple occurrences, at least not within the same conte
On 2/17/23 10:09 AM, Tom Lane wrote:
"Jonathan S. Katz" writes:
Good catch. Do you have thoughts on how we can adjust the naming logic
to handle cases like this?
I think it's perfectly fine that ruleutils decided to use different
aliases for the two different occurrences of "t1": the statemen
"Jonathan S. Katz" writes:
> Good catch. Do you have thoughts on how we can adjust the naming logic
> to handle cases like this?
I think it's perfectly fine that ruleutils decided to use different
aliases for the two different occurrences of "t1": the statement is
quite confusing as written. Th
On 2/17/23 5:22 AM, vignesh C wrote:
Hi,
The output sql generated by pg_dump for the below function refers to a
modified table name:
create table t1 (c1 int);
create table t2 (c1 int);
CREATE OR REPLACE FUNCTION test_fun(c1 int)
RETURNS void
LANGUAGE SQL
BEGIN ATOMIC
WITH delete_t1 AS (
Hi,
The output sql generated by pg_dump for the below function refers to a
modified table name:
create table t1 (c1 int);
create table t2 (c1 int);
CREATE OR REPLACE FUNCTION test_fun(c1 int)
RETURNS void
LANGUAGE SQL
BEGIN ATOMIC
WITH delete_t1 AS (
DELETE FROM t1 WHERE c1 = $1
