Re: Outer join differences

From: Yuva Chandolu <ychandolu(at)ebates(dot)com>
To: 'Tom Lane' <tgl(at)sss(dot)pgh(dot)pa(dot)us>
Cc: "'pgsql-hackers(at)postgresql(dot)org'" <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: Outer join differences
Date: 2002-07-31 04:29:48
Message-ID: A0F24737FCB34F489EC955D143BDD8510173E0F1@exchange-sf1.corp.ebates.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Hi Tom,

Thanks for your prompt reply, after second thought(before receiving your
reply) I realized that postgres is doing more logically - i.e if the outer
join condition returns false then replace by nulls for right table columns.
We may change our code accordingly :-(.

Thanks
Yuva

-----Original Message-----
From: Tom Lane [mailto:tgl(at)sss(dot)pgh(dot)pa(dot)us]
Sent: Tuesday, July 30, 2002 9:15 PM
To: Yuva Chandolu
Cc: 'pgsql-hackers(at)postgresql(dot)org'
Subject: Re: [HACKERS] Outer join differences

Yuva Chandolu <ychandolu(at)ebates(dot)com> writes:
> I see different results in Oracle and postgres for same outer join
queries.

I believe you are sending your bug report to the wrong database.

> When I run the query "select yt1_name, yt1_descr, yt2_name, yt2_descr from
> yuva_test1 left outer join yuva_test2 on yt1_id=yt2_id and yt2_name =
> '2-name2'" on postgres database I get the following results

> yt1_name yt1_descr yt2_name yt2_descr
> 1-name1 1-descr1
> 1-name2 1-descr2 2-name2 2-descr2
> 1-name3 1-descr3
> 1-name4 1-descr4
> 1-name5 1-descr5
> 1-name6 1-descr6

> But when I tried the same on Oracle(8.1.7) (the query is "select yt1_name,
> yt1_descr, yt2_name, yt2_descr from yuva_test1, yuva_test2 where
> yt1_id=yt2_id(+) and yt2_name = '2-name2'') I get the following results

> yt1_name yt1_descr yt2_name yt2_descr
> 1-name2 1-descr2 2-name2 2-descr2

According to the SQL spec, the output of a LEFT JOIN consists of those
joined rows where the join condition is true, plus those rows of the
left table for which no right-table row produced a true join condition
(substituting nulls for the right-table columns). Our output clearly
conforms to the spec.

I do not know what Oracle thinks is the correct output when one
condition is marked with (+) and the other is not --- it's not very
obvious what that corresponds to in the spec's terminology. But I
suggest you take it up with them, not us.

regards, tom lane

Browse pgsql-hackers by date

  From Date Subject
Next Message Curt Sampson 2002-07-31 04:31:15 Re: Why is MySQL more chosen over PostgreSQL?
Previous Message Tom Lane 2002-07-31 04:29:18 Re: Open 7.3 items