> You seem to be describing a straight reconciliation between two tables.
> My
> current means of doing this are programmatically but for the simple case
> pure SQL should be doable. The main thing is that you have to distinguish
> between "duplicate" records first and then match them up:
>
> Tabl
> You seem to be describing a straight reconciliation between two tables.
> My
> current means of doing this are programmatically but for the simple case
> pure SQL should be doable. The main thing is that you have to distinguish
> between "duplicate" records first and then match them up:
>
> Tabl
gt;
> Best,
> Oliver
>
> - Original Message -
> From: "Rihad"
> To:
> Sent: Thursday, June 21, 2012 6:48 PM
> Subject: [SQL] Need help building this query
>
>
>> Hi, folks. I currently need to join two tables that lack primary keys,
>> and
:50 PM
Subject: Re: [SQL] Need help building this query
For matching triples (foo, bar, baz) the date in table B shouldnt always
be after any date in table A, as table B contains complete operations?
Best,
Oliver
- Original Message -
From: "Rihad"
To:
Sent: Thursday, J
For matching triples (foo, bar, baz) the date in table B shouldnt always be
after any date in table A, as table B contains complete operations?
Best,
Oliver
- Original Message -
From: "Rihad"
To:
Sent: Thursday, June 21, 2012 6:48 PM
Subject: [SQL] Need help building
> -Original Message-
> From: pgsql-sql-ow...@postgresql.org [mailto:pgsql-sql-
> ow...@postgresql.org] On Behalf Of Rihad
> Sent: Thursday, June 21, 2012 1:49 PM
> To: pgsql-sql@postgresql.org
> Subject: [SQL] Need help building this query
>
> Hi, folks. I cu
Hi, folks. I currently need to join two tables that lack primary keys,
and columns used to distinguish each record can be duplicated. I need to
build statistics over the data in those tables. Consider this:
TableA:
row 1: foo: 123, bar: 456, baz: 789, amount: 10.99, date_of_op: date
row 2: foo
