Finding duplicate records using two columns

I'm trying to find duplicate records in a table. I know how to do that using
a single column value, but I need to do it using the combined values of two
columns, company_id and client_id. There multiple are records with the same
company_id value (e.g., 123) and multiple records with the same department_id
value (e.g., 456). But there should be only one record with a company_id
value of 123 and a department_id value of 456.
I've searched for data in one table that is not in another table on these
two columns using CAST to concatenate the values:
select * from company c
where cast(c.company_id as varchar(20)) + cast(c.client_id as varchar(20))
not in
(select cast(r.company_id as varchar(20)) + cast(r.client_id as varchar(20))
from client r)
But I can't seem to figure out how to use CAST to find duplicate records in
a single table. Should I be doing this a different way?
Any and all help would be appreciated.
JohnSELECT company_id, client_id, count(*) as Duplicates
FROM Company
GROUP BY company_id, client_id
HAVING COUNT(*) > 1
This works great with one column or ten columns.
Roy Harvey
Beacon Falls, CT
On Tue, 30 Oct 2007 12:56:05 -0700, John Steen
<moderndads(nospam)@.hotmail.com> wrote:
>I'm trying to find duplicate records in a table. I know how to do that using
>a single column value, but I need to do it using the combined values of two
>columns, company_id and client_id. There multiple are records with the same
>company_id value (e.g., 123) and multiple records with the same department_id
>value (e.g., 456). But there should be only one record with a company_id
>value of 123 and a department_id value of 456.
>I've searched for data in one table that is not in another table on these
>two columns using CAST to concatenate the values:
>select * from company c
>where cast(c.company_id as varchar(20)) + cast(c.client_id as varchar(20))
>not in
>(select cast(r.company_id as varchar(20)) + cast(r.client_id as varchar(20))
>from client r)
>But I can't seem to figure out how to use CAST to find duplicate records in
>a single table. Should I be doing this a different way?
>Any and all help would be appreciated.
>John|||Thank you, Roy! It worked perfectly.
John
"Roy Harvey (SQL Server MVP)" wrote:
> SELECT company_id, client_id, count(*) as Duplicates
> FROM Company
> GROUP BY company_id, client_id
> HAVING COUNT(*) > 1
> This works great with one column or ten columns.
> Roy Harvey
> Beacon Falls, CT
> On Tue, 30 Oct 2007 12:56:05 -0700, John Steen
> <moderndads(nospam)@.hotmail.com> wrote:
> >I'm trying to find duplicate records in a table. I know how to do that using
> >a single column value, but I need to do it using the combined values of two
> >columns, company_id and client_id. There multiple are records with the same
> >company_id value (e.g., 123) and multiple records with the same department_id
> >value (e.g., 456). But there should be only one record with a company_id
> >value of 123 and a department_id value of 456.
> >
> >I've searched for data in one table that is not in another table on these
> >two columns using CAST to concatenate the values:
> >
> >select * from company c
> >where cast(c.company_id as varchar(20)) + cast(c.client_id as varchar(20))
> >not in
> >(select cast(r.company_id as varchar(20)) + cast(r.client_id as varchar(20))
> >from client r)
> >
> >But I can't seem to figure out how to use CAST to find duplicate records in
> >a single table. Should I be doing this a different way?
> >
> >Any and all help would be appreciated.
> >
> >John
>