Mailing-List: contact user-help@hive.apache.org; run by ezmlm
Precedence: bulk
Reply-To: user@hive.apache.org
Received-SPF: pass (athena.apache.org: domain of edlinuxguru@gmail.com
 designates 209.85.223.176 as permitted sender)
MIME-Version: 1.0
In-Reply-To: <C155E15D0707204885EB0BA11E54B1DF0A019806@XMBVAG72.northgrum.com>
References: <C155E15D0707204885EB0BA11E54B1DF0A019806@XMBVAG72.northgrum.com>
Date: Thu, 1 Nov 2012 23:17:04 -0400
Message-ID: 
 <CAENxBwwAbU1hQ32eCv1M=9R-3kCNAtuRwFB+FpFvQyCMwmMyYw@mail.gmail.com>
Subject: Re: Group By Concatenation
From: Edward Capriolo <edlinuxguru@gmail.com>
To: user@hive.apache.org
Content-Type: text/plain; charset=windows-1252
Content-Transfer-Encoding: quoted-printable

Collect_set() is built into hive. If you want a version that does not
de-duplicate look here.
https://github.com/edwardcapriolo/hive-collect

Caution both of these functions can produce out of memory if the
results are later then a mapper can store in memory.

On Thu, Nov 1, 2012 at 2:27 PM, Ratner, Alan S (IS) <Alan.Ratner@ngc.com> w=
rote:
> Sorry to ask what is probably a very na=EFve Hive question but here goes:
>
>
>
> I have a table as follows:
>
> Col1   Col2
>
> K1       V1
>
> K1       V1
>
> K2       V1
>
> K3       V1
>
> K1       V2
>
> K1       V3
>
> K2       V2
>
>
>
> Now I have managed to SELECT Col1,COUNT(DISTINCT Col2) FROM =85 BY COL1; =
to
> obtain
>
> K1       3
>
> K2       2
>
> K3       1
>
>
>
> But what I want is a concatenated list of all the distinct Col2 values fo=
r
> each Col1 key.
>
> K1       V1       V2       V3
>
> K2       V1       V2
>
> K3       V1
>
>
>
> This is something absolutely trivial in MR but I cannot seem to find
> anything in Hive that will do this for me.  Do I have to write a UDF to
> accomplish this?
>
>
>
>
>
> Alan
>
>