Hi,

In my preput coprocessor I would like to get the old value of the row been
input. Now I am creating a HTable instance and using the get interface;
function wise it works fine. Given the row is physically in the same
region as the cp, is there any lightweight approach doing that?

Thanks!
Best Regards,
Wei

Hi
    To your prePut() method you are getting an ObserverContext argument. From this you can easily get a ref to the current region.
ctx.getEnvironment().getRegion()
Directly make calls to this region object. Creation of the HTable instance and making a get() will be too much unwanted overhead.

-Anoop-
________________________________________
From: Wei Tan [[EMAIL PROTECTED]]
Sent: Wednesday, October 10, 2012 2:05 AM
To: [EMAIL PROTECTED]
Subject: connect to the region from coprocessor

Hi,

In my preput coprocessor I would like to get the old value of the row been
input. Now I am creating a HTable instance and using the get interface;
function wise it works fine. Given the row is physically in the same
region as the cp, is there any lightweight approach doing that?

Thanks!
Best Regards,
Wei

But is it possible that the previous value for this row to be stored
in another region?

I mean, if cells are big, they might be on different regions / regionservers?

2012/10/10, Anoop Sam John <[EMAIL PROTECTED]>:
> Hi
>     To your prePut() method you are getting an ObserverContext argument.
> From this you can easily get a ref to the current region.
> ctx.getEnvironment().getRegion()
> Directly make calls to this region object. Creation of the HTable instance
> and making a get() will be too much unwanted overhead.
>
> -Anoop-
> ________________________________________
> From: Wei Tan [[EMAIL PROTECTED]]
> Sent: Wednesday, October 10, 2012 2:05 AM
> To: [EMAIL PROTECTED]
> Subject: connect to the region from coprocessor
>
> Hi,
>
> In my preput coprocessor I would like to get the old value of the row been
> input. Now I am creating a HTable instance and using the get interface;
> function wise it works fine. Given the row is physically in the same
> region as the cp, is there any lightweight approach doing that?
>
> Thanks!
>
>
> Best Regards,
> Wei

Here oldValue anyway will have the same rowkey right.  So it should be in the same region but as older version.

Regards
Ram

> -----Original Message-----
> From: Jean-Marc Spaggiari [mailto:[EMAIL PROTECTED]]
> Sent: Wednesday, October 10, 2012 5:24 PM
> To: [EMAIL PROTECTED]
> Subject: Re: connect to the region from coprocessor
>
> But is it possible that the previous value for this row to be stored
> in another region?
>
> I mean, if cells are big, they might be on different regions /
> regionservers?
>
> 2012/10/10, Anoop Sam John <[EMAIL PROTECTED]>:
> > Hi
> >     To your prePut() method you are getting an ObserverContext
> argument.
> > From this you can easily get a ref to the current region.
> > ctx.getEnvironment().getRegion()
> > Directly make calls to this region object. Creation of the HTable
> instance
> > and making a get() will be too much unwanted overhead.
> >
> > -Anoop-
> > ________________________________________
> > From: Wei Tan [[EMAIL PROTECTED]]
> > Sent: Wednesday, October 10, 2012 2:05 AM
> > To: [EMAIL PROTECTED]
> > Subject: connect to the region from coprocessor
> >
> > Hi,
> >
> > In my preput coprocessor I would like to get the old value of the row
> been
> > input. Now I am creating a HTable instance and using the get
> interface;
> > function wise it works fine. Given the row is physically in the same
> > region as the cp, is there any lightweight approach doing that?
> >
> > Thanks!
> >
> >
> > Best Regards,
> > Wei

>I mean, if cells are big, they might be on different regions / regionservers?
All the KVs [columns, versions] corresponding to one rowkey will be always there in one region. One row is not splittable

-Anoop-
________________________________________
From: Jean-Marc Spaggiari [[EMAIL PROTECTED]]
Sent: Wednesday, October 10, 2012 5:24 PM
To: [EMAIL PROTECTED]
Subject: Re: connect to the region from coprocessor

But is it possible that the previous value for this row to be stored
in another region?

I mean, if cells are big, they might be on different regions / regionservers?

2012/10/10, Anoop Sam John <[EMAIL PROTECTED]>:
> Hi
>     To your prePut() method you are getting an ObserverContext argument.
> From this you can easily get a ref to the current region.
> ctx.getEnvironment().getRegion()
> Directly make calls to this region object. Creation of the HTable instance
> and making a get() will be too much unwanted overhead.
>
> -Anoop-
> ________________________________________
> From: Wei Tan [[EMAIL PROTECTED]]
> Sent: Wednesday, October 10, 2012 2:05 AM
> To: [EMAIL PROTECTED]
> Subject: connect to the region from coprocessor
>
> Hi,
>
> In my preput coprocessor I would like to get the old value of the row been
> input. Now I am creating a HTable instance and using the get interface;
> function wise it works fine. Given the row is physically in the same
> region as the cp, is there any lightweight approach doing that?
>
> Thanks!
>
>
> Best Regards,
> Wei

Thank you Anoop and that is a very helpful suggestion.
Best Regards,
Wei

From:   Anoop Sam John <[EMAIL PROTECTED]>
To:     "[EMAIL PROTECTED]" <[EMAIL PROTECTED]>,
"[EMAIL PROTECTED]" <[EMAIL PROTECTED]>,
Date:   10/10/2012 05:22 AM
Subject:        RE: connect to the region from coprocessor

Hi
    To your prePut() method you are getting an ObserverContext argument.
>From this you can easily get a ref to the current region.
ctx.getEnvironment().getRegion()
Directly make calls to this region object. Creation of the HTable instance
and making a get() will be too much unwanted overhead.

-Anoop-
________________________________________
From: Wei Tan [[EMAIL PROTECTED]]
Sent: Wednesday, October 10, 2012 2:05 AM
To: [EMAIL PROTECTED]
Subject: connect to the region from coprocessor

Hi,

In my preput coprocessor I would like to get the old value of the row been
input. Now I am creating a HTable instance and using the get interface;
function wise it works fine. Given the row is physically in the same
region as the cp, is there any lightweight approach doing that?

Thanks!
Best Regards,
Wei