hbase-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Ted Yu <yuzhih...@gmail.com>
Subject Re: Fuzzy Row filter with ranges
Date Fri, 05 Feb 2016 15:39:03 GMT
In this case, it seems stop key should be 0xAB.

Rajeshkumar's email didn't mention prefix.
I assume he was looking for generic fuzzy row filter.

Cheers

On Fri, Feb 5, 2016 at 7:25 AM, Jean-Marc Spaggiari <jean-marc@spaggiari.org
> wrote:

> Hi,
>
> Be careful with you stop key here,  you might miss some values.
>
> Consider prefix 0xAA.
>
> Start=0xAA
> Stop=0xAA 0xFF
>
> You will miss 0xAA 0xFF 0xFF which is still a valid option.
>
> JMS
> Le 2016-02-05 3:21 AM, "Jameson Li" <hovlj.ei@gmail.com> a écrit :
>
> > 2016-02-01 19:08 GMT+08:00 Jean-Marc Spaggiari <jean-marc@spaggiari.org
> >:
> >
> > > example of what you are trying to d
> >
> >
> > like query prefixKey%, you can struct a query like this
> > startRow=Bytes.toBytes(prefixKey) ,
> > stopRow=Bytes.add(prefixKey, new byte[]{(byte)0xff})
> >
> > And be remember set setCacheBlocks(false), if the query will not query
> > repeat.
> >
> > And remember let a certain start and end, even if as a
> PrefixScan(sometimes
> > just write start but not end, it maybe scan the whole bottom half of the
> > table), and let the section smaller with query more faster.
> >
> >
> > --
> >
> >
> > Thanks & Regards,
> > 李剑 Jameson Li
> > Focus on Hadoop,Mysql
> >
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message