phoenix-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From James Taylor <jamestay...@apache.org>
Subject Re: Importing existing HBase table's rowkey
Date Thu, 23 Jul 2015 01:19:35 GMT
Yes, UDFs are as of 4.4.0. I've corrected the JIRA.
Thanks,
James

On Wed, Jul 22, 2015 at 6:15 PM, Anchal Agrawal <anchal@yahoo-inc.com>
wrote:

> Hi James,
>
> The long value isn't serialized that way. There's some internal
> serialization logic in the rowkey, so I'm writing a UDF to extract the long
> value. I looked at your tutorial
> <http://phoenix-hbase.blogspot.in/2013/04/how-to-add-your-own-built-in-function.html>
> for writing UDFs, and I'm also reading up various Phoenix functions here
> <https://github.com/apache/phoenix/tree/master/phoenix-core/src/main/java/org/apache/phoenix/expression/function>
> .
>
> Quick question - if I'm using v4.4.0, do I need to include the UDF in the
> Phoenix jar? JIRA [PHOENIX-538]
> <https://issues.apache.org/jira/browse/PHOENIX-538> says that the fix
> version is 5.0.0 but the Phoenix UDF page says that v4.4.0 supports UDFs
> with the CREATE FUNCTION query.
>
> Thank you,
> Anchal
>
>
>
>   On Wednesday, July 22, 2015 5:56 PM, James Taylor <
> jamestaylor@apache.org> wrote:
>
>
> If it leads with a long that was serialized using Bytes.toBytes(long),
> then you can map that to the UNSIGNED_LONG type in Phoenix. What's the rest
> of your row key look like?
>
> On Wed, Jul 22, 2015 at 5:54 PM, Anchal Agrawal <anchal@yahoo-inc.com>
> wrote:
>
> Anil and Krishna,
>
> Thanks for your replies. My rowkey is made up of a LONG value with some
> other parameters added in. Krishna, that's what I was looking for! I've
> mapped the HBase table's rowkey to the Phoenix table, and I'm writing a UDF
> to extract fields from it.
>
> - Anchal
>
>
>
>   On Wednesday, July 22, 2015 4:04 PM, Krishna <research800@gmail.com>
> wrote:
>
>
> You can map HBase composite row key to Phoenix primary key only if
> serialization used for HBase matches with Phoenix. Ex: leading 1 byte for
> bucket, 0-byte char for separating columns, etc.
>
> If you used a different mechanism to serialize rowkey in HBase, you can
> still map it Phoenix table but declare PK as VARBINARY and see if you can
> create a UDF to separate columns.
>
> On Tuesday, July 21, 2015, Anchal Agrawal <anchal@yahoo-inc.com> wrote:
>
> Hi,
>
> I'm trying to map an existing HBase table to Phoenix. Can the existing
> HBase table's rowkey be imported as the rowkey of the Phoenix table? On
> this page (
> https://phoenix.apache.org/faq.html#How_I_map_Phoenix_table_to_an_existing_HBase_table),
> there's an example:
>
> CREATE VIEW t1 ( pk VARCHAR PRIMARY KEY, f1.val VARCHAR )
>
> Here, is pk the HBase table's column that's being used as the primary
> key, or is it a Phoenix keyword/placeholder to refer to the HBase table's
> rowkey? My table's rowkey is made up of several fields that are not stored
> as columns in that table. If I could just import the rowkey into Phoenix,
> that'd be great.
>
> Thank you!
>
> Sincerely,
> Anchal
>
>
>
>
>
>
>

Mime
View raw message