Home | About | Sematext search-lucene.com search-hadoop.com
 Search Hadoop and all its subprojects:

Switch to Plain View
Hive >> mail # user >> how to split the value of hbase into hive


+
ylyy-1985 2012-04-18, 07:36
+
Nitin Pawar 2012-04-18, 07:42
+
ylyy-1985 2012-04-18, 08:16
Copy link to this message
-
Re: Re: how to split the value of hbase into hive
you can checkout hive trunk and do a build
it has implementation of accessing hbase tables directly through hive with
HBaseStorageHandler.

That will solve plenty of other problems which you might face during
getting data from hbase to hive tables

You can search on hive mailing lists. Vishal had solved this problem
already in another mail thread with topic "Hive query on existing HBase
table"
On Wed, Apr 18, 2012 at 1:46 PM, ylyy-1985 <[EMAIL PROTECTED]> wrote:

> **
> hi,Nitin and other friends:
>
>  I see the UDF function is suited for me, but I don't really understand
> how to use it. Right now, I have put a record into Hbase, and scan it:
>
> hbase(main):015:0> scan 'table_wyl'
> ROW                              COLUMN+CELL
>  r1                    column=key:value, timestamp=1334765068428,
> value=010,020,030
>
> then how can I get the array in hive? and insert it into a hive table?
> 2012-04-18
>  ------------------------------
>  **
> Best Regards
> Andes
>
> **
>  ------------------------------
>  *发件人:*Nitin Pawar
> *发送时间:*2012-04-18 15:43
> *主题:*Re: how to split the value of hbase into hive
> *收件人:*"user"<[EMAIL PROTECTED]>
> *抄送:*
>
> you can use the inbuilt udf split(string str, string pat) to split the
> string with the separator you want. It returns the array and you can access
> the array and insert array elements in the hive table
>
> On Wed, Apr 18, 2012 at 1:06 PM, ylyy-1985 <[EMAIL PROTECTED]> wrote:
>
>> **
>> **
>> hi,all;
>>
>> I have a problem. I have an hbase table, with key-value pair like
>> "user1001:12,1223,string". And now how can I split the value into three
>> fields "ID,phone,and memo" and then insert into hive table? does hive
>> support this?
>>
>> thanks
>>
>> 2012-04-18
>> ------------------------------
>>  Andes
>> **
>>
>
>
>
> --
> Nitin Pawar
>
>
--
Nitin Pawar
+
ylyy-1985 2012-04-18, 07:48