[
https://issues.apache.org/jira/browse/CARBONDATA-855?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15956410#comment-15956410
]
sehriff commented on CARBONDATA-855:
------------------------------------
import org.apache.spark.sql.CarbonContext
val cc = new CarbonContext(sc, "hdfs:///user/e_carbon/carbon.store")
cc.sql("CREATE TABLE IF NOT EXISTS e_carbon.test_table (id string, name string,
city string, age Int) STORED BY 'carbondata'")
cc.sql("LOAD DATA INPATH 'hdfs:///user/e_carbon/sample.csv' INTO TABLE
e_carbon.test_table")
cc.sql("update e_carbon.test_table set (name,age)=('jerry',25) where
city='wuhan'").show
> Can't update successfully.
> --------------------------
>
> Key: CARBONDATA-855
> URL: https://issues.apache.org/jira/browse/CARBONDATA-855
> Project: CarbonData
> Issue Type: Bug
> Environment: spark1.6.0 carbon1.0.0
> Reporter: sehriff
> Attachments: metadataupdate.txt, updatefail1.txt
>
>
> can't update carbondata table neither using cc.sql("update...).show nor hive
> table under hive shell(update hive table set ... where...).Most part of
> executing cc.sql logs【attachment updatefail1.txt】 is INFO,whick seems
> normal,but actually it doesn't update successfully,values should be updated
> remains unchanged.
> Because carboncontext extends from hivecontext,so I was wondering if I should
> change hive configurations to make updating hive table under hive shell
> working then i can try update carbondata table successfully.
> Also,I didn't see any updatedelta files in hdfs but tableupdatestatus files
> under metadata directory【attachment metadataupdate.txt】,maybe there's
> something configurations should be configured in hdfs?
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
