[
https://issues.apache.org/jira/browse/HBASE-26491?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17451790#comment-17451790
]
mengdou commented on HBASE-26491:
---------------------------------
[~zhangduo] May I have a look at the issues list of hbase-connectors-1.0.1 ?
During usage of this connector recently, I encountered other problems. For
example, the property of "hbase.spark.pushdown.columnfilter" is unsupported,
because it is available only when it appears in the options of a table
definition rather than spark configuration.
!image-2021-12-01-20-52-11-664.png!
But, in the javadoc of class HBaseSparkConf, it says we can set it in both
SparkConf and table options. Is it as expected?
!image-2021-12-01-20-53-05-405.png!
> I can't read/write a Hbase table by spark-hbase connector when the table is
> in non-default namespace
> -----------------------------------------------------------------------------------------------------
>
> Key: HBASE-26491
> URL: https://issues.apache.org/jira/browse/HBASE-26491
> Project: HBase
> Issue Type: Bug
> Components: hbase-connectors
> Affects Versions: 1.0.0
> Reporter: mengdou
> Priority: Minor
> Attachments: image-2021-11-26-17-32-53-507.png,
> image-2021-12-01-20-52-11-664.png, image-2021-12-01-20-53-05-405.png
>
>
> I found I can't read/write a Hbase table by spark-hbase connector when the
> hbase table is in a non-default namespace.
>
> Because when spark opens a table(related to a hbase table), it creates a
> HBaseRelation instance first, and initializes a HBaseTableCatalog from the
> table definition saved in spark catalog. But in the function 'convert' the
> field 'tableCatalog' is constructed from a string template, in which the
> namespace is set as 'default', leading to a wrong namespace. This namespace
> is not the one defined when user created the table before.
>
> Pls have a look:
> !image-2021-11-26-17-32-53-507.png!
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
