[ 
https://issues.apache.org/jira/browse/IMPALA-7140?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16775425#comment-16775425
 ] 

ASF subversion and git services commented on IMPALA-7140:
---------------------------------------------------------

Commit 4a279e1bc430b5204e59e73b3b8fb68a0b5e64ab in impala's branch 
refs/heads/2.x from Todd Lipcon
[ https://gitbox.apache.org/repos/asf?p=impala.git;h=4a279e1 ]

IMPALA-7140 (part 3): load partitions for FS tables

This implements the loading of partitions from the HMS for FS-based
tables. The LocalPartitionSpec class implements PrunablePartition based
on parsing the partition names returned from the HMS. After partitions
are pruned, the remaining partitions can be loaded by name.

With this patch, I am able to connect to an impalad running with
--use_local_catalog and issue 'show partitions functional.alltypes' with
the expected results.

A new unit test is added which shares some code with CatalogTest to
ensure that partitions can be loaded and parsed properly.

Testing partition pruning via end-to-end planning is not quite
supported yet: we need to implement 'toThriftDescriptor()' before we can
run those end-to-end tests.

Change-Id: Iddf2edbd6bdc0684560b2ecca9c4c6b6819ef1d3
Reviewed-on: http://gerrit.cloudera.org:8080/10713
Reviewed-by: Vuk Ercegovac <[email protected]>
Tested-by: Impala Public Jenkins <[email protected]>


> Build out support for HDFS tables and views in LocalCatalog
> -----------------------------------------------------------
>
>                 Key: IMPALA-7140
>                 URL: https://issues.apache.org/jira/browse/IMPALA-7140
>             Project: IMPALA
>          Issue Type: Sub-task
>          Components: Catalog, Frontend
>            Reporter: Todd Lipcon
>            Assignee: Todd Lipcon
>            Priority: Major
>             Fix For: Impala 3.1.0
>
>
> This subtask tracks the work to build out basic read-only support for HDFS 
> tables and views in the LocalCatalog implementation:
> - loading table schemas
> - loading partitions
> - loading file information from HDFS
> This work will be broken up into a number of patches to keep each piece 
> reviewable. Once this subtask is complete we should be able to plan most 
> simple read-only queries.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to