[
https://issues.apache.org/jira/browse/HUDI-658?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17053659#comment-17053659
]
Vinoth Chandar commented on HUDI-658:
-------------------------------------
May be.. but I have a concern with how we approaching this multi engine
refactoring.. we need a top down design and probably proof of concept few
things at the high level first, before we can go down the list to smaller items
like this.. We cannot do this class by class IMO.. lets hash out a high level
plan first?
> Make ClientUtils spark-free
> ---------------------------
>
> Key: HUDI-658
> URL: https://issues.apache.org/jira/browse/HUDI-658
> Project: Apache Hudi (incubating)
> Issue Type: Sub-task
> Reporter: vinoyang
> Assignee: vinoyang
> Priority: Major
>
> {{ClientUtils#createMetaClient}} require {{JavaSparkContext}} only for
> getting the hadoop configuration obejct. We can pass the {{Configuration}}
> object directly so that we can make {{ClientUtils}} spark-free.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
