[
https://issues.apache.org/jira/browse/SQOOP-1450?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14102741#comment-14102741
]
Pratik Khadloya commented on SQOOP-1450:
----------------------------------------
Oh i just figured that sqoop already writes the schema file to a temp
directory, so we may not need this jira.
I see it being created on my local machine at
/tmp/sqoop-pkhadloya/compile/6b19077ea09b86e6b88ccd9c0740894d/campaign.avsc
> Copy the avro schema file to hdfs for AVRO based import
> -------------------------------------------------------
>
> Key: SQOOP-1450
> URL: https://issues.apache.org/jira/browse/SQOOP-1450
> Project: Sqoop
> Issue Type: New Feature
> Affects Versions: 1.4.5
> Reporter: Pratik Khadloya
> Assignee: Pratik Khadloya
> Priority: Critical
> Labels: newbie
>
> Currently sqoop imports the avro file directly to hdfs.
> The schema of the avro files is very important to have. Currently the only
> way to get the schema from the avro file is to follow the steps listed in
> http://dev.gbif.org/wiki/display/DEV/Sqoop+to+Hive+using+Avro
> Apart from being able to programmatically use the schema,it also enables us
> to convert the avro file to Parquet too using the avro2parquet converter.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
