[
https://issues.apache.org/jira/browse/FLINK-1303?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14322116#comment-14322116
]
Yiannis Gkoufas edited comment on FLINK-1303 at 2/15/15 7:23 PM:
-----------------------------------------------------------------
Actually, my intention was to benchmark Flink with working with some
Lzo-compressed csv files that I have.
Is there something more needed than including the lzo jar in the lib folder of
flink? Do I need to set the path for the native libraries as well?
Also, I am getting this error with the latest 0.9 version for the above line:
could not find implicit value for parameter tpe:
org.apache.flink.api.common.typeinfo.TypeInformation[(org.apache.hadoop.io.LongWritable,
org.apache.hadoop.io.Text)]
was (Author: johngouf):
Actually, my intention was to benchmark Flink with working with some
Lzo-compressed csv files that I have.
Is there something more needed than including the lzo jar in the lib folder of
flink? Do I need to set the path for the native libraries as well?
> HadoopInputFormat does not work with Scala API
> ----------------------------------------------
>
> Key: FLINK-1303
> URL: https://issues.apache.org/jira/browse/FLINK-1303
> Project: Flink
> Issue Type: Sub-task
> Components: Scala API
> Reporter: Aljoscha Krettek
> Assignee: Aljoscha Krettek
> Fix For: 0.9, 0.8.1
>
>
> It fails because the HadoopInputFormat uses the Flink Tuple2 type. For this,
> type extraction fails at runtime.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
