[
https://issues.apache.org/jira/browse/TOREE-345?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15540765#comment-15540765
]
Hollin Wilkins commented on TOREE-345:
--------------------------------------
It would be nice if toree used the same code paths to add dependencies as Spark
does, as Spark already handles excluding dependencies that interfere quite
nicely. Then toree can add the dependencies for iteself with the exclusions it
knows about, and Spark can add the dependencies with the exclusions it knows
about.
> Spark 2.0.0 Fails when trying to use spark-avro
> -----------------------------------------------
>
> Key: TOREE-345
> URL: https://issues.apache.org/jira/browse/TOREE-345
> Project: TOREE
> Issue Type: Bug
> Environment: Spark 2.0.0, Scala 2.11, Hadoop 2.7, Toree Git SHA:
> 7c1bfb6df7130477c558e69bbb518b0af364e06a
> Reporter: Hollin Wilkins
>
> When trying to use the spark-avro project to load Avro files from Jupyter, we
> get errors.
> First:
> {code}
> %AddDeps com.databricks spark-avro_2.11 3.0.1 --transitive --trace
> {code}
> Then try to load an Avro file and show it:
> {code}
> spark.sqlContext.read.format("com.databricks.spark.avro").load("/tmp/test.avro").show()
> {code}
> And we get an error. I will attach the trace as a file.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
