[
https://issues.apache.org/jira/browse/CRUNCH-442?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14055335#comment-14055335
]
Gabriel Reid commented on CRUNCH-442:
-------------------------------------
Ok, that would be great if you could post if that works or not.
The reason the snippet of code can't just be taken and put into a unit test is
that the whole issue revolves around how the hadoop command itself sets up its
class loaders. There are actually quite a few places in the existing
integration tests where Avro specific records are read via materialize, and
they all work, because all of the Avro specific classes are on the root class
path. When a job is run with the hadoop command, the Avro specific classes are
loaded from the job jar using (I think) the context class loader.
> Trying to read specific avros, getting ClassCastException
> ---------------------------------------------------------
>
> Key: CRUNCH-442
> URL: https://issues.apache.org/jira/browse/CRUNCH-442
> Project: Crunch
> Issue Type: Bug
> Components: IO
> Affects Versions: 0.10.0
> Reporter: Jason Gauci
>
> I have code that looks like this:
> ArrayList<MySpecificAvro> data =
> Lists.newArrayList(pipeline.read(From.avroFile(
> "/user/jgauci/data.avro",
>
> Avros.specifics(MySpecificAvro.class))).materialize());
> for (MySpecificAvro msa : data) {
> System.out.println(msa.getName());
> }
> And I get this error:
> Exception in thread "main" java.lang.ClassCastException:
> org.apache.avro.generic.GenericData$Record cannot be cast to MySpecificAvro
> The error fires inside the for loop after the materialize has completed.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
