Hi James,

You can work around the error by building with Java 7. You could also build
your Scala object model as a separate project and depend on the released
version of Parquet. That way you don't need to build it at all.

I'm assuming that you want to build an object model rather than a native
Scala implementation. Is that right? If so, then I recommend taking a look
at parquet-avro, which is the Avro object model on top of Parquet. That's
what Spark's implementation was modeled on as well. The dependency for
parquet-avro is on parquet-hadoop. You don't have to pull in any modules
beyond that.

rb

On Tue, Jun 14, 2016 at 9:51 AM, James M Decker <[email protected]> wrote:

> Hi there,
>
>
> I'm looking to implement a Parquet reader in Scala. I've found the
> parquet-mr project, but it doesn't seem to build for me. I found a few
> Stackoverflow posts (
> http://stackoverflow.com/questions/31229445/build-failure-apache-parquet-mr-source-mvn-install-failure,
> and http://stackoverflow.com/questions/33492447/parquet-file-reader?lq=1),
> where other people are reporting the same issue.
>
>
> Either way, my main question is this: In implementing a Parquet reader, is
> it really required that I add the entire parquet-mr project, or is there a
> subset I could use for just the reading capabilities?
>
>
> Thank you,
>
>
> James Decker
>
> [email protected]
>
>
>


-- 
Ryan Blue
Software Engineer
Netflix

Reply via email to