You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@zeppelin.apache.org by Furcy Pin <fu...@flaminem.com> on 2015/01/13 18:46:35 UTC

SparkHiveInterpreter

Hi Zeppelin team,


first, let me congratulate you for this awesome project!
I heard about it this morning, tried and installed it without too much
hassle
(except to use it with Yarn) and this is a really great interface!

I wanted to try using it to query directly Hive tables, so I did some quick
hacking
to try it, and replaced every occurrence of SQLContext with HiveContext.
After solving a few maven voodoo issue, it finally worked.

In short, the question I wanted to ask, is :

Are you planning to implement soon a SparkHiveInterpreter or something
similar?
It could work with a %hql or a %hive prefix, and would not be hard to do
implement
starting from the SparkSQLInterpreter.

If you are interested, I would be glad to contribute and submit an
implementation.

What do you think?

Cheers,

Furcy

Re: SparkHiveInterpreter

Posted by "Kevin (Sangwoo) Kim" <ke...@apache.org>.
Hi Furcy,
Glad that we have another great developer here!

I like your plan and %hive looks better to me.
And also you can look at this PR
https://github.com/NFLabs/zeppelin/pull/235
for better understanding for our plan!

Regards,
Kevin

On 2015년 1월 14일 (수) 02:48 Furcy Pin <fu...@flaminem.com> wrote:

> Hi Zeppelin team,
>
>
> first, let me congratulate you for this awesome project!
> I heard about it this morning, tried and installed it without too much
> hassle
> (except to use it with Yarn) and this is a really great interface!
>
> I wanted to try using it to query directly Hive tables, so I did some quick
> hacking
> to try it, and replaced every occurrence of SQLContext with HiveContext.
> After solving a few maven voodoo issue, it finally worked.
>
> In short, the question I wanted to ask, is :
>
> Are you planning to implement soon a SparkHiveInterpreter or something
> similar?
> It could work with a %hql or a %hive prefix, and would not be hard to do
> implement
> starting from the SparkSQLInterpreter.
>
> If you are interested, I would be glad to contribute and submit an
> implementation.
>
> What do you think?
>
> Cheers,
>
> Furcy
>