[Mondrian] Skip the metadata lookup
tom at analytical-labs.com
Tue Oct 27 19:59:08 EDT 2015
Hmm, i'll go back and check again, but I saw similar when I was testing
with Spark a few months ago as well, unless you did a CREATE TABLE and
store the stuff in the Hive Metastore, at which point suddenly the same
schema started working.
Thanks for the clarification anyway Julian.
On 27 October 2015 at 23:53, Julian Hyde <julianhyde at gmail.com> wrote:
> I don’t recall what the code does, but it was definitely my intention when
> I was designing PhysicalSchema that a schema designer could provide enough
> metadata that you could validate a schema “disconnected”. (I don’t mean
> literally disconnected, but relying on the metadata in the schema file.)
> That would include explicit lists of the columns in each table, and the
> types of those columns.
> And if the actual types in the JDBC source don’t match, presumably there
> would be a way to reconcile. E.g. ignore what the JDBC source says and
> apply casts everywhere.
> In other words, what you’re asking for is reasonable.
> > On Oct 27, 2015, at 3:26 AM, Tom Barber <tom at analytical-labs.com> wrote:
> > Hello folks
> > Here's a question for you. I've been playing with some NOSQL stuff
> recently and I'm interested in sitting Mondrian over a few different tools,
> Drill for example would be great.
> > One problem though seems to be that Mondrian 4 reads the database
> metadata and doesn't just trust the developer when it comes to defining the
> schema. But in drill for example unless you store your table in Hive, show
> tables and metadata lookup doesn't exist because whilst the table exists..
> it doesn't, but of course select x,y,z from tablex does work.
> > So..... in Mondrian 4, is there currently any way of bypassing the
> schema validation? Or do I need to put some effort into allowing that?
> > Thanks
> > Tom
> > _______________________________________________
> > Mondrian mailing list
> > Mondrian at pentaho.org
> > http://lists.pentaho.org/mailman/listinfo/mondrian
> Mondrian mailing list
> Mondrian at pentaho.org
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Mondrian