Sanjay Kaluskar 2013-03-22, 02:27
Ted Dunning 2013-03-22, 06:59
Reto Matter 2013-03-22, 07:27
Ted Dunning 2013-03-22, 08:38
Jacques Nadeau 2013-03-23, 18:15
As Jacques says, I am working on the SQL-to-logical translator. It's best that I work single-threaded for a while, until I get the main operators (scan, filter, project, join, union, group by) implemented. That will only take a few more weeks. Then I could use some help making the translation more robust. (Note that I am not even addressing the issue of logical-to-physical translation, and query optimization, just yet.)
Two areas you could work on in the mean time. First, kick the tires of the JDBC driver. Try it in one or two common JDBC interfaces. I'd like people to be able to connect to Drill via JDBC and have it work with a minimum of headaches.
Second, You could add a test data set and some test queries. Right now we just have donuts.json. We could use, say, an employees.json and a departments.json, and a dozen or so queries that show off the key relational operations.
On Mar 23, 2013, at 11:15 AM, Jacques Nadeau <[EMAIL PROTECTED]> wrote:
> Hello Sanjay,
> Welcome. I took a quick look at your background and seems like you have
> lots to of expertise the community could benefit from. One of the things
> that we're working on right now is the SQL query parsing. Julian Hyde is
> leading that effort under https://issues.apache.org/jira/browse/DRILL-34.
> Maybe you could help out with his work? I know he maintains a WIP
> repository @ https://github.com/julianhyde/incubator-drill. Julian, are
> the some particular pieces of the SQL parsing work that Sanjay might be
> able to help out with?
> On Thu, Mar 21, 2013 at 7:27 PM, Sanjay Kaluskar <
> [EMAIL PROTECTED]> wrote:
>> I am interested in helping with this project. I have worked on a few
>> products in the areas of databases, cep, data integration and data quality.
>> I currently work for 7 inc and we deal with the challenges that drill
>> is trying to overcome.
Timothy Chen 2013-03-25, 17:10