i am interested in plugins/triggers. it seems like a good way to solve
the problem of streaming data out of cassandra into, say a hadoop
cluster or some other BI backend
On 02/08/2011 02:25 PM, Jeremy Hanna wrote:
I think the plugins/triggers/coprocessors is a great discussion to get going
again if there is interest. It's been on the to-do list for a long time and
just from what I've heard anecdotally, there is a lot of interest.
On Feb 8, 2011, at 3:57 PM, David G. Boney wrote:
Hi All,
I am also interested a stored procedure feature. I spoke with Matthew Dennis at
DataStax about a month ago about some of the issues involved with implementing
a stored procedure feature. If anyone is in the Austin area and wants to get
together for a cup of coffee to discuss this feature, drop me a line.
-------------
Sincerely,
David G. Boney
dbon...@semanticartifacts.com
http://www.semanticartifacts.com
On Feb 8, 2011, at 3:47 PM, Miguel Verde wrote:
See whatever preceded https://issues.apache.org/jira/browse/CASSANDRA-380 and
the current https://issues.apache.org/jira/browse/CASSANDRA-1016 for
potentially similar ideas, as well as
https://issues.apache.org/jira/browse/CASSANDRA-1215 and
https://issues.apache.org/jira/browse/CASSANDRA-1311 for related issues.
On Tue, Feb 8, 2011 at 2:56 PM, Morten Wegelbye Nissen<m...@monit.dk> wrote:
Hello mighty developers of Cassandra,
I have been thinking of creating a feature like stored procedures for
Cassandra.
Concept is actually pretty simple add one of the javascript compilers. (
Mozilla Rhino or one alike ). Save js source in a CF in the system keyspace.
Add feature to thrift to invoke the code. Return just like get_slice.
Needless to say that the execution environment needs access to the
keyspaces and needs to be sandboxed. (ie. no access to filesystem etc. )
On the cli it would be something like;> invoke myProc param1, param2,
param3
The alternative where the expansions, like the existing once, is done by
implementing interfaces. Would require a rather complex distribution of
jars.
Now I might have the option to get this done as a educational project,
where I after the project would like to release the code to freedom.
Would a feature like that ever make it to the core of Cassandra?
./Morten
