Also tried: client.multiget_slice(ksName, mKeyList, parent, predicate, ConsistencyLevel.ONE);
Gets the same result, if I put the super_column as null or empty array, it returns nothing, but if I give the correct super_column value, it returns expected 3 columns. I'm using the Keyspace1.Super1. Regards, Dop -----Original Message----- From: Dop Sun [mailto:su...@dopsun.com] Sent: Monday, April 12, 2010 9:03 PM To: user@cassandra.apache.org Subject: RE: FW: Is there any way to enable the multiple super column names at a time? Thanks for your quick reply. Actually, I tried to leave it as null or byte[0], nothing returned, and if I put right super column value, it returns 3 items actually (in my simple test case). I'm using get_range_slices. I'm calling two times with all other arguments same. I'm using the 0.6 rc1 via Thrift API. Regards, Dop -----Original Message----- From: Jonathan Ellis [mailto:jbel...@gmail.com] Sent: Monday, April 12, 2010 8:32 PM To: user@cassandra.apache.org Subject: Re: FW: Is there any way to enable the multiple super column names at a time? The supercolumn parameter to ColumnParent is optional precisely so you can do this. http://wiki.apache.org/cassandra/API On Mon, Apr 12, 2010 at 6:37 AM, Dop Sun <su...@dopsun.com> wrote: > I guess I'm putting in the wrong email list. Sorry for this. > > My question in short is: whether I can get a certain range of super column > in a single query? > > I have checked the following APIs, but looks they don't work for this: > get: for a single column > get_slice: contains column_parent as parameter, means a single CF + SC > multiget_slice: contains column_parent as parameter, means a single CF + SC > get_range_slices: contains column_parent as parameter, means a single CF + > SC > > Regards, > Dop > > -----Original Message----- > From: Dop Sun [mailto:su...@dopsun.com] > Sent: Monday, April 12, 2010 6:41 AM > To: d...@cassandra.apache.org > Subject: Is there any way to enable the multiple super column names at a > time? > > Hi, > > I don't know whether I have missed anything here, but can we get columns for > multiple Super Column Names at a time, like what we can do for Keys (by key > range or list of keys) or columns (by column range or column list)? > > While I'm doing the work with Cassandra, I usually think the structure like > this: > > KS - CF - Key - Super Column - Column > > But looks like I should change it to something like: > > KS - CF - Super Column - Key - Column > > Since CF-SuperColumn are combined within a single parameter: ColumnParent, > and they are usually unique (single) in a query. > > Regards, > Dop > > > > >
