I have created a jira for this: APEXMALHAR-1985 <https://issues.apache.org/jira/browse/APEXMALHAR-1985>
On Wed, Jan 27, 2016 at 4:13 PM, Sandeep Deshmukh <[email protected]> wrote: > I think it should be token(%s). > > Regards, > Sandeep > > On Mon, Jan 25, 2016 at 5:49 PM, Priyanka Gugale <[email protected]> > wrote: > > > Hi, > > > > The CassandraPOJOInputOperator in Malhar repository seems to scan the > table > > from start row provided by user configuration. The query is accepted in > the > > form "select * from %t where token(%p) > %s limit %l" where %p is primary > > key, %s is start value. My question is: > > > > *Why %s is start value?*, it should be token(start value). Here we should > > compare against the hash value of the key and not the key value itself. > > Please refer datastax discussion thread on how to scan table > > < > > > http://mail-archives.apache.org/mod_mbox/cassandra-user/201308.mbox/%3cca+vsrlpezqsv_85yqqr-99f-q2pmv2laj825wzrl7meaf_j...@mail.gmail.com%3E > > > > > . > > > > Can someone please confirm if %s is set to primary key value for some > > reason or this was mistaken? > > > > -Priyanka > > >
