Maybe it's better to try bucketed hive table? https://cwiki.apache.org/confluence/display/Hive/LanguageManual+DDL+BucketedTables
2013/11/18 Ruslan Al-Fakikh <[email protected]> > Hi soniya, > > In you example you are hard-coding the ID to 2 in your filter statement. > You could also hard-code it in STORE: > Store A into '/main/abc'/2 ; > If you want to separate the rows by a value of a field then you could try > MultiStorage from piggybank. > > Thanks > > > On Mon, Nov 18, 2013 at 6:01 AM, soniya B <[email protected]> > wrote: > > > Hi friends, > > > > I have a sample script below > > > > AA = .... > > A= filter AA by ID = 2; > > B= foreach A generate ID,name,description; > > Store A into '/main/abc' ; > > > > > > I want to store the ID value in the end of filepath. i.e 'main/abc/ID' . > > > > here, i have to get filepath as '/main/abc/2' . > > > > Could you please help me? if is not possible in pig latin then how to do > it > > in the shell script? > > > > Thanks > > Soniya > > >
