asterix-app/src/test/resources/runtimets/queries/dapd/q3.aql

drop dataverse test if exists;
create dataverse test;
use dataverse test;


create type AddressType as closed {
 street: string,
 city: string,
 zip: string,
 latlong: point2d
}

create type UserType as open{
 name: string,
 interests: {{string}},
 address: AddressType,
 member_of: {{
  {
    sig_id: int32,
    chapter_name: string,
    member_since: date
  }
}}
}


create nodegroup group1  if not exists on nc1, nc2;

create dataset User(UserType)
 partitioned by key name on group1;

load dataset User 
using "edu.uci.ics.asterix.external.dataset.adapter.NCFileSystemAdapter"
(("path"="nc1://data/events/tiny/user.adm"),("format"="json")) pre-sorted;


write output to nc1:'rttest/dapd_q3.adm';

set simthreshold '.5f';

for $user in dataset('User')
let $similar_users :=
  for $similar_user in dataset('User')
  where $user.name != $similar_user.name
       and $user.interests ~= $similar_user.interests
  let $sim := similarity-jaccard($user.interests, $similar_user.interests)
  order by $sim desc
  limit 10
  return { "user_name": $similar_user.name }
order by $user.name  
return { "user_name" : $user.name, "similar_users" : $similar_users }