blob: ca2827de343a8ce91b688aeec015ae6f7e549937 [file] [log] [blame]
/*
* Description : Fuzzy joins two datasets, DBLP and CSX, based on the edit-distance function of their authors.
* DBLP has a 3-gram index on authors, and we expect the join to be transformed into an indexed nested-loop join.
* We expect the top-level equi join introduced because of surrogate optimization to be removed, since it is not necessary.
* Success : Yes
*/
drop dataverse test if exists;
create dataverse test;
use dataverse test;
create type DBLPType as closed {
id: int32,
dblpid: string,
title: string,
authors: string,
misc: string
}
create type CSXType as closed {
id: int32,
csxid: string,
title: string,
authors: string,
misc: string
}
create dataset DBLP(DBLPType) primary key id;
create dataset CSX(CSXType) primary key id;
create index ngram_index on DBLP(authors) type ngram(3);
write output to nc1:"rttest/inverted-index-join-noeqjoin_ngram-edit-distance.adm";
for $a in dataset('DBLP')
for $b in dataset('CSX')
where edit-distance($a.authors, $b.authors) < 3 and $a.id < $b.id
return {"aauthors": $a.authors, "bauthors": $b.authors}