Reintegrated asterix-fuzzy.
git-svn-id: https://asterixdb.googlecode.com/svn/branches/asterix_stabilization@437 eaa15691-b419-025a-1212-ee371bd00084
diff --git a/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ngram-contains-panic.aql b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ngram-contains-panic.aql
new file mode 100644
index 0000000..66ef43b
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ngram-contains-panic.aql
@@ -0,0 +1,33 @@
+/*
+ * Description : Tests whether an ngram_index is applied to optimize a selection query using the contains function.
+ * The index should *not* be applied (see below).
+ * 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 dataset DBLP(DBLPType) partitioned by key id;
+
+load dataset DBLP
+using "edu.uci.ics.asterix.external.dataset.adapter.NCFileSystemAdapter"
+(("path"="nc1://data/dblp-small/dblp-small-id.txt"),("format"="delimited-text"),("delimiter"=":")) pre-sorted;
+
+create index ngram_index on DBLP(title) type ngram(3);
+
+write output to nc1:"rttest/inverted-index-basic_ngram-contains-panic.adm";
+
+// Cannot optimize this query because the string constant is shorter than the gram length.
+for $o in dataset('DBLP')
+where contains($o.title, "Mu")
+order by $o.id
+return $o
\ No newline at end of file
diff --git a/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ngram-contains.aql b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ngram-contains.aql
new file mode 100644
index 0000000..0aa8056
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ngram-contains.aql
@@ -0,0 +1,32 @@
+/*
+ * Description : Tests whether an ngram_index is applied to optimize a selection query using the contains function.
+ * The index should be applied.
+ * 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 dataset DBLP(DBLPType) partitioned by key id;
+
+load dataset DBLP
+using "edu.uci.ics.asterix.external.dataset.adapter.NCFileSystemAdapter"
+(("path"="nc1://data/dblp-small/dblp-small-id.txt"),("format"="delimited-text"),("delimiter"=":")) pre-sorted;
+
+create index ngram_index on DBLP(title) type ngram(3);
+
+write output to nc1:"rttest/inverted-index-basic_ngram-contains.adm";
+
+for $o in dataset('DBLP')
+where contains($o.title, "Multimedia")
+order by $o.id
+return $o
\ No newline at end of file
diff --git a/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ngram-edit-distance-check-panic.aql b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ngram-edit-distance-check-panic.aql
new file mode 100644
index 0000000..21e41f8
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ngram-edit-distance-check-panic.aql
@@ -0,0 +1,33 @@
+/*
+ * Description : Tests whether an ngram_index is applied to optimize a selection query using the edit-distance-check function on strings.
+ * The index should *not* be applied (see below).
+ * 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 dataset DBLP(DBLPType) partitioned by key id;
+
+load dataset DBLP
+using "edu.uci.ics.asterix.external.dataset.adapter.NCFileSystemAdapter"
+(("path"="nc1://data/dblp-small/dblp-small-id.txt"),("format"="delimited-text"),("delimiter"=":")) pre-sorted;
+
+create index ngram_index on DBLP(authors) type ngram(3);
+
+write output to nc1:"rttest/inverted-index-basic_ngram-edit-distance-check-panic.adm";
+
+// This query cannot be optimized with an index, based on the high edit distance.
+for $o in dataset('DBLP')
+let $ed := edit-distance-check($o.authors, "Amihay Motro", 5)
+where $ed[0]
+return $o
\ No newline at end of file
diff --git a/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ngram-edit-distance-check.aql b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ngram-edit-distance-check.aql
new file mode 100644
index 0000000..ff3728d
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ngram-edit-distance-check.aql
@@ -0,0 +1,31 @@
+/*
+ * Description : Tests whether an ngram_index is applied to optimize a selection query using the edit-distance-check function on strings.
+ * The index should be applied.
+ * 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 dataset DBLP(DBLPType) partitioned by key id;
+
+load dataset DBLP
+using "edu.uci.ics.asterix.external.dataset.adapter.NCFileSystemAdapter"
+(("path"="nc1://data/dblp-small/dblp-small-id.txt"),("format"="delimited-text"),("delimiter"=":")) pre-sorted;
+
+create index ngram_index on DBLP(authors) type ngram(3);
+
+write output to nc1:"rttest/inverted-index-basic_ngram-edit-distance-check.adm";
+
+for $o in dataset('DBLP')
+where edit-distance-check($o.authors, "Amihay Motro", 1)[0]
+return $o
\ No newline at end of file
diff --git a/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ngram-edit-distance-panic.aql b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ngram-edit-distance-panic.aql
new file mode 100644
index 0000000..c993c57
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ngram-edit-distance-panic.aql
@@ -0,0 +1,32 @@
+/*
+ * Description : Tests whether an ngram_index is applied to optimize a selection query using the edit-distance function on strings.
+ * The index should *not* be applied (see below).
+ * 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 dataset DBLP(DBLPType) partitioned by key id;
+
+load dataset DBLP
+using "edu.uci.ics.asterix.external.dataset.adapter.NCFileSystemAdapter"
+(("path"="nc1://data/dblp-small/dblp-small-id.txt"),("format"="delimited-text"),("delimiter"=":")) pre-sorted;
+
+create index ngram_index on DBLP(authors) type ngram(3);
+
+write output to nc1:"rttest/inverted-index-basic_ngram-edit-distance-panic.adm";
+
+// This query cannot be optimized with an index, based on the high edit distance.
+for $o in dataset('DBLP')
+where edit-distance($o.authors, "Amihay Motro") <= 5
+return $o
\ No newline at end of file
diff --git a/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ngram-edit-distance.aql b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ngram-edit-distance.aql
new file mode 100644
index 0000000..a143cde
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ngram-edit-distance.aql
@@ -0,0 +1,31 @@
+/*
+ * Description : Tests whether an ngram_index is applied to optimize a selection query using the edit-distance function on strings.
+ * The index should be applied.
+ * 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 dataset DBLP(DBLPType) partitioned by key id;
+
+load dataset DBLP
+using "edu.uci.ics.asterix.external.dataset.adapter.NCFileSystemAdapter"
+(("path"="nc1://data/dblp-small/dblp-small-id.txt"),("format"="delimited-text"),("delimiter"=":")) pre-sorted;
+
+create index ngram_index on DBLP(authors) type ngram(3);
+
+write output to nc1:"rttest/inverted-index-basic_ngram-edit-distance.adm";
+
+for $o in dataset('DBLP')
+where edit-distance($o.authors, "Amihay Motro") <= 1
+return $o
\ No newline at end of file
diff --git a/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ngram-fuzzyeq-edit-distance.aql b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ngram-fuzzyeq-edit-distance.aql
new file mode 100644
index 0000000..87de1fb
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ngram-fuzzyeq-edit-distance.aql
@@ -0,0 +1,34 @@
+/*
+ * Description : Tests whether an ngram_index is applied to optimize a selection query with ~= using edit-distance on strings.
+ * The index should be applied.
+ * 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 dataset DBLP(DBLPType) partitioned by key id;
+
+load dataset DBLP
+using "edu.uci.ics.asterix.external.dataset.adapter.NCFileSystemAdapter"
+(("path"="nc1://data/dblp-small/dblp-small-id.txt"),("format"="delimited-text"),("delimiter"=":")) pre-sorted;
+
+create index ngram_index on DBLP(authors) type ngram(3);
+
+write output to nc1:"rttest/inverted-index-basic_ngram-fuzzyeq-edit-distance.adm";
+
+set simfunction 'edit-distance';
+set simthreshold '1';
+
+for $o in dataset('DBLP')
+where $o.authors ~= "Amihay Motro"
+return $o
\ No newline at end of file
diff --git a/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ngram-fuzzyeq-jaccard.aql b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ngram-fuzzyeq-jaccard.aql
new file mode 100644
index 0000000..3db4c92
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ngram-fuzzyeq-jaccard.aql
@@ -0,0 +1,34 @@
+/*
+ * Description : Tests whether an ngram_index is applied to optimize a selection query with ~= using Jaccard on 3-gram tokens.
+ * The index should be applied.
+ * 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 dataset DBLP(DBLPType) partitioned by key id;
+
+load dataset DBLP
+using "edu.uci.ics.asterix.external.dataset.adapter.NCFileSystemAdapter"
+(("path"="nc1://data/dblp-small/dblp-small-id.txt"),("format"="delimited-text"),("delimiter"=":")) pre-sorted;
+
+create index ngram_index on DBLP(title) type ngram(3);
+
+write output to nc1:"rttest/inverted-index-basic_ngram-fuzzyeq-jaccard.adm";
+
+set simfunction 'jaccard';
+set simthreshold '0.8f';
+
+for $o in dataset('DBLP')
+where gram-tokens($o.title, 3, false) ~= gram-tokens("Transactions for Cooperative Environments", 3, false)
+return $o
diff --git a/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ngram-jaccard-check.aql b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ngram-jaccard-check.aql
new file mode 100644
index 0000000..c39d531
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ngram-jaccard-check.aql
@@ -0,0 +1,31 @@
+/*
+ * Description : Tests whether an ngram_index is applied to optimize a selection query using the similarity-jaccard-check function on 3-gram tokens.
+ * The index should be applied.
+ * 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 dataset DBLP(DBLPType) partitioned by key id;
+
+load dataset DBLP
+using "edu.uci.ics.asterix.external.dataset.adapter.NCFileSystemAdapter"
+(("path"="nc1://data/dblp-small/dblp-small-id.txt"),("format"="delimited-text"),("delimiter"=":")) pre-sorted;
+
+create index ngram_index on DBLP(title) type ngram(3);
+
+write output to nc1:"rttest/inverted-index-basic_ngram-jaccard-check.adm";
+
+for $o in dataset('DBLP')
+where similarity-jaccard-check(gram-tokens($o.title, 3, false), gram-tokens("Transactions for Cooperative Environments", 3, false), 0.5f)[0]
+return $o
\ No newline at end of file
diff --git a/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ngram-jaccard.aql b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ngram-jaccard.aql
new file mode 100644
index 0000000..3855ade
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ngram-jaccard.aql
@@ -0,0 +1,31 @@
+/*
+ * Description : Tests whether an ngram_index is applied to optimize a selection query using the similarity-jaccard function on 3-gram tokens.
+ * The index should be applied.
+ * 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 dataset DBLP(DBLPType) partitioned by key id;
+
+load dataset DBLP
+using "edu.uci.ics.asterix.external.dataset.adapter.NCFileSystemAdapter"
+(("path"="nc1://data/dblp-small/dblp-small-id.txt"),("format"="delimited-text"),("delimiter"=":")) pre-sorted;
+
+create index ngram_index on DBLP(title) type ngram(3);
+
+write output to nc1:"rttest/inverted-index-basic_ngram-jaccard.adm";
+
+for $o in dataset('DBLP')
+where similarity-jaccard(gram-tokens($o.title, 3, false), gram-tokens("Transactions for Cooperative Environments", 3, false)) >= 0.5f
+return $o
diff --git a/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/olist-edit-distance-check-panic.aql b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/olist-edit-distance-check-panic.aql
new file mode 100644
index 0000000..6a6700a
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/olist-edit-distance-check-panic.aql
@@ -0,0 +1,40 @@
+/*
+ * Description : Tests whether a keyword index is applied to optimize a selection query using the edit-distance-check function on lists.
+ * The index should *not* be applied (see below).
+ * Success : Yes
+ */
+
+drop dataverse test if exists;
+create dataverse test;
+use dataverse test;
+
+create type AddressType as closed {
+ number: int32,
+ street: string,
+ city: string
+}
+
+create type CustomerType as closed {
+ cid: int32,
+ name: string,
+ age: int32?,
+ address: AddressType?,
+ interests: [string],
+ children: [ { name: string, age: int32? } ]
+}
+
+create dataset Customers(CustomerType) partitioned by key cid;
+
+load dataset Customers
+using "edu.uci.ics.asterix.external.dataset.adapter.NCFileSystemAdapter"
+(("path"="nc1://data/semistructured/co1k_olist/customer.adm"),("format"="adm"));
+
+create index interests_index on Customers(interests) type keyword;
+
+write output to nc1:"rttest/inverted-index-basic_olist-edit-distance-check-panic.adm";
+
+// Index should not be applied because all list elements can be modified by 3 edit operations.
+for $c in dataset('Customers')
+where edit-distance-check($c.interests, ["computers", "wine", "walking"], 3)[0]
+order by $c.cid
+return $c
diff --git a/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/olist-edit-distance-check.aql b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/olist-edit-distance-check.aql
new file mode 100644
index 0000000..f705a4a
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/olist-edit-distance-check.aql
@@ -0,0 +1,39 @@
+/*
+ * Description : Tests whether a keyword index is applied to optimize a selection query using the edit-distance-check function on lists.
+ * The index should be applied.
+ * Success : Yes
+ */
+
+drop dataverse test if exists;
+create dataverse test;
+use dataverse test;
+
+create type AddressType as closed {
+ number: int32,
+ street: string,
+ city: string
+}
+
+create type CustomerType as closed {
+ cid: int32,
+ name: string,
+ age: int32?,
+ address: AddressType?,
+ interests: [string],
+ children: [ { name: string, age: int32? } ]
+}
+
+create dataset Customers(CustomerType) partitioned by key cid;
+
+load dataset Customers
+using "edu.uci.ics.asterix.external.dataset.adapter.NCFileSystemAdapter"
+(("path"="nc1://data/semistructured/co1k_olist/customer.adm"),("format"="adm"));
+
+create index interests_index on Customers(interests) type keyword;
+
+write output to nc1:"rttest/inverted-index-basic_olist-edit-distance-check.adm";
+
+for $c in dataset('Customers')
+where edit-distance-check($c.interests, ["computers", "wine", "walking"], 1)[0]
+order by $c.cid
+return $c
diff --git a/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/olist-edit-distance-panic.aql b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/olist-edit-distance-panic.aql
new file mode 100644
index 0000000..1fc7906
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/olist-edit-distance-panic.aql
@@ -0,0 +1,40 @@
+/*
+ * Description : Tests whether a keyword index is applied to optimize a selection query using the edit-distance function on lists.
+ * The index should *not* be applied (see below).
+ * Success : Yes
+ */
+
+drop dataverse test if exists;
+create dataverse test;
+use dataverse test;
+
+create type AddressType as closed {
+ number: int32,
+ street: string,
+ city: string
+}
+
+create type CustomerType as closed {
+ cid: int32,
+ name: string,
+ age: int32?,
+ address: AddressType?,
+ interests: [string],
+ children: [ { name: string, age: int32? } ]
+}
+
+create dataset Customers(CustomerType) partitioned by key cid;
+
+load dataset Customers
+using "edu.uci.ics.asterix.external.dataset.adapter.NCFileSystemAdapter"
+(("path"="nc1://data/semistructured/co1k_olist/customer.adm"),("format"="adm"));
+
+create index interests_index on Customers(interests) type keyword;
+
+write output to nc1:"rttest/inverted-index-basic_olist-edit-distance-panic.adm";
+
+// Index should not be applied because all list elements can be modified by 3 edit operations.
+for $c in dataset('Customers')
+where edit-distance($c.interests, ["computers", "wine", "walking"]) <= 3
+order by $c.cid
+return $c
diff --git a/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/olist-edit-distance.aql b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/olist-edit-distance.aql
new file mode 100644
index 0000000..70268f5
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/olist-edit-distance.aql
@@ -0,0 +1,39 @@
+/*
+ * Description : Tests whether a keyword index is applied to optimize a selection query using the edit-distance function on lists.
+ * The index should be applied.
+ * Success : Yes
+ */
+
+drop dataverse test if exists;
+create dataverse test;
+use dataverse test;
+
+create type AddressType as closed {
+ number: int32,
+ street: string,
+ city: string
+}
+
+create type CustomerType as closed {
+ cid: int32,
+ name: string,
+ age: int32?,
+ address: AddressType?,
+ interests: [string],
+ children: [ { name: string, age: int32? } ]
+}
+
+create dataset Customers(CustomerType) partitioned by key cid;
+
+load dataset Customers
+using "edu.uci.ics.asterix.external.dataset.adapter.NCFileSystemAdapter"
+(("path"="nc1://data/semistructured/co1k_olist/customer.adm"),("format"="adm"));
+
+create index interests_index on Customers(interests) type keyword;
+
+write output to nc1:"rttest/inverted-index-basic_olist-edit-distance.adm";
+
+for $c in dataset('Customers')
+where edit-distance($c.interests, ["computers", "wine", "walking"]) <= 1
+order by $c.cid
+return $c
diff --git a/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/olist-fuzzyeq-edit-distance.aql b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/olist-fuzzyeq-edit-distance.aql
new file mode 100644
index 0000000..90ce266
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/olist-fuzzyeq-edit-distance.aql
@@ -0,0 +1,42 @@
+/*
+ * Description : Tests whether a keyword index is applied to optimize a selection query with ~= using edit-distance on lists.
+ * The index should be applied.
+ * Success : Yes
+ */
+
+drop dataverse test if exists;
+create dataverse test;
+use dataverse test;
+
+create type AddressType as closed {
+ number: int32,
+ street: string,
+ city: string
+}
+
+create type CustomerType as closed {
+ cid: int32,
+ name: string,
+ age: int32?,
+ address: AddressType?,
+ interests: [string],
+ children: [ { name: string, age: int32? } ]
+}
+
+create dataset Customers(CustomerType) partitioned by key cid;
+
+load dataset Customers
+using "edu.uci.ics.asterix.external.dataset.adapter.NCFileSystemAdapter"
+(("path"="nc1://data/semistructured/co1k_olist/customer.adm"),("format"="adm"));
+
+create index interests_index on Customers(interests) type keyword;
+
+write output to nc1:"rttest/inverted-index-basic_olist-fuzzyeq-edit-distance.adm";
+
+set simfunction 'edit-distance';
+set simthreshold '1';
+
+for $c in dataset('Customers')
+where $c.interests ~= ["computers", "wine", "walking"]
+order by $c.cid
+return $c
diff --git a/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/olist-fuzzyeq-jaccard.aql b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/olist-fuzzyeq-jaccard.aql
new file mode 100644
index 0000000..09da02e
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/olist-fuzzyeq-jaccard.aql
@@ -0,0 +1,41 @@
+/*
+ * Description : Tests whether a keyword is applied to optimize a selection query with ~= using Jaccard on lists.
+ * The index should be applied.
+ * Success : Yes
+ */
+
+drop dataverse test if exists;
+create dataverse test;
+use dataverse test;
+
+create type AddressType as closed {
+ number: int32,
+ street: string,
+ city: string
+}
+
+create type CustomerType as closed {
+ cid: int32,
+ name: string,
+ age: int32?,
+ address: AddressType?,
+ interests: [string],
+ children: [ { name: string, age: int32? } ]
+}
+
+create dataset Customers(CustomerType) partitioned by key cid;
+
+load dataset Customers
+using "edu.uci.ics.asterix.external.dataset.adapter.NCFileSystemAdapter"
+(("path"="nc1://data/semistructured/co1k_olist/customer.adm"),("format"="adm"));
+
+create index interests_index on Customers(interests) type keyword;
+
+write output to nc1:"rttest/inverted-index-basic_olist-fuzzyeq-jaccard.adm";
+
+set simfunction 'jaccard';
+set simthreshold '0.8f';
+
+for $c in dataset('Customers')
+where $c.interests ~= ["databases", "computers", "wine"]
+return $c
diff --git a/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/olist-jaccard-check.aql b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/olist-jaccard-check.aql
new file mode 100644
index 0000000..18721c5
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/olist-jaccard-check.aql
@@ -0,0 +1,38 @@
+/*
+ * Description : Tests whether a keyword index is applied to optimize a selection query using the similarity-jaccard-check function on lists.
+ * The index should be applied.
+ * Success : Yes
+ */
+
+drop dataverse test if exists;
+create dataverse test;
+use dataverse test;
+
+create type AddressType as closed {
+ number: int32,
+ street: string,
+ city: string
+}
+
+create type CustomerType as closed {
+ cid: int32,
+ name: string,
+ age: int32?,
+ address: AddressType?,
+ interests: [string],
+ children: [ { name: string, age: int32? } ]
+}
+
+create dataset Customers(CustomerType) partitioned by key cid;
+
+load dataset Customers
+using "edu.uci.ics.asterix.external.dataset.adapter.NCFileSystemAdapter"
+(("path"="nc1://data/semistructured/co1k_olist/customer.adm"),("format"="adm"));
+
+create index interests_index on Customers(interests) type keyword;
+
+write output to nc1:"rttest/inverted-index-basic_olist-jaccard-check.adm";
+
+for $c in dataset('Customers')
+where similarity-jaccard-check($c.interests, ["databases", "computers", "wine"], 0.7f)[0]
+return $c
diff --git a/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/olist-jaccard.aql b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/olist-jaccard.aql
new file mode 100644
index 0000000..fa2291c
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/olist-jaccard.aql
@@ -0,0 +1,38 @@
+/*
+ * Description : Tests whether a keyword index is applied to optimize a selection query using the similarity-jaccard function on lists.
+ * The index should be applied.
+ * Success : Yes
+ */
+
+drop dataverse test if exists;
+create dataverse test;
+use dataverse test;
+
+create type AddressType as closed {
+ number: int32,
+ street: string,
+ city: string
+}
+
+create type CustomerType as closed {
+ cid: int32,
+ name: string,
+ age: int32?,
+ address: AddressType?,
+ interests: [string],
+ children: [ { name: string, age: int32? } ]
+}
+
+create dataset Customers(CustomerType) partitioned by key cid;
+
+load dataset Customers
+using "edu.uci.ics.asterix.external.dataset.adapter.NCFileSystemAdapter"
+(("path"="nc1://data/semistructured/co1k_olist/customer.adm"),("format"="adm"));
+
+create index interests_index on Customers(interests) type keyword;
+
+write output to nc1:"rttest/inverted-index-basic_olist-jaccard.adm";
+
+for $c in dataset('Customers')
+where similarity-jaccard($c.interests, ["databases", "computers", "wine"]) >= 0.7f
+return $c
diff --git a/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ulist-fuzzyeq-jaccard.aql b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ulist-fuzzyeq-jaccard.aql
new file mode 100644
index 0000000..89c2dac
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ulist-fuzzyeq-jaccard.aql
@@ -0,0 +1,41 @@
+/*
+ * Description : Tests whether a keyword is applied to optimize a selection query with ~= using Jaccard on sets.
+ * The index should be applied.
+ * Success : Yes
+ */
+
+drop dataverse test if exists;
+create dataverse test;
+use dataverse test;
+
+create type AddressType as closed {
+ number: int32,
+ street: string,
+ city: string
+}
+
+create type CustomerType as closed {
+ cid: int32,
+ name: string,
+ age: int32?,
+ address: AddressType?,
+ interests: {{string}},
+ children: [ { name: string, age: int32? } ]
+}
+
+create dataset Customers(CustomerType) partitioned by key cid;
+
+load dataset Customers
+using "edu.uci.ics.asterix.external.dataset.adapter.NCFileSystemAdapter"
+(("path"="nc1://data/semistructured/co1k/customer.adm"),("format"="adm"));
+
+create index interests_index on Customers(interests) type keyword;
+
+write output to nc1:"rttest/inverted-index-basic_ulist-fuzzyeq-jaccard.adm";
+
+set simfunction 'jaccard';
+set simthreshold '0.8f';
+
+for $c in dataset('Customers')
+where $c.interests ~= ["databases", "computers", "wine"]
+return $c
diff --git a/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ulist-jaccard-check.aql b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ulist-jaccard-check.aql
new file mode 100644
index 0000000..8089d6d
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ulist-jaccard-check.aql
@@ -0,0 +1,38 @@
+/*
+ * Description : Tests whether a keyword index is applied to optimize a selection query using the similarity-jaccard-check function on sets.
+ * The index should be applied.
+ * Success : Yes
+ */
+
+drop dataverse test if exists;
+create dataverse test;
+use dataverse test;
+
+create type AddressType as closed {
+ number: int32,
+ street: string,
+ city: string
+}
+
+create type CustomerType as closed {
+ cid: int32,
+ name: string,
+ age: int32?,
+ address: AddressType?,
+ interests: {{string}},
+ children: [ { name: string, age: int32? } ]
+}
+
+create dataset Customers(CustomerType) partitioned by key cid;
+
+load dataset Customers
+using "edu.uci.ics.asterix.external.dataset.adapter.NCFileSystemAdapter"
+(("path"="nc1://data/semistructured/co1k/customer.adm"),("format"="adm"));
+
+create index interests_index on Customers(interests) type keyword;
+
+write output to nc1:"rttest/inverted-index-basic_ulist-jaccard.adm";
+
+for $c in dataset('Customers')
+where similarity-jaccard-check($c.interests, ["databases", "computers", "wine"], 0.7f)[0]
+return $c
diff --git a/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ulist-jaccard.aql b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ulist-jaccard.aql
new file mode 100644
index 0000000..547f7e2
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ulist-jaccard.aql
@@ -0,0 +1,38 @@
+/*
+ * Description : Tests whether a keyword index is applied to optimize a selection query using the similarity-jaccard function on sets.
+ * The index should be applied.
+ * Success : Yes
+ */
+
+drop dataverse test if exists;
+create dataverse test;
+use dataverse test;
+
+create type AddressType as closed {
+ number: int32,
+ street: string,
+ city: string
+}
+
+create type CustomerType as closed {
+ cid: int32,
+ name: string,
+ age: int32?,
+ address: AddressType?,
+ interests: {{string}},
+ children: [ { name: string, age: int32? } ]
+}
+
+create dataset Customers(CustomerType) partitioned by key cid;
+
+load dataset Customers
+using "edu.uci.ics.asterix.external.dataset.adapter.NCFileSystemAdapter"
+(("path"="nc1://data/semistructured/co1k/customer.adm"),("format"="adm"));
+
+create index interests_index on Customers(interests) type keyword;
+
+write output to nc1:"rttest/inverted-index-basic_ulist-jaccard.adm";
+
+for $c in dataset('Customers')
+where similarity-jaccard($c.interests, ["databases", "computers", "wine"]) >= 0.7f
+return $c
diff --git a/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/word-contains.aql b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/word-contains.aql
new file mode 100644
index 0000000..0b5367b
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/word-contains.aql
@@ -0,0 +1,33 @@
+/*
+ * Description : Tests whether a keyword index is applied to optimize a selection query using the contains function.
+ * The index should *not* be applied (see below).
+ * 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 dataset DBLP(DBLPType) partitioned by key id;
+
+load dataset DBLP
+using "edu.uci.ics.asterix.external.dataset.adapter.NCFileSystemAdapter"
+(("path"="nc1://data/dblp-small/dblp-small-id.txt"),("format"="delimited-text"),("delimiter"=":")) pre-sorted;
+
+create index keyword_index on DBLP(title) type keyword;
+
+write output to nc1:"rttest/inverted-index-basic_word-contains.adm";
+
+// Contains cannot be answered with a word inverted index.
+for $o in dataset('DBLP')
+where contains($o.title, "Multimedia")
+order by $o.id
+return $o
diff --git a/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/word-fuzzyeq-jaccard.aql b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/word-fuzzyeq-jaccard.aql
new file mode 100644
index 0000000..39b247d
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/word-fuzzyeq-jaccard.aql
@@ -0,0 +1,34 @@
+/*
+ * Description : Tests whether a keyword is applied to optimize a selection query with ~= using Jaccard on word tokens.
+ * The index should be applied.
+ * 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 dataset DBLP(DBLPType) partitioned by key id;
+
+load dataset DBLP
+using "edu.uci.ics.asterix.external.dataset.adapter.NCFileSystemAdapter"
+(("path"="nc1://data/dblp-small/dblp-small-id.txt"),("format"="delimited-text"),("delimiter"=":")) pre-sorted;
+
+create index keyword_index on DBLP(title) type keyword;
+
+write output to nc1:"rttest/inverted-index-basic_word-fuzzyeq-jaccard.adm";
+
+set simfunction 'jaccard';
+set simthreshold '0.5f';
+
+for $o in dataset('DBLP')
+where word-tokens($o.title) ~= word-tokens("Transactions for Cooperative Environments")
+return $o
diff --git a/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/word-jaccard-check.aql b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/word-jaccard-check.aql
new file mode 100644
index 0000000..f1d3d4b
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/word-jaccard-check.aql
@@ -0,0 +1,32 @@
+/*
+ * Description : Tests whether a keyword index is applied to optimize a selection query using the similarity-jaccard-check function on word tokens.
+ * The index should be applied.
+ * 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 dataset DBLP(DBLPType) partitioned by key id;
+
+load dataset DBLP
+using "edu.uci.ics.asterix.external.dataset.adapter.NCFileSystemAdapter"
+(("path"="nc1://data/dblp-small/dblp-small-id.txt"),("format"="delimited-text"),("delimiter"=":")) pre-sorted;
+
+create index keyword_index on DBLP(title) type keyword;
+
+write output to nc1:"rttest/inverted-index-basic_word-jaccard-check.adm";
+
+for $o in dataset('DBLP')
+where similarity-jaccard-check(word-tokens($o.title), word-tokens("Transactions for Cooperative Environments"), 0.5f)[0]
+return $o
+
diff --git a/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/word-jaccard.aql b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/word-jaccard.aql
new file mode 100644
index 0000000..16fb8d3
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/word-jaccard.aql
@@ -0,0 +1,32 @@
+/*
+ * Description : Tests whether a keyword index is applied to optimize a selection query using the similarity-jaccard function on word tokens.
+ * The index should be applied.
+ * 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 dataset DBLP(DBLPType) partitioned by key id;
+
+load dataset DBLP
+using "edu.uci.ics.asterix.external.dataset.adapter.NCFileSystemAdapter"
+(("path"="nc1://data/dblp-small/dblp-small-id.txt"),("format"="delimited-text"),("delimiter"=":")) pre-sorted;
+
+create index keyword_index on DBLP(title) type keyword;
+
+write output to nc1:"rttest/inverted-index-basic_word-jaccard.adm";
+
+for $o in dataset('DBLP')
+where similarity-jaccard(word-tokens($o.title), word-tokens("Transactions for Cooperative Environments")) >= 0.5f
+return $o
+