Merged asterix_lsm_stabilization upto r1547
git-svn-id: https://asterixdb.googlecode.com/svn/trunk/asterix@1622 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..d193238
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ngram-contains-panic.aql
@@ -0,0 +1,29 @@
+/*
+ * 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) primary key id;
+
+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..137b1c7
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ngram-contains.aql
@@ -0,0 +1,28 @@
+/*
+ * 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) primary key id;
+
+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..6f7f38c
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ngram-edit-distance-check-panic.aql
@@ -0,0 +1,29 @@
+/*
+ * 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) primary key id;
+
+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..0d9db19
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ngram-edit-distance-check.aql
@@ -0,0 +1,27 @@
+/*
+ * 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) primary key id;
+
+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..7decdb5
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ngram-edit-distance-panic.aql
@@ -0,0 +1,28 @@
+/*
+ * 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) primary key id;
+
+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..c7a9ba4
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ngram-edit-distance.aql
@@ -0,0 +1,27 @@
+/*
+ * 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) primary key id;
+
+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..3a3d7bb
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ngram-fuzzyeq-edit-distance.aql
@@ -0,0 +1,30 @@
+/*
+ * 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) primary key id;
+
+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..5e9ac0e
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ngram-fuzzyeq-jaccard.aql
@@ -0,0 +1,30 @@
+/*
+ * 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) primary key id;
+
+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..dbb9c71
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ngram-jaccard-check.aql
@@ -0,0 +1,27 @@
+/*
+ * 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) primary key id;
+
+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..f25fb8b
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ngram-jaccard.aql
@@ -0,0 +1,27 @@
+/*
+ * 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) primary key id;
+
+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..c105eab
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/olist-edit-distance-check-panic.aql
@@ -0,0 +1,36 @@
+/*
+ * 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) primary key cid;
+
+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..b12037d1
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/olist-edit-distance-check.aql
@@ -0,0 +1,35 @@
+/*
+ * 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) primary key cid;
+
+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..84bcbb9
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/olist-edit-distance-panic.aql
@@ -0,0 +1,36 @@
+/*
+ * 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) primary key cid;
+
+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..afafa37
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/olist-edit-distance.aql
@@ -0,0 +1,35 @@
+/*
+ * 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) primary key cid;
+
+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..86b6474
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/olist-fuzzyeq-edit-distance.aql
@@ -0,0 +1,38 @@
+/*
+ * 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) primary key cid;
+
+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..64812db
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/olist-fuzzyeq-jaccard.aql
@@ -0,0 +1,37 @@
+/*
+ * 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) primary key cid;
+
+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..4f8b8fe
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/olist-jaccard-check.aql
@@ -0,0 +1,34 @@
+/*
+ * 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) primary key cid;
+
+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..62d6806
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/olist-jaccard.aql
@@ -0,0 +1,34 @@
+/*
+ * 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) primary key cid;
+
+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..4ac02c8
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ulist-fuzzyeq-jaccard.aql
@@ -0,0 +1,37 @@
+/*
+ * 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) primary key cid;
+
+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..80e8aef
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ulist-jaccard-check.aql
@@ -0,0 +1,34 @@
+/*
+ * 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) primary key cid;
+
+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..588ca14
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/ulist-jaccard.aql
@@ -0,0 +1,34 @@
+/*
+ * 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) primary key cid;
+
+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..8d9cb02
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/word-contains.aql
@@ -0,0 +1,29 @@
+/*
+ * 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) primary key id;
+
+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..006398e
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/word-fuzzyeq-jaccard.aql
@@ -0,0 +1,30 @@
+/*
+ * 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) primary key id;
+
+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..4c814e4
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/word-jaccard-check.aql
@@ -0,0 +1,28 @@
+/*
+ * 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) primary key id;
+
+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..53a5bbb
--- /dev/null
+++ b/asterix-app/src/test/resources/optimizerts/queries/inverted-index-basic/word-jaccard.aql
@@ -0,0 +1,28 @@
+/*
+ * 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) primary key id;
+
+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
+