use of edu.uci.ics.textdb.api.field.StringField in project textdb by TextDB.
the class FuzzyTokenMatcherTest method TestFuzzyTokenMatcherWithLimitOffset.
@Test
public void TestFuzzyTokenMatcherWithLimitOffset() throws Exception {
String query = "Twelve Angry Men";
// The ratio of tokens that need to be matched
double threshold = 0.5;
ArrayList<String> attributeNames = new ArrayList<>();
attributeNames.add(TestConstants.DESCRIPTION);
Schema schema = Utils.addAttributeToSchema(TestConstants.SCHEMA_PEOPLE, RESULTS_ATTR);
List<Span> list = new ArrayList<>();
Span span = new Span(TestConstants.DESCRIPTION, 5, 10, "angry", "Angry", 1);
list.add(span);
IField[] fields1 = { new StringField("bruce"), new StringField("john Lee"), new IntegerField(46), new DoubleField(5.50), new DateField(new SimpleDateFormat("MM-dd-yyyy").parse("01-14-1970")), new TextField("Tall Angry"), new ListField<>(list) };
list = new ArrayList<>();
span = new Span(TestConstants.DESCRIPTION, 6, 11, "angry", "Angry", 1);
list.add(span);
IField[] fields2 = { new StringField("brad lie angelina"), new StringField("pitt"), new IntegerField(44), new DoubleField(6.10), new DateField(new SimpleDateFormat("MM-dd-yyyy").parse("01-12-1972")), new TextField("White Angry"), new ListField<>(list) };
list = new ArrayList<>();
span = new Span(TestConstants.DESCRIPTION, 40, 45, "angry", "Angry", 8);
list.add(span);
IField[] fields3 = { new StringField("george lin lin"), new StringField("lin clooney"), new IntegerField(43), new DoubleField(6.06), new DateField(new SimpleDateFormat("MM-dd-yyyy").parse("01-13-1973")), new TextField("Lin Clooney is Short and lin clooney is Angry"), new ListField<>(list) };
list = new ArrayList<>();
span = new Span(TestConstants.DESCRIPTION, 6, 11, "angry", "angry", 1);
list.add(span);
IField[] fields4 = { new StringField("Mary brown"), new StringField("Lake Forest"), new IntegerField(42), new DoubleField(5.99), new DateField(new SimpleDateFormat("MM-dd-yyyy").parse("01-13-1974")), new TextField("Short angry"), new ListField<>(list) };
Tuple tuple1 = new Tuple(schema, fields1);
Tuple tuple2 = new Tuple(schema, fields2);
Tuple tuple3 = new Tuple(schema, fields3);
Tuple tuple4 = new Tuple(schema, fields4);
List<Tuple> expectedResultList = new ArrayList<>();
expectedResultList.add(tuple1);
expectedResultList.add(tuple2);
expectedResultList.add(tuple3);
expectedResultList.add(tuple4);
List<Tuple> results = FuzzyTokenMatcherTestHelper.getQueryResults(PEOPLE_TABLE, query, threshold, attributeNames, 2, 1);
Assert.assertEquals(expectedResultList.size(), 4);
Assert.assertEquals(results.size(), 2);
Assert.assertTrue(TestUtils.containsAll(expectedResultList, results));
}
use of edu.uci.ics.textdb.api.field.StringField in project textdb by TextDB.
the class FuzzyTokenMatcherTest method TestFuzzyTokenMatcherWithThresholdVariation.
@Test
public void TestFuzzyTokenMatcherWithThresholdVariation() throws Exception {
String query = "Twelve Angry Men Cafe";
double threshold = 0.25;
ArrayList<String> attributeNames = new ArrayList<>();
attributeNames.add(TestConstants.DESCRIPTION);
Schema schema = Utils.addAttributeToSchema(TestConstants.SCHEMA_PEOPLE, RESULTS_ATTR);
List<Span> spanList1 = Arrays.asList(new Span(TestConstants.DESCRIPTION, 5, 10, "angry", "Angry", 1));
IField[] fields1 = { new StringField("bruce"), new StringField("john Lee"), new IntegerField(46), new DoubleField(5.50), new DateField(new SimpleDateFormat("MM-dd-yyyy").parse("01-14-1970")), new TextField("Tall Angry"), new ListField<Span>(spanList1) };
List<Span> spanList2 = Arrays.asList(new Span(TestConstants.DESCRIPTION, 6, 11, "angry", "Angry", 1));
IField[] fields2 = { new StringField("brad lie angelina"), new StringField("pitt"), new IntegerField(44), new DoubleField(6.10), new DateField(new SimpleDateFormat("MM-dd-yyyy").parse("01-12-1972")), new TextField("White Angry"), new ListField<Span>(spanList2) };
List<Span> spanList3 = Arrays.asList(new Span(TestConstants.DESCRIPTION, 40, 45, "angry", "Angry", 8));
IField[] fields3 = { new StringField("george lin lin"), new StringField("lin clooney"), new IntegerField(43), new DoubleField(6.06), new DateField(new SimpleDateFormat("MM-dd-yyyy").parse("01-13-1973")), new TextField("Lin Clooney is Short and lin clooney is Angry"), new ListField<Span>(spanList3) };
List<Span> spanList4 = Arrays.asList(new Span(TestConstants.DESCRIPTION, 6, 11, "angry", "angry", 1));
IField[] fields4 = { new StringField("Mary brown"), new StringField("Lake Forest"), new IntegerField(42), new DoubleField(5.99), new DateField(new SimpleDateFormat("MM-dd-yyyy").parse("01-13-1974")), new TextField("Short angry"), new ListField<Span>(spanList4) };
Tuple tuple1 = new Tuple(schema, fields1);
Tuple tuple2 = new Tuple(schema, fields2);
Tuple tuple3 = new Tuple(schema, fields3);
Tuple tuple4 = new Tuple(schema, fields4);
List<Tuple> expectedResultList = new ArrayList<>();
expectedResultList.add(tuple1);
expectedResultList.add(tuple2);
expectedResultList.add(tuple3);
expectedResultList.add(tuple4);
List<Tuple> results = FuzzyTokenMatcherTestHelper.getQueryResults(PEOPLE_TABLE, query, threshold, attributeNames);
boolean contains = TestUtils.equals(expectedResultList, results);
Assert.assertTrue(contains);
}
use of edu.uci.ics.textdb.api.field.StringField in project textdb by TextDB.
the class FuzzyTokenMatcherTest method TestFuzzyTokenMatcherWithLimit.
@Test
public void TestFuzzyTokenMatcherWithLimit() throws Exception {
String query = "Twelve Angry Men";
// The ratio of tokens that need to be matched
double threshold = 0.5;
ArrayList<String> attributeNames = new ArrayList<>();
attributeNames.add(TestConstants.DESCRIPTION);
Schema schema = Utils.addAttributeToSchema(TestConstants.SCHEMA_PEOPLE, RESULTS_ATTR);
List<Span> list = new ArrayList<>();
Span span = new Span(TestConstants.DESCRIPTION, 5, 10, "angry", "Angry", 1);
list.add(span);
IField[] fields1 = { new StringField("bruce"), new StringField("john Lee"), new IntegerField(46), new DoubleField(5.50), new DateField(new SimpleDateFormat("MM-dd-yyyy").parse("01-14-1970")), new TextField("Tall Angry"), new ListField<>(list) };
list = new ArrayList<>();
span = new Span(TestConstants.DESCRIPTION, 6, 11, "angry", "Angry", 1);
list.add(span);
IField[] fields2 = { new StringField("brad lie angelina"), new StringField("pitt"), new IntegerField(44), new DoubleField(6.10), new DateField(new SimpleDateFormat("MM-dd-yyyy").parse("01-12-1972")), new TextField("White Angry"), new ListField<>(list) };
list = new ArrayList<>();
span = new Span(TestConstants.DESCRIPTION, 40, 45, "angry", "Angry", 8);
list.add(span);
IField[] fields3 = { new StringField("george lin lin"), new StringField("lin clooney"), new IntegerField(43), new DoubleField(6.06), new DateField(new SimpleDateFormat("MM-dd-yyyy").parse("01-13-1973")), new TextField("Lin Clooney is Short and lin clooney is Angry"), new ListField<>(list) };
list = new ArrayList<>();
span = new Span(TestConstants.DESCRIPTION, 6, 11, "angry", "angry", 1);
list.add(span);
IField[] fields4 = { new StringField("Mary brown"), new StringField("Lake Forest"), new IntegerField(42), new DoubleField(5.99), new DateField(new SimpleDateFormat("MM-dd-yyyy").parse("01-13-1974")), new TextField("Short angry"), new ListField<>(list) };
Tuple tuple1 = new Tuple(schema, fields1);
Tuple tuple2 = new Tuple(schema, fields2);
Tuple tuple3 = new Tuple(schema, fields3);
Tuple tuple4 = new Tuple(schema, fields4);
List<Tuple> expectedResultList = new ArrayList<>();
expectedResultList.add(tuple1);
expectedResultList.add(tuple2);
expectedResultList.add(tuple3);
expectedResultList.add(tuple4);
List<Tuple> results = FuzzyTokenMatcherTestHelper.getQueryResults(PEOPLE_TABLE, query, threshold, attributeNames, 2, 0);
Assert.assertEquals(expectedResultList.size(), 4);
Assert.assertEquals(results.size(), 2);
Assert.assertTrue(TestUtils.containsAll(expectedResultList, results));
}
use of edu.uci.ics.textdb.api.field.StringField in project textdb by TextDB.
the class FuzzyTokenMatcherTest method TestFuzzyTokenMatcher2.
@Test
public void TestFuzzyTokenMatcher2() throws Exception {
String query = "Twelve Angry Men";
// The ratio of tokens that need to be matched
double threshold = 0.5;
ArrayList<String> attributeNames = new ArrayList<>();
attributeNames.add(TestConstants.DESCRIPTION);
Schema schema = Utils.addAttributeToSchema(TestConstants.SCHEMA_PEOPLE, RESULTS_ATTR);
List<Span> spanList1 = Arrays.asList(new Span(TestConstants.DESCRIPTION, 5, 10, "angry", "Angry", 1));
IField[] fields1 = { new StringField("bruce"), new StringField("john Lee"), new IntegerField(46), new DoubleField(5.50), new DateField(new SimpleDateFormat("MM-dd-yyyy").parse("01-14-1970")), new TextField("Tall Angry"), new ListField<Span>(spanList1) };
List<Span> spanList2 = Arrays.asList(new Span(TestConstants.DESCRIPTION, 6, 11, "angry", "Angry", 1));
IField[] fields2 = { new StringField("brad lie angelina"), new StringField("pitt"), new IntegerField(44), new DoubleField(6.10), new DateField(new SimpleDateFormat("MM-dd-yyyy").parse("01-12-1972")), new TextField("White Angry"), new ListField<Span>(spanList2) };
List<Span> spanList3 = Arrays.asList(new Span(TestConstants.DESCRIPTION, 40, 45, "angry", "Angry", 8));
IField[] fields3 = { new StringField("george lin lin"), new StringField("lin clooney"), new IntegerField(43), new DoubleField(6.06), new DateField(new SimpleDateFormat("MM-dd-yyyy").parse("01-13-1973")), new TextField("Lin Clooney is Short and lin clooney is Angry"), new ListField<Span>(spanList3) };
List<Span> spanList4 = Arrays.asList(new Span(TestConstants.DESCRIPTION, 6, 11, "angry", "angry", 1));
IField[] fields4 = { new StringField("Mary brown"), new StringField("Lake Forest"), new IntegerField(42), new DoubleField(5.99), new DateField(new SimpleDateFormat("MM-dd-yyyy").parse("01-13-1974")), new TextField("Short angry"), new ListField<Span>(spanList4) };
Tuple tuple1 = new Tuple(schema, fields1);
Tuple tuple2 = new Tuple(schema, fields2);
Tuple tuple3 = new Tuple(schema, fields3);
Tuple tuple4 = new Tuple(schema, fields4);
List<Tuple> expectedResultList = new ArrayList<>();
expectedResultList.add(tuple1);
expectedResultList.add(tuple2);
expectedResultList.add(tuple3);
expectedResultList.add(tuple4);
List<Tuple> results = FuzzyTokenMatcherTestHelper.getQueryResults(PEOPLE_TABLE, query, threshold, attributeNames);
boolean contains = TestUtils.equals(expectedResultList, results);
Assert.assertTrue(contains);
}
use of edu.uci.ics.textdb.api.field.StringField in project textdb by TextDB.
the class JoinDistanceTest method testSpansOverlapAndWithinThreshold.
/*
* This case tests for the spans to be joined have some overlap and both
* |(span 1 spanStartIndex) - (span 2 spanStartIndex)|,
* |(span 1 spanEndIndex) - (span 2 spanEndIndex)| are within threshold.
*
* e.g.
* [<75, 97>]
* [<92, 109>]
* threshold = 20 (within threshold)
* result: [<75, 109>]
*
* Test result: The list contains a tuple with all the fields and a span
* list consisting of the joined span. The joined span is made up of the
* field name, start and stop index (computed as <min(span1 spanStartIndex,
* span2 spanStartIndex), max(span1 spanEndIndex, span2 spanEndIndex)>)
* key (combination of span1 key and span2 key) and value (combination of
* span1 value and span2 value).
*/
@Test
public void testSpansOverlapAndWithinThreshold() throws Exception {
JoinTestHelper.insertToTable(BOOK_TABLE, JoinTestConstants.bookGroup1.get(0));
KeywordMatcherSourceOperator keywordSourceOuter = JoinTestHelper.getKeywordSource(BOOK_TABLE, "gastrointestinal tract", phrase);
KeywordMatcherSourceOperator keywordSourceInner = JoinTestHelper.getKeywordSource(BOOK_TABLE, "tract interesting", phrase);
List<Tuple> resultList = JoinTestHelper.getJoinDistanceResults(keywordSourceInner, keywordSourceOuter, new JoinDistancePredicate(JoinTestConstants.REVIEW, 20), Integer.MAX_VALUE, 0);
Schema resultSchema = Utils.createSpanSchema(JoinTestConstants.BOOK_SCHEMA);
List<Span> spanList = new ArrayList<>();
Span span1 = new Span(JoinTestConstants.REVIEW, 75, 109, "gastrointestinal tract_" + "tract interesting", "gastrointestinal " + "tract interesting");
spanList.add(span1);
IField[] book1 = { new IntegerField(52), new StringField("Mary Roach"), new StringField("Grunt: The Curious Science of Humans at War"), new IntegerField(288), new TextField("It takes a special kind " + "of writer to make topics ranging from death to our " + "gastrointestinal tract interesting (sometimes " + "hilariously so), and pop science writer Mary Roach is " + "always up to the task."), new ListField<>(spanList) };
Tuple expectedTuple = new Tuple(resultSchema, book1);
List<Tuple> expectedResult = new ArrayList<>();
expectedResult.add(expectedTuple);
Assert.assertEquals(1, resultList.size());
Assert.assertTrue(TestUtils.equals(expectedResult, resultList));
}
Aggregations