Examples with Span edu.uci.ics.textdb.api.span.Span used on opensource projects

Example 46 with Span

use of edu.uci.ics.textdb.api.span.Span in project textdb by TextDB.

the class FuzzyTokenMatcherTest method TestFuzzyTokenMatcherWithLargeTokens.

@Test
public void TestFuzzyTokenMatcherWithLargeTokens() throws Exception {
    String query = "Twelve Angry Men Came Cafe Have Coffee Eat Chocolate Burger Fries SandWidch Cool Food Drinks American drama film elements film noir adapted teleplay same name Reginald Rose Written Rose directed  Sidney Lumet trial film tells story jury made deliberate guilt acquittal defendant basis reasonable doubt United States verdict most criminal ";
    double threshold = 0.02;
    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);
}

Example 47 with Span

use of edu.uci.ics.textdb.api.span.Span in project textdb by TextDB.

the class JoinDistanceTest method testIdsMatchFieldsMatchSpanWithinThreshold.

/*
     * This case tests for the scenario when the difference of keyword spans
     *  is within the given span threshold.
     *  e.g.
     *  [<11, 18>]
     *  [<27, 33>]
     *  threshold = 20 (within threshold)
     *  result: [<11, 33>]
     *  
     * 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 testIdsMatchFieldsMatchSpanWithinThreshold() throws Exception {
    JoinTestHelper.insertToTable(BOOK_TABLE, JoinTestConstants.bookGroup1.get(0));
    KeywordMatcherSourceOperator keywordSourceOuter = JoinTestHelper.getKeywordSource(BOOK_TABLE, "special", conjunction);
    KeywordMatcherSourceOperator keywordSourceInner = JoinTestHelper.getKeywordSource(BOOK_TABLE, "writer", conjunction);
    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, 11, 33, "special_writer", "special kind of " + "writer");
    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));
}

Example 48 with Span

use of edu.uci.ics.textdb.api.span.Span in project textdb by TextDB.

the class JoinDistanceTest method testBothOperatorsMultipleTuplesSpanWithinThreshold.

// --------------------<END of single tuple test cases>--------------------
/*
     * This case tests for the scenario when both the operators' have multiple
     * tuples and spans are within threshold.
     * 
     * Test result: Join should result in a list containing tuples with spans.
     * The number of tuples is equal to the number of tuples with spans within threshold.
     */
@Test
public void testBothOperatorsMultipleTuplesSpanWithinThreshold() throws Exception {
    List<Tuple> tuples = new ArrayList<>();
    tuples.add(JoinTestConstants.bookGroup1.get(3));
    tuples.add(JoinTestConstants.bookGroup2.get(2));
    tuples.add(JoinTestConstants.bookGroup2.get(4));
    JoinTestHelper.insertToTable(BOOK_TABLE, tuples);
    KeywordMatcherSourceOperator keywordSourceOuter = JoinTestHelper.getKeywordSource(BOOK_TABLE, "review", conjunction);
    KeywordMatcherSourceOperator keywordSourceInner = JoinTestHelper.getKeywordSource(BOOK_TABLE, "book", conjunction);
    List<Tuple> resultList = JoinTestHelper.getJoinDistanceResults(keywordSourceInner, keywordSourceOuter, new JoinDistancePredicate(JoinTestConstants.REVIEW, 12), Integer.MAX_VALUE, 0);
    Schema resultSchema = Utils.createSpanSchema(JoinTestConstants.BOOK_SCHEMA);
    List<Span> spanList = new ArrayList<>();
    Span span1 = new Span(JoinTestConstants.REVIEW, 0, 16, "review_book", "Review of a " + "Book");
    spanList.add(span1);
    Span span2 = new Span(JoinTestConstants.REVIEW, 62, 73, "review_book", "book review");
    spanList.add(span2);
    Span span3 = new Span(JoinTestConstants.REVIEW, 235, 246, "review_book", "book review");
    spanList.add(span3);
    IField[] book1 = { new IntegerField(54), new StringField("Andria Williams"), new StringField("The Longest Night: A Novel"), new IntegerField(400), new TextField("Review of a Book. This is a typical " + "review. This is a test. A book review " + "test. A test to test queries without " + "actually using actual review. From " + "here onwards, we can pretend this to " + "be actually a review even if it is not " + "your typical book review."), new ListField<>(spanList) };
    IField[] book2 = { new IntegerField(65), new StringField("Sharon Guskin"), new StringField("The Forgetting Time: A Novel"), new IntegerField(368), new TextField("Review of a Book. This is a typical " + "review. This is a test. A book review " + "test. A test to test queries without " + "actually using actual review. From " + "here onwards, we can pretend this to " + "be actually a review even if it is not " + "your typical book review."), new ListField<>(spanList) };
    IField[] book3 = { new IntegerField(63), new StringField("Paul Kalanithi"), new StringField("When Breath Becomes Air"), new IntegerField(256), new TextField("Review of a Book. This is a typical " + "review. This is a test. A book review " + "test. A test to test queries without " + "actually using actual review. From " + "here onwards, we can pretend this to " + "be actually a review even if it is not " + "your typical book review."), new ListField<>(spanList) };
    Tuple expectedTuple1 = new Tuple(resultSchema, book1);
    Tuple expectedTuple2 = new Tuple(resultSchema, book2);
    Tuple expectedTuple3 = new Tuple(resultSchema, book3);
    List<Tuple> expectedResult = new ArrayList<>();
    expectedResult.add(expectedTuple1);
    expectedResult.add(expectedTuple2);
    expectedResult.add(expectedTuple3);
    Assert.assertEquals(3, resultList.size());
    Assert.assertTrue(TestUtils.equals(expectedResult, resultList));
}

Example 49 with Span

use of edu.uci.ics.textdb.api.span.Span in project textdb by TextDB.

the class JoinDistanceTest method testForLimitWhenLimitIsGreaterThanActualNumberOfResults.

/*
     * This case tests for the scenario when limit is some integer greater than
     * 0 and greater than the actual number of results and offset is 0 and join
     * is performed.
     * Test result: A list of tuples with number of tuples equal to the maximum
     * number of tuples operator can generate (which is lesser than limit.)
     */
@Test
public void testForLimitWhenLimitIsGreaterThanActualNumberOfResults() throws Exception {
    List<Tuple> tuples = JoinTestConstants.bookGroup1.subList(1, 5);
    JoinTestHelper.insertToTable(BOOK_TABLE, tuples);
    KeywordMatcherSourceOperator keywordSourceOuter = JoinTestHelper.getKeywordSource(BOOK_TABLE, "typical", conjunction);
    KeywordMatcherSourceOperator keywordSourceInner = JoinTestHelper.getKeywordSource(BOOK_TABLE, "actually", conjunction);
    List<Tuple> resultList = JoinTestHelper.getJoinDistanceResults(keywordSourceInner, keywordSourceOuter, new JoinDistancePredicate(JoinTestConstants.REVIEW, 90), 10, 0);
    Schema resultSchema = Utils.createSpanSchema(JoinTestConstants.BOOK_SCHEMA);
    List<Span> spanList = new ArrayList<>();
    Span span1 = new Span(JoinTestConstants.REVIEW, 28, 119, "typical_actually", "typical review. " + "This is a test. A book review test. " + "A test to test queries without actually");
    spanList.add(span1);
    Span span2 = new Span(JoinTestConstants.REVIEW, 186, 234, "typical_actually", "actually a review " + "even if it is not your typical");
    spanList.add(span2);
    IField[] book1 = { new IntegerField(51), new StringField("author unknown"), new StringField("typical"), new IntegerField(300), new TextField("Review of a Book. This is a typical " + "review. This is a test. A book review " + "test. A test to test queries without " + "actually using actual review. From " + "here onwards, we can pretend this to " + "be actually a review even if it is not " + "your typical book review."), new ListField<>(spanList) };
    IField[] book2 = { new IntegerField(53), new StringField("Noah Hawley"), new StringField("Before the Fall"), new IntegerField(400), new TextField("Review of a Book. This is a typical " + "review. This is a test. A book review " + "test. A test to test queries without " + "actually using actual review. From " + "here onwards, we can pretend this to " + "be actually a review even if it is not " + "your typical book review."), new ListField<>(spanList) };
    IField[] book3 = { new IntegerField(54), new StringField("Andria Williams"), new StringField("The Longest Night: A Novel"), new IntegerField(400), new TextField("Review of a Book. This is a typical " + "review. This is a test. A book review " + "test. A test to test queries without " + "actually using actual review. From " + "here onwards, we can pretend this to " + "be actually a review even if it is not " + "your typical book review."), new ListField<>(spanList) };
    IField[] book4 = { new IntegerField(55), new StringField("Matti Friedman"), new StringField("Pumpkinflowers: A Soldier's " + "Story"), new IntegerField(256), new TextField("Review of a Book. This is a typical " + "review. This is a test. A book review " + "test. A test to test queries without " + "actually using actual review. From " + "here onwards, we can pretend this to " + "be actually a review even if it is not " + "your typical book review."), new ListField<>(spanList) };
    Tuple expectedTuple1 = new Tuple(resultSchema, book1);
    Tuple expectedTuple2 = new Tuple(resultSchema, book2);
    Tuple expectedTuple3 = new Tuple(resultSchema, book3);
    Tuple expectedTuple4 = new Tuple(resultSchema, book4);
    List<Tuple> expectedResult = new ArrayList<>(5);
    expectedResult.add(expectedTuple1);
    expectedResult.add(expectedTuple2);
    expectedResult.add(expectedTuple3);
    expectedResult.add(expectedTuple4);
    Assert.assertEquals(4, resultList.size());
    Assert.assertTrue(TestUtils.equals(expectedResult, resultList));
}

Example 50 with Span

use of edu.uci.ics.textdb.api.span.Span in project textdb by TextDB.

the class DictionaryMatcherTest method testStopWordsInQueryUsingPhrase.

/**
     * Scenario S-14:verifies: Query with Stop Words match corresponding phrases
     * in the document using PHRASE OPERATOR.
     */
@Test
public void testStopWordsInQueryUsingPhrase() throws Exception {
    ArrayList<String> names = new ArrayList<String>(Arrays.asList("lin and is angry"));
    Dictionary dictionary = new Dictionary(names);
    // create a data tuple first
    List<Span> list = new ArrayList<>();
    Span span = new Span("description", 25, 45, "lin and is angry", "lin clooney is Angry");
    list.add(span);
    Attribute[] schemaAttributes = new Attribute[TestConstants.ATTRIBUTES_PEOPLE.length + 1];
    for (int count = 0; count < schemaAttributes.length - 1; count++) {
        schemaAttributes[count] = TestConstants.ATTRIBUTES_PEOPLE[count];
    }
    schemaAttributes[schemaAttributes.length - 1] = RESULTS_ATTRIBUTE;
    IField[] fields1 = { 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>(list) };
    Tuple tuple1 = new Tuple(new Schema(schemaAttributes), fields1);
    List<Tuple> expectedResults = new ArrayList<Tuple>();
    expectedResults.add(tuple1);
    List<String> attributeNames = Arrays.asList(TestConstants.FIRST_NAME, TestConstants.LAST_NAME, TestConstants.DESCRIPTION);
    List<Tuple> returnedResults = DictionaryMatcherTestHelper.getQueryResults(PEOPLE_TABLE, dictionary, attributeNames, KeywordMatchingType.PHRASE_INDEXBASED);
    boolean contains = TestUtils.equals(expectedResults, returnedResults);
    Assert.assertTrue(contains);
}