Search in sources :

Example 1 with KeywordMatcherSourceOperator

use of edu.uci.ics.texera.dataflow.keywordmatcher.KeywordMatcherSourceOperator 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 = new Schema.Builder().add(JoinTestConstants.BOOK_SCHEMA).add(SchemaConstants.SPAN_LIST_ATTRIBUTE).build();
    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));
}
Also used : Schema(edu.uci.ics.texera.api.schema.Schema) ArrayList(java.util.ArrayList) IntegerField(edu.uci.ics.texera.api.field.IntegerField) IField(edu.uci.ics.texera.api.field.IField) JoinDistancePredicate(edu.uci.ics.texera.dataflow.join.JoinDistancePredicate) Span(edu.uci.ics.texera.api.span.Span) KeywordMatcherSourceOperator(edu.uci.ics.texera.dataflow.keywordmatcher.KeywordMatcherSourceOperator) StringField(edu.uci.ics.texera.api.field.StringField) TextField(edu.uci.ics.texera.api.field.TextField) Tuple(edu.uci.ics.texera.api.tuple.Tuple) Test(org.junit.Test)

Example 2 with KeywordMatcherSourceOperator

use of edu.uci.ics.texera.dataflow.keywordmatcher.KeywordMatcherSourceOperator 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 = new Schema.Builder().add(JoinTestConstants.BOOK_SCHEMA).add(SchemaConstants.SPAN_LIST_ATTRIBUTE).build();
    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));
}
Also used : Schema(edu.uci.ics.texera.api.schema.Schema) ArrayList(java.util.ArrayList) IntegerField(edu.uci.ics.texera.api.field.IntegerField) IField(edu.uci.ics.texera.api.field.IField) JoinDistancePredicate(edu.uci.ics.texera.dataflow.join.JoinDistancePredicate) Span(edu.uci.ics.texera.api.span.Span) KeywordMatcherSourceOperator(edu.uci.ics.texera.dataflow.keywordmatcher.KeywordMatcherSourceOperator) StringField(edu.uci.ics.texera.api.field.StringField) TextField(edu.uci.ics.texera.api.field.TextField) Tuple(edu.uci.ics.texera.api.tuple.Tuple) Test(org.junit.Test)

Example 3 with KeywordMatcherSourceOperator

use of edu.uci.ics.texera.dataflow.keywordmatcher.KeywordMatcherSourceOperator in project textdb by TextDB.

the class JoinDistanceTest method testOneSpanEncompassesOtherAndDifferenceLessThanThreshold.

// This case tests for the scenario when one of the spans to be joined encompasses the other span
// and both |(span 1 spanStartIndex) - (span 2 spanStartIndex)|,
// |(span 1 spanEndIndex) - (span 2 spanEndIndex)| are within threshold.
// e.g.
// [<11, 18>]
// [<3, 33>]
// threshold = 20 (within threshold)
// Test result: The bigger span should be returned.
// [<3, 33>]
@Test
public void testOneSpanEncompassesOtherAndDifferenceLessThanThreshold() throws Exception {
    JoinTestHelper.insertToTable(BOOK_TABLE, JoinTestConstants.bookGroup1.get(0));
    KeywordMatcherSourceOperator keywordSourceOuter = JoinTestHelper.getKeywordSource(BOOK_TABLE, "special", conjunction);
    KeywordMatcherSourceOperator keywordSourceInner = JoinTestHelper.getKeywordSource(BOOK_TABLE, "takes a special kind of writer", phrase);
    List<Tuple> resultList = JoinTestHelper.getJoinDistanceResults(keywordSourceInner, keywordSourceOuter, new JoinDistancePredicate(JoinTestConstants.REVIEW, 20), Integer.MAX_VALUE, 0);
    Schema resultSchema = new Schema.Builder().add(JoinTestConstants.BOOK_SCHEMA).add(SchemaConstants.SPAN_LIST_ATTRIBUTE).build();
    List<Span> spanList = new ArrayList<>();
    Span span1 = new Span(JoinTestConstants.REVIEW, 3, 33, "special_takes a special " + "kind of writer", "takes a 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));
}
Also used : Schema(edu.uci.ics.texera.api.schema.Schema) ArrayList(java.util.ArrayList) IntegerField(edu.uci.ics.texera.api.field.IntegerField) IField(edu.uci.ics.texera.api.field.IField) JoinDistancePredicate(edu.uci.ics.texera.dataflow.join.JoinDistancePredicate) Span(edu.uci.ics.texera.api.span.Span) KeywordMatcherSourceOperator(edu.uci.ics.texera.dataflow.keywordmatcher.KeywordMatcherSourceOperator) StringField(edu.uci.ics.texera.api.field.StringField) TextField(edu.uci.ics.texera.api.field.TextField) Tuple(edu.uci.ics.texera.api.tuple.Tuple) Test(org.junit.Test)

Example 4 with KeywordMatcherSourceOperator

use of edu.uci.ics.texera.dataflow.keywordmatcher.KeywordMatcherSourceOperator in project textdb by TextDB.

the class JoinDistanceTest method testForLimitWhenLimitIsLesserThanActualNumberOfResults.

// ---------------------<Limit and offset test cases.>---------------------
/*
     * This case tests for the scenario when limit is some integer greater than
     * 0 and less 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 limit.
     */
@Test
public void testForLimitWhenLimitIsLesserThanActualNumberOfResults() 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), 3, 0);
    Schema resultSchema = new Schema.Builder().add(JoinTestConstants.BOOK_SCHEMA).add(SchemaConstants.SPAN_LIST_ATTRIBUTE).build();
    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<>(3);
    expectedResult.add(expectedTuple1);
    expectedResult.add(expectedTuple2);
    expectedResult.add(expectedTuple3);
    expectedResult.add(expectedTuple4);
    Assert.assertEquals(3, resultList.size());
    Assert.assertTrue(TestUtils.containsAll(expectedResult, resultList));
}
Also used : Schema(edu.uci.ics.texera.api.schema.Schema) ArrayList(java.util.ArrayList) IntegerField(edu.uci.ics.texera.api.field.IntegerField) IField(edu.uci.ics.texera.api.field.IField) JoinDistancePredicate(edu.uci.ics.texera.dataflow.join.JoinDistancePredicate) Span(edu.uci.ics.texera.api.span.Span) KeywordMatcherSourceOperator(edu.uci.ics.texera.dataflow.keywordmatcher.KeywordMatcherSourceOperator) StringField(edu.uci.ics.texera.api.field.StringField) TextField(edu.uci.ics.texera.api.field.TextField) Tuple(edu.uci.ics.texera.api.tuple.Tuple) Test(org.junit.Test)

Example 5 with KeywordMatcherSourceOperator

use of edu.uci.ics.texera.dataflow.keywordmatcher.KeywordMatcherSourceOperator in project textdb by TextDB.

the class JoinDistanceTest method testOffsetGreaterThanNumberOfResults.

/*
     * This case tests for the scenario when offset is some integer greater 
     * than 0 and greater than the actual number of results and join is 
     * performed.
     * Test result: An empty list.
     */
@Test
public void testOffsetGreaterThanNumberOfResults() 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), 1, 10);
    Assert.assertEquals(0, resultList.size());
}
Also used : JoinDistancePredicate(edu.uci.ics.texera.dataflow.join.JoinDistancePredicate) Tuple(edu.uci.ics.texera.api.tuple.Tuple) KeywordMatcherSourceOperator(edu.uci.ics.texera.dataflow.keywordmatcher.KeywordMatcherSourceOperator) Test(org.junit.Test)

Aggregations

KeywordMatcherSourceOperator (edu.uci.ics.texera.dataflow.keywordmatcher.KeywordMatcherSourceOperator)26 Test (org.junit.Test)22 Tuple (edu.uci.ics.texera.api.tuple.Tuple)20 JoinDistancePredicate (edu.uci.ics.texera.dataflow.join.JoinDistancePredicate)19 Span (edu.uci.ics.texera.api.span.Span)11 ArrayList (java.util.ArrayList)11 IField (edu.uci.ics.texera.api.field.IField)9 IntegerField (edu.uci.ics.texera.api.field.IntegerField)9 StringField (edu.uci.ics.texera.api.field.StringField)9 TextField (edu.uci.ics.texera.api.field.TextField)9 Schema (edu.uci.ics.texera.api.schema.Schema)9 KeywordSourcePredicate (edu.uci.ics.texera.dataflow.keywordmatcher.KeywordSourcePredicate)4 IOperator (edu.uci.ics.texera.api.dataflow.IOperator)3 ISink (edu.uci.ics.texera.api.dataflow.ISink)3 Plan (edu.uci.ics.texera.api.engine.Plan)3 Join (edu.uci.ics.texera.dataflow.join.Join)3 RegexMatcher (edu.uci.ics.texera.dataflow.regexmatcher.RegexMatcher)3 TupleSink (edu.uci.ics.texera.dataflow.sink.tuple.TupleSink)3 OneToNBroadcastConnector (edu.uci.ics.texera.dataflow.connector.OneToNBroadcastConnector)2 ConnectorOutputOperator (edu.uci.ics.texera.dataflow.connector.OneToNBroadcastConnector.ConnectorOutputOperator)2