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

Example 96 with Span

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

the class RegexMatcherPerformanceTest method matchRegex.

/*
     *         This function does match for a list of regex queries
     */
public static void matchRegex(List<String> regexes, String tableName) throws TextDBException, IOException {
    List<String> attributeNames = Arrays.asList(MedlineIndexWriter.ABSTRACT);
    for (String regex : regexes) {
        // analyzer should generate grams all in lower case to build a lower
        // case index.
        RegexSourcePredicate predicate = new RegexSourcePredicate(regex, attributeNames, tableName, null);
        RegexMatcherSourceOperator regexSource = new RegexMatcherSourceOperator(predicate);
        long startMatchTime = System.currentTimeMillis();
        regexSource.open();
        int counter = 0;
        Tuple nextTuple = null;
        while ((nextTuple = regexSource.getNextTuple()) != null) {
            ListField<Span> spanListField = nextTuple.getField(SchemaConstants.SPAN_LIST);
            List<Span> spanList = spanListField.getValue();
            counter += spanList.size();
        }
        regexSource.close();
        long endMatchTime = System.currentTimeMillis();
        double matchTime = (endMatchTime - startMatchTime) / 1000.0;
        totalMatchingTime += matchTime;
        totalRegexResultCount += counter;
    }
}

Example 97 with Span

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

the class ExcelSinkTest method writeSampleExcelFile.

/**
     * Create two tuples, write into a excel file. Need to manually delete the generated file.
     * @throws ParseException
     */
@Test
public void writeSampleExcelFile() throws Exception {
    ArrayList<String> attributeNames = new ArrayList<>();
    attributeNames.add(TestConstants.FIRST_NAME);
    attributeNames.add(TestConstants.LAST_NAME);
    attributeNames.add(TestConstants.DESCRIPTION);
    // Prepare the expected result list
    List<Span> list = new ArrayList<>();
    Span span1 = new Span("firstName", 0, 5, "bruce", "bruce");
    Span span2 = new Span("lastnName", 0, 5, "jacki", "jacki");
    list.add(span1);
    list.add(span2);
    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] = SchemaConstants.SPAN_LIST_ATTRIBUTE;
    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) };
    IField[] fields2 = { new StringField("test"), new StringField("jackie chan"), new IntegerField(0), new DoubleField(6.0), new DateField(new SimpleDateFormat("MM-dd-yyyy").parse("09-18-1994")), new TextField("Angry Bird"), new ListField<>(list) };
    Tuple tuple1 = new Tuple(new Schema(schemaAttributes), fields1);
    Tuple tuple2 = new Tuple(new Schema(schemaAttributes), fields2);
    IOperator inputOperator = Mockito.mock(IOperator.class);
    Mockito.when(inputOperator.getOutputSchema()).thenReturn(new Schema(schemaAttributes)).thenReturn(null);
    Mockito.when(inputOperator.getNextTuple()).thenReturn(tuple1).thenReturn(tuple2).thenReturn(null);
    excelSink = new ExcelSink(new ExcelSinkPredicate());
    excelSink.setInputOperator(inputOperator);
    excelSink.open();
    excelSink.collectAllTuples();
    excelSink.close();
    Files.deleteIfExists(Paths.get(excelSink.getFilePath()));
}

Example 98 with Span

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

the class FuzzyTokenMatcherPerformanceTest method match.

/*
     * This function does match for a list of queries
     */
public static void match(ArrayList<String> queryList, double threshold, String luceneAnalyzerStr, String tableName, boolean bool) throws TextDBException, IOException {
    List<String> attributeNames = Arrays.asList(MedlineIndexWriter.ABSTRACT);
    for (String query : queryList) {
        FuzzyTokenSourcePredicate predicate = new FuzzyTokenSourcePredicate(query, attributeNames, luceneAnalyzerStr, threshold, tableName, null);
        FuzzyTokenMatcherSourceOperator fuzzyTokenSource = new FuzzyTokenMatcherSourceOperator(predicate);
        long startMatchTime = System.currentTimeMillis();
        fuzzyTokenSource.open();
        int counter = 0;
        Tuple nextTuple = null;
        while ((nextTuple = fuzzyTokenSource.getNextTuple()) != null) {
            ListField<Span> spanListField = nextTuple.getField(SchemaConstants.SPAN_LIST);
            List<Span> spanList = spanListField.getValue();
            counter += spanList.size();
        }
        fuzzyTokenSource.close();
        long endMatchTime = System.currentTimeMillis();
        double matchTime = (endMatchTime - startMatchTime) / 1000.0;
        timeResults.add(Double.parseDouble(String.format("%.4f", matchTime)));
        totalResultCount += counter;
    }
}

Example 99 with Span

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

the class DictionaryMatcherTest method testSingleWordQueryInTextFieldUsingScan.

/**
     * Scenario S-5:verifies GetNextTuple of DictionaryMatcher and single word
     * queries in Text Field using SCANOPERATOR
     */
@Test
public void testSingleWordQueryInTextFieldUsingScan() throws Exception {
    ArrayList<String> names = new ArrayList<String>(Arrays.asList("tall"));
    Dictionary dictionary = new Dictionary(names);
    // create a data tuple first
    List<Span> list = new ArrayList<Span>();
    Span span = new Span("description", 0, 4, "tall", "Tall");
    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("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>(list) };
    IField[] fields2 = { new StringField("christian john wayne"), new StringField("rock bale"), new IntegerField(42), new DoubleField(5.99), new DateField(new SimpleDateFormat("MM-dd-yyyy").parse("01-13-1974")), new TextField("Tall Fair"), new ListField<Span>(list) };
    Tuple tuple1 = new Tuple(new Schema(schemaAttributes), fields1);
    Tuple tuple2 = new Tuple(new Schema(schemaAttributes), fields2);
    List<Tuple> expectedResults = new ArrayList<Tuple>();
    expectedResults.add(tuple1);
    expectedResults.add(tuple2);
    List<String> attributeNames = Arrays.asList(TestConstants.FIRST_NAME, TestConstants.LAST_NAME, TestConstants.DESCRIPTION);
    List<Tuple> returnedResults = DictionaryMatcherTestHelper.getQueryResults(PEOPLE_TABLE, dictionary, attributeNames, KeywordMatchingType.SUBSTRING_SCANBASED);
    boolean contains = TestUtils.equals(expectedResults, returnedResults);
    Assert.assertTrue(contains);
}

Example 100 with Span

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

the class DictionaryMatcherTest method testWordInMultipleFieldsQueryUsingScan.

/**
     * Scenario S-11:verifies: data source has multiple attributes, and an
     * entity can appear in all the fields and multiple times using SUBSTRING_SCANBASE
     * OPERATOR.
     */
@Test
public void testWordInMultipleFieldsQueryUsingScan() throws Exception {
    ArrayList<String> names = new ArrayList<String>(Arrays.asList("lin clooney"));
    Dictionary dictionary = new Dictionary(names);
    // create a data tuple first
    List<Span> list = new ArrayList<Span>();
    Span span1 = new Span("lastName", 0, 11, "lin clooney", "lin clooney");
    Span span2 = new Span("description", 0, 11, "lin clooney", "Lin Clooney");
    Span span3 = new Span("description", 25, 36, "lin clooney", "lin clooney");
    list.add(span1);
    list.add(span2);
    list.add(span3);
    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.SUBSTRING_SCANBASED);
    boolean contains = TestUtils.equals(expectedResults, returnedResults);
    Assert.assertTrue(contains);
}