Search in sources :

Example 1 with BlockKeyDefinition

use of org.talend.dataquality.rules.BlockKeyDefinition in project tdq-studio-se by Talend.

the class ExecuteMatchRuleHandlerTest method testExecute4.

/**
 * Test method for
 * {@link org.talend.dq.analysis.ExecuteMatchRuleHandler#execute(java.util.Map, org.talend.dataquality.indicators.columnset.RecordMatchingIndicator, java.util.List, org.talend.dataquality.indicators.columnset.BlockKeyIndicator)}
 * .
 *
 * same to case 3 but the match rule order is exchange
 */
@Test
public void testExecute4() {
    Map<MetadataColumn, String> columnMap = new HashMap<MetadataColumn, String>();
    MetadataColumn col0 = ConnectionFactory.eINSTANCE.createMetadataColumn();
    col0.setName(columnName0);
    // $NON-NLS-1$
    columnMap.put(col0, "0");
    MetadataColumn col1 = ConnectionFactory.eINSTANCE.createMetadataColumn();
    col1.setName(columnName1);
    // $NON-NLS-1$
    columnMap.put(col1, "1");
    MetadataColumn col2 = ConnectionFactory.eINSTANCE.createMetadataColumn();
    col2.setName(columnName2);
    // $NON-NLS-1$
    columnMap.put(col2, "2");
    MetadataColumn col3 = ConnectionFactory.eINSTANCE.createMetadataColumn();
    col3.setName(columnName3);
    // $NON-NLS-1$
    columnMap.put(col3, "3");
    RecordMatchingIndicator recordMatchingIndicator = ColumnsetFactory.eINSTANCE.createRecordMatchingIndicator();
    MatchRuleDefinition matchRuleDef = RulesPackage.eINSTANCE.getRulesFactory().createMatchRuleDefinition();
    recordMatchingIndicator.setBuiltInMatchRuleDefinition(matchRuleDef);
    // create match rule
    MatchRule matchRule2 = RulesFactory.eINSTANCE.createMatchRule();
    MatchKeyDefinition createMatchKeyDefinition2 = RulesFactory.eINSTANCE.createMatchKeyDefinition();
    matchRule2.getMatchKeys().add(createMatchKeyDefinition2);
    createMatchKeyDefinition2.setColumn(columnName3);
    createMatchKeyDefinition2.setConfidenceWeight(1);
    // $NON-NLS-1$
    createMatchKeyDefinition2.setName("rule1.matchkey1");
    createMatchKeyDefinition2.setHandleNull(HandleNullEnum.NULL_MATCH_NULL.getValue());
    AlgorithmDefinition createAlgorithmDefinition2 = RulesFactory.eINSTANCE.createAlgorithmDefinition();
    createAlgorithmDefinition2.setAlgorithmType(AttributeMatcherType.EXACT.name());
    createMatchKeyDefinition2.setAlgorithm(createAlgorithmDefinition2);
    matchRuleDef.getMatchRules().add(matchRule2);
    // create match rule
    MatchRule matchRule1 = RulesFactory.eINSTANCE.createMatchRule();
    MatchKeyDefinition createMatchKeyDefinition1 = RulesFactory.eINSTANCE.createMatchKeyDefinition();
    matchRule1.getMatchKeys().add(createMatchKeyDefinition1);
    createMatchKeyDefinition1.setColumn(columnName2);
    createMatchKeyDefinition1.setConfidenceWeight(1);
    // $NON-NLS-1$
    createMatchKeyDefinition1.setName("rule1.matchkey1");
    createMatchKeyDefinition1.setHandleNull(HandleNullEnum.NULL_MATCH_NULL.getValue());
    AlgorithmDefinition createAlgorithmDefinition1 = RulesFactory.eINSTANCE.createAlgorithmDefinition();
    createAlgorithmDefinition1.setAlgorithmType(AttributeMatcherType.EXACT.name());
    createMatchKeyDefinition1.setAlgorithm(createAlgorithmDefinition1);
    matchRuleDef.getMatchRules().add(matchRule1);
    // create block key
    BlockKeyDefinition createBlockKeyDefinition = RulesFactory.eINSTANCE.createBlockKeyDefinition();
    createBlockKeyDefinition.setColumn(columnName1);
    // $NON-NLS-1$
    createBlockKeyDefinition.setName("blockKey1");
    // setPreAlgorithm
    AlgorithmDefinition blockPreAlgorithm = RulesFactory.eINSTANCE.createAlgorithmDefinition();
    blockPreAlgorithm.setAlgorithmType(BlockingKeyPreAlgorithmEnum.NON_ALGO.getValue());
    createBlockKeyDefinition.setPreAlgorithm(blockPreAlgorithm);
    // setAlgorithm
    AlgorithmDefinition blockAlgorithm = RulesFactory.eINSTANCE.createAlgorithmDefinition();
    blockAlgorithm.setAlgorithmType(BlockingKeyAlgorithmEnum.EXACT.getValue());
    createBlockKeyDefinition.setAlgorithm(blockAlgorithm);
    // setPostAlgorithm
    AlgorithmDefinition blockPostAlgorithm = RulesFactory.eINSTANCE.createAlgorithmDefinition();
    blockPostAlgorithm.setAlgorithmType(BlockingKeyPostAlgorithmEnum.NON_ALGO.getValue());
    createBlockKeyDefinition.setPostAlgorithm(blockPostAlgorithm);
    matchRuleDef.getBlockKeys().add(createBlockKeyDefinition);
    List<Object[]> matchRows = new ArrayList<Object[]>();
    // $NON-NLS-1$ //$NON-NLS-2$ //$NON-NLS-3$ //$NON-NLS-4$
    matchRows.add(new String[] { "id1", "name1", "number1", "date1" });
    // $NON-NLS-1$ //$NON-NLS-2$ //$NON-NLS-3$ //$NON-NLS-4$
    matchRows.add(new String[] { "id2", "name1", "number2", "date1" });
    // $NON-NLS-1$ //$NON-NLS-2$ //$NON-NLS-3$ //$NON-NLS-4$
    matchRows.add(new String[] { "id3", "name2", "number2", "date3" });
    // $NON-NLS-1$ //$NON-NLS-2$ //$NON-NLS-3$ //$NON-NLS-4$
    matchRows.add(new String[] { "id4", "name2", "number2", "date1" });
    BlockKeyIndicator blockKeyIndicator = ColumnsetFactory.eINSTANCE.createBlockKeyIndicator();
    ExecuteMatchRuleHandler execHandler = new ExecuteMatchRuleHandler();
    MatchGroupResultConsumer matchResultConsumer = createMatchGroupResultConsumer(columnMap, recordMatchingIndicator);
    TypedReturnCode<MatchGroupResultConsumer> executeResult = execHandler.execute(columnMap, recordMatchingIndicator, matchRows, blockKeyIndicator, matchResultConsumer);
    Assert.assertTrue(executeResult.isOk());
    Assert.assertTrue(executeResult.getMessage() == null);
    Assert.assertTrue(executeResult.getObject() != null);
    MatchGroupResultConsumer ResultConsumer = executeResult.getObject();
    List<Object[]> fullMatchResult = ResultConsumer.getFullMatchResult();
    Assert.assertTrue(fullMatchResult.size() == 4);
    for (int i = 0; i < fullMatchResult.size(); i++) {
        Object[] objectArray = fullMatchResult.get(i);
        Object masterValue = objectArray[7];
        Object idValue = objectArray[0];
        // id2 is because of matchRule1 id4 is because of matchRule2
        if ("id2".equals(idValue) || "id4".equals(idValue)) {
            // $NON-NLS-1$ //$NON-NLS-2$
            Assert.assertFalse(Boolean.parseBoolean(masterValue.toString()));
        } else {
            Assert.assertTrue(Boolean.parseBoolean(masterValue.toString()));
        }
    }
}
Also used : ExecuteMatchRuleHandler(org.talend.dq.analysis.match.ExecuteMatchRuleHandler) BlockKeyIndicator(org.talend.dataquality.indicators.columnset.BlockKeyIndicator) HashMap(java.util.HashMap) ArrayList(java.util.ArrayList) MatchRuleDefinition(org.talend.dataquality.rules.MatchRuleDefinition) MatchRule(org.talend.dataquality.rules.MatchRule) RecordMatchingIndicator(org.talend.dataquality.indicators.columnset.RecordMatchingIndicator) MetadataColumn(org.talend.core.model.metadata.builder.connection.MetadataColumn) MatchGroupResultConsumer(org.talend.dataquality.record.linkage.grouping.MatchGroupResultConsumer) MatchKeyDefinition(org.talend.dataquality.rules.MatchKeyDefinition) AlgorithmDefinition(org.talend.dataquality.rules.AlgorithmDefinition) BlockKeyDefinition(org.talend.dataquality.rules.BlockKeyDefinition) Test(org.junit.Test)

Example 2 with BlockKeyDefinition

use of org.talend.dataquality.rules.BlockKeyDefinition in project tdq-studio-se by Talend.

the class AnalysisRecordGroupingUtils method createAppliedBlockKeyByGenKey.

/**
 * By default for analysis, the applied blocking key will be the key from key generation definition. This will be
 * refined when there is a need to define the applied blocking key manually by user later.
 *
 * @param recordMatchingIndicator
 */
public static void createAppliedBlockKeyByGenKey(RecordMatchingIndicator recordMatchingIndicator) {
    List<AppliedBlockKey> appliedBlockKeys = recordMatchingIndicator.getBuiltInMatchRuleDefinition().getAppliedBlockKeys();
    appliedBlockKeys.clear();
    List<BlockKeyDefinition> blockKeyDefs = recordMatchingIndicator.getBuiltInMatchRuleDefinition().getBlockKeys();
    if (blockKeyDefs != null && blockKeyDefs.size() > 0) {
        AppliedBlockKey appliedBlockKey = RulesPackage.eINSTANCE.getRulesFactory().createAppliedBlockKey();
        appliedBlockKey.setColumn(PluginConstant.BLOCK_KEY);
        appliedBlockKey.setName(PluginConstant.BLOCK_KEY);
        appliedBlockKeys.add(appliedBlockKey);
    }
}
Also used : BlockKeyDefinition(org.talend.dataquality.rules.BlockKeyDefinition) AppliedBlockKey(org.talend.dataquality.rules.AppliedBlockKey)

Example 3 with BlockKeyDefinition

use of org.talend.dataquality.rules.BlockKeyDefinition in project tdq-studio-se by Talend.

the class MatchRuleDefinitionWriter method copy.

/**
 * copy the block/match keys from matchRule to ruleDefinition.
 *
 * @param ruleDefinition: created one
 * @param matchRule: exported one from the match analysis
 */
public void copy(MatchRuleDefinition ruleDefinition, MatchRuleDefinition matchRule) {
    // copy blocking keys
    ruleDefinition.getBlockKeys().clear();
    if (matchRule.getBlockKeys() != null && matchRule.getBlockKeys().size() > 0) {
        for (BlockKeyDefinition blockKey : matchRule.getBlockKeys()) {
            BlockKeyDefinition copy = EcoreUtil.copy(blockKey);
            // should not empty the column value when export it
            ruleDefinition.getBlockKeys().add(copy);
        }
    }
    // copy match keys in each match rules
    ruleDefinition.getMatchRules().clear();
    if (matchRule.getMatchRules() != null && matchRule.getMatchRules().size() > 0) {
        for (MatchRule oneMatchRule : matchRule.getMatchRules()) {
            ruleDefinition.getMatchRules().add(EObjectHelper.deepCopy(oneMatchRule));
        }
    }
}
Also used : BlockKeyDefinition(org.talend.dataquality.rules.BlockKeyDefinition) MatchRule(org.talend.dataquality.rules.MatchRule)

Example 4 with BlockKeyDefinition

use of org.talend.dataquality.rules.BlockKeyDefinition in project tdq-studio-se by Talend.

the class BlockingKeySection method removeTableItem.

/*
     * (non-Javadoc)
     * 
     * @see org.talend.dataquality.record.linkage.ui.section.AbstractMatchAnaysisTableSection#removeTableItem()
     */
@Override
public void removeTableItem() {
    boolean success = false;
    ISelection selectItems = tableComposite.getSelectItems();
    if (selectItems instanceof StructuredSelection) {
        Iterator<BlockKeyDefinition> iterator = ((StructuredSelection) selectItems).iterator();
        while (iterator.hasNext()) {
            BlockKeyDefinition next = iterator.next();
            removeBlockingKey(next);
            success = true;
        }
        if (success) {
            listeners.firePropertyChange(MatchAnalysisConstant.ISDIRTY_PROPERTY, true, false);
            listeners.firePropertyChange(MatchAnalysisConstant.MATCH_RULE_TAB_SWITCH, true, false);
        }
    }
}
Also used : ISelection(org.eclipse.jface.viewers.ISelection) StructuredSelection(org.eclipse.jface.viewers.StructuredSelection) BlockKeyDefinition(org.talend.dataquality.rules.BlockKeyDefinition)

Example 5 with BlockKeyDefinition

use of org.talend.dataquality.rules.BlockKeyDefinition in project tdq-studio-se by Talend.

the class BlockingKeySection method isKeyDefinitionAdded.

/*
     * (non-Javadoc)
     * 
     * @see
     * org.talend.dataquality.record.linkage.ui.section.AbstractMatchAnaysisTableSection#isKeyDefinitionAdded(java.lang
     * .String)
     */
@Override
public Boolean isKeyDefinitionAdded(String columnName) {
    Boolean isAdded = Boolean.FALSE;
    RecordMatchingIndicator recordMatchingIndicator = MatchRuleAnlaysisUtils.getRecordMatchIndicatorFromAna(analysis);
    List<BlockKeyDefinition> keyDefs = recordMatchingIndicator.getBuiltInMatchRuleDefinition().getBlockKeys();
    for (KeyDefinition keyDef : keyDefs) {
        // the key's name can NOT be same, the column can be same
        if (StringUtils.equals(columnName, keyDef.getName())) {
            isAdded = Boolean.TRUE;
            break;
        }
    }
    return isAdded;
}
Also used : BlockKeyDefinition(org.talend.dataquality.rules.BlockKeyDefinition) RecordMatchingIndicator(org.talend.dataquality.indicators.columnset.RecordMatchingIndicator) BlockKeyDefinition(org.talend.dataquality.rules.BlockKeyDefinition) KeyDefinition(org.talend.dataquality.rules.KeyDefinition)

Aggregations

BlockKeyDefinition (org.talend.dataquality.rules.BlockKeyDefinition)21 ArrayList (java.util.ArrayList)11 RecordMatchingIndicator (org.talend.dataquality.indicators.columnset.RecordMatchingIndicator)7 HashMap (java.util.HashMap)6 MatchKeyDefinition (org.talend.dataquality.rules.MatchKeyDefinition)6 MetadataColumn (org.talend.core.model.metadata.builder.connection.MetadataColumn)5 AlgorithmDefinition (org.talend.dataquality.rules.AlgorithmDefinition)5 MatchRule (org.talend.dataquality.rules.MatchRule)5 KeyDefinition (org.talend.dataquality.rules.KeyDefinition)4 MatchRuleDefinition (org.talend.dataquality.rules.MatchRuleDefinition)4 Map (java.util.Map)3 ISelection (org.eclipse.jface.viewers.ISelection)3 StructuredSelection (org.eclipse.jface.viewers.StructuredSelection)3 Test (org.junit.Test)3 BlockKeyIndicator (org.talend.dataquality.indicators.columnset.BlockKeyIndicator)3 MatchGroupResultConsumer (org.talend.dataquality.record.linkage.grouping.MatchGroupResultConsumer)3 ExecuteMatchRuleHandler (org.talend.dq.analysis.match.ExecuteMatchRuleHandler)3 AppliedBlockKey (org.talend.dataquality.rules.AppliedBlockKey)2 GridData (org.eclipse.swt.layout.GridData)1 GridLayout (org.eclipse.swt.layout.GridLayout)1