use of org.talend.dataquality.indicators.DuplicateCountIndicator in project tdq-studio-se by Talend.
the class ColumnSetMultiValueIndicatorImpl method basicSetDuplicateCountIndicator.
/**
* <!-- begin-user-doc --> <!-- end-user-doc -->
* @generated
*/
public NotificationChain basicSetDuplicateCountIndicator(DuplicateCountIndicator newDuplicateCountIndicator, NotificationChain msgs) {
DuplicateCountIndicator oldDuplicateCountIndicator = duplicateCountIndicator;
duplicateCountIndicator = newDuplicateCountIndicator;
if (eNotificationRequired()) {
ENotificationImpl notification = new ENotificationImpl(this, Notification.SET, ColumnsetPackage.COLUMN_SET_MULTI_VALUE_INDICATOR__DUPLICATE_COUNT_INDICATOR, oldDuplicateCountIndicator, newDuplicateCountIndicator);
if (msgs == null)
msgs = notification;
else
msgs.add(notification);
}
return msgs;
}
use of org.talend.dataquality.indicators.DuplicateCountIndicator in project tdq-studio-se by Talend.
the class DelimitedFileIndicatorEvaluator method executeSqlQuery.
@Override
protected ReturnCode executeSqlQuery(String sqlStatement) {
ReturnCode returnCode = new ReturnCode(true);
if (delimitedFileconnection == null) {
delimitedFileconnection = (DelimitedFileConnection) analysis.getContext().getConnection();
}
if (delimitedFileconnection.isContextMode()) {
IRepositoryContextService service = CoreRuntimePlugin.getInstance().getRepositoryContextService();
delimitedFileconnection = (DelimitedFileConnection) service.cloneOriginalValueConnection(delimitedFileconnection);
}
String path = JavaSqlFactory.getURL(delimitedFileconnection);
IPath iPath = new Path(path);
File file = iPath.toFile();
if (!file.exists()) {
// $NON-NLS-1$
returnCode.setReturnCode(Messages.getString("DelimitedFileIndicatorEvaluator.CanNotFindFile"), false);
return returnCode;
}
List<ModelElement> analysisElementList = this.analysis.getContext().getAnalysedElements();
EMap<Indicator, AnalyzedDataSet> indicToRowMap = analysis.getResults().getIndicToRowMap();
indicToRowMap.clear();
List<MetadataColumn> columnElementList = new ArrayList<MetadataColumn>();
for (int i = 0; i < analysisElementList.size(); i++) {
MetadataColumn mColumn = (MetadataColumn) analysisElementList.get(i);
MetadataTable mTable = ColumnHelper.getColumnOwnerAsMetadataTable(mColumn);
columnElementList = mTable == null ? columnElementList : mTable.getColumns();
if (!columnElementList.isEmpty()) {
break;
}
}
ReturnCode readDataReturnCode = new ReturnCode(true);
// use CsvReader to parse.
if (Escape.CSV.equals(delimitedFileconnection.getEscapeType())) {
readDataReturnCode = useCsvReader(file, analysisElementList, columnElementList, indicToRowMap);
} else {
readDataReturnCode = useDelimitedReader(analysisElementList, columnElementList, indicToRowMap);
}
// handle error message
if (!readDataReturnCode.isOk()) {
Display.getDefault().asyncExec(new Runnable() {
public void run() {
MessageDialog.openWarning(PlatformUI.getWorkbench().getActiveWorkbenchWindow().getShell(), // $NON-NLS-1$
Messages.getString("DelimitedFileIndicatorEvaluator.badlyForm.Title"), // $NON-NLS-1$
Messages.getString("DelimitedFileIndicatorEvaluator.badlyForm.Message"));
}
});
}
// Added yyin 20120608 TDQ-3589
for (MetadataColumn col : columnElementList) {
List<Indicator> indicators = getIndicators(col.getLabel());
for (Indicator indicator : indicators) {
if (indicator instanceof DuplicateCountIndicator) {
AnalyzedDataSet analyzedDataSet = indicToRowMap.get(indicator);
if (analyzedDataSet == null) {
analyzedDataSet = AnalysisFactory.eINSTANCE.createAnalyzedDataSet();
indicToRowMap.put(indicator, analyzedDataSet);
analyzedDataSet.setDataCount(analysis.getParameters().getMaxNumberRows());
analyzedDataSet.setRecordSize(0);
}
// indicator.finalizeComputation();
addResultToIndicatorToRowMap(indicator, indicToRowMap);
}
}
}
return returnCode;
}
use of org.talend.dataquality.indicators.DuplicateCountIndicator in project tdq-studio-se by Talend.
the class DelimitedFileIndicatorEvaluator method handleByARow.
private ReturnCode handleByARow(String[] rowValues, long currentRow, List<ModelElement> analysisElementList, List<MetadataColumn> columnElementList, EMap<Indicator, AnalyzedDataSet> indicToRowMap) {
ReturnCode returnCode = new ReturnCode(true);
Object object = null;
int maxNumberRows = analysis.getParameters().getMaxNumberRows();
int recordIncrement = 0;
element: for (int i = 0; i < analysisElementList.size(); i++) {
MetadataColumn mColumn = (MetadataColumn) analysisElementList.get(i);
Integer position = ColumnHelper.getColumnIndex(mColumn);
// warning with a file of badly form
if (position == null || position >= rowValues.length) {
log.warn(// $NON-NLS-1$
Messages.getString(// $NON-NLS-1$
"DelimitedFileIndicatorEvaluator.incorrectData", mColumn.getLabel(), currentRow, delimitedFileconnection.getFilePath()));
returnCode.setOk(false);
continue;
}
object = TalendTypeConvert.convertToObject(mColumn.getTalendType(), rowValues[position], mColumn.getPattern());
List<Indicator> indicators = getIndicators(mColumn.getLabel());
for (Indicator indicator : indicators) {
if (!continueRun()) {
break element;
}
// bug 19036,to irregularly data,still compute for RowCountIndicator
if (object == null && !(indicator instanceof RowCountIndicator)) {
continue element;
}
// Added yyin 20120608 TDQ-3589
if (indicator instanceof DuplicateCountIndicator) {
((DuplicateCountIndicator) indicator).handle(object, rowValues);
} else {
// ~
indicator.handle(object);
}
AnalyzedDataSet analyzedDataSet = indicToRowMap.get(indicator);
if (analyzedDataSet == null) {
analyzedDataSet = AnalysisFactory.eINSTANCE.createAnalyzedDataSet();
indicToRowMap.put(indicator, analyzedDataSet);
analyzedDataSet.setDataCount(maxNumberRows);
analyzedDataSet.setRecordSize(0);
}
// see IndicatorEvaluator line 166, the logic is almost the same
if (analysis.getParameters().isStoreData()) {
if (indicator.mustStoreRow()) {
List<Object[]> valueObjectList = initDataSet(indicator, indicToRowMap, object);
recordIncrement = valueObjectList.size();
List<Object> inputRowList = new ArrayList<Object>();
for (int j = 0; j < rowValues.length; j++) {
Object newobject = rowValues[j];
if (indicator.isUsedMapDBMode()) {
inputRowList.add(newobject == null ? PluginConstant.NULL_STRING : newobject);
continue;
} else {
if (recordIncrement < maxNumberRows) {
if (recordIncrement < valueObjectList.size()) {
valueObjectList.get(recordIncrement)[j] = newobject;
} else {
Object[] valueObject = new Object[rowValues.length];
valueObject[j] = newobject;
valueObjectList.add(valueObject);
}
} else {
break;
}
}
}
if (indicator.isUsedMapDBMode()) {
MapDBUtils.handleDrillDownData(object, inputRowList, indicator);
}
} else if (indicator instanceof UniqueCountIndicator && analysis.getResults().getIndicToRowMap().get(indicator).getData() != null) {
List<Object[]> removeValueObjectList = analysis.getResults().getIndicToRowMap().get(indicator).getData();
if (columnElementList.size() == 0) {
continue;
}
int offsetting = columnElementList.indexOf(indicator.getAnalyzedElement());
for (Object[] dataObject : removeValueObjectList) {
// Added yyin 20120611 TDQ5279
if (object instanceof Integer) {
if (object.equals(Integer.parseInt((String) dataObject[offsetting]))) {
removeValueObjectList.remove(dataObject);
break;
}
}
// ~
if (dataObject[offsetting].equals(object)) {
removeValueObjectList.remove(dataObject);
break;
}
}
}
}
}
}
return returnCode;
}
use of org.talend.dataquality.indicators.DuplicateCountIndicator in project tdq-studio-se by Talend.
the class CountsIndicatorImpl method basicSetDuplicateCountIndicator.
/**
* <!-- begin-user-doc --> <!-- end-user-doc -->
* @generated
*/
public NotificationChain basicSetDuplicateCountIndicator(DuplicateCountIndicator newDuplicateCountIndicator, NotificationChain msgs) {
DuplicateCountIndicator oldDuplicateCountIndicator = duplicateCountIndicator;
duplicateCountIndicator = newDuplicateCountIndicator;
if (eNotificationRequired()) {
ENotificationImpl notification = new ENotificationImpl(this, Notification.SET, IndicatorsPackage.COUNTS_INDICATOR__DUPLICATE_COUNT_INDICATOR, oldDuplicateCountIndicator, newDuplicateCountIndicator);
if (msgs == null)
msgs = notification;
else
msgs.add(notification);
}
return msgs;
}
use of org.talend.dataquality.indicators.DuplicateCountIndicator in project tdq-studio-se by Talend.
the class IndicatorEvaluationMain method main.
/**
* DOC scorreia Comment method "main".
*
* @param args
*/
public static void main(String[] args) {
TypedProperties connectionParams = PropertiesLoader.getProperties(IndicatorEvaluator.class, "db.properties");
String driverClassName = connectionParams.getProperty("driver");
String dbUrl = connectionParams.getProperty("url");
try {
TimeTracer tt = new TimeTracer("Indicator evaluation", null);
tt.start();
// create connection
Connection connection = ConnectionUtils.createConnection(dbUrl, driverClassName, connectionParams);
String database = "test";
String tableName = "my_test";
// --- columns to analyze
String[] columnsArray = new String[] { // 0
"my_int", // 1
"my_double", // 2
"my_text", // 4
"my_date", // 3
"my_string", // 5
"my_int_null" };
List<String> columns = Arrays.asList(columnsArray);
// store in file
File file = new File("out/columnTest_0.1.ana");
EMFUtil util = new EMFUtil();
Resource resource = util.getResourceSet().createResource(URI.createFileURI(file.getAbsolutePath()));
rContents = resource.getContents();
evaluator.setConnection(connection);
// --- create indicators
RowCountIndicator rowCountIndicator = IndicatorsFactory.eINSTANCE.createRowCountIndicator();
NullCountIndicator nullCountIndicator = IndicatorsFactory.eINSTANCE.createNullCountIndicator();
DistinctCountIndicator distinctCountIndicator = IndicatorsFactory.eINSTANCE.createDistinctCountIndicator();
DistinctCountIndicator distinctCountIndicator2 = IndicatorsFactory.eINSTANCE.createDistinctCountIndicator();
UniqueCountIndicator uniqueCountIndicator = IndicatorsFactory.eINSTANCE.createUniqueCountIndicator();
DuplicateCountIndicator duplicateCountIndicator = IndicatorsFactory.eINSTANCE.createDuplicateCountIndicator();
BlankCountIndicator blankCountIndicator = IndicatorsFactory.eINSTANCE.createBlankCountIndicator();
MinLengthIndicator minLengthIndicator = IndicatorsFactory.eINSTANCE.createMinLengthIndicator();
MaxLengthIndicator maxLengthIndicator = IndicatorsFactory.eINSTANCE.createMaxLengthIndicator();
AverageLengthIndicator averageLengthIndicator = IndicatorsFactory.eINSTANCE.createAverageLengthIndicator();
AverageLengthIndicator averageLengthIndicator2 = IndicatorsFactory.eINSTANCE.createAverageLengthIndicator();
ModeIndicator modeIndicator = IndicatorsFactory.eINSTANCE.createModeIndicator();
FrequencyIndicator textFrequencyIndicator = IndicatorsFactory.eINSTANCE.createFrequencyIndicator();
// store in freq indic
// textFrequencyIndicator.setDistinctCountIndicator(distinctCountIndicator);
// textFrequencyIndicator.setDistinctCountIndicator(distinctCountIndicator2);
// textFrequencyIndicator.setUniqueCountIndicator(uniqueCountIndicator);
// textFrequencyIndicator.setDuplicateCountIndicator(duplicateCountIndicator);
// textFrequencyIndicator.setModeIndicator(modeIndicator);
MeanIndicator doubleMeanIndicator = IndicatorsFactory.eINSTANCE.createMeanIndicator();
MeanIndicator integerMeanIndicator = IndicatorsFactory.eINSTANCE.createMeanIndicator();
MedianIndicator medianIndicator = IndicatorsFactory.eINSTANCE.createMedianIndicator();
SumIndicator integerSumIndicator = IndicatorsFactory.eINSTANCE.createSumIndicator();
addIndicator(columnsArray[0], medianIndicator);
addIndicator(columnsArray[1], doubleMeanIndicator);
addIndicator(columnsArray[2], blankCountIndicator);
addIndicator(columnsArray[5], nullCountIndicator);
// addIndicator(columnsArray[2], textFrequencyIndicator);
// addIndicator(columnsArray[2], distinctCountIndicator); // probably not useful?
// addIndicator(columnsArray[2], uniqueCountIndicator); // probably not useful?
// addIndicator(columnsArray[2], duplicateCountIndicator); // probably not useful?
// addIndicator(columnsArray[2], modeIndicator); // probably not useful?
addIndicator(columnsArray[3], rowCountIndicator);
addIndicator(columnsArray[5], integerSumIndicator);
addIndicator(columnsArray[5], integerMeanIndicator);
addIndicator(columnsArray[2], averageLengthIndicator);
addIndicator(columnsArray[3], averageLengthIndicator2);
addIndicator(columnsArray[3], minLengthIndicator);
addIndicator(columnsArray[3], maxLengthIndicator);
// build query on columns
// TODO scorreia add filter somewhere here...
String selectCols = sqlSelectColumns(database, tableName, columns);
// --- create a description of the column set
QueryExpression queryExpression = DatatypesFactory.eINSTANCE.createQueryExpression();
queryExpression.setBody(selectCols);
// TODO scorreia externalize this as a constant
queryExpression.setLanguage("SQL");
tt.start("compute");
evaluator.setFetchSize(10000);
evaluator.evaluateIndicators(selectCols, true);
tt.end("compute");
// Print indicators the median
System.out.println("Median=" + medianIndicator.getMedian());
System.out.println("# Unique values= " + textFrequencyIndicator.getUniqueValueCount());
System.out.println("# Distinct values= " + textFrequencyIndicator.getDistinctValueCount());
for (String col : columns) {
printIndicators(evaluator.getIndicators(col));
}
tt.start("save");
util.save();
tt.end("saved in " + file.getAbsolutePath());
tt.end();
CwmResource cwmR = (CwmResource) resource;
String id = cwmR.getID(medianIndicator);
System.out.println("ecore util.getId= " + EcoreUtil.getID(medianIndicator));
System.out.println("uuId= " + id);
// test reload this file
// LoadSerialData.main(args);
} catch (SQLException e) {
// TODO Auto-generated catch block
log.error(e, e);
} catch (InstantiationException e) {
// TODO Auto-generated catch block
log.error(e, e);
} catch (IllegalAccessException e) {
// TODO Auto-generated catch block
log.error(e, e);
} catch (ClassNotFoundException e) {
// TODO Auto-generated catch block
log.error(e, e);
}
}
Aggregations