use of org.apache.hadoop.hive.metastore.api.ColumnStatisticsData._Fields in project hive by apache.
the class TestHBaseStore method doubleTableStatistics.
@Test
public void doubleTableStatistics() throws Exception {
createMockTable(DOUBLE_TYPE);
// Add a double table stats for DOUBLE_COL to DB
// Because of the way our mock implementation works we actually need to not create the table
// before we set statistics on it.
ColumnStatistics stats = new ColumnStatistics();
// Get a default ColumnStatisticsDesc for table level stats
ColumnStatisticsDesc desc = getMockTblColStatsDesc();
stats.setStatsDesc(desc);
// Get one of the pre-created ColumnStatisticsObj
ColumnStatisticsObj obj = doubleColStatsObjs.get(0);
DoubleColumnStatsData doubleData = obj.getStatsData().getDoubleStats();
// Add to DB
stats.addToStatsObj(obj);
store.updateTableColumnStatistics(stats);
// Get from DB
ColumnStatistics statsFromDB = store.getTableColumnStatistics(DB, TBL, Arrays.asList(DOUBLE_COL));
// Compare ColumnStatisticsDesc
Assert.assertEquals(desc.getLastAnalyzed(), statsFromDB.getStatsDesc().getLastAnalyzed());
Assert.assertEquals(DB, statsFromDB.getStatsDesc().getDbName());
Assert.assertEquals(TBL, statsFromDB.getStatsDesc().getTableName());
Assert.assertTrue(statsFromDB.getStatsDesc().isIsTblLevel());
// Compare ColumnStatisticsObj
Assert.assertEquals(1, statsFromDB.getStatsObjSize());
ColumnStatisticsObj objFromDB = statsFromDB.getStatsObj().get(0);
ColumnStatisticsData dataFromDB = objFromDB.getStatsData();
// Compare ColumnStatisticsData
Assert.assertEquals(ColumnStatisticsData._Fields.DOUBLE_STATS, dataFromDB.getSetField());
// Compare DoubleColumnStatsData
DoubleColumnStatsData doubleDataFromDB = dataFromDB.getDoubleStats();
Assert.assertEquals(doubleData.getHighValue(), doubleDataFromDB.getHighValue(), 0.01);
Assert.assertEquals(doubleData.getLowValue(), doubleDataFromDB.getLowValue(), 0.01);
Assert.assertEquals(doubleData.getNumNulls(), doubleDataFromDB.getNumNulls());
Assert.assertEquals(doubleData.getNumDVs(), doubleDataFromDB.getNumDVs());
}
use of org.apache.hadoop.hive.metastore.api.ColumnStatisticsData._Fields in project hive by apache.
the class MetaDataFormatUtils method extractColumnValues.
public static String[] extractColumnValues(FieldSchema col, boolean isColStatsAvailable, ColumnStatisticsObj columnStatisticsObj) {
List<String> ret = new ArrayList<>();
ret.add(col.getName());
ret.add(col.getType());
if (isColStatsAvailable) {
if (columnStatisticsObj != null) {
ColumnStatisticsData csd = columnStatisticsObj.getStatsData();
// @formatter:off
if (csd.isSetBinaryStats()) {
BinaryColumnStatsData bcsd = csd.getBinaryStats();
ret.addAll(Lists.newArrayList("", "", "" + bcsd.getNumNulls(), "", "" + bcsd.getAvgColLen(), "" + bcsd.getMaxColLen(), "", "", convertToString(bcsd.getBitVectors())));
} else if (csd.isSetStringStats()) {
StringColumnStatsData scsd = csd.getStringStats();
ret.addAll(Lists.newArrayList("", "", "" + scsd.getNumNulls(), "" + scsd.getNumDVs(), "" + scsd.getAvgColLen(), "" + scsd.getMaxColLen(), "", "", convertToString(scsd.getBitVectors())));
} else if (csd.isSetBooleanStats()) {
BooleanColumnStatsData bcsd = csd.getBooleanStats();
ret.addAll(Lists.newArrayList("", "", "" + bcsd.getNumNulls(), "", "", "", "" + bcsd.getNumTrues(), "" + bcsd.getNumFalses(), convertToString(bcsd.getBitVectors())));
} else if (csd.isSetDecimalStats()) {
DecimalColumnStatsData dcsd = csd.getDecimalStats();
ret.addAll(Lists.newArrayList(convertToString(dcsd.getLowValue()), convertToString(dcsd.getHighValue()), "" + dcsd.getNumNulls(), "" + dcsd.getNumDVs(), "", "", "", "", convertToString(dcsd.getBitVectors())));
} else if (csd.isSetDoubleStats()) {
DoubleColumnStatsData dcsd = csd.getDoubleStats();
ret.addAll(Lists.newArrayList("" + dcsd.getLowValue(), "" + dcsd.getHighValue(), "" + dcsd.getNumNulls(), "" + dcsd.getNumDVs(), "", "", "", "", convertToString(dcsd.getBitVectors())));
} else if (csd.isSetLongStats()) {
LongColumnStatsData lcsd = csd.getLongStats();
ret.addAll(Lists.newArrayList("" + lcsd.getLowValue(), "" + lcsd.getHighValue(), "" + lcsd.getNumNulls(), "" + lcsd.getNumDVs(), "", "", "", "", convertToString(lcsd.getBitVectors())));
} else if (csd.isSetDateStats()) {
DateColumnStatsData dcsd = csd.getDateStats();
ret.addAll(Lists.newArrayList(convertToString(dcsd.getLowValue()), convertToString(dcsd.getHighValue()), "" + dcsd.getNumNulls(), "" + dcsd.getNumDVs(), "", "", "", "", convertToString(dcsd.getBitVectors())));
}
// @formatter:on
} else {
ret.addAll(Lists.newArrayList("", "", "", "", "", "", "", "", ""));
}
}
ret.add(getComment(col));
return ret.toArray(new String[] {});
}
use of org.apache.hadoop.hive.metastore.api.ColumnStatisticsData._Fields in project hive by apache.
the class ColumnStatisticsObjTranslator method unpackPrimitiveObject.
private static void unpackPrimitiveObject(ObjectInspector oi, Object o, String fieldName, ColumnStatisticsObj statsObj) throws UnsupportedDoubleException {
if (o == null) {
return;
}
// First infer the type of object
if (fieldName.equals("columntype")) {
PrimitiveObjectInspector poi = (PrimitiveObjectInspector) oi;
String s = ((StringObjectInspector) poi).getPrimitiveJavaObject(o);
ColumnStatisticsData statsData = new ColumnStatisticsData();
if (s.equalsIgnoreCase("long")) {
LongColumnStatsDataInspector longStats = new LongColumnStatsDataInspector();
statsData.setLongStats(longStats);
statsObj.setStatsData(statsData);
} else if (s.equalsIgnoreCase("double")) {
DoubleColumnStatsDataInspector doubleStats = new DoubleColumnStatsDataInspector();
statsData.setDoubleStats(doubleStats);
statsObj.setStatsData(statsData);
} else if (s.equalsIgnoreCase("string")) {
StringColumnStatsDataInspector stringStats = new StringColumnStatsDataInspector();
statsData.setStringStats(stringStats);
statsObj.setStatsData(statsData);
} else if (s.equalsIgnoreCase("boolean")) {
BooleanColumnStatsData booleanStats = new BooleanColumnStatsData();
statsData.setBooleanStats(booleanStats);
statsObj.setStatsData(statsData);
} else if (s.equalsIgnoreCase("binary")) {
BinaryColumnStatsData binaryStats = new BinaryColumnStatsData();
statsData.setBinaryStats(binaryStats);
statsObj.setStatsData(statsData);
} else if (s.equalsIgnoreCase("decimal")) {
DecimalColumnStatsDataInspector decimalStats = new DecimalColumnStatsDataInspector();
statsData.setDecimalStats(decimalStats);
statsObj.setStatsData(statsData);
} else if (s.equalsIgnoreCase("date")) {
DateColumnStatsDataInspector dateStats = new DateColumnStatsDataInspector();
statsData.setDateStats(dateStats);
statsObj.setStatsData(statsData);
}
} else {
// invoke the right unpack method depending on data type of the column
if (statsObj.getStatsData().isSetBooleanStats()) {
unpackBooleanStats(oi, o, fieldName, statsObj);
} else if (statsObj.getStatsData().isSetLongStats()) {
unpackLongStats(oi, o, fieldName, statsObj);
} else if (statsObj.getStatsData().isSetDoubleStats()) {
unpackDoubleStats(oi, o, fieldName, statsObj);
} else if (statsObj.getStatsData().isSetStringStats()) {
unpackStringStats(oi, o, fieldName, statsObj);
} else if (statsObj.getStatsData().isSetBinaryStats()) {
unpackBinaryStats(oi, o, fieldName, statsObj);
} else if (statsObj.getStatsData().isSetDecimalStats()) {
unpackDecimalStats(oi, o, fieldName, statsObj);
} else if (statsObj.getStatsData().isSetDateStats()) {
unpackDateStats(oi, o, fieldName, statsObj);
}
}
}
use of org.apache.hadoop.hive.metastore.api.ColumnStatisticsData._Fields in project hive by apache.
the class BinaryColumnStatsAggregator method aggregate.
@Override
public ColumnStatisticsObj aggregate(List<ColStatsObjWithSourceInfo> colStatsWithSourceInfo, List<String> partNames, boolean areAllPartsFound) throws MetaException {
ColumnStatisticsObj statsObj = null;
String colType = null;
String colName = null;
BinaryColumnStatsData aggregateData = null;
for (ColStatsObjWithSourceInfo csp : colStatsWithSourceInfo) {
ColumnStatisticsObj cso = csp.getColStatsObj();
if (statsObj == null) {
colName = cso.getColName();
colType = cso.getColType();
statsObj = ColumnStatsAggregatorFactory.newColumnStaticsObj(colName, colType, cso.getStatsData().getSetField());
}
BinaryColumnStatsData newData = cso.getStatsData().getBinaryStats();
if (aggregateData == null) {
aggregateData = newData.deepCopy();
} else {
aggregateData.setMaxColLen(Math.max(aggregateData.getMaxColLen(), newData.getMaxColLen()));
aggregateData.setAvgColLen(Math.max(aggregateData.getAvgColLen(), newData.getAvgColLen()));
aggregateData.setNumNulls(aggregateData.getNumNulls() + newData.getNumNulls());
}
}
ColumnStatisticsData columnStatisticsData = new ColumnStatisticsData();
columnStatisticsData.setBinaryStats(aggregateData);
statsObj.setStatsData(columnStatisticsData);
return statsObj;
}
use of org.apache.hadoop.hive.metastore.api.ColumnStatisticsData._Fields in project hive by apache.
the class ColumnStatsAggregatorFactory method newColumnStaticsObj.
public static ColumnStatisticsObj newColumnStaticsObj(String colName, String colType, _Fields type) {
ColumnStatisticsObj cso = new ColumnStatisticsObj();
ColumnStatisticsData csd = new ColumnStatisticsData();
cso.setColName(colName);
cso.setColType(colType);
switch(type) {
case BOOLEAN_STATS:
csd.setBooleanStats(new BooleanColumnStatsData());
break;
case LONG_STATS:
csd.setLongStats(new LongColumnStatsDataInspector());
break;
case DATE_STATS:
csd.setDateStats(new DateColumnStatsDataInspector());
break;
case DOUBLE_STATS:
csd.setDoubleStats(new DoubleColumnStatsDataInspector());
break;
case STRING_STATS:
csd.setStringStats(new StringColumnStatsDataInspector());
break;
case BINARY_STATS:
csd.setBinaryStats(new BinaryColumnStatsData());
break;
case DECIMAL_STATS:
csd.setDecimalStats(new DecimalColumnStatsDataInspector());
break;
default:
throw new RuntimeException("Woh, bad. Unknown stats type!");
}
cso.setStatsData(csd);
return cso;
}
Aggregations