use of org.apache.hadoop.hive.ql.exec.vector.expressions.VectorExpression in project hive by apache.
the class TestVectorBetweenIn method doVectorBetweenInTest.
private boolean doVectorBetweenInTest(TypeInfo typeInfo, BetweenInVariation betweenInVariation, List<Object> compareList, List<String> columns, String[] columnNames, TypeInfo[] typeInfos, DataTypePhysicalVariation[] dataTypePhysicalVariations, List<ExprNodeDesc> children, GenericUDF udf, ExprNodeGenericFuncDesc exprDesc, BetweenInTestMode betweenInTestMode, VectorRandomBatchSource batchSource, ObjectInspector objectInspector, TypeInfo outputTypeInfo, Object[] resultObjects) throws Exception {
HiveConf hiveConf = new HiveConf();
if (betweenInTestMode == BetweenInTestMode.ADAPTOR) {
hiveConf.setBoolVar(HiveConf.ConfVars.HIVE_TEST_VECTOR_ADAPTOR_OVERRIDE, true);
}
final boolean isFilter = betweenInVariation.isFilter;
VectorizationContext vectorizationContext = new VectorizationContext("name", columns, Arrays.asList(typeInfos), Arrays.asList(dataTypePhysicalVariations), hiveConf);
VectorExpression vectorExpression = vectorizationContext.getVectorExpression(exprDesc, (isFilter ? VectorExpressionDescriptor.Mode.FILTER : VectorExpressionDescriptor.Mode.PROJECTION));
vectorExpression.transientInit(hiveConf);
if (betweenInTestMode == BetweenInTestMode.VECTOR_EXPRESSION) {
String vecExprString = vectorExpression.toString();
if (vectorExpression instanceof VectorUDFAdaptor) {
System.out.println("*NO NATIVE VECTOR EXPRESSION* typeInfo " + typeInfo.toString() + " betweenInTestMode " + betweenInTestMode + " betweenInVariation " + betweenInVariation + " vectorExpression " + vecExprString);
} else if (dataTypePhysicalVariations[0] == DataTypePhysicalVariation.DECIMAL_64) {
final String nameToCheck = vectorExpression.getClass().getSimpleName();
if (!nameToCheck.contains("Decimal64")) {
System.out.println("*EXPECTED DECIMAL_64 VECTOR EXPRESSION* typeInfo " + typeInfo.toString() + " betweenInTestMode " + betweenInTestMode + " betweenInVariation " + betweenInVariation + " vectorExpression " + vecExprString);
}
}
}
// System.out.println("*VECTOR EXPRESSION* " + vectorExpression.getClass().getSimpleName());
/*
System.out.println(
"*DEBUG* typeInfo " + typeInfo.toString() +
" betweenInTestMode " + betweenInTestMode +
" betweenInVariation " + betweenInVariation +
" vectorExpression " + vectorExpression.toString());
*/
VectorRandomRowSource rowSource = batchSource.getRowSource();
VectorizedRowBatchCtx batchContext = new VectorizedRowBatchCtx(columnNames, rowSource.typeInfos(), rowSource.dataTypePhysicalVariations(), /* dataColumnNums */
null, /* partitionColumnCount */
0, /* virtualColumnCount */
0, /* neededVirtualColumns */
null, vectorizationContext.getScratchColumnTypeNames(), vectorizationContext.getScratchDataTypePhysicalVariations());
VectorizedRowBatch batch = batchContext.createVectorizedRowBatch();
VectorExtractRow resultVectorExtractRow = null;
Object[] scrqtchRow = null;
if (!isFilter) {
resultVectorExtractRow = new VectorExtractRow();
final int outputColumnNum = vectorExpression.getOutputColumnNum();
resultVectorExtractRow.init(new TypeInfo[] { outputTypeInfo }, new int[] { outputColumnNum });
scrqtchRow = new Object[1];
}
boolean copySelectedInUse = false;
int[] copySelected = new int[VectorizedRowBatch.DEFAULT_SIZE];
batchSource.resetBatchIteration();
int rowIndex = 0;
while (true) {
if (!batchSource.fillNextBatch(batch)) {
break;
}
final int originalBatchSize = batch.size;
if (isFilter) {
copySelectedInUse = batch.selectedInUse;
if (batch.selectedInUse) {
System.arraycopy(batch.selected, 0, copySelected, 0, originalBatchSize);
}
}
// In filter mode, the batch size can be made smaller.
vectorExpression.evaluate(batch);
if (!isFilter) {
extractResultObjects(batch, rowIndex, resultVectorExtractRow, scrqtchRow, objectInspector, resultObjects);
} else {
final int currentBatchSize = batch.size;
if (copySelectedInUse && batch.selectedInUse) {
int selectIndex = 0;
for (int i = 0; i < originalBatchSize; i++) {
final int originalBatchIndex = copySelected[i];
final boolean booleanResult;
if (selectIndex < currentBatchSize && batch.selected[selectIndex] == originalBatchIndex) {
booleanResult = true;
selectIndex++;
} else {
booleanResult = false;
}
resultObjects[rowIndex + i] = new BooleanWritable(booleanResult);
}
} else if (batch.selectedInUse) {
int selectIndex = 0;
for (int i = 0; i < originalBatchSize; i++) {
final boolean booleanResult;
if (selectIndex < currentBatchSize && batch.selected[selectIndex] == i) {
booleanResult = true;
selectIndex++;
} else {
booleanResult = false;
}
resultObjects[rowIndex + i] = new BooleanWritable(booleanResult);
}
} else if (currentBatchSize == 0) {
// Whole batch got zapped.
for (int i = 0; i < originalBatchSize; i++) {
resultObjects[rowIndex + i] = new BooleanWritable(false);
}
} else {
// Every row kept.
for (int i = 0; i < originalBatchSize; i++) {
resultObjects[rowIndex + i] = new BooleanWritable(true);
}
}
}
rowIndex += originalBatchSize;
}
return true;
}
use of org.apache.hadoop.hive.ql.exec.vector.expressions.VectorExpression in project hive by apache.
the class TestVectorArithmetic method doVectorArithmeticTest.
private void doVectorArithmeticTest(TypeInfo typeInfo1, TypeInfo typeInfo2, List<String> columns, String[] columnNames, TypeInfo[] typeInfos, DataTypePhysicalVariation[] dataTypePhysicalVariations, List<ExprNodeDesc> children, ExprNodeGenericFuncDesc exprDesc, Arithmetic arithmetic, ArithmeticTestMode arithmeticTestMode, ColumnScalarMode columnScalarMode, VectorRandomBatchSource batchSource, ObjectInspector objectInspector, TypeInfo outputTypeInfo, Object[] resultObjects) throws Exception {
HiveConf hiveConf = new HiveConf();
if (arithmeticTestMode == ArithmeticTestMode.ADAPTOR) {
hiveConf.setBoolVar(HiveConf.ConfVars.HIVE_TEST_VECTOR_ADAPTOR_OVERRIDE, true);
// Don't use DECIMAL_64 with the VectorUDFAdaptor.
dataTypePhysicalVariations = null;
}
VectorizationContext vectorizationContext = new VectorizationContext("name", columns, Arrays.asList(typeInfos), dataTypePhysicalVariations == null ? null : Arrays.asList(dataTypePhysicalVariations), hiveConf);
VectorExpression vectorExpression = vectorizationContext.getVectorExpression(exprDesc);
vectorExpression.transientInit(hiveConf);
if (arithmeticTestMode == ArithmeticTestMode.VECTOR_EXPRESSION && vectorExpression instanceof VectorUDFAdaptor) {
System.out.println("*NO NATIVE VECTOR EXPRESSION* typeInfo1 " + typeInfo1.toString() + " typeInfo2 " + typeInfo2.toString() + " arithmeticTestMode " + arithmeticTestMode + " columnScalarMode " + columnScalarMode + " vectorExpression " + vectorExpression.toString());
}
String[] outputScratchTypeNames = vectorizationContext.getScratchColumnTypeNames();
DataTypePhysicalVariation[] outputDataTypePhysicalVariations = vectorizationContext.getScratchDataTypePhysicalVariations();
VectorizedRowBatchCtx batchContext = new VectorizedRowBatchCtx(columnNames, typeInfos, dataTypePhysicalVariations, /* dataColumnNums */
null, /* partitionColumnCount */
0, /* virtualColumnCount */
0, /* neededVirtualColumns */
null, outputScratchTypeNames, outputDataTypePhysicalVariations);
VectorizedRowBatch batch = batchContext.createVectorizedRowBatch();
VectorExtractRow resultVectorExtractRow = new VectorExtractRow();
resultVectorExtractRow.init(new TypeInfo[] { outputTypeInfo }, new int[] { vectorExpression.getOutputColumnNum() });
Object[] scrqtchRow = new Object[1];
// System.out.println("*VECTOR EXPRESSION* " + vectorExpression.getClass().getSimpleName());
/*
System.out.println(
"*DEBUG* typeInfo1 " + typeInfo1.toString() +
" typeInfo2 " + typeInfo2.toString() +
" arithmeticTestMode " + arithmeticTestMode +
" columnScalarMode " + columnScalarMode +
" vectorExpression " + vectorExpression.toString());
*/
batchSource.resetBatchIteration();
int rowIndex = 0;
while (true) {
if (!batchSource.fillNextBatch(batch)) {
break;
}
vectorExpression.evaluate(batch);
extractResultObjects(batch, rowIndex, resultVectorExtractRow, scrqtchRow, objectInspector, resultObjects);
rowIndex += batch.size;
}
}
use of org.apache.hadoop.hive.ql.exec.vector.expressions.VectorExpression in project hive by apache.
the class VectorMapJoinInnerBigOnlyStringOperator method processBatch.
@Override
public void processBatch(VectorizedRowBatch batch) throws HiveException {
try {
// For inner joins, we may apply the filter(s) now.
for (VectorExpression ve : bigTableFilterExpressions) {
ve.evaluate(batch);
}
final int inputLogicalSize = batch.size;
if (inputLogicalSize == 0) {
return;
}
// Perform any key expressions. Results will go into scratch columns.
if (bigTableKeyExpressions != null) {
for (VectorExpression ve : bigTableKeyExpressions) {
ve.evaluate(batch);
}
}
// We rebuild in-place the selected array with rows destine to be forwarded.
int numSel = 0;
/*
* Single-Column String specific declarations.
*/
// The one join column for this specialized class.
BytesColumnVector joinColVector = (BytesColumnVector) batch.cols[singleJoinColumn];
byte[][] vector = joinColVector.vector;
int[] start = joinColVector.start;
int[] length = joinColVector.length;
/*
* Single-Column String check for repeating.
*/
// Check single column for repeating.
boolean allKeyInputColumnsRepeating = joinColVector.isRepeating;
if (allKeyInputColumnsRepeating) {
/*
* Repeating.
*/
// All key input columns are repeating. Generate key once. Lookup once.
// Since the key is repeated, we must use entry 0 regardless of selectedInUse.
/*
* Single-Column String specific repeated lookup.
*/
JoinUtil.JoinResult joinResult;
if (!joinColVector.noNulls && joinColVector.isNull[0]) {
joinResult = JoinUtil.JoinResult.NOMATCH;
} else {
byte[] keyBytes = vector[0];
int keyStart = start[0];
int keyLength = length[0];
joinResult = hashMultiSet.contains(keyBytes, keyStart, keyLength, hashMultiSetResults[0]);
}
if (LOG.isDebugEnabled()) {
LOG.debug(CLASS_NAME + " batch #" + batchCounter + " repeated joinResult " + joinResult.name());
}
finishInnerBigOnlyRepeated(batch, joinResult, hashMultiSetResults[0]);
} else {
if (LOG.isDebugEnabled()) {
LOG.debug(CLASS_NAME + " batch #" + batchCounter + " non-repeated");
}
// We remember any matching rows in matchs / matchSize. At the end of the loop,
// selected / batch.size will represent both matching and non-matching rows for outer join.
// Only deferred rows will have been removed from selected.
int[] selected = batch.selected;
boolean selectedInUse = batch.selectedInUse;
int hashMultiSetResultCount = 0;
int allMatchCount = 0;
int equalKeySeriesCount = 0;
int spillCount = 0;
/*
* Single-Column String specific variables.
*/
int saveKeyBatchIndex = -1;
// We optimize performance by only looking up the first key in a series of equal keys.
boolean haveSaveKey = false;
JoinUtil.JoinResult saveJoinResult = JoinUtil.JoinResult.NOMATCH;
// Logical loop over the rows in the batch since the batch may have selected in use.
for (int logical = 0; logical < inputLogicalSize; logical++) {
int batchIndex = (selectedInUse ? selected[logical] : logical);
/*
* Single-Column String get key.
*/
// Implicit -- use batchIndex.
boolean isNull = !joinColVector.noNulls && joinColVector.isNull[batchIndex];
if (isNull || !haveSaveKey || StringExpr.equal(vector[saveKeyBatchIndex], start[saveKeyBatchIndex], length[saveKeyBatchIndex], vector[batchIndex], start[batchIndex], length[batchIndex]) == false) {
if (haveSaveKey) {
// Move on with our counts.
switch(saveJoinResult) {
case MATCH:
// We have extracted the count from the hash multi-set result, so we don't keep it.
equalKeySeriesCount++;
break;
case SPILL:
// We keep the hash multi-set result for its spill information.
hashMultiSetResultCount++;
break;
case NOMATCH:
break;
}
}
if (isNull) {
saveJoinResult = JoinUtil.JoinResult.NOMATCH;
haveSaveKey = false;
} else {
// Regardless of our matching result, we keep that information to make multiple use
// of it for a possible series of equal keys.
haveSaveKey = true;
/*
* Single-Column String specific save key.
*/
saveKeyBatchIndex = batchIndex;
/*
* Single-Column String specific lookup key.
*/
byte[] keyBytes = vector[batchIndex];
int keyStart = start[batchIndex];
int keyLength = length[batchIndex];
saveJoinResult = hashMultiSet.contains(keyBytes, keyStart, keyLength, hashMultiSetResults[hashMultiSetResultCount]);
}
switch(saveJoinResult) {
case MATCH:
equalKeySeriesValueCounts[equalKeySeriesCount] = hashMultiSetResults[hashMultiSetResultCount].count();
equalKeySeriesAllMatchIndices[equalKeySeriesCount] = allMatchCount;
equalKeySeriesDuplicateCounts[equalKeySeriesCount] = 1;
allMatchs[allMatchCount++] = batchIndex;
// VectorizedBatchUtil.debugDisplayOneRow(batch, batchIndex, CLASS_NAME + " MATCH isSingleValue " + equalKeySeriesIsSingleValue[equalKeySeriesCount] + " currentKey " + currentKey);
break;
case SPILL:
spills[spillCount] = batchIndex;
spillHashMapResultIndices[spillCount] = hashMultiSetResultCount;
spillCount++;
break;
case NOMATCH:
// VectorizedBatchUtil.debugDisplayOneRow(batch, batchIndex, CLASS_NAME + " NOMATCH" + " currentKey " + currentKey);
break;
}
} else {
switch(saveJoinResult) {
case MATCH:
equalKeySeriesDuplicateCounts[equalKeySeriesCount]++;
allMatchs[allMatchCount++] = batchIndex;
// VectorizedBatchUtil.debugDisplayOneRow(batch, batchIndex, CLASS_NAME + " MATCH duplicate");
break;
case SPILL:
spills[spillCount] = batchIndex;
spillHashMapResultIndices[spillCount] = hashMultiSetResultCount;
spillCount++;
break;
case NOMATCH:
// VectorizedBatchUtil.debugDisplayOneRow(batch, batchIndex, CLASS_NAME + " NOMATCH duplicate");
break;
}
}
}
if (haveSaveKey) {
// Update our counts for the last key.
switch(saveJoinResult) {
case MATCH:
// We have extracted the count from the hash multi-set result, so we don't keep it.
equalKeySeriesCount++;
break;
case SPILL:
// We keep the hash multi-set result for its spill information.
hashMultiSetResultCount++;
break;
case NOMATCH:
break;
}
}
if (LOG.isDebugEnabled()) {
LOG.debug(CLASS_NAME + " allMatchs " + intArrayToRangesString(allMatchs, allMatchCount) + " equalKeySeriesValueCounts " + longArrayToRangesString(equalKeySeriesValueCounts, equalKeySeriesCount) + " equalKeySeriesAllMatchIndices " + intArrayToRangesString(equalKeySeriesAllMatchIndices, equalKeySeriesCount) + " equalKeySeriesDuplicateCounts " + intArrayToRangesString(equalKeySeriesDuplicateCounts, equalKeySeriesCount) + " spills " + intArrayToRangesString(spills, spillCount) + " spillHashMapResultIndices " + intArrayToRangesString(spillHashMapResultIndices, spillCount) + " hashMapResults " + Arrays.toString(Arrays.copyOfRange(hashMultiSetResults, 0, hashMultiSetResultCount)));
}
finishInnerBigOnly(batch, allMatchCount, equalKeySeriesCount, spillCount, (VectorMapJoinHashTableResult[]) hashMultiSetResults, hashMultiSetResultCount);
}
if (batch.size > 0) {
// Forward any remaining selected rows.
forwardBigTableBatch(batch);
}
} catch (IOException e) {
throw new HiveException(e);
} catch (Exception e) {
throw new HiveException(e);
}
}
use of org.apache.hadoop.hive.ql.exec.vector.expressions.VectorExpression in project hive by apache.
the class VectorMapJoinInnerLongOperator method processBatch.
@Override
public void processBatch(VectorizedRowBatch batch) throws HiveException {
try {
// Do the per-batch setup for an inner join.
innerPerBatchSetup(batch);
// For inner joins, we may apply the filter(s) now.
for (VectorExpression ve : bigTableFilterExpressions) {
ve.evaluate(batch);
}
final int inputLogicalSize = batch.size;
if (inputLogicalSize == 0) {
return;
}
// Perform any key expressions. Results will go into scratch columns.
if (bigTableKeyExpressions != null) {
for (VectorExpression ve : bigTableKeyExpressions) {
ve.evaluate(batch);
}
}
/*
* Single-Column Long specific declarations.
*/
// The one join column for this specialized class.
LongColumnVector joinColVector = (LongColumnVector) batch.cols[singleJoinColumn];
long[] vector = joinColVector.vector;
/*
* Single-Column Long check for repeating.
*/
// Check single column for repeating.
boolean allKeyInputColumnsRepeating = joinColVector.isRepeating;
if (allKeyInputColumnsRepeating) {
/*
* Repeating.
*/
// All key input columns are repeating. Generate key once. Lookup once.
// Since the key is repeated, we must use entry 0 regardless of selectedInUse.
/*
* Single-Column Long specific repeated lookup.
*/
JoinUtil.JoinResult joinResult;
if (!joinColVector.noNulls && joinColVector.isNull[0]) {
joinResult = JoinUtil.JoinResult.NOMATCH;
} else {
long key = vector[0];
if (useMinMax && (key < min || key > max)) {
// Out of range for whole batch.
joinResult = JoinUtil.JoinResult.NOMATCH;
} else {
joinResult = hashMap.lookup(key, hashMapResults[0]);
}
}
if (LOG.isDebugEnabled()) {
LOG.debug(CLASS_NAME + " batch #" + batchCounter + " repeated joinResult " + joinResult.name());
}
finishInnerRepeated(batch, joinResult, hashMapResults[0]);
} else {
if (LOG.isDebugEnabled()) {
LOG.debug(CLASS_NAME + " batch #" + batchCounter + " non-repeated");
}
// We remember any matching rows in matchs / matchSize. At the end of the loop,
// selected / batch.size will represent both matching and non-matching rows for outer join.
// Only deferred rows will have been removed from selected.
int[] selected = batch.selected;
boolean selectedInUse = batch.selectedInUse;
int hashMapResultCount = 0;
int allMatchCount = 0;
int equalKeySeriesCount = 0;
int spillCount = 0;
/*
* Single-Column Long specific variables.
*/
long saveKey = 0;
// We optimize performance by only looking up the first key in a series of equal keys.
boolean haveSaveKey = false;
JoinUtil.JoinResult saveJoinResult = JoinUtil.JoinResult.NOMATCH;
// Logical loop over the rows in the batch since the batch may have selected in use.
for (int logical = 0; logical < inputLogicalSize; logical++) {
int batchIndex = (selectedInUse ? selected[logical] : logical);
/*
* Single-Column Long get key.
*/
long currentKey;
boolean isNull;
if (!joinColVector.noNulls && joinColVector.isNull[batchIndex]) {
currentKey = 0;
isNull = true;
} else {
currentKey = vector[batchIndex];
isNull = false;
}
if (isNull || !haveSaveKey || currentKey != saveKey) {
if (haveSaveKey) {
// Move on with our counts.
switch(saveJoinResult) {
case MATCH:
hashMapResultCount++;
equalKeySeriesCount++;
break;
case SPILL:
hashMapResultCount++;
break;
case NOMATCH:
break;
}
}
if (isNull) {
saveJoinResult = JoinUtil.JoinResult.NOMATCH;
haveSaveKey = false;
} else {
// Regardless of our matching result, we keep that information to make multiple use
// of it for a possible series of equal keys.
haveSaveKey = true;
/*
* Single-Column Long specific save key.
*/
saveKey = currentKey;
if (useMinMax && (currentKey < min || currentKey > max)) {
// Key out of range for whole hash table.
saveJoinResult = JoinUtil.JoinResult.NOMATCH;
} else {
saveJoinResult = hashMap.lookup(currentKey, hashMapResults[hashMapResultCount]);
}
}
switch(saveJoinResult) {
case MATCH:
equalKeySeriesHashMapResultIndices[equalKeySeriesCount] = hashMapResultCount;
equalKeySeriesAllMatchIndices[equalKeySeriesCount] = allMatchCount;
equalKeySeriesIsSingleValue[equalKeySeriesCount] = hashMapResults[hashMapResultCount].isSingleRow();
equalKeySeriesDuplicateCounts[equalKeySeriesCount] = 1;
allMatchs[allMatchCount++] = batchIndex;
// VectorizedBatchUtil.debugDisplayOneRow(batch, batchIndex, CLASS_NAME + " MATCH isSingleValue " + equalKeySeriesIsSingleValue[equalKeySeriesCount] + " currentKey " + currentKey);
break;
case SPILL:
spills[spillCount] = batchIndex;
spillHashMapResultIndices[spillCount] = hashMapResultCount;
spillCount++;
break;
case NOMATCH:
// VectorizedBatchUtil.debugDisplayOneRow(batch, batchIndex, CLASS_NAME + " NOMATCH" + " currentKey " + currentKey);
break;
}
} else {
switch(saveJoinResult) {
case MATCH:
equalKeySeriesDuplicateCounts[equalKeySeriesCount]++;
allMatchs[allMatchCount++] = batchIndex;
// VectorizedBatchUtil.debugDisplayOneRow(batch, batchIndex, CLASS_NAME + " MATCH duplicate");
break;
case SPILL:
spills[spillCount] = batchIndex;
spillHashMapResultIndices[spillCount] = hashMapResultCount;
spillCount++;
break;
case NOMATCH:
// VectorizedBatchUtil.debugDisplayOneRow(batch, batchIndex, CLASS_NAME + " NOMATCH duplicate");
break;
}
}
}
if (haveSaveKey) {
// Update our counts for the last key.
switch(saveJoinResult) {
case MATCH:
hashMapResultCount++;
equalKeySeriesCount++;
break;
case SPILL:
hashMapResultCount++;
break;
case NOMATCH:
break;
}
}
if (LOG.isDebugEnabled()) {
LOG.debug(CLASS_NAME + " allMatchs " + intArrayToRangesString(allMatchs, allMatchCount) + " equalKeySeriesHashMapResultIndices " + intArrayToRangesString(equalKeySeriesHashMapResultIndices, equalKeySeriesCount) + " equalKeySeriesAllMatchIndices " + intArrayToRangesString(equalKeySeriesAllMatchIndices, equalKeySeriesCount) + " equalKeySeriesIsSingleValue " + Arrays.toString(Arrays.copyOfRange(equalKeySeriesIsSingleValue, 0, equalKeySeriesCount)) + " equalKeySeriesDuplicateCounts " + Arrays.toString(Arrays.copyOfRange(equalKeySeriesDuplicateCounts, 0, equalKeySeriesCount)) + " spills " + intArrayToRangesString(spills, spillCount) + " spillHashMapResultIndices " + intArrayToRangesString(spillHashMapResultIndices, spillCount) + " hashMapResults " + Arrays.toString(Arrays.copyOfRange(hashMapResults, 0, hashMapResultCount)));
}
finishInner(batch, allMatchCount, equalKeySeriesCount, spillCount, hashMapResultCount);
}
if (batch.size > 0) {
// Forward any remaining selected rows.
forwardBigTableBatch(batch);
}
} catch (IOException e) {
throw new HiveException(e);
} catch (Exception e) {
throw new HiveException(e);
}
}
use of org.apache.hadoop.hive.ql.exec.vector.expressions.VectorExpression in project hive by apache.
the class VectorMapJoinInnerStringOperator method processBatch.
@Override
public void processBatch(VectorizedRowBatch batch) throws HiveException {
try {
// Do the per-batch setup for an inner join.
innerPerBatchSetup(batch);
// For inner joins, we may apply the filter(s) now.
for (VectorExpression ve : bigTableFilterExpressions) {
ve.evaluate(batch);
}
final int inputLogicalSize = batch.size;
if (inputLogicalSize == 0) {
return;
}
// Perform any key expressions. Results will go into scratch columns.
if (bigTableKeyExpressions != null) {
for (VectorExpression ve : bigTableKeyExpressions) {
ve.evaluate(batch);
}
}
/*
* Single-Column String specific declarations.
*/
// The one join column for this specialized class.
BytesColumnVector joinColVector = (BytesColumnVector) batch.cols[singleJoinColumn];
byte[][] vector = joinColVector.vector;
int[] start = joinColVector.start;
int[] length = joinColVector.length;
/*
* Single-Column String check for repeating.
*/
// Check single column for repeating.
boolean allKeyInputColumnsRepeating = joinColVector.isRepeating;
if (allKeyInputColumnsRepeating) {
/*
* Repeating.
*/
// All key input columns are repeating. Generate key once. Lookup once.
// Since the key is repeated, we must use entry 0 regardless of selectedInUse.
/*
* Single-Column String specific repeated lookup.
*/
JoinUtil.JoinResult joinResult;
if (!joinColVector.noNulls && joinColVector.isNull[0]) {
joinResult = JoinUtil.JoinResult.NOMATCH;
} else {
byte[] keyBytes = vector[0];
int keyStart = start[0];
int keyLength = length[0];
joinResult = hashMap.lookup(keyBytes, keyStart, keyLength, hashMapResults[0]);
}
if (LOG.isDebugEnabled()) {
LOG.debug(CLASS_NAME + " batch #" + batchCounter + " repeated joinResult " + joinResult.name());
}
finishInnerRepeated(batch, joinResult, hashMapResults[0]);
} else {
if (LOG.isDebugEnabled()) {
LOG.debug(CLASS_NAME + " batch #" + batchCounter + " non-repeated");
}
// We remember any matching rows in matchs / matchSize. At the end of the loop,
// selected / batch.size will represent both matching and non-matching rows for outer join.
// Only deferred rows will have been removed from selected.
int[] selected = batch.selected;
boolean selectedInUse = batch.selectedInUse;
int hashMapResultCount = 0;
int allMatchCount = 0;
int equalKeySeriesCount = 0;
int spillCount = 0;
/*
* Single-Column String specific variables.
*/
int saveKeyBatchIndex = -1;
// We optimize performance by only looking up the first key in a series of equal keys.
boolean haveSaveKey = false;
JoinUtil.JoinResult saveJoinResult = JoinUtil.JoinResult.NOMATCH;
// Logical loop over the rows in the batch since the batch may have selected in use.
for (int logical = 0; logical < inputLogicalSize; logical++) {
int batchIndex = (selectedInUse ? selected[logical] : logical);
/*
* Single-Column String get key.
*/
// Implicit -- use batchIndex.
boolean isNull = !joinColVector.noNulls && joinColVector.isNull[batchIndex];
if (isNull || !haveSaveKey || StringExpr.equal(vector[saveKeyBatchIndex], start[saveKeyBatchIndex], length[saveKeyBatchIndex], vector[batchIndex], start[batchIndex], length[batchIndex]) == false) {
if (haveSaveKey) {
// Move on with our counts.
switch(saveJoinResult) {
case MATCH:
hashMapResultCount++;
equalKeySeriesCount++;
break;
case SPILL:
hashMapResultCount++;
break;
case NOMATCH:
break;
}
}
if (isNull) {
saveJoinResult = JoinUtil.JoinResult.NOMATCH;
haveSaveKey = false;
} else {
// Regardless of our matching result, we keep that information to make multiple use
// of it for a possible series of equal keys.
haveSaveKey = true;
/*
* Single-Column String specific save key.
*/
saveKeyBatchIndex = batchIndex;
/*
* Single-Column String specific lookup key.
*/
byte[] keyBytes = vector[batchIndex];
int keyStart = start[batchIndex];
int keyLength = length[batchIndex];
saveJoinResult = hashMap.lookup(keyBytes, keyStart, keyLength, hashMapResults[hashMapResultCount]);
}
switch(saveJoinResult) {
case MATCH:
equalKeySeriesHashMapResultIndices[equalKeySeriesCount] = hashMapResultCount;
equalKeySeriesAllMatchIndices[equalKeySeriesCount] = allMatchCount;
equalKeySeriesIsSingleValue[equalKeySeriesCount] = hashMapResults[hashMapResultCount].isSingleRow();
equalKeySeriesDuplicateCounts[equalKeySeriesCount] = 1;
allMatchs[allMatchCount++] = batchIndex;
// VectorizedBatchUtil.debugDisplayOneRow(batch, batchIndex, CLASS_NAME + " MATCH isSingleValue " + equalKeySeriesIsSingleValue[equalKeySeriesCount] + " currentKey " + currentKey);
break;
case SPILL:
spills[spillCount] = batchIndex;
spillHashMapResultIndices[spillCount] = hashMapResultCount;
spillCount++;
break;
case NOMATCH:
// VectorizedBatchUtil.debugDisplayOneRow(batch, batchIndex, CLASS_NAME + " NOMATCH" + " currentKey " + currentKey);
break;
}
} else {
switch(saveJoinResult) {
case MATCH:
equalKeySeriesDuplicateCounts[equalKeySeriesCount]++;
allMatchs[allMatchCount++] = batchIndex;
// VectorizedBatchUtil.debugDisplayOneRow(batch, batchIndex, CLASS_NAME + " MATCH duplicate");
break;
case SPILL:
spills[spillCount] = batchIndex;
spillHashMapResultIndices[spillCount] = hashMapResultCount;
spillCount++;
break;
case NOMATCH:
// VectorizedBatchUtil.debugDisplayOneRow(batch, batchIndex, CLASS_NAME + " NOMATCH duplicate");
break;
}
}
}
if (haveSaveKey) {
// Update our counts for the last key.
switch(saveJoinResult) {
case MATCH:
hashMapResultCount++;
equalKeySeriesCount++;
break;
case SPILL:
hashMapResultCount++;
break;
case NOMATCH:
break;
}
}
if (LOG.isDebugEnabled()) {
LOG.debug(CLASS_NAME + " allMatchs " + intArrayToRangesString(allMatchs, allMatchCount) + " equalKeySeriesHashMapResultIndices " + intArrayToRangesString(equalKeySeriesHashMapResultIndices, equalKeySeriesCount) + " equalKeySeriesAllMatchIndices " + intArrayToRangesString(equalKeySeriesAllMatchIndices, equalKeySeriesCount) + " equalKeySeriesIsSingleValue " + Arrays.toString(Arrays.copyOfRange(equalKeySeriesIsSingleValue, 0, equalKeySeriesCount)) + " equalKeySeriesDuplicateCounts " + Arrays.toString(Arrays.copyOfRange(equalKeySeriesDuplicateCounts, 0, equalKeySeriesCount)) + " spills " + intArrayToRangesString(spills, spillCount) + " spillHashMapResultIndices " + intArrayToRangesString(spillHashMapResultIndices, spillCount) + " hashMapResults " + Arrays.toString(Arrays.copyOfRange(hashMapResults, 0, hashMapResultCount)));
}
finishInner(batch, allMatchCount, equalKeySeriesCount, spillCount, hashMapResultCount);
}
if (batch.size > 0) {
// Forward any remaining selected rows.
forwardBigTableBatch(batch);
}
} catch (IOException e) {
throw new HiveException(e);
} catch (Exception e) {
throw new HiveException(e);
}
}
Aggregations