use of org.apache.hadoop.hive.ql.parse.GenTezUtils.DynamicPartitionPrunerContext in project hive by apache.
the class DynamicPartitionPruningOptimization method process.
@Override
public Object process(Node nd, Stack<Node> stack, NodeProcessorCtx procCtx, Object... nodeOutputs) throws SemanticException {
ParseContext parseContext;
if (procCtx instanceof OptimizeTezProcContext) {
parseContext = ((OptimizeTezProcContext) procCtx).parseContext;
} else if (procCtx instanceof OptimizeSparkProcContext) {
parseContext = ((OptimizeSparkProcContext) procCtx).getParseContext();
} else {
throw new IllegalArgumentException("expected parseContext to be either " + "OptimizeTezProcContext or OptimizeSparkProcContext, but found " + procCtx.getClass().getName());
}
FilterOperator filter = (FilterOperator) nd;
FilterDesc desc = filter.getConf();
if (!parseContext.getConf().getBoolVar(ConfVars.TEZ_DYNAMIC_PARTITION_PRUNING) && !parseContext.getConf().isSparkDPPAny()) {
// nothing to do when the optimization is off
return null;
}
TableScanOperator ts = null;
if (filter.getParentOperators().size() == 1 && filter.getParentOperators().get(0) instanceof TableScanOperator) {
ts = (TableScanOperator) filter.getParentOperators().get(0);
}
if (LOG.isDebugEnabled()) {
LOG.debug("Parent: " + filter.getParentOperators().get(0));
LOG.debug("Filter: " + desc.getPredicateString());
LOG.debug("TableScan: " + ts);
}
DynamicPartitionPrunerContext removerContext = new DynamicPartitionPrunerContext();
// collect the dynamic pruning conditions
removerContext.dynLists.clear();
GenTezUtils.collectDynamicPruningConditions(desc.getPredicate(), removerContext);
if (ts == null) {
// Replace the synthetic predicate with true and bail out
for (DynamicListContext ctx : removerContext) {
ExprNodeDesc constNode = new ExprNodeConstantDesc(ctx.parent.getTypeInfo(), true);
replaceExprNode(ctx, desc, constNode);
}
return false;
}
boolean semiJoin = parseContext.getConf().getBoolVar(ConfVars.TEZ_DYNAMIC_SEMIJOIN_REDUCTION);
if (HiveConf.getVar(parseContext.getConf(), HiveConf.ConfVars.HIVE_EXECUTION_ENGINE).equals("spark")) {
// TODO HIVE-16862: Implement a similar feature like "hive.tez.dynamic.semijoin.reduction" in hive on spark
semiJoin = false;
}
List<ExprNodeDesc> newBetweenNodes = new ArrayList<>();
List<ExprNodeDesc> newBloomFilterNodes = new ArrayList<>();
for (DynamicListContext ctx : removerContext) {
if (ctx.desc.getTypeInfo().getCategory() != ObjectInspector.Category.PRIMITIVE) {
// https://issues.apache.org/jira/browse/HIVE-24988
continue;
}
String column = ExprNodeDescUtils.extractColName(ctx.parent);
boolean semiJoinAttempted = false;
ExprNodeDesc constNode = new ExprNodeConstantDesc(ctx.parent.getTypeInfo(), true);
if (column != null) {
// Need unique IDs to refer to each min/max key value in the DynamicValueRegistry
String keyBaseAlias = "";
Table table = ts.getConf().getTableMetadata();
boolean nonEquiJoin = isNonEquiJoin(ctx.parent);
if (table != null && table.isPartitionKey(column) && !nonEquiJoin) {
String columnType = table.getPartColByName(column).getType();
String alias = ts.getConf().getAlias();
PrunedPartitionList plist = parseContext.getPrunedPartitions(alias, ts);
if (LOG.isDebugEnabled()) {
LOG.debug("alias: " + alias);
LOG.debug("pruned partition list: ");
if (plist != null) {
for (Partition p : plist.getPartitions()) {
LOG.debug(p.getCompleteName());
}
}
}
// have been already filtered
if (plist == null || plist.getPartitions().size() != 0) {
LOG.info("Dynamic partitioning: " + table.getCompleteName() + "." + column);
generateEventOperatorPlan(ctx, parseContext, ts, column, columnType, null);
} else {
// all partitions have been statically removed
LOG.debug("No partition pruning necessary.");
}
} else if (table.isNonNative() && table.getStorageHandler().addDynamicSplitPruningEdge(table, ctx.parent)) {
generateEventOperatorPlan(ctx, parseContext, ts, column, table.getCols().stream().filter(e -> e.getName().equals(column)).map(e -> e.getType()).findFirst().get(), ctx.parent);
} else {
// semijoin
LOG.debug("Column " + column + " is not a partition column");
if (semiJoin && !disableSemiJoinOptDueToExternalTable(parseContext.getConf(), ts, ctx) && ts.getConf().getFilterExpr() != null && !nonEquiJoin) {
LOG.debug("Initiate semijoin reduction for " + column + " (" + ts.getConf().getFilterExpr().getExprString());
StringBuilder internalColNameBuilder = new StringBuilder();
StringBuilder colNameBuilder = new StringBuilder();
// Apply best effort to fetch the correct table alias. If not
// found, fallback to old logic.
StringBuilder tabAliasBuilder = new StringBuilder();
if (getColumnInfo(ctx, internalColNameBuilder, colNameBuilder, tabAliasBuilder)) {
String colName = colNameBuilder.toString();
String tableAlias;
if (tabAliasBuilder.length() > 0) {
tableAlias = tabAliasBuilder.toString();
} else {
// falling back
Operator<?> op = ctx.generator;
while (!(op == null || op instanceof TableScanOperator)) {
op = op.getParentOperators().get(0);
}
tableAlias = (op == null ? "" : ((TableScanOperator) op).getConf().getAlias());
}
// Use the tableAlias to generate keyBaseAlias
keyBaseAlias = ctx.generator.getOperatorId() + "_" + tableAlias + "_" + colName;
Map<String, List<SemiJoinHint>> hints = parseContext.getSemiJoinHints();
if (hints != null) {
// Create semijoin optimizations ONLY for hinted columns
semiJoinAttempted = processSemiJoinHints(parseContext, ctx, hints, tableAlias, internalColNameBuilder.toString(), colName, ts, keyBaseAlias);
} else {
// fallback to regular logic
semiJoinAttempted = generateSemiJoinOperatorPlan(ctx, parseContext, ts, keyBaseAlias, internalColNameBuilder.toString(), colName, null);
}
}
}
}
// we always remove the condition by replacing it with "true"
if (semiJoinAttempted) {
List<ExprNodeDesc> betweenArgs = new ArrayList<ExprNodeDesc>();
// Do not invert between result
betweenArgs.add(new ExprNodeConstantDesc(Boolean.FALSE));
// add column expression here
betweenArgs.add(ctx.parent.getChildren().get(0));
betweenArgs.add(new ExprNodeDynamicValueDesc(new DynamicValue(keyBaseAlias + "_min", ctx.desc.getTypeInfo())));
betweenArgs.add(new ExprNodeDynamicValueDesc(new DynamicValue(keyBaseAlias + "_max", ctx.desc.getTypeInfo())));
ExprNodeDesc betweenNode = ExprNodeGenericFuncDesc.newInstance(FunctionRegistry.getFunctionInfo("between").getGenericUDF(), betweenArgs);
// add column expression for bloom filter
List<ExprNodeDesc> bloomFilterArgs = new ArrayList<ExprNodeDesc>();
bloomFilterArgs.add(ctx.parent.getChildren().get(0));
bloomFilterArgs.add(new ExprNodeDynamicValueDesc(new DynamicValue(keyBaseAlias + "_bloom_filter", TypeInfoFactory.binaryTypeInfo)));
ExprNodeDesc bloomFilterNode = ExprNodeGenericFuncDesc.newInstance(FunctionRegistry.getFunctionInfo("in_bloom_filter").getGenericUDF(), bloomFilterArgs);
newBetweenNodes.add(betweenNode);
newBloomFilterNodes.add(bloomFilterNode);
}
}
replaceExprNode(ctx, desc, constNode);
}
if (!newBetweenNodes.isEmpty()) {
// We need to add the new nodes: first the between nodes, then the bloom filters
if (FunctionRegistry.isOpAnd(desc.getPredicate())) {
// AND
desc.getPredicate().getChildren().addAll(newBetweenNodes);
desc.getPredicate().getChildren().addAll(newBloomFilterNodes);
} else {
List<ExprNodeDesc> andArgs = new ArrayList<>();
andArgs.add(desc.getPredicate());
andArgs.addAll(newBetweenNodes);
andArgs.addAll(newBloomFilterNodes);
ExprNodeGenericFuncDesc andExpr = ExprNodeGenericFuncDesc.newInstance(FunctionRegistry.getFunctionInfo("and").getGenericUDF(), andArgs);
// Also pass in filter as tableScan filterExpr
ts.getConf().setFilterExpr(andExpr);
desc.setPredicate(andExpr);
}
}
// if we pushed the predicate into the table scan we need to remove the
// synthetic conditions there.
cleanTableScanFilters(ts);
return false;
}
use of org.apache.hadoop.hive.ql.parse.GenTezUtils.DynamicPartitionPrunerContext in project hive by apache.
the class DynamicPartitionPruningOptimization method cleanTableScanFilters.
private void cleanTableScanFilters(TableScanOperator ts) throws SemanticException {
if (ts == null || ts.getConf() == null || ts.getConf().getFilterExpr() == null) {
// nothing to do
return;
}
DynamicPartitionPrunerContext removerContext = new DynamicPartitionPrunerContext();
// collect the dynamic pruning conditions
removerContext.dynLists.clear();
GenTezUtils.collectDynamicPruningConditions(ts.getConf().getFilterExpr(), removerContext);
for (DynamicListContext ctx : removerContext) {
// remove the condition by replacing it with "true"
ExprNodeDesc constNode = new ExprNodeConstantDesc(ctx.parent.getTypeInfo(), true);
if (ctx.grandParent == null) {
// we're the only node, just clear out the expression
ts.getConf().setFilterExpr(null);
} else {
int i = ctx.grandParent.getChildren().indexOf(ctx.parent);
ctx.grandParent.getChildren().remove(i);
ctx.grandParent.getChildren().add(i, constNode);
}
}
}
Aggregations