Many resources are needed to download a project. Please understand that we have to compensate our server costs. Thank you in advance. Project price only 1 $
You can buy this project and download/modify it how often you want.
/* This file is part of VoltDB.
* Copyright (C) 2008-2020 VoltDB Inc.
*
* This program is free software: you can redistribute it and/or modify
* it under the terms of the GNU Affero General Public License as
* published by the Free Software Foundation, either version 3 of the
* License, or (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU Affero General Public License for more details.
*
* You should have received a copy of the GNU Affero General Public License
* along with VoltDB. If not, see .
*/
package org.voltdb.plannodes;
import java.util.Collection;
import org.json_voltpatches.JSONException;
import org.json_voltpatches.JSONObject;
import org.json_voltpatches.JSONStringer;
import org.voltdb.catalog.Database;
import org.voltdb.compiler.DatabaseEstimates;
import org.voltdb.compiler.ScalarValueHints;
import org.voltdb.exceptions.ValidationError;
import org.voltdb.expressions.AbstractExpression;
import org.voltdb.expressions.AbstractSubqueryExpression;
import org.voltdb.expressions.TupleValueExpression;
import org.voltdb.types.PlanNodeType;
import org.voltdb.types.SortDirectionType;
public class NestLoopIndexPlanNode extends AbstractJoinPlanNode {
public NestLoopIndexPlanNode() {
super();
}
@Override
public PlanNodeType getPlanNodeType() {
return PlanNodeType.NESTLOOPINDEX;
}
@Override
public void generateOutputSchema(Database db) {
// Important safety tip regarding this inlined
// index scan and ITS inlined projection:
// That projection is currently only used/usable as
// a means to narrow the set of columns from the
// indexscan's target table that make it into the
// rest of the plan. the expressions that are
// given to the projection are currently not ever used
IndexScanPlanNode inlineScan =
(IndexScanPlanNode) m_inlineNodes.get(PlanNodeType.INDEXSCAN);
assert(inlineScan != null);
inlineScan.generateOutputSchema(db);
assert(m_children.size() == 1);
m_children.get(0).generateOutputSchema(db);
// Join the schema together to form the output schema
// The child subplan's output is the outer table
// The inlined node's output is the inner table.
//
// Note that the inner table's contribution to the join_tuple doesn't include
// all the columns from the inner table---just the ones needed as determined by
// the inlined scan's own inlined projection, as described above.
m_outputSchemaPreInlineAgg =
m_children.get(0).getOutputSchema().
join(inlineScan.getOutputSchema()).copyAndReplaceWithTVE();
m_hasSignificantOutputSchema = true;
generateRealOutputSchema(db);
// Generate the output schema for subqueries
Collection subqueryExpressions = findAllSubquerySubexpressions();
for (AbstractExpression subqueryExpression : subqueryExpressions) {
assert(subqueryExpression instanceof AbstractSubqueryExpression);
((AbstractSubqueryExpression) subqueryExpression).generateOutputSchema(db);
}
}
@Override
public void resolveColumnIndexes() {
IndexScanPlanNode inlineScan =
(IndexScanPlanNode) m_inlineNodes.get(PlanNodeType.INDEXSCAN);
assert (m_children.size() == 1 && inlineScan != null);
for (AbstractPlanNode child : m_children) {
child.resolveColumnIndexes();
}
LimitPlanNode limit = (LimitPlanNode)getInlinePlanNode(PlanNodeType.LIMIT);
if (limit != null) {
// output schema of limit node has not been used
limit.m_outputSchema = m_outputSchemaPreInlineAgg;
limit.m_hasSignificantOutputSchema = false;
}
// We need the schema from the target table from the inlined index
final NodeSchema completeInnerTableSchema = inlineScan.getTableSchema();
// We need the output schema from the child node
final NodeSchema outerSchema = m_children.get(0).getOutputSchema();
// pull every expression out of the inlined index scan
// and resolve all of the TVEs against our two input schema from above.
//
// Tickets ENG-9389, ENG-9533: we use the complete schema for the inner
// table (rather than the smaller schema from the inlined index scan's
// inlined project node) because the inlined scan has no temp table,
// so predicates will be accessing the index-scanned table directly.
resolvePredicate(inlineScan.getPredicate(), outerSchema, completeInnerTableSchema);
resolvePredicate(inlineScan.getEndExpression(), outerSchema, completeInnerTableSchema);
resolvePredicate(inlineScan.getInitialExpression(), outerSchema, completeInnerTableSchema);
resolvePredicate(inlineScan.getSkipNullPredicate(), outerSchema, completeInnerTableSchema);
resolvePredicate(inlineScan.getSearchKeyExpressions(), outerSchema, completeInnerTableSchema);
resolvePredicate(m_preJoinPredicate, outerSchema, completeInnerTableSchema);
resolvePredicate(m_joinPredicate, outerSchema, completeInnerTableSchema);
resolvePredicate(m_wherePredicate, outerSchema, completeInnerTableSchema);
// Resolve subquery expression indexes
resolveSubqueryColumnIndexes();
// Resolve TVE indexes for each schema column.
for (int i = 0; i < m_outputSchemaPreInlineAgg.size(); ++i) {
SchemaColumn col = m_outputSchemaPreInlineAgg.getColumn(i);
// These are all TVEs.
assert(col.getExpression() instanceof TupleValueExpression);
TupleValueExpression tve = (TupleValueExpression)col.getExpression();
int index;
int tableIdx;
if (i < outerSchema.size()) {
tableIdx = 0; // 0 for outer table
index = outerSchema.getIndexOfTve(tve);
if (index >= 0) {
tve.setColumnIndex(index);
}
} else {
tableIdx = 1; // 1 for inner table
index = tve.setColumnIndexUsingSchema(completeInnerTableSchema);
}
if (index == -1) {
throw new RuntimeException("Unable to find index for column: " +
col.toString());
}
tve.setTableIndex(tableIdx);
}
// We want the output columns to be ordered like [outer table columns][inner table columns],
// and further ordered by TVE index within the left- and righthand sides.
// generateOutputSchema already places outer columns on the left and inner on the right,
// so we just need to order the left- and righthand sides by TVE index separately.
m_outputSchemaPreInlineAgg.sortByTveIndex(0, outerSchema.size());
m_outputSchemaPreInlineAgg.sortByTveIndex(outerSchema.size(), m_outputSchemaPreInlineAgg.size());
m_hasSignificantOutputSchema = true;
resolveRealOutputSchema();
}
@Override
public void resolveSortDirection() {
super.resolveSortDirection();
// special treatment for NLIJ, when the outer table is a materialized scan node
// the sort direction from the outer table should be the same as the that in the inner table
// (because we set when building this NLIJ)
if (m_children.get(0).getPlanNodeType() == PlanNodeType.MATERIALIZEDSCAN) {
IndexScanPlanNode ispn = (IndexScanPlanNode) m_inlineNodes.get(PlanNodeType.INDEXSCAN);
assert (((MaterializedScanPlanNode)(m_children.get(0))).getSortDirection() == ispn.getSortDirection());
m_sortDirection = ispn.getSortDirection();
}
}
@Override
public void validate() {
super.validate();
// Check that we have an inline IndexScanPlanNode
if (m_inlineNodes.isEmpty()) {
throw new ValidationError("No inline PlanNodes are set for %s", toString());
} else if (!m_inlineNodes.containsKey(PlanNodeType.INDEXSCAN)) {
throw new ValidationError("No inline PlanNode with type '%s' was set for %s",
PlanNodeType.INDEXSCAN, toString());
}
}
/**
* Does the (sub)plan guarantee an identical result/effect when "replayed"
* against the same database state, such as during replication or CL recovery.
* @return
*/
@Override
public boolean isOrderDeterministic() {
if (! super.isOrderDeterministic()) {
return false;
}
IndexScanPlanNode index_scan = getInlineIndexScan();
if (! index_scan.isOrderDeterministic()) {
m_nondeterminismDetail = index_scan.m_nondeterminismDetail;
return false;
}
return true;
}
@Override
public boolean hasInlinedIndexScanOfTable(String tableName) {
IndexScanPlanNode index_scan = getInlineIndexScan();
if (index_scan.getTargetTableName().equals(tableName)) {
return true;
} else {
return getChild(0).hasInlinedIndexScanOfTable(tableName);
}
}
@Override
public void computeCostEstimates(long childOutputTupleCountEstimate, DatabaseEstimates estimates, ScalarValueHints[] paramHints) {
// Add the cost of the inlined index scan to the cost of processing the input tuples.
// This isn't really a fair representation of what's going on, as the index is scanned once
// per input tuple, but I think it will still cause the plan selector to pick the join
// order with the lowest total access cost.
IndexScanPlanNode indexScan = getInlineIndexScan();
m_estimatedOutputTupleCount = indexScan.getEstimatedOutputTupleCount() + childOutputTupleCountEstimate;
// Discount outer child estimates based on the number of its filters
m_estimatedProcessedTupleCount = indexScan.getEstimatedProcessedTupleCount() +
discountEstimatedProcessedTupleCount(m_children.get(0));
}
public IndexScanPlanNode getInlineIndexScan() {
IndexScanPlanNode indexScan =
(IndexScanPlanNode) getInlinePlanNode(PlanNodeType.INDEXSCAN);
assert(indexScan != null);
return indexScan;
}
@Override
protected String explainPlanForNode(String indent) {
return "NESTLOOP INDEX " + this.m_joinType.toString() + " JOIN" +
(m_sortDirection == SortDirectionType.INVALID ? "" : " (" + m_sortDirection + ")") +
explainFilters(indent);
}
@Override
public void toJSONString(JSONStringer stringer) throws JSONException {
super.toJSONString(stringer);
if (m_sortDirection != SortDirectionType.INVALID) {
stringer.keySymbolValuePair(Members.SORT_DIRECTION.name(),
m_sortDirection.toString());
}
}
@Override
public void loadFromJSONObject(JSONObject jobj, Database db) throws JSONException {
super.loadFromJSONObject(jobj, db);
if ( ! jobj.isNull(Members.SORT_DIRECTION.name())) {
m_sortDirection = SortDirectionType.get(
jobj.getString(Members.SORT_DIRECTION.name()));
}
}
}
