GroupByNode.java
| Index Score | ||
|---|---|---|
 |
 |
org.apache.derby.impl.sql.compile |
 |
 |
Apache Derby |
View: Reasons, Metrics, Source Code
These are the metrics that contribute to the Enerjy Score for this file, ranked by impact. So the metrics listed at the top influence the score to a greater extent that the metrics listed at the bottom.
| Metric | Description | |
|---|---|---|
| EXEC_COMMENTS | Comments in executable code |  |
| LINE_COMMENT | Number of line comments | |
| COMMENTS | Comment lines | |
| SIZE | Size of the file in bytes | |
| EXITS | Procedure exits | |
| LINES | Number of lines in the source file | |
| BLOCK_COMMENT | Number of block comment lines | |
| JAVA0145 | JAVA0145 Tab character used in source file | |
| JAVA0177 | JAVA0177 Variable declaration missing initializer | |
| DOC_COMMENT | Number of javadoc comment lines | |
| ELOC | Effective lines of code | |
| LOC | Lines of code | |
| LOGICAL_LINES | Number of statements | |
| RETURNS | Number of return points from functions | |
| OPERANDS | Number of operands | |
| PROGRAM_LENGTH | Halstead program length | |
| UNIQUE_OPERANDS | Number of unique operands | |
| OPERATORS | Number of operators | |
| PROGRAM_VOCAB | Halstead program vocabulary | |
| DECL_COMMENTS | Comments in declarations | |
| JAVA0075 | JAVA0075 Method parameter hides field | |
| INTERFACE_COMPLEXITY | Interface complexity | |
| JAVA0115 | JAVA0115 Incorrect javadoc: no @throws or @exception tag for 'exception' | |
| LOOPS | Number of loops | |
| CYCLOMATIC | Cyclomatic complexity | |
| JAVA0173 | JAVA0173 Unused method parameter | |
| NEST_DEPTH | Maximum nesting depth | |
| JAVA0108 | JAVA0108 Incorrect javadoc: no @param tag for 'parameter' | |
| JAVA0035 | JAVA0035 Missing braces in for statement | |
| JAVA0179 | JAVA0179 Local variable hides visible field | |
| JAVA0285 | JAVA0285 Dereference of potentially null variable | |
| BLOCKS | Number of blocks | |
| JAVA0049 | JAVA0049 Nested block at depth N (maximum: M) | |
| JAVA0110 | JAVA0110 Incorrect javadoc: no @return tag | |
| PARAMS | Number of formal parameter declarations | |
| COMPARISONS | Number of comparison operators | |
| JAVA0117 | JAVA0117 Missing javadoc: method 'method' | |
| UNIQUE_OPERATORS | Number of unique operators | |
| PROGRAM_VOLUME | Halstead program volume | |
| JAVA0136 | JAVA0136 N methods defined in class (maximum: M) | |
| JAVA0160 | JAVA0160 Method does not throw specified exception | |
| JAVA0126 | JAVA0126 Method declares unchecked exception in throws | |
| JAVA0171 | JAVA0171 Unused local variable | |
| JAVA0112 | JAVA0112 Incorrect javadoc: no exception 'exception' in throws | |
| JAVA0100 | JAVA0100 Class contains N non-final fields (maximum: M) | |
| WHITESPACE | Number of whitespace lines | |
| JAVA0254 | JAVA0254 Use enhanced for loop construct instead of Iterator | |
| JAVA0270 | JAVA0270 Use Java 5.0 enhanced for loop construct to iterate over all elements in an array | |
/*
Derby - Class org.apache.derby.impl.sql.compile.GroupByNode
Licensed to the Apache Software Foundation (ASF) under one or more
contributor license agreements. See the NOTICE file distributed with
this work for additional information regarding copyright ownership.
The ASF licenses this file to you under the Apache License, Version 2.0
(the "License"); you may not use this file except in compliance with
the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
*/
package org.apache.derby.impl.sql.compile;
import java.util.Iterator;
import java.util.Vector;
import java.util.ArrayList;
import java.util.Comparator;
import java.util.Collections;
import org.apache.derby.catalog.IndexDescriptor;
import org.apache.derby.iapi.error.StandardException;
import org.apache.derby.iapi.reference.ClassName;
import org.apache.derby.iapi.reference.SQLState;
import org.apache.derby.iapi.services.classfile.VMOpcode;
import org.apache.derby.iapi.services.compiler.MethodBuilder;
import org.apache.derby.iapi.services.io.FormatableArrayHolder;
import org.apache.derby.iapi.services.sanity.SanityManager;
import org.apache.derby.iapi.sql.LanguageFactory;
import org.apache.derby.iapi.sql.ResultColumnDescriptor;
import org.apache.derby.iapi.sql.compile.AccessPath;
import org.apache.derby.iapi.sql.compile.C_NodeTypes;
import org.apache.derby.iapi.sql.compile.CostEstimate;
import org.apache.derby.iapi.sql.compile.Optimizable;
import org.apache.derby.iapi.sql.compile.OptimizablePredicate;
import org.apache.derby.iapi.sql.compile.OptimizablePredicateList;
import org.apache.derby.iapi.sql.compile.Optimizer;
import org.apache.derby.iapi.sql.compile.RequiredRowOrdering;
import org.apache.derby.iapi.sql.compile.RowOrdering;
import org.apache.derby.iapi.sql.dictionary.ConglomerateDescriptor;
import org.apache.derby.iapi.sql.dictionary.DataDictionary;
import org.apache.derby.iapi.store.access.ColumnOrdering;
import org.apache.derby.impl.sql.execute.AggregatorInfo;
import org.apache.derby.impl.sql.execute.AggregatorInfoList;
/**
* A GroupByNode represents a result set for a grouping operation
* on a select. Note that this includes a SELECT with aggregates
* and no grouping columns (in which case the select list is null)
* It has the same description as its input result set.
* <p>
* For the most part, it simply delegates operations to its bottomPRSet,
* which is currently expected to be a ProjectRestrictResultSet generated
* for a SelectNode.
* <p>
* NOTE: A GroupByNode extends FromTable since it can exist in a FromList.
* <p>
* There is a lot of room for optimizations here: <UL>
* <LI> agg(distinct x) group by x => agg(x) group by x (for min and max) </LI>
* <LI> min()/max() use index scans if possible, no sort may
* be needed. </LI>
* </UL>
*
*
*/
public class GroupByNode extends SingleChildResultSetNode
{
/**
* The GROUP BY list
*/
GroupByList groupingList;
/**
* The list of all aggregates in the query block
* that contains this group by.
*/
Vector aggregateVector;
/**
* Information that is used at execution time to
* process aggregates.
*/
private AggregatorInfoList aggInfo;
/**
* The parent to the GroupByNode. If we need to
* generate a ProjectRestrict over the group by
* then this is set to that node. Otherwise it
* is null.
*/
FromTable parent;
private boolean addDistinctAggregate;
private boolean singleInputRowOptimization;
private int addDistinctAggregateColumnNum;
// Is the source in sorted order
private boolean isInSortedOrder;
private ValueNode havingClause;
private SubqueryList havingSubquerys;
/**
* Intializer for a GroupByNode.
*
* @param bottomPR The child FromTable
* @param groupingList The groupingList
* @param aggregateVector The vector of aggregates from
* the query block. Since aggregation is done
* at the same time as grouping, we need them
* here.
* @param havingClause The having clause.
* @param havingSubquerys subqueries in the having clause.
* @param tableProperties Properties list associated with the table
* @param nestingLevel nestingLevel of this group by node. This is used for
* error checking of group by queries with having clause.
* @exception StandardException Thrown on error
*/
public void init(
Object bottomPR,
Object groupingList,
Object aggregateVector,
Object havingClause,
Object havingSubquerys,
Object tableProperties,
Object nestingLevel)
throws StandardException
{
super.init(bottomPR, tableProperties);
setLevel(((Integer)nestingLevel).intValue());
this.havingClause = (ValueNode)havingClause;
this.havingSubquerys = (SubqueryList)havingSubquerys;
/* Group by without aggregates gets xformed into distinct */
if (SanityManager.DEBUG)
{
// Aggregage vector can be null if we have a having clause.
// select c1 from t1 group by c1 having c1 > 1;
// SanityManager.ASSERT(((Vector) aggregateVector).size() > 0,
// "aggregateVector expected to be non-empty");
if (!(childResult instanceof Optimizable))
{
SanityManager.THROWASSERT("childResult, " + childResult.getClass().getName() +
", expected to be instanceof Optimizable");
}
if (!(childResult instanceof FromTable))
{
SanityManager.THROWASSERT("childResult, " + childResult.getClass().getName() +
", expected to be instanceof FromTable");
}
}
ResultColumnList newBottomRCL;
this.groupingList = (GroupByList) groupingList;
this.aggregateVector = (Vector) aggregateVector;
this.parent = this;
/*
** The first thing we do is put ourselves on
** top of the SELECT. The select becomes the
** childResult. So our RCL becomes its RCL (so
** nodes above it now point to us). Map our
** RCL to its columns.
*/
newBottomRCL = childResult.getResultColumns().copyListAndObjects();
resultColumns = childResult.getResultColumns();
childResult.setResultColumns(newBottomRCL);
/*
** We have aggregates, so we need to add
** an extra PRNode and we also have to muck around
** with our trees a might.
*/
addAggregates();
/* We say that the source is never in sorted order if there is a distinct aggregate.
* (Not sure what happens if it is, so just skip it for now.)
* Otherwise, we check to see if the source is in sorted order on any permutation
* of the grouping columns.)
*/
if (! addDistinctAggregate && groupingList != null)
{
ColumnReference[] crs =
new ColumnReference[this.groupingList.size()];
// Now populate the CR array and see if ordered
int glSize = this.groupingList.size();
int index;
for (index = 0; index < glSize; index++)
{
GroupByColumn gc =
(GroupByColumn) this.groupingList.elementAt(index);
if (gc.getColumnExpression() instanceof ColumnReference)
{
crs[index] = (ColumnReference)gc.getColumnExpression();
}
else
{
isInSortedOrder = false;
break;
}
}
if (index == glSize) {
isInSortedOrder = childResult.isOrderedOn(crs, true, (Vector)null);
}
}
}
/**
* Get whether or not the source is in sorted order.
*
* @return Whether or not the source is in sorted order.
*/
boolean getIsInSortedOrder()
{
return isInSortedOrder;
}
/**
* Add the extra result columns required by the aggregates
* to the result list.
*
* @exception standard exception
*/
private void addAggregates()
throws StandardException
{
addNewPRNode();
addNewColumnsForAggregation();
addDistinctAggregatesToOrderBy();
}
/**
* Add any distinct aggregates to the order by list.
* Asserts that there are 0 or more distincts.
*/
private void addDistinctAggregatesToOrderBy()
{
int numDistinct = numDistinctAggregates(aggregateVector);
if (numDistinct != 0)
{
if (SanityManager.DEBUG)
{
SanityManager.ASSERT(numDistinct == 1,
"Should not have more than 1 distinct aggregate per Group By node");
}
AggregatorInfo agg = null;
int count = aggInfo.size();
for (int i = 0; i < count; i++)
{
agg = (AggregatorInfo) aggInfo.elementAt(i);
if (agg.isDistinct())
{
break;
}
}
if (SanityManager.DEBUG)
{
SanityManager.ASSERT(agg != null && agg.isDistinct());
}
addDistinctAggregate = true;
addDistinctAggregateColumnNum = agg.getInputColNum();
}
}
/**
* Add a new PR node for aggregation. Put the
* new PR under the sort.
*
* @exception standard exception
*/
private void addNewPRNode()
throws StandardException
{
/*
** Get the new PR, put above the GroupBy.
*/
ResultColumnList rclNew = (ResultColumnList)getNodeFactory().getNode(
C_NodeTypes.RESULT_COLUMN_LIST,
getContextManager());
int sz = resultColumns.size();
for (int i = 0; i < sz; i++)
{
ResultColumn rc = (ResultColumn) resultColumns.elementAt(i);
if (!rc.isGenerated()) {
rclNew.addElement(rc);
}
}
// if any columns in the source RCL were generated for an order by
// remember it in the new RCL as well. After the sort is done it will
// have to be projected out upstream.
rclNew.copyOrderBySelect(resultColumns);
parent = (FromTable) getNodeFactory().getNode(
C_NodeTypes.PROJECT_RESTRICT_NODE,
this, // child
rclNew,
null, //havingClause,
null, // restriction list
null, // project subqueries
havingSubquerys,
tableProperties,
getContextManager());
/*
** Reset the bottom RCL to be empty.
*/
childResult.setResultColumns((ResultColumnList)
getNodeFactory().getNode(
C_NodeTypes.RESULT_COLUMN_LIST,
getContextManager()));
/*
** Set the group by RCL to be empty
*/
resultColumns = (ResultColumnList) getNodeFactory().getNode(
C_NodeTypes.RESULT_COLUMN_LIST,
getContextManager());
}
/**
* In the query rewrite for group by, add the columns on which
* we are doing the group by.
* @see #addNewColumnsForAggregation
*/
private void addUnAggColumns() throws StandardException
{
ResultColumnList bottomRCL = childResult.getResultColumns();
ResultColumnList groupByRCL = resultColumns;
ArrayList referencesToSubstitute = new ArrayList();
ArrayList havingRefsToSubstitute = null;
if (havingClause != null)
havingRefsToSubstitute = new ArrayList();
int sz = groupingList.size();
for (int i = 0; i < sz; i++)
{
GroupByColumn gbc = (GroupByColumn) groupingList.elementAt(i);
ResultColumn newRC = (ResultColumn) getNodeFactory().getNode(
C_NodeTypes.RESULT_COLUMN,
"##UnaggColumn",
gbc.getColumnExpression(),
getContextManager());
// add this result column to the bottom rcl
bottomRCL.addElement(newRC);
newRC.markGenerated();
newRC.bindResultColumnToExpression();
newRC.setVirtualColumnId(bottomRCL.size());
// now add this column to the groupbylist
ResultColumn gbRC = (ResultColumn) getNodeFactory().getNode(
C_NodeTypes.RESULT_COLUMN,
"##UnaggColumn",
gbc.getColumnExpression(),
getContextManager());
groupByRCL.addElement(gbRC);
gbRC.markGenerated();
gbRC.bindResultColumnToExpression();
gbRC.setVirtualColumnId(groupByRCL.size());
/*
** Reset the original node to point to the
** Group By result set.
*/
VirtualColumnNode vc = (VirtualColumnNode) getNodeFactory().getNode(
C_NodeTypes.VIRTUAL_COLUMN_NODE,
this, // source result set.
gbRC,
new Integer(groupByRCL.size()),
getContextManager());
// we replace each group by expression
// in the projection list with a virtual column node
// that effectively points to a result column
// in the result set doing the group by
//
// Note that we don't perform the replacements
// immediately, but instead we accumulate them
// until the end of the loop. This allows us to
// sort the expressions and process them in
// descending order of complexity, necessary
// because a compound expression may contain a
// reference to a simple grouped column, but in
// such a case we want to process the expression
// as an expression, not as individual column
// references. E.g., if the statement was:
// SELECT ... GROUP BY C1, C1 * (C2 / 100), C3
// then we don't want the replacement of the
// simple column reference C1 to affect the
// compound expression C1 * (C2 / 100). DERBY-3094.
//
ValueNode vn = gbc.getColumnExpression();
SubstituteExpressionVisitor vis =
new SubstituteExpressionVisitor(vn, vc,
AggregateNode.class);
referencesToSubstitute.add(vis);
// Since we always need a PR node on top of the GB
// node to perform projection we can use it to perform
// the having clause restriction as well.
// To evaluate the having clause correctly, we need to
// convert each aggregate and expression to point
// to the appropriate result column in the group by node.
// This is no different from the transformations we do to
// correctly evaluate aggregates and expressions in the
// projection list.
//
//
// For this query:
// SELECT c1, SUM(c2), MAX(c3)
// FROM t1
// HAVING c1+max(c3) > 0;
// PRSN RCL -> (ptr(gbn:rcl[0]), ptr(gbn:rcl[1]), ptr(gbn:rcl[4]))
// Restriction: (> (+ ptr(gbn:rcl[0]) ptr(gbn:rcl[4])) 0)
// |
// GBN (RCL) -> (C1, SUM(C2), <input>, <aggregator>, MAX(C3), <input>, <aggregator>
// |
// FBT (C1, C2)
if (havingClause != null)
{
SubstituteExpressionVisitor havingSE =
new SubstituteExpressionVisitor(vn,vc,null);
havingRefsToSubstitute.add(havingSE);
}
gbc.setColumnPosition(bottomRCL.size());
}
Comparator sorter = new ExpressionSorter();
Collections.sort(referencesToSubstitute,sorter);
for (int r = 0; r < referencesToSubstitute.size(); r++)
parent.getResultColumns().accept(
(SubstituteExpressionVisitor)referencesToSubstitute.get(r));
if (havingRefsToSubstitute != null)
{
Collections.sort(havingRefsToSubstitute,sorter);
for (int r = 0; r < havingRefsToSubstitute.size(); r++)
havingClause.accept(
(SubstituteExpressionVisitor)havingRefsToSubstitute.get(r));
}
}
/**
* Add a whole slew of columns needed for
* aggregation. Basically, for each aggregate we add
* 3 columns: the aggregate input expression
* and the aggregator column and a column where the aggregate
* result is stored. The input expression is
* taken directly from the aggregator node. The aggregator
* is the run time aggregator. We add it to the RC list
* as a new object coming into the sort node.
* <P>
* At this point this is invoked, we have the following
* tree: <UL>
* PR - (PARENT): RCL is the original select list
* |
* PR - GROUP BY: RCL is empty
* |
* PR - FROM TABLE: RCL is empty </UL> <P>
*
* For each ColumnReference in PR RCL <UL>
* <LI> clone the ref </LI>
* <LI> create a new RC in the bottom RCL and set it
* to the col ref </LI>
* <LI> create a new RC in the GROUPBY RCL and set it to
* point to the bottom RC </LI>
* <LI> reset the top PR ref to point to the new GROUPBY
* RC</LI></UL>
*
* For each aggregate in aggregateVector <UL>
* <LI> create RC in FROM TABLE. Fill it with
* aggs Operator.
* <LI> create RC in FROM TABLE for agg result</LI>
* <LI> create RC in FROM TABLE for aggregator</LI>
* <LI> create RC in GROUPBY for agg input, set it
* to point to FROM TABLE RC </LI>
* <LI> create RC in GROUPBY for agg result</LI>
* <LI> create RC in GROUPBY for aggregator</LI>
* <LI> replace Agg with reference to RC for agg result </LI></UL>.
* <P>
* For a query like,
* <pre>
select c1, sum(c2), max(c3)
from t1
group by c1;
</pre>
* the query tree ends up looking like this:
<pre>
ProjectRestrictNode RCL -> (ptr to GBN(column[0]), ptr to GBN(column[1]), ptr to GBN(column[4]))
|
GroupByNode RCL->(C1, SUM(C2), <agg-input>, <aggregator>, MAX(C3), <agg-input>, <aggregator>)
|
ProjectRestrict RCL->(C1, C2, C3)
|
FromBaseTable
</pre>
*
* The RCL of the GroupByNode contains all the unagg (or grouping columns)
* followed by 3 RC's for each aggregate in this order: the final computed
* aggregate value, the aggregate input and the aggregator function.
* <p>
* The Aggregator function puts the results in the first of the 3 RC's
* and the PR resultset in turn picks up the value from there.
* <p>
* The notation (ptr to GBN(column[0])) basically means that it is
* a pointer to the 0th RC in the RCL of the GroupByNode.
* <p>
* The addition of these unagg and agg columns to the GroupByNode and
* to the PRN is performed in addUnAggColumns and addAggregateColumns.
* <p>
* Note that that addition of the GroupByNode is done after the
* query is optimized (in SelectNode#modifyAccessPaths) which means a
* fair amount of patching up is needed to account for generated group by columns.
* @exception standard exception
*/
private void addNewColumnsForAggregation()
throws StandardException
{
aggInfo = new AggregatorInfoList();
if (groupingList != null)
{
addUnAggColumns();
}
if (havingClause != null) {
// we have replaced group by expressions in the having clause.
// there should be no column references in the having clause
// referencing this table. Skip over aggregate nodes.
// select a, sum(b) from t group by a having a+c > 1
// is not valid because of column c.
//
// it is allright to have columns from parent or child subqueries;
// select * from p where p.p1 in
// (select c.c1 from c group by c.c1 having count(*) = p.p2
CollectNodesVisitor collectNodesVisitor =
new CollectNodesVisitor(ColumnReference.class, AggregateNode.class);
havingClause.accept(collectNodesVisitor);
for (Iterator it = collectNodesVisitor.getList().iterator();
it.hasNext(); )
{
ColumnReference cr = (ColumnReference)it.next();
if (!cr.getGeneratedToReplaceAggregate() &&
cr.getSourceLevel() == level) {
throw StandardException.newException(
SQLState.LANG_INVALID_COL_HAVING_CLAUSE,
cr.getSQLColumnName());
}
}
}
addAggregateColumns();
}
/**
* In the query rewrite involving aggregates, add the columns for
* aggregation.
*
* @see #addNewColumnsForAggregation
*/
private void addAggregateColumns() throws StandardException
{
DataDictionary dd = getDataDictionary();
AggregateNode aggregate = null;
ColumnReference newColumnRef;
ResultColumn newRC;
ResultColumn tmpRC;
ResultColumn aggInputRC;
ResultColumnList bottomRCL = childResult.getResultColumns();
ResultColumnList groupByRCL = resultColumns;
ResultColumnList aggRCL;
int aggregatorVColId;
int aggInputVColId;
int aggResultVColId;
/*
** Now process all of the aggregates. Replace
** every aggregate with an RC. We toss out
** the list of RCs, we need to get each RC
** as we process its corresponding aggregate.
*/
LanguageFactory lf = getLanguageConnectionContext().getLanguageFactory();
ReplaceAggregatesWithCRVisitor replaceAggsVisitor =
new ReplaceAggregatesWithCRVisitor(
(ResultColumnList) getNodeFactory().getNode(
C_NodeTypes.RESULT_COLUMN_LIST,
getContextManager()),
((FromTable) childResult).getTableNumber(),
ResultSetNode.class);
parent.getResultColumns().accept(replaceAggsVisitor);
if (havingClause != null)
{
// replace aggregates in the having clause with column references.
replaceAggsVisitor = new ReplaceAggregatesWithCRVisitor(
(ResultColumnList) getNodeFactory().getNode(
C_NodeTypes.RESULT_COLUMN_LIST,
getContextManager()),
((FromTable)childResult).getTableNumber());
havingClause.accept(replaceAggsVisitor);
// make having clause a restriction list in the parent
// project restrict node.
ProjectRestrictNode parentPRSN = (ProjectRestrictNode)parent;
parentPRSN.setRestriction(havingClause);
}
/*
** For each aggregate
*/
int alSize = aggregateVector.size();
for (int index = 0; index < alSize; index++)
{
aggregate = (AggregateNode) aggregateVector.elementAt(index);
/*
** AGG RESULT: Set the aggregate result to null in the
** bottom project restrict.
*/
newRC = (ResultColumn) getNodeFactory().getNode(
C_NodeTypes.RESULT_COLUMN,
"##aggregate result",
aggregate.getNewNullResultExpression(),
getContextManager());
newRC.markGenerated();
newRC.bindResultColumnToExpression();
bottomRCL.addElement(newRC);
newRC.setVirtualColumnId(bottomRCL.size());
aggResultVColId = newRC.getVirtualColumnId();
/*
** Set the GB aggregrate result column to
** point to this. The GB aggregate result
** was created when we called
** ReplaceAggregatesWithColumnReferencesVisitor()
*/
newColumnRef = (ColumnReference) getNodeFactory().getNode(
C_NodeTypes.COLUMN_REFERENCE,
newRC.getName(),
null,
getContextManager());
newColumnRef.setSource(newRC);
newColumnRef.setNestingLevel(this.getLevel());
newColumnRef.setSourceLevel(this.getLevel());
tmpRC = (ResultColumn) getNodeFactory().getNode(
C_NodeTypes.RESULT_COLUMN,
newRC.getColumnName(),
newColumnRef,
getContextManager());
tmpRC.markGenerated();
tmpRC.bindResultColumnToExpression();
groupByRCL.addElement(tmpRC);
tmpRC.setVirtualColumnId(groupByRCL.size());
/*
** Set the column reference to point to
** this.
*/
newColumnRef = aggregate.getGeneratedRef();
newColumnRef.setSource(tmpRC);
/*
** AGG INPUT: Create a ResultColumn in the bottom
** project restrict that has the expression that is
** to be aggregated
*/
newRC = aggregate.getNewExpressionResultColumn(dd);
newRC.markGenerated();
newRC.bindResultColumnToExpression();
bottomRCL.addElement(newRC);
newRC.setVirtualColumnId(bottomRCL.size());
aggInputVColId = newRC.getVirtualColumnId();
aggInputRC = newRC;
/*
** Add a reference to this column into the
** group by columns.
*/
tmpRC = getColumnReference(newRC, dd);
groupByRCL.addElement(tmpRC);
tmpRC.setVirtualColumnId(groupByRCL.size());
/*
** AGGREGATOR: Add a getAggregator method call
** to the bottom result column list.
*/
newRC = aggregate.getNewAggregatorResultColumn(dd);
newRC.markGenerated();
newRC.bindResultColumnToExpression();
bottomRCL.addElement(newRC);
newRC.setVirtualColumnId(bottomRCL.size());
aggregatorVColId = newRC.getVirtualColumnId();
/*
** Add a reference to this column in the Group By result
** set.
*/
tmpRC = getColumnReference(newRC, dd);
groupByRCL.addElement(tmpRC);
tmpRC.setVirtualColumnId(groupByRCL.size());
/*
** Piece together a fake one column rcl that we will use
** to generate a proper result description for input
** to this agg if it is a user agg.
*/
aggRCL = (ResultColumnList) getNodeFactory().getNode(
C_NodeTypes.RESULT_COLUMN_LIST,
getContextManager());
aggRCL.addElement(aggInputRC);
/*
** Note that the column ids in the row are 0 based
** so we have to subtract 1.
*/
aggInfo.addElement(new AggregatorInfo(
aggregate.getAggregateName(),
aggregate.getAggregatorClassName(),
aggInputVColId - 1, // aggregate input column
aggResultVColId -1, // the aggregate result column
aggregatorVColId - 1, // the aggregator column
aggregate.isDistinct(),
lf.getResultDescription(aggRCL.makeResultDescriptors(), "SELECT")
));
}
}
/**
* Return the parent node to this one, if there is
* one. It will return 'this' if there is no generated
* node above this one.
*
* @return the parent node
*/
public FromTable getParent()
{
return parent;
}
/*
* Optimizable interface
*/
/**
* @see Optimizable#optimizeIt
*
* @exception StandardException Thrown on error
*/
public CostEstimate optimizeIt(
Optimizer optimizer,
OptimizablePredicateList predList,
CostEstimate outerCost,
RowOrdering rowOrdering)
throws StandardException
{
// RESOLVE: NEED TO FACTOR IN THE COST OF GROUPING (SORTING) HERE
CostEstimate childCost = ((Optimizable) childResult).optimizeIt(
optimizer,
predList,
outerCost,
rowOrdering);
CostEstimate retval = super.optimizeIt(
optimizer,
predList,
outerCost,
rowOrdering
);
return retval;
}
/**
* @see Optimizable#estimateCost
*
* @exception StandardException Thrown on error
*/
public CostEstimate estimateCost(OptimizablePredicateList predList,
ConglomerateDescriptor cd,
CostEstimate outerCost,
Optimizer optimizer,
RowOrdering rowOrdering
)
throws StandardException
{
// RESOLVE: NEED TO FACTOR IN THE COST OF GROUPING (SORTING) HERE
//
CostEstimate childCost = ((Optimizable) childResult).estimateCost(
predList,
cd,
outerCost,
optimizer,
rowOrdering);
CostEstimate costEstimate = getCostEstimate(optimizer);
costEstimate.setCost(childCost.getEstimatedCost(),
childCost.rowCount(),
childCost.singleScanRowCount());
return costEstimate;
}
/**
* @see org.apache.derby.iapi.sql.compile.Optimizable#pushOptPredicate
*
* @exception StandardException Thrown on error
*/
public boolean pushOptPredicate(OptimizablePredicate optimizablePredicate)
throws StandardException
{
return ((Optimizable) childResult).pushOptPredicate(optimizablePredicate);
}
/**
* Convert this object to a String. See comments in QueryTreeNode.java
* for how this should be done for tree printing.
*
* @return This object as a String
*/
public String toString()
{
if (SanityManager.DEBUG)
{
return "singleInputRowOptimization: " + singleInputRowOptimization + "\n" +
childResult.toString() + "\n" + super.toString();
}
else
{
return "";
}
}
/**
* Evaluate whether or not the subquery in a FromSubquery is flattenable.
* Currently, a FSqry is flattenable if all of the following are true:
* o Subquery is a SelectNode.
* o It contains no top level subqueries. (RESOLVE - we can relax this)
* o It does not contain a group by or having clause
* o It does not contain aggregates.
*
* @param fromList The outer from list
*
* @return boolean Whether or not the FromSubquery is flattenable.
*/
public boolean flattenableInFromSubquery(FromList fromList)
{
/* Can't flatten a GroupByNode */
return false;
}
/**
* Optimize this GroupByNode.
*
* @param dataDictionary The DataDictionary to use for optimization
* @param predicates The PredicateList to optimize. This should
* be a join predicate.
* @param outerRows The number of outer joining rows
*
* @return ResultSetNode The top of the optimized subtree
*
* @exception StandardException Thrown on error
*/
public ResultSetNode optimize(DataDictionary dataDictionary,
PredicateList predicates,
double outerRows)
throws StandardException
{
/* We need to implement this method since a PRN can appear above a
* SelectNode in a query tree.
*/
childResult = (FromTable) childResult.optimize(
dataDictionary,
predicates,
outerRows);
Optimizer optimizer = getOptimizer(
(FromList) getNodeFactory().getNode(
C_NodeTypes.FROM_LIST,
getNodeFactory().doJoinOrderOptimization(),
getContextManager()),
predicates,
dataDictionary,
(RequiredRowOrdering) null);
// RESOLVE: NEED TO FACTOR IN COST OF SORTING AND FIGURE OUT HOW
// MANY ROWS HAVE BEEN ELIMINATED.
costEstimate = optimizer.newCostEstimate();
costEstimate.setCost(childResult.getCostEstimate().getEstimatedCost(),
childResult.getCostEstimate().rowCount(),
childResult.getCostEstimate().singleScanRowCount());
return this;
}
ResultColumnDescriptor[] makeResultDescriptors()
{
return childResult.makeResultDescriptors();
}
/**
* Return whether or not the underlying ResultSet tree will return
* a single row, at most.
* This is important for join nodes where we can save the extra next
* on the right side if we know that it will return at most 1 row.
*
* @return Whether or not the underlying ResultSet tree will return a single row.
* @exception StandardException Thrown on error
*/
public boolean isOneRowResultSet() throws StandardException
{
// Only consider scalar aggregates for now
return ((groupingList == null) || (groupingList.size() == 0));
}
/**
* generate the sort result set operating over the source
* resultset. Adds distinct aggregates to the sort if
* necessary.
*
* @exception StandardException Thrown on error
*/
public void generate(ActivationClassBuilder acb,
MethodBuilder mb)
throws StandardException
{
int orderingItem = 0;
int aggInfoItem = 0;
FormatableArrayHolder orderingHolder;
/* Get the next ResultSet#, so we can number this ResultSetNode, its
* ResultColumnList and ResultSet.
*/
assignResultSetNumber();
// Get the final cost estimate from the child.
costEstimate = childResult.getFinalCostEstimate();
/*
** Get the column ordering for the sort. Note that
** for a scalar aggegate we may not have any ordering
** columns (if there are no distinct aggregates).
** WARNING: if a distinct aggregate is passed to
** SortResultSet it assumes that the last column
** is the distinct one. If this assumption changes
** then SortResultSet will have to change.
*/
orderingHolder = acb.getColumnOrdering(groupingList);
if (addDistinctAggregate)
{
orderingHolder = acb.addColumnToOrdering(
orderingHolder,
addDistinctAggregateColumnNum);
}
if (SanityManager.DEBUG)
{
if (SanityManager.DEBUG_ON("AggregateTrace"))
{
StringBuffer s = new StringBuffer();
s.append("Group by column ordering is (");
ColumnOrdering[] ordering =
(ColumnOrdering[])orderingHolder.getArray(ColumnOrdering.class);
for (int i = 0; i < ordering.length; i++)
{
s.append(ordering[i].getColumnId());
s.append(" ");
}
s.append(")");
SanityManager.DEBUG("AggregateTrace", s.toString());
}
}
orderingItem = acb.addItem(orderingHolder);
/*
** We have aggregates, so save the aggInfo
** struct in the activation and store the number
*/
if (SanityManager.DEBUG)
{
SanityManager.ASSERT(aggInfo != null,
"aggInfo not set up as expected");
}
aggInfoItem = acb.addItem(aggInfo);
acb.pushGetResultSetFactoryExpression(mb);
// Generate the child ResultSet
childResult.generate(acb, mb);
mb.push(isInSortedOrder);
mb.push(aggInfoItem);
mb.push(orderingItem);
resultColumns.generateHolder(acb, mb);
mb.push(resultColumns.getTotalColumnSize());
mb.push(resultSetNumber);
/* Generate a (Distinct)ScalarAggregateResultSet if scalar aggregates */
if ((groupingList == null) || (groupingList.size() == 0))
{
genScalarAggregateResultSet(acb, mb);
}
/* Generate a (Distinct)GroupedAggregateResultSet if grouped aggregates */
else
{
genGroupedAggregateResultSet(acb, mb);
}
}
/**
* Generate the code to evaluate scalar aggregates.
*
*/
private void genScalarAggregateResultSet(ActivationClassBuilder acb,
MethodBuilder mb)
{
/* Generate the (Distinct)ScalarAggregateResultSet:
* arg1: childExpress - Expression for childResult
* arg2: isInSortedOrder - true if source result set in sorted order
* arg3: aggregateItem - entry in saved objects for the aggregates,
* arg4: orderItem - entry in saved objects for the ordering
* arg5: Activation
* arg6: rowAllocator - method to construct rows for fetching
* from the sort
* arg7: row size
* arg8: resultSetNumber
* arg9: Whether or not to perform min optimization.
*/
String resultSet = (addDistinctAggregate) ? "getDistinctScalarAggregateResultSet" : "getScalarAggregateResultSet";
mb.push(singleInputRowOptimization);
mb.push(costEstimate.rowCount());
mb.push(costEstimate.getEstimatedCost());
mb.callMethod(VMOpcode.INVOKEINTERFACE, (String) null, resultSet,
ClassName.NoPutResultSet, 10);
}
/**
* Generate the code to evaluate grouped aggregates.
*
*/
private void genGroupedAggregateResultSet(ActivationClassBuilder acb,
MethodBuilder mb)
throws StandardException
{
/* Generate the (Distinct)GroupedAggregateResultSet:
* arg1: childExpress - Expression for childResult
* arg2: isInSortedOrder - true if source result set in sorted order
* arg3: aggregateItem - entry in saved objects for the aggregates,
* arg4: orderItem - entry in saved objects for the ordering
* arg5: Activation
* arg6: rowAllocator - method to construct rows for fetching
* from the sort
* arg7: row size
* arg8: resultSetNumber
*/
String resultSet = (addDistinctAggregate) ? "getDistinctGroupedAggregateResultSet" : "getGroupedAggregateResultSet";
mb.push(costEstimate.rowCount());
mb.push(costEstimate.getEstimatedCost());
mb.callMethod(VMOpcode.INVOKEINTERFACE, (String) null, resultSet,
ClassName.NoPutResultSet, 9);
}
///////////////////////////////////////////////////////////////
//
// UTILITIES
//
///////////////////////////////////////////////////////////////
/**
* Method for creating a new result column referencing
* the one passed in.
*
* @param targetRC the source
* @param dd
*
* @return the new result column
*
* @exception StandardException on error
*/
private ResultColumn getColumnReference(ResultColumn targetRC,
DataDictionary dd)
throws StandardException
{
ColumnReference tmpColumnRef;
ResultColumn newRC;
tmpColumnRef = (ColumnReference) getNodeFactory().getNode(
C_NodeTypes.COLUMN_REFERENCE,
targetRC.getName(),
null,
getContextManager());
tmpColumnRef.setSource(targetRC);
tmpColumnRef.setNestingLevel(this.getLevel());
tmpColumnRef.setSourceLevel(this.getLevel());
newRC = (ResultColumn) getNodeFactory().getNode(
C_NodeTypes.RESULT_COLUMN,
targetRC.getColumnName(),
tmpColumnRef,
getContextManager());
newRC.markGenerated();
newRC.bindResultColumnToExpression();
return newRC;
}
/**
* Consider any optimizations after the optimizer has chosen a plan.
* Optimizations include:
* o min optimization for scalar aggregates
* o max optimization for scalar aggregates
*
* @param selectHasPredicates true if SELECT containing this
* vector/scalar aggregate has a restriction
*
* @exception StandardException on error
*/
void considerPostOptimizeOptimizations(boolean selectHasPredicates)
throws StandardException
{
/* Consider the optimization for min with asc index on that column or
* max with desc index on that column:
* o No group by
* o One of:
* o min/max(ColumnReference) is only aggregate && source is
* ordered on the ColumnReference
* o min/max(ConstantNode)
* The optimization of the other way around (min with desc index or
* max with asc index) has the same restrictions with the additional
* temporary restriction of no qualifications at all (because
* we don't have true backward scans).
*/
if (groupingList == null)
{
if (aggregateVector.size() == 1)
{
AggregateNode an = (AggregateNode) aggregateVector.elementAt(0);
AggregateDefinition ad = an.getAggregateDefinition();
if (ad instanceof MaxMinAggregateDefinition)
{
if (an.getOperand() instanceof ColumnReference)
{
/* See if the underlying ResultSet tree
* is ordered on the ColumnReference.
*/
ColumnReference[] crs = new ColumnReference[1];
crs[0] = (ColumnReference) an.getOperand();
Vector tableVector = new Vector();
boolean minMaxOptimizationPossible = isOrderedOn(crs, false, tableVector);
if (SanityManager.DEBUG)
{
SanityManager.ASSERT(tableVector.size() <= 1, "bad number of FromBaseTables returned by isOrderedOn() -- "+tableVector.size());
}
if (minMaxOptimizationPossible)
{
boolean ascIndex = true;
int colNum = crs[0].getColumnNumber();
/* Check if we have an access path, this will be
* null in a join case (See Beetle 4423)
*/
AccessPath accessPath= getTrulyTheBestAccessPath();
if (accessPath == null)
return;
IndexDescriptor id = accessPath.
getConglomerateDescriptor().
getIndexDescriptor();
int[] keyColumns = id.baseColumnPositions();
boolean[] isAscending = id.isAscending();
for (int i = 0; i < keyColumns.length; i++)
{
/* in such a query: select min(c3) from
* tab1 where c1 = 2 and c2 = 5, if prefix keys
* have equality operator, then we can still use
* the index. The checking of equality operator
* has been done in isStrictlyOrderedOn.
*/
if (colNum == keyColumns[i])
{
if (! isAscending[i])
ascIndex = false;
break;
}
}
FromBaseTable fbt = (FromBaseTable)tableVector.firstElement();
MaxMinAggregateDefinition temp = (MaxMinAggregateDefinition)ad;
/* MAX ASC NULLABLE
* ---- ----------
* TRUE TRUE TRUE/FALSE = Special Last Key Scan (ASC Index Last key with null skips)
* TRUE FALSE TRUE/FALSE = JustDisableBulk(DESC index 1st key with null skips)
* FALSE TRUE TRUE/FALSE = JustDisableBulk(ASC index 1st key)
* FALSE FALSE TRUE/FALSE = Special Last Key Scan(Desc Index Last Key)
*/
if (((!temp.isMax()) && ascIndex) ||
((temp.isMax()) && !ascIndex))
{
fbt.disableBulkFetch();
singleInputRowOptimization = true;
}
/*
** Max optimization with asc index or min with
** desc index is currently more
** restrictive than otherwise.
** We are getting the store to return the last
** row from an index (for the time being, the
** store cannot do real backward scans). SO
** we cannot do this optimization if we have
** any predicates at all.
*/
else if (!selectHasPredicates &&
((temp.isMax() && ascIndex) ||
(!temp.isMax() && !ascIndex )))
{
fbt.disableBulkFetch();
fbt.doSpecialMaxScan();
singleInputRowOptimization = true;
}
}
}
else if (an.getOperand() instanceof ConstantNode)
{
singleInputRowOptimization = true;
}
}
}
}
}
/**
* Comparator class for GROUP BY expression substitution.
*
* This class enables the sorting of a collection of
* SubstituteExpressionVisitor instances. We sort the visitors
* during the tree manipulation processing in order to process
* expressions of higher complexity prior to expressions of
* lower complexity. Processing the expressions in this order ensures
* that we choose the best match for an expression, and thus avoids
* problems where we substitute a sub-expression instead of the
* full expression. For example, if the statement is:
* ... GROUP BY a+b, a, a*(a+b), a+b+c
* we'll process those expressions in the order: a*(a+b),
* a+b+c, a+b, then a.
*/
private static final class ExpressionSorter implements Comparator
{
public int compare(Object o1, Object o2)
{
try {
ValueNode v1 = ((SubstituteExpressionVisitor)o1).getSource();
ValueNode v2 = ((SubstituteExpressionVisitor)o2).getSource();
int refCount1, refCount2;
CollectNodesVisitor vis = new CollectNodesVisitor(
ColumnReference.class);
v1.accept(vis);
refCount1 = vis.getList().size();
vis = new CollectNodesVisitor(ColumnReference.class);
v2.accept(vis);
refCount2 = vis.getList().size();
// The ValueNode with the larger number of refs
// should compare lower. That way we are sorting
// the expressions in descending order of complexity.
return refCount2 - refCount1;
}
catch (StandardException e)
{
throw new RuntimeException(e);
}
}
}
}
The table below shows all metrics for GroupByNode.java.
| Metric | Value | Description | |
|---|---|---|---|
| BLOCKS | 69.00 | Number of blocks | |
| BLOCK_COMMENT | 189.00 | Number of block comment lines | |
| COMMENTS | 541.00 | Comment lines | |
| COMMENT_DENSITY | 1.05 | Comment density | |
| COMPARISONS | 45.00 | Number of comparison operators | |
| CYCLOMATIC | 84.00 | Cyclomatic complexity | |
| DECL_COMMENTS | 31.00 | Comments in declarations | |
| DOC_COMMENT | 272.00 | Number of javadoc comment lines | |
| ELOC | 513.00 | Effective lines of code | |
| EXEC_COMMENTS | 55.00 | Comments in executable code | |
| EXITS | 129.00 | Procedure exits | |
| FUNCTIONS | 23.00 | Number of function declarations | |
| HALSTEAD_DIFFICULTY | 87.24 | Halstead difficulty | |
| HALSTEAD_EFFORT | 0.00 | Halstead effort | |
| INTERFACE_COMPLEXITY | 110.00 | Interface complexity | |
| JAVA0001 | 0.00 | JAVA0001 Package name does not contain only lower case letters | |
| JAVA0002 | 0.00 | JAVA0002 Package name does not begin with a top level domain name or country code | |
| JAVA0003 | 0.00 | JAVA0003 Minimize use of on-demand (.*) imports | |
| JAVA0004 | 0.00 | JAVA0004 Unnecessary import from java.lang | |
| JAVA0005 | 1.00 | JAVA0005 Imports not in specified order | |
| JAVA0006 | 0.00 | JAVA0006 Empty finally block | |
| JAVA0007 | 0.00 | JAVA0007 Should not declare public field | |
| JAVA0008 | 0.00 | JAVA0008 Empty catch block | |
| JAVA0009 | 0.00 | JAVA0009 Protected member in final class | |
| JAVA0010 | 0.00 | JAVA0010 Non-instantiable class does not contain a non-private static member | |
| JAVA0011 | 0.00 | JAVA0011 Abstract class does not contain an abstract method | |
| JAVA0012 | 0.00 | JAVA0012 Non-constructor method with same name as declaring class | |
| JAVA0013 | 0.00 | JAVA0013 Non-blank final field is not static | |
| JAVA0014 | 0.00 | JAVA0014 Class with only static members has non-private constructor | |
| JAVA0015 | 0.00 | JAVA0015 Package class contains public nested type | |
| JAVA0016 | 0.00 | JAVA0016 Abstract class contains public constructor | |
| JAVA0017 | 0.00 | JAVA0017 Class name does not have required form | |
| JAVA0018 | 0.00 | JAVA0018 Method name does not have required form | |
| JAVA0019 | 0.00 | JAVA0019 Interface name does not have required form | |
| JAVA0020 | 0.00 | JAVA0020 Field name does not have required form | |
| JAVA0021 | 0.00 | JAVA0021 Interface method name does not have required form | |
| JAVA0022 | 0.00 | JAVA0022 Static final field name does not have required form | |
| JAVA0023 | 0.00 | JAVA0023 Empty finalize method | |
| JAVA0024 | 0.00 | JAVA0024 Empty class | |
| JAVA0025 | 0.00 | JAVA0025 Method override is empty | |
| JAVA0026 | 0.00 | JAVA0026 Finalize method with parameters | |
| JAVA0029 | 0.00 | JAVA0029 Private method not used | |
| JAVA0030 | 0.00 | JAVA0030 Private field not used | |
| JAVA0031 | 0.00 | JAVA0031 Case statement not properly closed | |
| JAVA0032 | 0.00 | JAVA0032 Switch statement missing default | |
| JAVA0033 | 0.00 | JAVA0033 default: not last case in switch statement | |
| JAVA0034 | 3.00 | JAVA0034 Missing braces in if statement | |
| JAVA0035 | 2.00 | JAVA0035 Missing braces in for statement | |
| JAVA0036 | 0.00 | JAVA0036 Missing braces in while statement | |
| JAVA0038 | 0.00 | JAVA0038 Non-case label in switch statement | |
| JAVA0039 | 0.00 | JAVA0039 Break statement with label | |
| JAVA0040 | 0.00 | JAVA0040 Switch statement contains N cases (maximum: M) | |
| JAVA0041 | 0.00 | JAVA0041 Nested synchronized block | |
| JAVA0042 | 0.00 | JAVA0042 Empty synchronized statement | |
| JAVA0043 | 0.00 | JAVA0043 Inner class does not use outer class | |
| JAVA0044 | 0.00 | JAVA0044 Serializable class with no instance variables | |
| JAVA0045 | 0.00 | JAVA0045 Serializable class with only transient fields | |
| JAVA0046 | 0.00 | JAVA0046 Name of class not derived from Exception ends with 'Exception' | |
| JAVA0047 | 0.00 | JAVA0047 Serializable class derives from invalid base class | |
| JAVA0048 | 0.00 | JAVA0048 Name of class derived from Exception does not end with 'Exception' | |
| JAVA0049 | 2.00 | JAVA0049 Nested block at depth N (maximum: M) | |
| JAVA0050 | 0.00 | JAVA0050 Class derives from java.lang.Error | |
| JAVA0051 | 0.00 | JAVA0051 Class derives from java.lang.RuntimeException | |
| JAVA0052 | 0.00 | JAVA0052 Class derives from java.lang.Throwable | |
| JAVA0053 | 0.00 | JAVA0053 Unused label | |
| JAVA0054 | 1.00 | JAVA0054 Inheritance depth N exceeds maximum M | |
| JAVA0055 | 0.00 | JAVA0055 Class should be interface | |
| JAVA0056 | 0.00 | JAVA0056 Unnecessary abstract modifier for interface or annotation | |
| JAVA0057 | 0.00 | JAVA0057 Unnecessary default constructor | |
| JAVA0058 | 0.00 | JAVA0058 Constructor calls super() | |
| JAVA0059 | 0.00 | JAVA0059 Method override only calls super() | |
| JAVA0061 | 0.00 | JAVA0061 Inaccessible member in anonymous class | |
| JAVA0062 | 0.00 | JAVA0062 Public class missing public member or protected constructor | |
| JAVA0063 | 0.00 | JAVA0063 Identifier name should not contain '$' | |
| JAVA0064 | 0.00 | JAVA0064 N variations of identifier name (maximum: M) | |
| JAVA0065 | 0.00 | JAVA0065 Unnecessary final modifier for method in final class | |
| JAVA0066 | 0.00 | JAVA0066 Unnecessary modifier for interface nested type | |
| JAVA0067 | 0.00 | JAVA0067 Array descriptor on identifier name | |
| JAVA0068 | 0.00 | JAVA0068 Modifiers not declared in recommended order | |
| JAVA0071 | 0.00 | JAVA0071 Strings compared with == | |
| JAVA0073 | 0.00 | JAVA0073 Integer division in floating-point context | |
| JAVA0074 | 0.00 | JAVA0074 Use of Object.notify() | |
| JAVA0075 | 7.00 | JAVA0075 Method parameter hides field | |
| JAVA0076 | 0.00 | JAVA0076 Use of magic number | |
| JAVA0077 | 0.00 | JAVA0077 Private field not used in declaring class | |
| JAVA0078 | 0.00 | JAVA0078 Floating point values compared with == | |
| JAVA0079 | 0.00 | JAVA0079 Use of instance to reference static member | |
| JAVA0080 | 0.00 | JAVA0080 Import declaration not used | |
| JAVA0081 | 0.00 | JAVA0081 Boolean literal in comparison | |
| JAVA0082 | 0.00 | JAVA0082 Unnecessary widening cast | |
| JAVA0083 | 0.00 | JAVA0083 Unnecessary instanceof test | |
| JAVA0084 | 0.00 | JAVA0084 Should use compound assignment operator | |
| JAVA0085 | 0.00 | JAVA0085 Use of sun.* class | |
| JAVA0087 | 0.00 | JAVA0087 Use of Thread.sleep() | |
| JAVA0089 | 0.00 | JAVA0089 Use of restricted package | |
| JAVA0092 | 0.00 | JAVA0092 Use of restricted type | |
| JAVA0093 | 0.00 | JAVA0093 Redundant assignment | |
| JAVA0094 | 0.00 | JAVA0094 Field hides a superclass field | |
| JAVA0095 | 0.00 | JAVA0095 Uninitialized private field | |
| JAVA0096 | 0.00 | JAVA0096 Field in nested class hides outer field | |
| JAVA0098 | 1.00 | JAVA0098 Minimize use of implicit field initializers | |
| JAVA0100 | 1.00 | JAVA0100 Class contains N non-final fields (maximum: M) | |
| JAVA0101 | 0.00 | JAVA0101 Unnecessary modifier for field in interface | |
| JAVA0102 | 0.00 | JAVA0102 Last statement in finalize() not super.finalize() | |
| JAVA0103 | 0.00 | JAVA0103 Explicit call to finalize() | |
| JAVA0104 | 0.00 | JAVA0104 finalize() only calls super.finalize() | |
| JAVA0105 | 0.00 | JAVA0105 Duplicate import declaration | |
| JAVA0106 | 0.00 | JAVA0106 Unnecessary import from current package | |
| JAVA0108 | 4.00 | JAVA0108 Incorrect javadoc: no @param tag for 'parameter' | |
| JAVA0109 | 0.00 | JAVA0109 Incorrect javadoc: no parameter 'parameter' | |
| JAVA0110 | 3.00 | JAVA0110 Incorrect javadoc: no @return tag | |
| JAVA0111 | 0.00 | JAVA0111 Incorrect javadoc: @return tag for void method | |
| JAVA0112 | 3.00 | JAVA0112 Incorrect javadoc: no exception 'exception' in throws | |
| JAVA0113 | 1.00 | JAVA0113 Incorrect javadoc: no @author tag | |
| JAVA0114 | 1.00 | JAVA0114 Incorrect javadoc: no @version tag | |
| JAVA0115 | 6.00 | JAVA0115 Incorrect javadoc: no @throws or @exception tag for 'exception' | |
| JAVA0116 | 0.00 | JAVA0116 Missing javadoc: field 'field' | |
| JAVA0117 | 0.00 | JAVA0117 Missing javadoc: method 'method' | |
| JAVA0118 | 0.00 | JAVA0118 Missing javadoc: type 'type' | |
| JAVA0119 | 0.00 | JAVA0119 Control variable changed within body of for loop | |
| JAVA0123 | 0.00 | JAVA0123 Use all three components of for loop | |
| JAVA0125 | 0.00 | JAVA0125 Continue statement with label | |
| JAVA0126 | 0.00 | JAVA0126 Method declares unchecked exception in throws | |
| JAVA0128 | 0.00 | JAVA0128 Public constructor in non-public class | |
| JAVA0130 | 0.00 | JAVA0130 Non-static method does not use instance fields | |
| JAVA0131 | 0.00 | JAVA0131 Compatible method does not override base | |
| JAVA0132 | 0.00 | JAVA0132 Method overload with compatible signature | |
| JAVA0133 | 0.00 | JAVA0133 Non-synchronized method overrides synchronized method | |
| JAVA0135 | 0.00 | JAVA0135 Only one of Object.equals and Object.hashCode defined: missing 'method' | |
| JAVA0136 | 1.00 | JAVA0136 N methods defined in class (maximum: M) | |
| JAVA0137 | 1.00 | JAVA0137 Non-abstract class missing constructor | |
| JAVA0138 | 1.00 | JAVA0138 N parameters defined for method (maximum: M) | |
| JAVA0139 | 0.00 | JAVA0139 Definition of main other than public static void main(java.lang.String[]) | |
| JAVA0141 | 0.00 | JAVA0141 Unnecessary modifier for method in interface | |
| JAVA0143 | 0.00 | JAVA0143 Synchronized method | |
| JAVA0144 | 1.00 | JAVA0144 Line exceeds maximum M characters | |
| JAVA0145 | 3648.00 | JAVA0145 Tab character used in source file | |
| JAVA0150 | 0.00 | JAVA0150 java.lang.Error (or subclass) thrown | |
| JAVA0153 | 0.00 | JAVA0153 Inefficient conversion of integer to string | |
| JAVA0159 | 0.00 | JAVA0159 Inefficient conversion of string to integer | |
| JAVA0160 | 1.00 | JAVA0160 Method does not throw specified exception | |
| JAVA0161 | 0.00 | JAVA0161 Conditional wait() not in loop | |
| JAVA0163 | 0.00 | JAVA0163 Empty statement | |
| JAVA0165 | 0.00 | JAVA0165 Conflicting return statement in finally block | |
| JAVA0166 | 0.00 | JAVA0166 Generic exception caught | |
| JAVA0167 | 0.00 | JAVA0167 ThreadDeath not rethrown | |
| JAVA0169 | 0.00 | JAVA0169 Unnecessary catch block: exception 'exception' | |
| JAVA0170 | 0.00 | JAVA0170 Caught exception not derived from java.lang.Exception | |
| JAVA0171 | 1.00 | JAVA0171 Unused local variable | |
| JAVA0173 | 3.00 | JAVA0173 Unused method parameter | |
| JAVA0174 | 0.00 | JAVA0174 Assigned local variable never used | |
| JAVA0175 | 0.00 | JAVA0175 Successive assignment to variable | |
| JAVA0176 | 0.00 | JAVA0176 Local variable name does not have required form | |
| JAVA0177 | 15.00 | JAVA0177 Variable declaration missing initializer | |
| JAVA0179 | 2.00 | JAVA0179 Local variable hides visible field | |
| JAVA0233 | 0.00 | JAVA0233 Definition of serialVersionUID other than 'private static final long serialVersionUID' | |
| JAVA0234 | 0.00 | JAVA0234 Class is Serializable but does not define serialVersionUID | |
| JAVA0235 | 0.00 | JAVA0235 Class defines serialVersionUID but does not implement Serializable | |
| JAVA0236 | 0.00 | JAVA0236 Attempt to clone an object which does not implement Cloneable | |
| JAVA0237 | 0.00 | JAVA0237 Class implements Cloneable but does not have public clone method | |
| JAVA0238 | 0.00 | JAVA0238 Clone method does not call super.clone() | |
| JAVA0239 | 0.00 | JAVA0239 Class declares 'readObject' or 'writeObject' but does not implement Serializable | |
| JAVA0240 | 0.00 | JAVA0240 Serializable class which declares readObject or writeObject but not both | |
| JAVA0241 | 0.00 | JAVA0241 'readObject' or 'writeObject' should be declared private in Serializable class | |
| JAVA0242 | 0.00 | JAVA0242 Transient field in non-Serializable class | |
| JAVA0243 | 0.00 | JAVA0243 'readResolve' or 'writeReplace' should be declared private or protected | |
| JAVA0244 | 0.00 | JAVA0244 Field or method name in subclass differs only by case from inherited field or method | |
| JAVA0245 | 0.00 | JAVA0245 JUnit TestCase with non-trivial constructor | |
| JAVA0246 | 0.00 | JAVA0246 JUnit assertXXX statement missing message parameter | |
| JAVA0247 | 0.00 | JAVA0247 JUnit 'setUp()' and 'tearDown()' should call super method | |
| JAVA0248 | 0.00 | JAVA0248 JUnit method 'setUp' or 'tearDown' with incorrect signature | |
| JAVA0249 | 0.00 | JAVA0249 JUnit TestCase 'suite()' should be declared static | |
| JAVA0250 | 0.00 | JAVA0250 JUnit TestCase declares testXXX method with incorrect signature | |
| JAVA0251 | 0.00 | JAVA0251 Use '%n' for line breaks in printf/format for platform independence | |
| JAVA0252 | 0.00 | JAVA0252 'enum' is a Java 1.5 reserved word | |
| JAVA0253 | 0.00 | JAVA0253 Not all enum constants consumed in switch statement | |
| JAVA0254 | 1.00 | JAVA0254 Use enhanced for loop construct instead of Iterator | |
| JAVA0255 | 0.00 | JAVA0255 Result of method invocation not used | |
| JAVA0256 | 0.00 | JAVA0256 Assignment of external collection/array to field | |
| JAVA0257 | 0.00 | JAVA0257 Use of 'Constant Interface' anti-pattern | |
| JAVA0258 | 0.00 | JAVA0258 Implement Iterable for foreach compatibility | |
| JAVA0259 | 0.00 | JAVA0259 Return of collection/array field | |
| JAVA0260 | 0.00 | JAVA0260 Use 'enum' instead of Enumerated Type pattern | |
| JAVA0261 | 0.00 | JAVA0261 Use specialized Enum collection types | |
| JAVA0262 | 0.00 | JAVA0262 Use of char in integer context | |
| JAVA0263 | 0.00 | JAVA0263 Long literal ends with 'l' instead of 'L' | |
| JAVA0264 | 0.00 | JAVA0264 Integer math in long context - check for overflow | |
| JAVA0265 | 0.00 | JAVA0265 Use of Throwable.printStackTrace() | |
| JAVA0266 | 0.00 | JAVA0266 Use of System.out | |
| JAVA0267 | 0.00 | JAVA0267 Use of System.err | |
| JAVA0269 | 0.00 | JAVA0269 Contents of StringBuffer never used | |
| JAVA0270 | 1.00 | JAVA0270 Use Java 5.0 enhanced for loop construct to iterate over all elements in an array | |
| JAVA0271 | 0.00 | JAVA0271 Minimize use of on-demand (.*) static imports | |
| JAVA0272 | 0.00 | JAVA0272 Thread.run() called | |
| JAVA0273 | 0.00 | JAVA0273 Non-final derivative of Thread calls start() in constructor | |
| JAVA0274 | 0.00 | JAVA0274 Serializable class has a synchronized readObject() | |
| JAVA0275 | 0.00 | JAVA0275 Serializable class has a synchronized writeObject() and no other synchronized methods | |
| JAVA0276 | 0.00 | JAVA0276 Unnecessary use of String constructor | |
| JAVA0277 | 0.00 | JAVA0277 Iterator.next() implementation does not throw NoSuchElementException | |
| JAVA0278 | 0.00 | JAVA0278 Unnecessary use of Boolean constructor | |
| JAVA0279 | 0.00 | JAVA0279 Serialization method readObject or readObjectNoData calls an overridable method | |
| JAVA0280 | 0.00 | JAVA0280 IllegalMonitorStateException caught | |
| JAVA0281 | 0.00 | JAVA0281 Iterator.next() not called in loop | |
| JAVA0282 | 0.00 | JAVA0282 Call to Iterator.next() in loop which does not test Iterator.hasNext() | |
| JAVA0283 | 0.00 | JAVA0283 Control variable not updated in loop body | |
| JAVA0284 | 0.00 | JAVA0284 Explicit garbage collection | |
| JAVA0285 | 2.00 | JAVA0285 Dereference of potentially null variable | |
| JAVA0286 | 0.00 | JAVA0286 Dereference of null variable | |
| JAVA0287 | 0.00 | JAVA0287 Unnecessary null check | |
| JAVA0288 | 0.00 | JAVA0288 Inconsistent null check | |
| LINES | 1301.00 | Number of lines in the source file | |
| LINE_COMMENT | 80.00 | Number of line comments | |
| LOC | 651.00 | Lines of code | |
| LOGICAL_LINES | 299.00 | Number of statements | |
| LOOPS | 10.00 | Number of loops | |
| NEST_DEPTH | 8.00 | Maximum nesting depth | |
| OPERANDS | 1365.00 | Number of operands | |
| OPERATORS | 2419.00 | Number of operators | |
| PARAMS | 32.00 | Number of formal parameter declarations | |
| PROGRAM_LENGTH | 3784.00 | Halstead program length | |
| PROGRAM_VOCAB | 450.00 | Halstead program vocabulary | |
| PROGRAM_VOLUME | 0.00 | Halstead program volume | |
| RETURNS | 78.00 | Number of return points from functions | |
| SIZE | 41381.00 | Size of the file in bytes | |
| UNIQUE_OPERANDS | 399.00 | Number of unique operands | |
| UNIQUE_OPERATORS | 51.00 | Number of unique operators | |
| WHITESPACE | 109.00 | Number of whitespace lines | |