FileDocCategorySizeDatePackage
TableSorter.javaAPI DocSun JDK 1.4.2 Example12873Thu May 12 00:35:28 BST 2005None

TableSorter

public class TableSorter extends TableMap
A sorter for TableModels. The sorter has a model (conforming to TableModel) and itself implements TableModel. TableSorter does not store or copy the data in the TableModel, instead it maintains an array of integers which it keeps the same size as the number of rows in its model. When the model changes it notifies the sorter that something has changed eg. "rowsAdded" so that its internal array of integers can be reallocated. As requests are made of the sorter (like getValueAt(row, col) it redirects them to its model via the mapping array. That way the TableSorter appears to hold another copy of the table with the rows in a different order. The sorting algorthm used is stable which means that it does not move around rows when its comparison function returns 0 to denote that they are equivalent.
version
1.12 01/23/03
author
Philip Milne

Fields Summary
int[]
indexes
Vector
sortingColumns
boolean
ascending
int
compares
Constructors Summary
public TableSorter()


     
    
        indexes = new int[0]; // For consistency.        
    
public TableSorter(TableModel model)

        setModel(model);
    
Methods Summary
public voidaddMouseListenerToHeaderInTable(javax.swing.JTable table)

 
        final TableSorter sorter = this; 
        final JTable tableView = table; 
        tableView.setColumnSelectionAllowed(false); 
        MouseAdapter listMouseListener = new MouseAdapter() {
            public void mouseClicked(MouseEvent e) {
                TableColumnModel columnModel = tableView.getColumnModel();
                int viewColumn = columnModel.getColumnIndexAtX(e.getX()); 
                int column = tableView.convertColumnIndexToModel(viewColumn); 
                if(e.getClickCount() == 1 && column != -1) {
                    System.out.println("Sorting ..."); 
                    int shiftPressed = e.getModifiers()&InputEvent.SHIFT_MASK; 
                    boolean ascending = (shiftPressed == 0); 
                    sorter.sortByColumn(column, ascending); 
                }
             }
         };
        JTableHeader th = tableView.getTableHeader(); 
        th.addMouseListener(listMouseListener); 
    
public voidcheckModel()

        if (indexes.length != model.getRowCount()) {
            System.err.println("Sorter not informed of a change in model.");
        }
    
public intcompare(int row1, int row2)

        compares++;
        for(int level = 0; level < sortingColumns.size(); level++)
            {
                Integer column = (Integer)sortingColumns.elementAt(level);
                int result = compareRowsByColumn(row1, row2, column.intValue());
                if (result != 0)
                    return ascending ? result : -result;
            }
        return 0;
    
public intcompareRowsByColumn(int row1, int row2, int column)

        Class type = model.getColumnClass(column);
        TableModel data = model;

        // Check for nulls

        Object o1 = data.getValueAt(row1, column);
        Object o2 = data.getValueAt(row2, column); 

        // If both values are null return 0
        if (o1 == null && o2 == null) {
            return 0; 
        }
        else if (o1 == null) { // Define null less than everything. 
            return -1; 
        } 
        else if (o2 == null) { 
            return 1; 
        }

/* We copy all returned values from the getValue call in case
an optimised model is reusing one object to return many values.
The Number subclasses in the JDK are immutable and so will not be used in 
this way but other subclasses of Number might want to do this to save 
space and avoid unnecessary heap allocation. 
*/
        if (type.getSuperclass() == java.lang.Number.class)
            {
                Number n1 = (Number)data.getValueAt(row1, column);
                double d1 = n1.doubleValue();
                Number n2 = (Number)data.getValueAt(row2, column);
                double d2 = n2.doubleValue();

                if (d1 < d2)
                    return -1;
                else if (d1 > d2)
                    return 1;
                else
                    return 0;
            }
        else if (type == java.util.Date.class)
            {
                Date d1 = (Date)data.getValueAt(row1, column);
                long n1 = d1.getTime();
                Date d2 = (Date)data.getValueAt(row2, column);
                long n2 = d2.getTime();

                if (n1 < n2)
                    return -1;
                else if (n1 > n2)
                    return 1;
                else return 0;
            }
        else if (type == String.class)
            {
                String s1 = (String)data.getValueAt(row1, column);
                String s2    = (String)data.getValueAt(row2, column);
                int result = s1.compareTo(s2);

                if (result < 0)
                    return -1;
                else if (result > 0)
                    return 1;
                else return 0;
            }
        else if (type == Boolean.class)
            {
                Boolean bool1 = (Boolean)data.getValueAt(row1, column);
                boolean b1 = bool1.booleanValue();
                Boolean bool2 = (Boolean)data.getValueAt(row2, column);
                boolean b2 = bool2.booleanValue();

                if (b1 == b2)
                    return 0;
                else if (b1) // Define false < true
                    return 1;
                else
                    return -1;
            }
        else
            {
                Object v1 = data.getValueAt(row1, column);
                String s1 = v1.toString();
                Object v2 = data.getValueAt(row2, column);
                String s2 = v2.toString();
                int result = s1.compareTo(s2);

                if (result < 0)
                    return -1;
                else if (result > 0)
                    return 1;
                else return 0;
            }
    
public java.lang.ObjectgetValueAt(int aRow, int aColumn)

        checkModel();
        return model.getValueAt(indexes[aRow], aColumn);
    
public voidn2sort()

        for(int i = 0; i < getRowCount(); i++) {
            for(int j = i+1; j < getRowCount(); j++) {
                if (compare(indexes[i], indexes[j]) == -1) {
                    swap(i, j);
                }
            }
        }
    
public voidreallocateIndexes()

        int rowCount = model.getRowCount();

        // Set up a new array of indexes with the right number of elements
        // for the new data model.
        indexes = new int[rowCount];

        // Initialise with the identity mapping.
        for(int row = 0; row < rowCount; row++)
            indexes[row] = row;
    
public voidsetModel(javax.swing.table.TableModel model)

        super.setModel(model); 
        reallocateIndexes(); 
    
public voidsetValueAt(java.lang.Object aValue, int aRow, int aColumn)

        checkModel();
        model.setValueAt(aValue, indexes[aRow], aColumn);
    
public voidshuttlesort(int[] from, int[] to, int low, int high)

        if (high - low < 2) {
            return;
        }
        int middle = (low + high)/2;
        shuttlesort(to, from, low, middle);
        shuttlesort(to, from, middle, high);

        int p = low;
        int q = middle;

        /* This is an optional short-cut; at each recursive call,
        check to see if the elements in this subset are already
        ordered.  If so, no further comparisons are needed; the
        sub-array can just be copied.  The array must be copied rather
        than assigned otherwise sister calls in the recursion might
        get out of sinc.  When the number of elements is three they
        are partitioned so that the first set, [low, mid), has one
        element and and the second, [mid, high), has two. We skip the
        optimisation when the number of elements is three or less as
        the first compare in the normal merge will produce the same
        sequence of steps. This optimisation seems to be worthwhile
        for partially ordered lists but some analysis is needed to
        find out how the performance drops to Nlog(N) as the initial
        order diminishes - it may drop very quickly.  */

        if (high - low >= 4 && compare(from[middle-1], from[middle]) <= 0) {
            for (int i = low; i < high; i++) {
                to[i] = from[i];
            }
            return;
        }

        // A normal merge. 

        for(int i = low; i < high; i++) {
            if (q >= high || (p < middle && compare(from[p], from[q]) <= 0)) {
                to[i] = from[p++];
            }
            else {
                to[i] = from[q++];
            }
        }
    
public voidsort(java.lang.Object sender)

        checkModel();

        compares = 0;
        // n2sort();
        // qsort(0, indexes.length-1);
        shuttlesort((int[])indexes.clone(), indexes, 0, indexes.length);
        System.out.println("Compares: "+compares);
    
public voidsortByColumn(int column)

        sortByColumn(column, true);
    
public voidsortByColumn(int column, boolean ascending)

        this.ascending = ascending;
        sortingColumns.removeAllElements();
        sortingColumns.addElement(new Integer(column));
        sort(this);
        super.tableChanged(new TableModelEvent(this)); 
    
public voidswap(int i, int j)

        int tmp = indexes[i];
        indexes[i] = indexes[j];
        indexes[j] = tmp;
    
public voidtableChanged(javax.swing.event.TableModelEvent e)

	System.out.println("Sorter: tableChanged"); 
        reallocateIndexes();

        super.tableChanged(e);