Thread: What data type ie collection or logic to be used for data? An abstraction problem

Hi,


I have a pivot and my pivotTableModel holds data in this fashion...
eg I have a data containing task,Year and month field values


The struture of my PivotTableModel. Below the data is being show in PivotTableModel rows...
it has theses fields

Task	Year	Month	Values
Issued check			7
	2010		4
		Feb	3
		Mar	1
	2011		3
		Dec	3

Now I have to show this pivot table

2011 2010 Total
Task Feb Mar Dec
Issued Checks 3 1 3 7

ie

				Total
	2011		2010
	Feb	Mar	Dec
Task				100 (including other tasks)
Issued Checks	3	1	3	7

and so on...


How to go about it?

I have a logic which tells me that user has asked for one (or more ) columns on the left and one or more columns on the top

I am havng success to show and also sort values of the left ie values of the task. but the values on the top. If single set of values I am to show. no problem, it is sorting too. but when I have more than that on the top, the top header values stop sorting and also the values against them in the table are not getting put correctly.


There can be more than one Feb 2011 occuring against tasks that I have not shown. So I need to merge all occurences and show as one Feb 2011 on the x-axis and show other tasks as occuring against it...


Thank you

Do you have any specific java programming questions?

Yes I have...


Here is example code...


import java.util.*;
class PivotTableModel{


private String []columns ={"name","Year","Month","Date","pivotValue","Total s"};;
private Class []columnClass ={String.class,Integer.class,Integer.class,Double. class,Double.class};;

private Object [][]rows = new Object[][]{
new Object[]{"CENTRAL BANK ORDER PROCESS",null,null,null,128.0},
new Object[]{null,2010,null,null,null,null},
new Object[]{null,null,12,null,2.0,null},
new Object[]{null,null,null,27,2.0,null},
new Object[]{null,null,null,28,2.0,null},
new Object[]{null,2011,null,null,null,null},
new Object[]{null,null,3,null,2.0,null},
new Object[]{null,null,4,null,15.0,null},
new Object[]{null,null,5,null,41.0,null},
new Object[]{null,null,null,27,2.0,null},
new Object[]{null,null,6,null,41.0,null},
new Object[]{null,null,8,null,27.0,null},
};

public PivotTableModel(){}


public int getColumnCount() {
return (columns != null) ? columns.length : 0;
}

public String getColumnName(int columnIndex) {
return (columns != null) ? columns[columnIndex] : null;

}


public int getRowCount() {
return (rows != null) ? rows.length : 0;
}

public Object getValueAt(int rowIndex, int columnIndex) {
return (rows != null) ? rows[rowIndex][columnIndex] : null;
}


public Class getColumnClass(int columnIndex) {
return columnClass[columnIndex];
}

public static void main(String []args){

PivotTableModel ptm = new PivotTableModel();

System.out.println(ptm.getValueAt(0,0));
System.out.println(ptm.getColumnCount());
System.out.println(ptm.getRowCount());
System.out.println(ptm.getColumnName(3));
System.out.println(ptm.getColumnClass(3));


// the number of table column that doesnot cocern us for this problem and its index number in the above table.
int rowHeaderLen =1;
int rowHeaderLenIndex =1-1;

// index of first column of interest
int firstColumnHeaderIndex=1;
// the number of columns that are to be included. Total 3 ie top column and 2 children level columns
int columnHeaderLen = 3;

// something to hold our collections
List data = new LinkedList();
List columnTopHeader;

int valueStoredAt=0;
Vector subElement = null;

for(int row = 0; row< ptm.getRowCount();row++){
int put=0;
for(int col = 0; col< ptm.getColumnCount();col++){
if(col>rowHeaderLenIndex&&col<rowHeaderLen+columnH eaderLen)
{
if(ptm.getValueAt(row,col)!=null){
put = col;
}
if(put>rowHeaderLenIndex&&ptm.getValueAt(row,col)! =null){

if(put==firstColumnHeaderIndex){
System.out.println("Add new one at "+put+" for "+ptm.getColumnName(col));
columnTopHeader = new LinkedList();
data.add(columnTopHeader);
//subElement = new Vector();
//if(subElement != null)
//columnTopHeader.add(subElement);
}else if(put>firstColumnHeaderIndex){

if(col>valueStoredAt){
// New vector should be created and put in the vector that is currently in focus
System.out.println("New vector pos at "+put+" for "+ptm.getColumnName(col));
//subElement = new Vector();
//subElement.add(ptm.getValueAt(row,col));

}else if(col==valueStoredAt){
// Add new element in the this level vector
System.out.println("Add New element pos at "+put+" for "+ptm.getColumnName(col));
//subElement.add(ptm.getValueAt(row,col));
}else if(col<valueStoredAt){
// Add new element in the previous level vector
System.out.println("Add New element pos at "+put+" for "+ptm.getColumnName(col));
//subElement.add(ptm.getValueAt(row,col));
}



}
valueStoredAt = col;
System.out.println("last value Added at "+valueStoredAt);


}

System.out.print(ptm.getValueAt(row,col)+"\t");
}

}
System.out.println();
}

}

}



But alas! Only the print statements correctly indicate what behaviour needs to followed; how to achieve that through code? Well, that beats me!

Do I need to have Hashtable/HashMap to hold Super level values and their subsequest child? How to go about it? What I need to code?
Please ignore Vector subElement or other code that may seem not correct. What the code tells is what is needed to be done at specific points in the program.


Thanks in advance