itextsharp converting 1 column to 3 columns per page - .net-4.0

Question

I was wondering if i could get some help converting this into a 3 column (going down to left) per page report.
using System;
using System.IO;
using iTextSharp.text;
using iTextSharp.text.pdf;
namespace ConsoleApp
{
class Program
{
static void Main(string[] args)
{
Console.WriteLine("-> Creates a PDF file with a block of Text.");
Document document = new Document(PageSize.LETTER);
try
{
PdfWriter writer = PdfWriter.GetInstance(
document,
new FileStream(#"c:\\temp\\column_example.pdf", FileMode.Create));
document.Open();
PdfContentByte cb = writer.DirectContent;
float pos;
PdfPTable table;
PdfPCell cell = new PdfPCell(new Phrase(string.Empty));
Phrase phrase;
float columnWidth = (PageSize.LETTER.Width - 36);
ColumnText ct = GetColumn(cb, columnWidth);
int status = 0;
string line = "Line{0}";
for(int i=0; i<50; i++)
{
table = new PdfPTable(1);
table.SpacingAfter = 9F;
cell = new PdfPCell(new Phrase("Header for table " + i));
table.AddCell(cell);
for (int j = 0; j < (i%2 == 0 ? 5 : 7); j++)
{
phrase = new Phrase(string.Format(line, i));
cell = new PdfPCell(phrase);
table.AddCell(cell);
}
ct.AddElement(table);
pos = ct.YLine;
status = ct.Go(true);
Console.WriteLine("Lines written:" + ct.LinesWritten + " Y-position: " + pos + " - " + ct.YLine);
if (!ColumnText.HasMoreText(status))
{
ct.AddElement(table);
ct.YLine = pos;
ct.Go(false);
}
else
{
document.NewPage();
ct.SetText(null);
ct.AddElement(table);
ct.YLine = PageSize.LETTER.Height - 36;
ct.Go();
}
}
}
catch (DocumentException de)
{
Console.Error.WriteLine(de.Message);
}
catch (IOException ioe)
{
Console.Error.WriteLine(ioe.Message);
}
finally
{
document.Close();
}
Console.ReadLine();
}
private static ColumnText GetColumn(PdfContentByte cb, float columnWidth)
{
var ct = new ColumnText(cb);
ct.SetSimpleColumn(36, 36, columnWidth, PageSize.LETTER.Height - 36, 18, Element.ALIGN_JUSTIFIED);
return ct;
}
}
}
I'm really new with itextsharp and can't find any good examples on how to do this.
Thanks for any help

Answer 1

The easiest way to do that is to put your individual tables into a master 3-column table. Below is code that does that. You'll probably want to adjust margins, widths and borders but this should get you started at least.
Also, since you said you were new to iTextSharp I'm going to assume that you don't have a specific need for using DirectContent. DC is very powerful but most of what you need to do with iTextSharp you can do through specific objects instead. The code below has all DC stuff removed.
//(iText 5.1.1.0)
using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
using iTextSharp.text;
using iTextSharp.text.pdf;
using System.IO;
namespace ConsoleApplication1
{
class Program
{
static void Main(string[] args)
{
string filePath = Path.Combine(Environment.GetFolderPath(Environment.SpecialFolder.Desktop), "column_example.pdf");
Console.WriteLine("-> Creates a PDF file with a block of Text.");
Document document = new Document(PageSize.LETTER);
try
{
PdfWriter writer = PdfWriter.GetInstance(document, new FileStream(filePath, FileMode.Create));
document.Open();
//Create a master table with 3 columns
PdfPTable masterTable = new PdfPTable(3);
//Set the column widths, this should probably be adjusted
masterTable.SetWidths(new float[] { 200, 200, 200 });
PdfPTable table;
PdfPCell cell;
Phrase phrase;
string line = "Line{0}";
for (int i = 0; i < 50; i++)
{
table = new PdfPTable(1);
table.SpacingAfter = 9F;
cell = new PdfPCell(new Phrase("Header for table " + i));
table.AddCell(cell);
for (int j = 0; j < (i % 2 == 0 ? 5 : 7); j++)
{
phrase = new Phrase(string.Format(line, i));
cell = new PdfPCell(phrase);
table.AddCell(cell);
}
//Add the sub-table to our master table instead of the writer
masterTable.AddCell(table);
}
//Add the master table to our document
document.Add(masterTable);
}
catch (DocumentException de)
{
Console.Error.WriteLine(de.Message);
}
catch (IOException ioe)
{
Console.Error.WriteLine(ioe.Message);
}
finally
{
document.Close();
}
Console.ReadLine();
}
}
}
EDIT
Sorry, I didn't understand from your original post what you were looking for but do now. Unfortunately you are entering the realm of Math and Mod. I don't have to time (or the brain power this morning) to go through this completely but hopefully I can give you a start.
The entire programming world is based on left-to-right and then top-to-bottom, when you switch it around you tend to have to jump through giant hoops to do what you want (like making an HTML list into 3 columns alphabetized with A's in column 1, B's in column 2, etc.)
In order to do what you want you need to know the heights of the tables so that you can calculate how many vertically that you can get on the page. Unfortunately table height isn't known until render time. The solution (at least for me) is to draw each table to a temporary document which allows us to know the height, then we store the table in an array and throw away the document. Now we've got an array of tables with known heights that we can walk through.
The snippet below does all of this. I changed your row count rule to a random number from 2 to 9 just to get more variety in the sample data. Also, starting with iTextSharp 5.1 (I think that's the right version) many of the "big" objects support IDisposable so I'm using. If you are using an older version you'll need to drop the using and switch to normal variable declaration. Hopefully the comments make sense. You'll see that I pulled out some magic numbers into variables, too.
//Our array of tables
List<PdfPTable> Tables = new List<PdfPTable>();
//Create a random number of rows to get better sample data
int rowCount;
Random r = new Random();
string line = "Line {0}";
PdfPTable table;
//This is the horizontal padding between tables
float hSpace = 5;
//Total number of columns that we want
int columnCount = 3;
//Create a temporary document to write our table to so that their sizes can be calculated
using (Document tempDoc = new Document(PageSize.LETTER))
{
using (MemoryStream tempMS = new MemoryStream())
{
using (PdfWriter tempW = PdfWriter.GetInstance(tempDoc, tempMS))
{
tempDoc.Open();
//Calculate the table width which is the usable space minus the padding between tables divided by the column count
float documentUseableWidth = tempDoc.PageSize.Width - tempDoc.LeftMargin - tempDoc.RightMargin;
float totalTableHPadding = (hSpace * (columnCount - 1));
float tableWidth = (documentUseableWidth - totalTableHPadding) / columnCount;
for (int i = 0; i < 50; i++)
{
table = new PdfPTable(1);
table.AddCell(new PdfPCell(new Phrase("Header for table " + i)));
rowCount = r.Next(2, 10);
for (int j = 0; j < rowCount; j++)
{
table.AddCell(new PdfPCell(new Phrase(string.Format(line, i))));
}
//In order to use WriteSelectedRows you need to set the width of the table
table.SetTotalWidth(new float[] { tableWidth });
//Write the table to our temporary document in order to calculate the height
table.WriteSelectedRows(1, table.Rows.Count, 0, 0, tempW.DirectContent);
//Add the table to our array
Tables.Add(table);
}
tempDoc.Close();
}
}
}
Once you've got your array of tables you can loop through those and draw them using:
Tables[i].WriteSelectedRows(1, Tables[i].Rows.Count, curX, curY, writer.DirectContent);
Where i is your current table index and curX and curY are your current coordinates.
Hopefully this gets you going in the right direction. WriteSelectedRows does a great job of putting a table exactly where you want it.
One last thing to remember, the Y coordinate that it takes starts at the bottom of the document, not the top, so 0 is the bottom and 720 is "above" it and not below.