Changes in iTextSharp 3.0.6 (2005-08-18) 
- constructor Image(Image) didn’t copy one of the fields 
- Image.ScaledToFit() didn’t take rotation into account 
- Table.Width was not being overriden. The result was an exception in PdfDocument 
- added BaseFont.CorrectArabicAdvance() to make Arabic Diactrics (tashkeel) 
have zero advance. Some Microsoft fonts like Arial.ttf have non-zero advances 
that confuse the Arabic shaper. 
- in PdfDocument the waiting image in the last page was not being placed 
- PdfString didn’t output the BOM in Unicode 
- PdfStamper supports encryption in append mode 
- added verbose analysis of the permissions in PdfEncryptor 
- fixed a conversion bug in PdfPCell.NoWrap 
- PdfReader.SelectPages() is working again 
- PdfWriter.PrintScalingNone had the wrong value 
- fixed a conversion bug in SimpleBookmark.IterateOutlines() 
- MultiColumnText didn’t respect mirrored margins 
- in RTF added the option for chars above ‘z’ to not be encoded in Unicode 
to support older Word versions