I am trying to create a PDF document using iText (a Java library) and I found out that Indic script specificaly Malayalam Language Unicode is not displayed properly. Ligature substitution and reordering is not working at all. (See attached image1.png)
However, Open Office does display the text properly (see attached image2.png)
and when I export this document to a PDF document, the text in the PDF is fine (See attached image3.png).
I am suprised at this bevahiour. The font I am using in all the three case is FreeSerif.ttf, which has support for Malayalam language and has the Open Type tables that is required for ligature substitutions.
One difference I noted is this -
Document Property (Fonts) of the PDF with problem shows FreeSerif , True Type (CID), Embedded
and Document Property (Fonts) of the Open Office Exported PDF shows Free Serif, True Type, Embedded SubSet
You may try a forum for iText.
My doubt is that whether PDF can internally layout complex text using Open Type tables in the font or does the program generating PDF (like iText) should do the Text Layout and provide PDF with the final GlyphIds to render?