Itext How to set both UTF encoding and Font to a Paragraph - itext

Question

I want to set a font to a text that is in cyrillic.I successfully convert the text to cyrilic, but i cannot set a Font to the same text.
File fontFile = new File("arialuni.ttf");
Document document = new Document();
PdfWriter writer = PdfWriter.getInstance(document, new FileOutputStream(RESULT));
document.open();
writer.getAcroForm().setNeedAppearances(true);
Font boldFont = new Font(Font.FontFamily.TIMES_ROMAN, 18, Font.BOLD);
Font normalFont = new Font(Font.FontFamily.TIMES_ROMAN, 10, Font.ITALIC);
BaseFont unicode = BaseFont.createFont(fontFile.getAbsolutePath(), BaseFont.IDENTITY_H, BaseFont.EMBEDDED);
FontSelector fs = new FontSelector();
fs.addFont(new Font(unicode));
addContent(document,article.getTitle(),fs,boldFont);
private static void addContent(Document document,String paragraph,FontSelector fs,Font font) throws DocumentException {
Phrase phrase = fs.process(paragraph);
Paragraph p = new Paragraph(phrase.toString(),font);
document.add(p);
}

Answer 1

As #mkl indicates in the comments, you are mixing FontSelector functionality that gives you a Phrase that could use the appropriate unicode fonts (fonts with BaseFont.IDENTITY_H as encoding parameter), with creating a Paragraph with a simple font (Font.FontFamily.TIMES_ROMAN).
When you do fs.process(paragraph), you get a Phrase in which every Chunk has the correct font, but when you do phrase.toString(), you throw away all those fonts, and you replace them with Font.FontFamily.TIMES_ROMAN. That doesn't make any sense.
Why don't you replace this:
Phrase phrase = fs.process(paragraph);
Paragraph p = new Paragraph(phrase.toString(),font);
document.add(p);
with:
document.add(fs.process(paragraph));
Why does your addContent() method need a Font as parameter? Also, if you really need a Paragraph object you can also do this:
Paragraph p = new Paragraph();
p.add(fs.process(paragraph));
document.add(p);
Or even:
Paragraph p = new Paragraph(fs.process(paragraph));
document.add(p);
As long as you don't replace the correct fonts with the incorrect fonts by "flattening" the Phrase to a String, you're probably OK.
Note that you probably even don't need the FontSelector. There's nothing wrong with doing this:
BaseFont unicode = BaseFont.createFont(
fontFile.getAbsolutePath(), BaseFont.IDENTITY_H, BaseFont.EMBEDDED);
Font font = new Font(unicode, 12);
Paragraph p = new Paragraph(paragraph, font);
It seems to me that you are making things unnecessary complex.