PdfViewer: Add support for extracting text from composite fonts with no ToUnicode mapping

Unplanned

Last Updated: 10 Nov 2020 15:27 by ADMIN

Mingchang

Created on: 10 Nov 2020 15:22

Category: PDFViewer

Type: Feature Request

PdfViewer: Add support for extracting text from composite fonts with no ToUnicode mapping

When extracting text from a composite font the ToUnicode mapping should be used but when this mapping is missing the char codes must be mapped by constructing a secondary map from fonts ordering and registry.

A possible workaround could be to register a TrueType font using the FontsRepository.RegisterFont() method:

FontsRepository.RegisterFont(new FontFamily(fontName), FontStyles.Normal, FontWeights.Normal, File.ReadAllBytes(pathToFont));

and then to iterate the document content and change the specific font with the registered one:

FontsRepository.TryCreateFont(new FontFamily(fontName), FontStyles.Normal, FontWeights.Normal, out FontBase font);

foreach (RadFixedPage page in document.Pages)
{
	foreach (ContentElementBase contentElement in page.Content)
	{
		if (contentElement is TextFragment textFragment)
		{
			if (textFragment.Font.Name == fontName)
			{
				textFragment.Font = font;
			}
		}
	}
}