0

I regularly create documents that need Unicode characters above U+FFFF. Unfortunately, OpenOffice and LibreOffice are both unable to correctly export these characters when creating a PDF. The actual data gets mangled by a completely asinine algorithm, while the display just consists of various overlapping question mark boxes.

This is not a font issue. I embed all used fonts in the PDF and all characters below U+FFFF work perfectly fine.

Until now I have been working around this issue by mapping the glyphs I need to a custom PUA font. This solves the display problems, but obviously makes the actual content of the text unsearchable and quite fragile. I haven’t been able to find any settings that might affect the handling of Unicode characters in PDF.

Therefore I have three questions:

  1. Is there a way to make OpenOffice/LibreOffice handle astral characters correctly on PDF export?
  2. If not, is there an external tool that can convert .odt files to PDF while preserving astral characters?
  3. If not, is there another good rich-text editor using a different file format that can deal with astral characters in PDFs?
CharlotteBuff
  • 3,389
  • 1
  • 16
  • 18
  • I can only partially reproduce your problem. I was able to create a PDF from an .odt document containing . It looks alright, but the character can't be searched for. Copy-pasting from the PDF gives me the replacement character (here the viewer and/or the OS might be interfering). I used LibreOffice 5.1.6.2 (Writer) on Ubuntu 16.04. – lenz Jul 27 '17 at 22:08
  • I’m using Windows 10. Maybe that is the problem? – CharlotteBuff Jul 28 '17 at 12:33
  • Win 10 is not the problem, its appear in ubuntu too.Its seems LO exported pdf can not render correctly those types of characters – Andrewboy Feb 04 '20 at 00:50

0 Answers0