Version: (using KDE 4.3.3) Installed from: SuSE RPMs Let's say you have a pdf with mixed languages (example: polish and russian). Here you can download one for testing: http://www.jezykiobce.net/jezykiobce_pdf/rosyjski_kurs_podst.pdf when you copy polish part or russian part (or both) you end up with garbage, like: êàíèêóëû only basic latin letters plus letter ó are copied correctly. It is not a matter of font used, because I can use for the same font intended letters: ęóąśł or яерсидоф
Given the problem is reproduceable (in the very same way) with: - acroread 9.2 - okular 0.9.2 + poppler 0.12.2 - evince 2.28.1 + poppler 0.12.2 I'm rather inclined to conclude the document might be badly encoded. (Note: what you see in a PDF is not what you copy as text.)
Pino, thank you for explanation. I assumed that by definition pdf has to be properly encoded (text I mean).