I was poking around at bug 2246 for some reason and thought it might be more fun to step back a bit. The result is a perl fragment that does the basic work of creating a pdf which has a barcode and text in Greek, Cyrillic, all the latins, Japanese, Thai, Hindi and Gujarati (I'm not conversant in most of those languages so I might have some silly mistakes there) all on one page. It uses PDF::API2 instead of PDF::Reuse. If you're curious, have a look at http://reedwade.net/koha/labels/ for the perl script and the pdf output. The enabling feature of interest is the ability in PDF::API2 to build a font from a set of ttf fonts and assigning different unicode code pages (or ranges) for each. It might be possible to rejigger C4/Creators/PDF.pm to use PDF::API2 instead of PDF::Reuse. This is something I predict I will not have time to pursue further anytime soon -- I'm hoping someone else might enjoy that? -reed