Q&A: Mixed Arabic and Roman script

Q One of the trickier tasks that I have to do is prepare labels containing mixed Arabic and Roman script. Clients normally send the text as image files, but sometimes it comes in Word documents; the latter become reversed when I open them in Word on my Mac, although our office PC displays them correctly. PDFs do not display correctly either. How can I correct this?

A This occurs because of two convergent issues.

First, Word and Acrobat have still not become thoroughly Unicode-based. For example most PDF text content is not stored in Unicode, as that remains only an option.

Handling documents containing mixed text using scripts running in different directions is the other problem, which is application-dependent and not part of the Unicode standard, thus open to vendor-specific kludges that do not work with all releases of their apps. In many cases, vendors offer localised versions of their apps in Arabic-based markets, which use native document formats that are not handled properly by non-localised versions.

Some word processors, such as Nisus, excel in mixed text work, or you can buy a localised version of Word. It may be simpler to stick with screenshots.

There is a series of article here considering some of these issues, starting here.

Updated from the original, which was first published in MacUser volume 29 issue 11, 2013.