Trouble: VDX and Unicode, non-Latin data

Hi,

I'm having the following troubles when producing a VDX file from a Unicode data file. My data contains Cyrillic characters, although it shouldn't matter since it's all in Unicode.

 

Trouble #1

If the first character in a variable text frame is a non-Latin character, it gets half cut in the VDX output. It works fine when outputting to PDF. There's a workaround, though: I put a space in front of the non-Latin character.

 

Trouble #2

Whenever a line of text in a variable text frame ends with a non-Latin character, its output gets cut off somewhere in the middle as if it did not fit in the frame. It works fine when outputting to PDF. This one is harder to find a workaround for, because some times a line has to end with a non-Latin character.

 

Am I doing anything wrong and this is a feature , or is this truly a bug?

I don't see how this question can possibly be answered in the abstract with so little information. Typesetting is a complicated business, and there are a lot of things which could affect the output, including the particular font you're using, the text you're trying to set, and other things particular to your job and your machine, not to mention the version of the software you're running (see my signature below). These complexities only multiply with Unicode. Therefore I advise you to please send the files for a job which demonstrates the problem to Support for analysis.