Hi all,

I am working on a program that outputting PostScript to printers. In the 
  Postscript output, the program encodes true type fonts into multiple 
type 42 fonts with a composite type 0 as parent. the program works fine 
on single-byte charsets, however, it does not work well if the font 
contains CJK characters, even on some quite new printers.

It seems that PostScript interpreters have problems dealing with large 
glyphs (>64K in size), even we encode the data into multiple strings in 
/sfnt array.

I am considering converting true type to type3 and it seems working 
except that I am not sure how to convert the encoding. Our goal is not 
to touch other part of the code, just encode true type fonts into type3.

Can anybody show a PostScript source code about this? I tried a couple 
of encoding choices, but failed to get one to work. I started to suspect 
if it is even possible in this case.

Your help is greatly appreciated.

John