Bug 692317

Summary: Type42 font converted with ps2write produces limitcheck error with Adobe Distiller 9.0
Product: Ghostscript Reporter: steve166
Component: PS WriterAssignee: Ken Sharp <ken.sharp>
Status: RESOLVED INVALID    
Severity: normal    
Priority: P4    
Version: 9.02   
Hardware: PC   
OS: Windows XP   
Customer: Word Size: ---
Attachments: ps testcase for ps2write distiller 9.0 font compatibility problem

Description steve166 2011-06-30 09:11:31 UTC
Created attachment 7635 [details]
ps testcase for ps2write distiller 9.0 font compatibility problem

Hello,

I have found another font which, after converted with ps2write, does not work
with Adobe distiller 9. The output works with distiller 5 but not with distiller 9.0 (both using the same compatibility setting). With Distiller 9.0 i get a limitcheck error (see attached log). The font is a type42 font generated from a truetype font.
I use the following gs commandline:
gswin32c -q -dBATCH -dNOPAUSE -sDEVICE=ps2write -dCompressPages=false
-sOutputFile=out.ps -dCompressFonts=false -dASCII85EncodePages=false distiller_ps2write_font_test3.ps

Best regards,
steve166
Comment 1 Ken Sharp 2011-09-03 11:32:49 UTC
(In reply to comment #0)

> I have found another font which, after converted with ps2write, does not work
> with Adobe distiller 9. The output works with distiller 5 but not with
> distiller 9.0 (both using the same compatibility setting).

I'm not at all sure that this is a Ghostscript bug. The output file from ps2write works when sent back through Ghostscript, and also another PostScript consumer, and you mention that it also works with an older version of Distiller. This leads me to believe that there is nothing instrinsically wrong with the file.

Distiller does do unusual things (ie contrary to the PostScript specification), so it is possible that this file triggers one of those. However its almost impossible for me to investigate this, not having access to the source code for Acrobat Distiller.

I would suggest that you report this as a bug to Adobe, and if they can provide information that explains what is invalid about this file (or if you can deduce this from other sources) then I will be happy to look at this again. In this case please reopen this bug with the additional information.