[reportlab-users] problem with pdf2ps

[David Hughes]

Last November, Jade Meskill wrote:

> I am having trouble with some of my ReportLab generated PDFs.  When I 
> convert them to postscript using pdf2ps i get the following error:
> 
> Error: /ioerror in --token--
  ...
> GNU Ghostscript 7.05: Unrecoverable error, exit code 1
> 
> This is only happening in a few of my documents, and I can't seem to 
> find a common thread between them.  If I open them up in Preview on Mac 
> OS X or in Acrobat Reader I can print it just fine, but printing using 
> lpr on linux causes invalid postscript to be sent to the printer.  Can 
> someone give me a hint as to where I can look to see what is causing 
> this error?

It's a long time ago, but there was no follow-up solution so you might 
still like to know that this is probably caused by the character pair ~> 
becoming split by a <CR><LF> in the pdf file, which produces the same 
intermittent error when I convert pdf to png using GNU Ghostscript 7.06.
Strictly this is not allowed, although Acrobat reader accepts it and, now, 
so does Ghostscript 8.1x but I don't think it has filtered down to GNU 
Ghostscript yet. I guess an interim solution would be to pipe the pdf 
through a Python script to strip out the offending characters.

Regards,

David Hughes
Forestfield Software Ltd
www.forestfield.co.uk