Description of problem: Hmm - this latest version of ghostscript can no longer read WordPerfect8-generated postscript files which were generated after around the summer of 2004. These later files most noticeably now have the top line "12345X@PJL ENTER LANGUAGE = POSTSCRIPT". Everything was working fine on FC4. Version-Release number of selected component (if applicable): ghostscript-8.15.2-1.1 How reproducible: Every time. Steps to Reproduce: 1. gv <file> or lpr <file> 2. 3. Actual results: ESP Ghostscript 815.02: Unrecoverable error, exit code 1 Expected results: Additional info: These newly-become problem postscipt files were all generated with WordPerfect 8. Other files, mostly graphic image postscript files, and postscipt recently generated from pdf, still seem to work normally. A "problem-file" is attached.
This sounds like a PJL-encapsulated PostScript job, so perhaps something like this is what's needed: http://www.rz.uni-karlsruhe.de/~andy/source/Local/pjlstrip-1.0/pjlstrip.c I didn't think that ghostscript intentionally stripped PJL before. Could you attach the file in question please? Thanks.
Did you get the test file I sent? This is still a problem for me since I have to print reports every month with WP.
No, there is no attachment to this bug report.
Created attachment 143781 [details] Document1.ps [sent by email]
I just had a look at this and it seems that it's the line containing setpagedevice and PreferredMediaSelection that fails. If u remove "PreferredMediaSelection true" from line 392, the document is working.
Yes! Thanks for looking at that. That difference seems to be consistent across other working and non-working postscript files I have. There are two WordPerfect files which contain that term - err, which is not "PreferredMediaSelection", but correctly "DeferredMediaSelection": shlib10/hpla4pps.prs shlib10/wp60ps01.us.all This was not a change between FC4 and FC5, but instead, my starting to use an HP LaserJet 4 Plus around that time, which apparently makes use of the shlib10/hpla4pps.prs file. The remaining question would be: Is there anything wrong with the way ghostscript is interpreting that line, or is that line invalid postscript from WordPerfect?
This seems to work fine now in Fedora 8 (ghostscript-8.61-5.fc8).