Trying to process a PDF file fails on Linux with "Error: /syntaxerror in pdfopen". Same file processes fine on the Windows XP platform. Command line is: gs -dNOPAUSE -dBATCH -r72 -sDEVICE=png16m -sOutputFile=test-%03d.png temp_48110.pdf
Created attachment 5303 [details] temp_48110.pdf Sample file
This is a regression, gs8.64 processes the file without error. The responsible change: r9826 | alexcher | 2009-06-29 17:06:33 -0700 (Mon, 29 Jun 2009) | 6 lines Extract all documents from PDF collection to temporary files and process them in the order they are listed in the /Names array. Bug 690422, customer 531.
Unfortunately, I don't know how to tell PDF collections from ordinary PDF files with attachments. One possible approach is to test whether the attached file is PDF by searching for the PDF header.
I misunderstood the spec. Collection attribute is marked in the spec as optional because a PDF file need not to be a portable collection but every collection has /Collection attribute. So this patch processes embedded files as a portable collection only when /Collection attribute is defined in the document root and ignores them otherwise. The following patch has been committed as a rev. 10008. http://ghostscript.com/pipermail/gs-cvs/2009-August/009714.html Regression testing shows no differences. Regarding Windows vs Linux differences, the patch is platform-independent and the bug shows up on Windows too. Most likely, you are using an old version of Ghostscript on Windows.
re: Windows vs. Linux You are correct, Windows version was 8.64, Linux was 8.70.