Customer reports and I've verified that Ghostscript produces an error, "/undefined in --get--", when reading the attached PostScript file. The command line I used: bin/gs -sDEVICE=ppmraw -sOutputFile=test.ppm ./Example.ps The error is the same with gs8.50 through gshead (r8044). Adobe Photoshop and Mac OS X Preview both open the file without error.
Created attachment 3014 [details] Example.ps
The sample file includes the following fragment without a stopped context or any other means to avoid errors. currentsystemparams /MaxImageBuffer get 4194304 lt { << /MaxImageBuffer 4194304 >> setsystemparams } if currentsystemparams /JobTimeout get 0 ne { << /JobTimeout 0 >> setsystemparams } if currentuserparams /MaxSuperScreen get 4194304 lt { << /MaxSuperScreen 4194304 >> setuserparams } if We don't have /MaxImageBuffer or /JobTimeout parameters.
*** Bug 689287 has been marked as a duplicate of this bug. ***
The problem is that the PPD used to make this file on the Mac with the LaserWriter 8 driver appears to be flawed because it does not protect some of the "Feature" setup by putting it in a 'stopped' context as is customary with PPD options. The offending section is: %%BeginFeature: *JobPatchFile 0 currentsystemparams/MaxImageBuffer get 4194304 lt{ << /MaxImageBuffer 4194304 >> setsystemparams}if currentsystemparams/JobTimeout get 0 ne{ << /JobTimeout 0 >> setsystemparams}if currentuserparams/MaxSuperScreen get 4194304 lt{ << /MaxSuperScreen 4194304 >> setuserparams}if %%EndFeature This gets an error on Ghostscript since we don't have the system param "MaxImageBuffer", thus throwing an 'undefined' error. I recommend that we investigate Adobe implementations and add any systemparams that are 'standard' in Adobe RIP products and distiller.
Created attachment 3041 [details] Acrobat7_systemparams.pbm
Created attachment 3042 [details] CPSI_systemparams.png
Created attachment 3043 [details] Tek360_systemparams.png
Created attachment 3181 [details] patch Add dummy system parameters for compatibility with Adobe products.
The patch is committed as a rev. 8119. There's no differences in Comparefiles collection. Cases 8-4-3, 27-4-1, 27-09-2 of the CET suite shows differences in the files that enumerate or test system parameters. Cases 27-7c-2, 29-7c-4 depend on the directory enumeration order in the page device processing. This is a bug in the page device code.