Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!iuvax!uxc.cso.uiuc.edu!csd4.milw.wisc.edu!bionet!ames!haven!umd5!zben From: zben@umd5.umd.edu (Ben Cranston) Newsgroups: comp.sys.mac Subject: Re: how to print a Postscript file from a Mac on a Laserwriter? Summary: Possible bug in Apple Macintosh 6.0 prolog Keywords: postscript, Macintosh, laser printer Message-ID: <5114@umd5.umd.edu> Date: 20 Jul 89 01:21:08 GMT References: <2215@cerberus.oakhill.UUCP> <26426@agate.BERKELEY.EDU> <1001@adobe.UUCP> Reply-To: zben@umd5.umd.edu (Ben Cranston) Distribution: usa Organization: University of Maryland, College Park Lines: 41 Expires: The good news is that in 6.0 they changed the flushfile into an explicit search for an explicit %endeexec marker. This keeps the job from being flushed, but does not address the problem of the printer's password being changed and there being password-dependant code in the prolog, for example: ...{statusdict begin 9 setsccinteractive 3 ne exch 0 ne or { 9 0 3 setsccinteractive }if Not to mention what mucking around with the scc parameters might do to the communications path for a spooled printer... Anyway, I wanted to bring to Apple's (and other people's) attention what looks like a bug in the 6.0 prolog. In the "open document" procedure "od" we find this code (indentation added :-): statusdict /processcolors known { %ifelse statusdict begin processcolors end dup 4 eq { %ifelse /4colors true def pop }{ 3 eq { %ifelse /3colors true def }{ /2color true def .................................^^^^^^ } ifelse } ifelse }{ /2colors true def .................^^^^^^^ } ifelse } bdf One doesn't have to know the prolog in great detail to know that one of these variable names is spelled wrong. Then again, a grep through the prolog shows nobody even USES these flags. Maybe some application will output them... Caveat Coder... -- Sig DS.L ('ZBen') ; Ben Cranston * Computer Science Center Network Infrastructures Group * University of Maryland at College Park