Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!uwm.edu!bionet!agate!usenet
From: raymond@math.berkeley.edu (Raymond Chen)
Newsgroups: comp.lang.postscript
Subject: Re: PS->EPS filter
Message-ID: <1991Feb19.011453.5797@agate.berkeley.edu>
Date: 19 Feb 91 01:14:53 GMT
References: <KRISHNA.91Feb18182501@bucasd.bu.edu>
Sender: usenet@agate.berkeley.edu (USENET Administrator)
Reply-To: raymond@math.berkeley.edu (Raymond Chen)
Organization: U.C. Berkeley
Lines: 28
In-Reply-To: krishna@bucasd.bu.edu (Krishna Govindarajan)

In article <KRISHNA.91Feb18182501@bucasd.bu.edu>, krishna@bucasd (Krishna Govindarajan) writes:
>Does anyone know of a filter that will convert Postscript to
>Encapsulated Postscript ?

Cannot be done in general.  Consider:

I write a Postscript program that performs several CPU hours of numerical
analysis, and outputs either

[1] nothing, if it failed to find a solution.
[2] one page of output containing the solution it found.

The only way to figure out what value to put in the %%Pages field in
the EPS header is by actually executing the program and seeing what
happens.

(A more mundane example is a PS program that violates one of the contraints
of EPS.  For example, if it redefines macros in midstream, or downloads
a font in the middle of the document, thereby ruining any possibility
of performing page-reversal.  Some dvi-to-Postscript converters are guilty
of the latter.)

Remember, Encapsulated Postscript is a restricted form of Postscript.

My favorite analogy is viewing Postscript as the English language, and
Encapsulated Postscript as a restricted form of the English language
where you are only allowed to use words that appear on a list of 10,000
`permitted' words.