Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!purdue!decwrl!hannah.dec.com!batcheldern
From: batcheldern@hannah.dec.com (Ned Batchelder, PostScript Eng.)
Newsgroups: comp.lang.postscript
Subject: Re: Compact data representations?
Message-ID: <8905021814.AA10980@decwrl.dec.com>
Date: 2 May 89 18:14:11 GMT
Organization: Digital Equipment Corporation
Lines: 24

In article <4876@pt.cs.cmu.edu> tsf@PROOF.ERGO.CS.CMU.EDU (Timothy Freeman)
writes:
>Well, you could use the same trick that the postscript text-printing
>programs use.  If a postscript program reads from its standard input,
>the text comes from the file that's being printed.  Your files could 
>look like this:
>
>/interpreter {...} def
>interpreter
>... encoded bytes ...

Please, please, remember to explicitly terminate your data.  Do not assume
that you can read to the end-of-file indicator on the standard input.  Not
only won't you be able to if your output is included as a figure, in
another document, but even as a stand-alone file, there may PostScript code
inserted at the end of the job by the spooler (to force out partially
completed sheets when doing 2-up, for example).  All you have to do is
reserve a code in your encoding that means end-of-data, and have the
interpreter function terminate when it reads that code.  It's very simple,
and the industry will thank you.

Ned Batchelder
Digital Equipment Corporation
BatchelderN@Hannah.DEC.com