[443] in Public-Access_Computer_Systems_Forum
PostScript translation?
daemon@ATHENA.MIT.EDU (Ernest Perez)
Mon Jun 8 12:24:59 1992
Date: Mon, 8 Jun 1992 10:44:51 CDT
Reply-To: Public-Access Computer Systems Forum <PACS-L%UHUPVM1.BITNET@RICEVM1.RICE.EDU>
From: Ernest Perez <EPEREZ%UTDALLAS.BITNET@RICEVM1.RICE.EDU>
To: Multiple recipients of list PACS-L <PACS-L%UHUPVM1.BITNET@RICEVM1.RICE.EDU>
----------------------------Original message----------------------------
There's been many questions and comments recently about useful files
that seem to be only available in Postscript format.
I sent a message dealing with this to an Internet-acquaintance of
mine, Mark Emmer, of Catspaw, Inc. Mark is the author and publisher of
multiple versions of SNOBOL4+, SPITBOL, and ICON, which are all
powerful text- and string-processing languages. His SPITBOL, for
instance, is available in 386, Unix workstations, and MacIntosh
versions.
I asked Mark if he happened to have done a utility program that would
convert Postscript to ASCII. Just figured he might have created one in
the past, since I'd seen a "Mac text --> MS-DOS ASCII" utility that he
wrote in SNOBOL. (It was only about 5 or 10 lines long, but it did the
job.)
His reply disabused me of the idea that it would be a similarly easy
task. I thought list members might find his comments to be informative...
====================
>Regarding Postscript to ASCII converter, that's an intriguing idea.
>I don't have such a program, and it's devilishly difficult. Here's
>why:
>
>If you're not familiar with the details of Postscript, it's logical
>to mistake it for another "standard format", like Rich Text Format,
>Wordstar, Word, etc. But it's not.
>
>Postscript is a _programming language_. When you look at a Postscript
>file, the first N lines (usually 100 - 1000 lines) are canned function
>definitions placed there by whoever created the file. Every word
>processor & desktop publisher uses their own unique functions & macros.
>What they do is define their own "language" in terms of Postscript.
>
>Following this header is a program written in terms of these functions
>and macros to compose each page. That is, there are a series of commands
>to select fonts, position around the page, spew out text, and draw
>graphics. The Postscript interpreter in the printer executes these
>commands, most of which are calls to the macros & functions defined
>earlier.
>
>Now, there's nothing that says the order of stuff in the Postscript
>file is linear -- left to right or top to bottom. You can sprinkle
>stuff around the page in any order, and finally do a "showpage" when
>you're done.
>
>I have friends who use desktop publishing programs to create newsletters,
>etc., and then bring the disk file over here for me to copy to my
>Postscript printer. Invariably, we find typos and mispositioned stuff
>when it's printed, and I try to edit the Postscript to fix their problems.
>Some are fairly straight-forward, many are a nightmare. I have to
>eyeball the function definitions, trying to figure out what they do,
>and then try to patch X,Y coordinates, etc.
>
>In order to do what you're asking, you'd have to create a mini-Postscript
>interpreter to "execute" the file, and then try to recreate the ASCII.
>It's an interesting concept, but nothing I'm going to tackle!
>
>s/Mark Emmer
======================
Mark's explanation certainly ties in with Andy Boze's recent message
about the GhostScript (GS) Postscript *interpreter*. If you noticed,
Andy mentioned that it is huge, taking up about 4.5 meg of disk space.
That space might be worth it, if you need the ASCII translations badly
enough.
..............................
Ernest Perez, Ph.D
Access Information Associates
2183 Buckingham, Suite 106
Richardson TX 75081
214-530-4800
INTERNET: eperez@utdallas.edu
BITNET: eperez@utdallas
..............................
