Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!usc!sdd.hp.com!caen!uwm.edu!uwvax!astroatc!vidiot!brown
From: brown@vidiot.UUCP (Vidiot)
Newsgroups: comp.lang.postscript
Subject: Re: postscript accounting
Message-ID: <1509@vidiot.UUCP>
Date: 19 Mar 91 16:45:12 GMT
References: <YXOC.91Mar8143905@fb14vax.cs.uni-sb.de> <1475@vidiot.UUCP> <1111@eba.eb.ele.tue.nl>
Reply-To: brown@vidiot.UUCP (Vidiot)
Distribution: comp
Organization: Vidiot's Hangout
Lines: 44

In article <1111@eba.eb.ele.tue.nl> wjw@eba.eb.ele.tue.nl (Willem Jan Withagen) writes:
<
<Well here you've got a little problem. I'm planing to hook a HP-LJ II with
<PS-cartridge to an Apollo-workstation. And this through the parallel port.
<This'll get me some performance for those wanting to use PCL and bitmaps.
<
<This make it imposible to get info from the printer. I know that this is
<stupid, since the printer could return info on print jobs. But then this
<is the way it's going to be.
<
<Knowing this, is there an Unix accounting filter for PS.

Actually, you have the problem.  Without being able to get at the number of
pages printed you are not going to be able to determine how many pages were
printed.

Here is the reason why.  With PostScript and a single line of PostScript
code, thousands of pages could be printed and you would have a hard time
determining that.  The code could look at some specific value in the printer
and use that a seed for a for loop.  If you were to look at the code you
wouldn't have the slightest idea of how many pages were printed.  You just
can't count on one showpage printing a single page.

Worse yet, the showpage operator is part of another definition for printing
the page, which is standard for many PostScript prologues.

The cure is to have a full PostScript interpreter that actually runs the
PostScript code and keeps track of the number of pages printed, at least
what it thinks are printed.  This will really get your users mad because now
there is this delay in getting the printout because you have to run the job
twice; once to count the pages and second to have the printer do it all over
again.

The cure: make sure that a banner page is printed and have someone physically
count the pages and charge the department.  Not exactly what you were looking
for.  The other cure is just as bad...get a real PostScript printer that
uses the serial port so that you can ask it how many pages were printed.  A
printer with a TCP/IP connection should also handle questions.

The above is, of course, personal opinion.
-- 
      harvard\     att!nicmad\          spool.cs.wisc.edu!astroatc!vidiot!brown
Vidiot  ucbvax!uwvax..........!astroatc!vidiot!brown
      rutgers/  decvax!nicmad/ INTERNET:vidiot!brown%astroatc@spool.cs.wisc.edu