🚀 go-pugleaf

	Thank you to the people who responded to my problem with the
LN03R Scriptprinter.  George Robbins of Commodore-Amiga solved it;
it was /usr/lib/lpf that was truncating lines, not the printer itself.

#define EAT_OWN_WORDS 1

	I told a few netters that the LN03R was at fault, and that it
was not capable of printing Postscript files with lines longer than 80
or 132 characters (depending on the phase of the moon).  I WAS WRONG.
THE PROBLEM WAS DUE TO OUR PRINT FILTER.
	Then again, the documentation on the "lpr" system is not
the world's greatest piece of literature either.

#undef EAT_OWN_WORDS

	Here is a working /etc/printcap entry that we are now using for our
Scriptprinter:

	ln03r|lnps|ps|ln03R postscript printer on a DECserver 200:\
		:br#9600:\
		:ts=CS2ATS:\
		:lp=/dev/tty14:\
		:df=/usr/local/bin/dvi2psfilter:\
		:of=/usr/lib/lpf:\
		:if=/usr/local/filters/psfilter:\
		:sh:\
		:op=PORT_2:\
		:sd=/usr/spool/ln03r:\
		:lf=/usr/spool/ln03r/messages:

The trick was to define an "accounting filter" with the "if" parameter.
/usr/local/filters/psfilter is nothing but this:

		main()
		{
			register char c;
			while ((c=getchar()) != EOF)
				putchar(c);
		}

Strangely enough, /bin/cat would not work as the accounting filter,
as George suggested it would.  Looking at the source code to /bin/cat
on our ULTRIX machine, I'm not surprised.  What a WEIRD piece of code!


                                                        Dan

 //////////////////////////////////////\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\
| Dan Barrett, Systems Administrator  --  barrett@cs.jhu.edu (128.220.13.4) |
| Dept. of Computer Science, Johns Hopkins University, Baltimore, MD  21218 |
| E-mail addresses:  ARPANET: barrett@cs.jhu.edu                            |
|                    BITNET:  ins_adjb@jhuvms.bitnet                        |
|                    UUCP:    barrett@jhunix.UUCP                      noog |
 \\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\/////////////////////////////////////