Skip site navigation (1)Skip section navigation (2)
Date:      Sun, 1 Oct 2006 15:12:57 -0700
From:      Gary Kline <kline@sage.thought.org>
To:        Ian Smith <smithi@nimnet.asn.au>
Cc:        Gary Kline <kline@sage.thought.org>, freebsd-questions@freebsd.org
Subject:   Re: triouble with my Deskjet 500
Message-ID:  <20061001221256.GA88740@thought.org>
In-Reply-To: <Pine.BSF.3.96.1061002021847.3783C-100000@gaia.nimnet.asn.au>
References:  <20061001012822.7EA2C16A591@hub.freebsd.org> <Pine.BSF.3.96.1061002021847.3783C-100000@gaia.nimnet.asn.au>

Next in thread | Previous in thread | Raw E-Mail | Index | Archive | Help
On Mon, Oct 02, 2006 at 02:49:56AM +1000, Ian Smith wrote:
> Re: freebsd-questions Digest, Vol 152, Issue 13
> On Sun, 1 Oct 2006 freebsd-questions-request@freebsd.org wrote:
>  > Message: 30
>  > Date: Sat, 30 Sep 2006 17:12:10 -0700
>  > From: Gary Kline <kline@sage.thought.org>
>  > Subject: triouble with my Deskjet 500
> 
> Hi Gary,
> 
>  > 	The trouble is that it only prints in ASCII Aand fails to fails
>  > 	to print xv images or anything else PostScript.  I'm playing
>  > 	around withthe following in /usr//local/libexec:
>  > 
	[[ ... ]]

>  > 	It's pretty obviously that I can toss the first several lines that
>  > 	came from the original hpif file; this was before I cared about
>  > 	graphics.  I lost the ghostscript part when my system had its 
>  > 	fatal trap.   Anybody out there who has an ancient hp djet500?
>  > 	or can help otherwise.  I'm out of ideas.
> 
> I ran into exactly this after upgrading 4.5-R to 5.4-R (now 5.5-S); that
> fancy stderr/stdout dance doesn't seem to work &/| be needed anymore.  I
> got sick of Mozilla crashing trying to print and dug up this fix, forget
> where/how.  Way over commented, but times like this it comes in handy :) 
> 
> #!/bin/sh
> #% /usr/local/libexec/if-lq850 smithi 6/4/3
> #% v2 29/6/6, maybe the 3>&1 stuff is what doesn't work in 5.4-R?
> #% as advised by FreeBSD handbook, from:
>     #  ifhp - Print Ghostscript-simulated PostScript on a DeskJet 500
>     #  as installed in /usr/local/libexec/hpif
> #% modified for printer canon bj10sx (configured as an epson lq850)
> 

	After fmessing around for an hour I found the bug that was
	introduced by the new/"improved"/(free) CUPS' lp*.  The *only*
	script change [[ at least that I can see ]] is that 
	"sOutputFile=" now fails to understand /dev/fd/3.  I got my old 
	"hpif working with only-ASCII and your version working with 
	only-[Gh|P]ostscript, or mostly since your file did print some
	ASCII.  But for me, it hung for some timeout or <???>.

	I dropped in your Postscript working lines, minus the
	re-direction stuff and voila!  Yes for good old fashioned ASCII 
	and Postscript both.

	Appended if my new hpif.   (I have four servers with
	/usr/local/libexec/hpif, so
	name-change rather than change four /etc/printcaps:-).

	thankee, thankee!


	gary

	PS: file appended. note lines 39-46... .

	PPS:  comments  appreciated; it's hard to *overcomment*, IMHO.


-- 
   Gary Kline     kline@thought.org   www.thought.org     Public service Unix


     1	#!/bin/sh 
     2	#
     3	# hpif - Simple text input filter for lpd for HP-PCL based printers
     4	# Installed in /usr/local/libexec/hpif
     5	#
     6	# Simply copies stdin to stdout.  Ignores all filter arguments.
     7	# Tells printer to treat LF as CR+LF. Writes a form feed character
     8	# after printing job.
     9	
    10	###printf "\033&k2G" && cat && printf "\f" && exit 0
    11	
    12	#
    13	#  ifhp - Print Ghostscript-simulated PostScript on a DeskJet 500
    14	#  Installed in /usr/local/libexec/ifhp
    15	
    16	#
    17	#  Treat LF as CR+LF:
    18	#
    19	printf "\033&k2G" || exit 2
    20	
    21	#
    22	#  Read first two characters of the file
    23	#
    24	IFS="" read -r first_line
    25	first_two_chars=`expr "$first_line" : '\(..\)'`
    26	
    27	if [ "$first_two_chars" = "%!" ]; then
    28	    #
    29	    #  It is PostScript; use Ghostscript to scan-convert and print it.
    30	    #
    31	    #  Note that PostScript files are actually interpreted programs,
    32	    #  and those programs are allowed to write to stdout, which will
    33	    #  mess up the printed output.  So, we redirect stdout to stderr
    34	    #  and then make descriptor 3 go to stdout, and have Ghostscript
    35	    #  write its output there.  Exercise for the clever reader:
    36	    #  capture the stderr output from Ghostscript and mail it back to
    37	    #  the user originating the print job.
    38	    #
    39	    ###exec 3>&1 1>&2
    40	    ###/usr/local/bin/gs -dSAFER -dNOPAUSE -q -sDEVICE=djet500 \
    41	        ###-sOutputFile=/dev/fd/3 - && exit 0
    42	
    43	
    44	
    45	    /usr/local/bin/gs -dSAFER -dNOPAUSE -q -sDEVICE=djet500 \
    46	        -sOutputFile=- - && exit 0
    47	
    48	else
    49	    #
    50	    #  Plain text or HP/PCL, so just print it directly; print a form feed
    51	    #  at the end to eject the last page.
    52	    #
    53	    echo "$first_line" && cat && printf "\033&l0H" && 
    54	exit 0
    55	fi
    56	
    57	exit 2



Want to link to this message? Use this URL: <http://docs.FreeBSD.org/cgi/mid.cgi?20061001221256.GA88740>