Using TI printer on Symbolics

[David Gadbois <AI.Gadbois@MCC.COM>]

    Date: Fri, 8 Sep 89 12:48 PDT
    From: Chucko@VERMITHRAX.SCH.Symbolics.COM (Charles R. Fry)

	Date: Tue, 5 Sep 89 08:20 EST
	From: PAT%atc.bendix.com@RELAY.CS.NET

	We have a 3650 and a TI OMNILASER 2108 and have been trying to get them
	to work together for a long time.  When I try to hardcopy to the printer
	I get the error

	  error : printer did not respond to an EOF marker

		  It was expected to return an acknowledging EOF marker.


	Symbolics doesn't have an idea whats going on.  Has anyone ever used
	a 2108 with a 3650?  If anyone can give me help I'd appreciate it.

	Thanks,

	Patrick Coleman
	pat@atc.bendix.com

    Try the following patch file, lifted from the development system.  I have not
    tested this myself, so please let me know how well it works.

     -- Chuck Fry
	Symbolics Software Support

    ;;; -*- Mode: LISP; Syntax: Common-Lisp; Package: USER; Base: 10; Patch-File: T -*-
    ;;; Patch file for Private version 0.0
    ;;; Reason: Function (FLAVOR:METHOD LGP::GET-POSTSCRIPT-VERSION LGP::LGP2-SERIAL-STREAM):  Devo version.
    ;;; Written by Chucko, 9/08/89 11:27:12
    ;;; while running on EUREKA from FEP0:>Inc-Statice-from-WSC-Base.ilod.1
    ;;; with Genera 7.4 Ivory, System 418.177, Logical Pathnames Translation Files NEWEST,
    ;;; RPC 407.10, MacIvory Support 412.40, Serial 402.15, Utilities 412.26,
    ;;; Server Utilities 410.2, Hardcopy 413.7, Zmail 410.8, LMFS 411.10, Tape 412.13,
    ;;; Nsage 407.7, RPC Development 401.4, MacIvory Development 401.5,
    ;;; Documentation Database 412.5, IP-TCP 413.3, NFS 405.0, DNA 407.0,
    ;;; Symbolics In-House 408.17, Statice 403.0, Unique-ID 402.0, UFS 404.3, DBFS 403.8,
    ;;; DBFS Utilities 405.0, Statice-Index 402.2, Statice-Record 401.0,
    ;;; Statice-Model 403.2, Experimental Statice Examples NEWEST, DBFS-DIR 401.1,
    ;;; Statice-Utilities 402.2, Tertiary Storage 405.2, DBFS Maintenance 405.3,
    ;;; Volume Librarian 403.1, Ivory Revision 2 (FPA enabled), FEP 311,
    ;;; FEP0:>I311-loaders.flod(14), FEP0:>I311-debug.flod(20), FEP0:>I311-lisp.flod(14),
    ;;; FEP0:>I311-info.flod(14), FEP0:>I311-kernel.fep(18), Boot PROM version 304,
    ;;; 1067x748 B&W Screen, Machine serial number 319,
    ;;; Default Dired spec to *.*.newest (from VTX:>chucko>hacks>zmacs-hacks-7-2.lisp.9),
    ;;; Provide font for FIX.CONDENSED.TINY (from VTX:>chucko>hacks>fix-for-brad-miller's-bug-reports.lisp.5),
    ;;; Smaller menus for Zmail sequences (from VTX:>chucko>hacks>smaller-zmail-menus.lisp.1).


    (NOTE-PRIVATE-PATCH "Fix for \"Printer xxx on host yyy did not respond to an EOF marker\"?")


    ;=====================================
    (SYSTEM-INTERNALS:BEGIN-PATCH-SECTION)
    ; From buffer postscript.lisp >sys>hardcopy Q: (1668)
    (SYSTEM-INTERNALS:PATCH-SECTION-ATTRIBUTES
      "-*- Syntax: Zetalisp; Mode: LISP; Base: 10;  Package: LGP; Lowercase: Yes -*-")

    (defmethod1 (get-postscript-version lgp2-serial-stream)0 ()
      (let ((string (response-string-from-command self "version =" "Read PostScript Version")))
	(ignore-errors
	  (scl:with-standard-io-environment
	    (let ((si:*suppress-read-eval* t))
	      (read-from-string string))))))


I think that this is only a partial solution to the problem.  Timeouts
(caused by who knows what) can occur at several points in the
communication with the LGP.  In particular, you have to make a similar
patch to (FLAVOR:METHOD :SYNC LGP::LGP2-SERIAL-STREAM).  This involves
ignoring a timeout error from the function LGP::WAIT-FOR-LGP2-EOF.  (The
restart handler "Retry synchronization handshake with LGP2 printer ~A"
doesn't work anyway.)

I am very curious about the binding of SI:*SUPPRESS-READ-EVAL*.  Surely
my LasewrWriter isn't sending any letter bombs over the serial line?

--David Gadbois