Subject: TeXhax Digest V90 #42
From: TeXhax Digest <TeXhax@cs.washington.edu>
Errors-To: TeXhax-request@cs.washington.edu
Maint-Path: TeXhax-request@cs.washington.edu
To: TeXhax-Distribution-List:;
Reply-To: TeXhax@cs.washington.edu

TeXhax Digest    Sunday,  April 29, 1990  Volume 90 : Issue 42

Moderators: Tiina Modisett and Pierre MacKay

%%% The TeXhax digest is brought to you as a service of the TeX Users Group %%%
%%%       in cooperation with the UnixTeX distribution service at the       %%%
%%%                      University of Washington                           %%%

Today's Topics:         

     Free dvi2ps running on MS-DOS using internal printer fonts?
                                  new web2c
                                Halmos symbol
               Re: TeX control words soak up spaces (V90 #39)
                            PostScript -> TeX ?
                             Memory limitations
                     Peter Flynn's TeXcalc spreadsheet
                                Page breaking
                         solutions to dvitovdu woes
            Re: TeXhax Digest V90 #38 (1799@shelby.Stanford.EDU)
                   Problem with TeXtyl on VMS and lasy fonts
                          Bug in Latex not corrected

-----------------------------------------------------------------------------

Date: Sat, 21 Apr 90 15:37:51 EDT
From: weil%husc4@harvard.harvard.edu (Philippe Weil)
Subject: Free dvi2ps running on MS-DOS using internal printer fonts?
Keywords: dviware

Does anyone know of a (free) DVI2PS running on MSDOS which uses 
*internal* printer fonts?  Where can the executables be found?  
 
 
Philippe Weil    

-----------------------------------------------------------------------------

Date: Sat, 21 Apr 90 07:43:48 EDT
From: Karl Berry <karl@aten.umb.edu>
Subject: new web2c
Keywrods: web2c

web2c version 5.0c is on ics.uci.edu.
It fixes a couple important bugs, and also some minor ones.
Have fun.

karl@cs.umb.edu

------------------------------------------------------------------------------

Date:  Wed, 18 Apr 90 08:51:34 EST
From: Jim Hefferon <HEFFERON%UCONNVM@UWAVM.ACS.WASHINGTON.EDU>
Subject: Halmos symbol
Keywords: LaTeX, Halmos symbol

I want to put a rectangular box at the end of proof in LaTeX.  I've tried
to define a new environment \begin{proof} ... \end{proof}, using the
following definition I've found in an old TeXhax:
  \def\qed{\unskip\nobreak\hfill\qedbox}
which (as I understand it) (i) eats spaces (ii) eats glue (iii) moves to
the right margin and (iv) writes the blob of ink.
  The problem is that when my proof ends in, say, ...\end{eqnarray*}\end{proof}
then the qedbox is put a line below the displayed equation.  That's not the
format I want (I'm basically stealing my format from Springer, so I think what
I want is correct in design terms).  Can I move back into the equation, or
is it boxed up forever?
  jim

----------------------------------------------------------------------------

Date: Tue, 17 Apr 90 14:55:07 cdt
From: mark@morgana.pubserv.com (inthe heat ofthe sun)
Subject:  Re: TeX control words soak up spaces (V90 #39)
Keywords: TeX, control words

  In TeXhax V90 #39, Philip Taylor presented a rather lengthy solution
to the problems of getting TeX to recognize the space after a control
sequence in text.  Mr. Taylor's solution involved changing the category
codes for "<" and ">" then coding as <key-word>.  I offer, as a shorter
alternative, a macro called \obeynexttoken. (in the spirit of \obeylines
and \obeyspaces).

\def\ONTcheck{\ifcat\next\space\let\next=\space\fi\aftergroup\next}
\def\obeynexttoken{\futurelet\next\ONTcheck}

   \obeynexttoken should be coded at the end of the replacement text
of the macro which may fall in text.  As an example,

	   def\keyword#1{<macro stuff here>\obeynexttoken}

then, when \keyword{} is called within text, the following space (or
punctuation mark, or control sequence) will be honored.  It will work,
too, in cases where the next token following \keyword{} is a \par.

   This method works for all macros which take an argument.  If the
contol sequence does not require an argument (i.e. \TeX), the
\obeynexttoken macro will fail to recognize spaces---punctuation marks
and other control sequences will still be recognized.  This happens
because the space following \TeX is abosorbed and lost by TeX when it
is reading the macro name. (Perhaps someone has a solution to this
problem?)

					Mark Moline
					Publication Services, Inc.
					mark@morgana.pubserv.com

-----------------------------------------------------------------------------

Date: Tue, 17 Apr 90 10:35 CDT
From: Ray 'StarSailer' Howard <C136ERH@UTARLG.ARL.UTEXAS.EDU>
Subject: PostScript -> TeX ?
Keywords: PostScript, TeX

    I have an interesting problem.  Some documentation that our site
    receives is in Postscript format.   We would like to be able to
    convert them into a format that can be edited to show site
    specific features.  The main thing that we want to reuse is the
    text portions, we can live with having to redo some of the
    figures/graphics.  Is there a Postscript converter/interpreter
    that will (or can be made to) output in TeX, ASCII or other
    format?  Any suggestions on writing such a program? 

    Thanks, Ray.

Ray Howard                                        //
Bitnet:  c136erh@utarlg                         \X/  #include <std.disclaimers>
Internet: c136erh@utarlg.arl.utexas.edu
UUCP:  ...!{ames,sun,texbell,<vertebra>}!utarlg.arl.utexas.edu!c136erh 

-------------------------------------------------------------------------------

Date: Tue, 17 Apr 90 13:19:34 MDT
From: najjar@purcell.CS.ColoState.EDU (walid najjar)
Subject: Memory limitations
Keywords: TeX, memory limitations

We are running TeX C version 2.93.  every so often TeX runs out of
memory when processing files that contain plots (done with epic.sty).
The behavior is not very predictable. Some files with more than 5
plots will compile while others with fewer and smaller plots do not.

Anyone has some insight on why TeX would run out of memory. I have not found
anything odd in these plots yet. The behavior is repeatable: i.e. a file that
does not compile will not, but the processing would stop at different points
in different runs.

Walid Najjar

-------------------------------------------------------------------------------

Date: Tue, 17 Apr 90 22:35 MET
From: "Johannes L. Braams" <JL_Braams%pttrnl.nl@UWAVM.ACS.WASHINGTON.EDU>
Subject: Peter Flynn's TeXcalc spreadsheet
Keywords: TeXcalc


    Hi all,

        Some of you may know Peter Flynn's TeXcalc spreadsheet,
        some of you might not. It's a wonderfull tool to help you
        in computing typographic dimensions. The orignal was written
        for InstaCalc(tm) from FormalSoft according to the documentation.
        As I do not hav that product I ported it into Lotus release 2
        (for my PC at home) and 2020 (for the VAX at work). I've sent
        UUencoded versions of both spreadsheets to Peter, if anyone
        wants to put it in one of the servers I will be happy to
        send him/her the files too.
        If you don't want to wait for that to happen, you might also
        try to ask me, but I would prefer it if they would become
        available on the servers where the original resides.

    Regards,

        Johannes Braams

PTT Research Neher Laboratorium,        P.O. box 421,
2260 AK Leidschendam,                   The Netherlands.
Phone    : +31 70 3325051               E-mail : JL_Braams@pttrnl.nl

E-mail was :
    EARN/BITnet : BRAAMS@HLSDNL5   UUCP        : hp4nl!dnlunx!johannes
    SURFnet     : DNLTS::BRAAMS    INTERnet    : BRAAMS%HLSDNL5@CUNYVM.cuny.edu
 PSS (DATAnet1) : +204 1170358::BRAAMS

-----------------------------------------------------------------------------

Date: Tue, 17 Apr 90 14:27:52 PDT
From: wagman@Csa1.LBL.Gov (Gary S. Wagman (415)486-6610)
Subject: Page breaking
Keywords: TeX, page break

I have a question about pagination. I have several documents over 20 pages in
length in which many \penalty's specify clues to TeX's page breaking algorithm.
Most of the time the page breaks are acceptable, but occasionally I manually
insert \break's or \nobreak's to make the final document perfect.  That's OK.

However, then I must run TeX on the same input document specifying a different
set of fonts that I tweaked from CMSS with METAFONT. The only difference
between my fonts and CMSS are that I redesigned the sixes and nines so the
tails don't curve so much, increased the size of dots (periods, commas, i-dots,
etc.), and increased the x-height. I swear to you that to the best of my
knowledge (hmmm...) none of the changes increased the vertical space
requirements beyond that of the baselineskip or struts. On the second pass, TeX
repaginated the document and breaks pages at different locations than in the
first pass.

WHY, WHY, WHY? I want identical without re-proofreading. (Humpf!)


Gary Wagman
Lawrence Berkeley Lab
LBL::Wagman
Wagman at LBL
Wagman@LBL.Gov

---------------------------------------------------------------------------

Date: Tue, 17 Apr 90 16:06:59 CDT
From: "Lee Schneider" <MATHPG2%UMCVMB.MISSOURI.EDU@UWAVM.ACS.WASHINGTON.EDU>
Subject: solutions to dvitovdu woes
Keywords: dvitovdu

Several weeks ago I sent a query to TeXhax regarding the following problem:
I wanted to use the dvitovdu program archived at june.cs.washington.edu,
but it came as Modula-2 sources, and I only had access to Pascal and
C compilers.  The following is a summary of what I learned:

1) A Pascal version of dvitovdu is archived at ymir.claremont.edu.  However,
   this version is said to contain code specific to Pyramid computers.
   I did not try to get this version running.

2) A Pascal to C conversion program "p2c" available by ftp from
   csvax.caltech.edu, may be able to translate Modula-2 to C.
   I haven't verified this, because it turned out that I didn't need to.

3) There is a C version of dvitovdu in the Aston TeX archive.  I haven't
   figured out how to access this archive, but someone sent me a copy
   of the dvitovdu files.  After a making a few modifications to support the
   PK-file naming scheme used by the Arbortext version of TeX, and to support
   MS-Kermit, which is a Textronix emulator we use here, we have dvitovdu
   up and running!

Several people have written me asking me to send a copy of the C-sources
when and if I found them.  I have sent them out to a few of those people.
However, I lost the names of a few others who also requested them.  I am
willing to send additional copies out occasionally;  However, I don't want
to have to spend a lot of time doing this.  Therefore, if any site wants
to volunteer to archive the C version of dvitovdu, I will be more than happy
to send a copy to that site.

Thanks to all who helped,

Lee Schneider
Department of Mathematics, University of Missouri-Columbia
MATHPG2@UMCVMB.BITNET or MATHPG2@UMCVMB.MISSOURI.EDU

---------------------------------------------------------------------------

Date: Sat, 14 Apr 90 12:34:26 EST
From: wilker@hopf.math.purdue.edu (Clarence Wilkerson)
Subject: Re: TeXhax Digest V90 #38 (1799@shelby.Stanford.EDU)
Keywords: dviware

I don't know how much hacking you want to do, but one place
to start is a source code for a dvi to dot matrix printer,
such as found in the Beebe archives at SCIENCE.UTAH.EDU. 
Most dot matrix printer drivers do prepare a bit-mapped image
of the page, while any reasonable laser driver downloads fonts
to the printer.
Clarence Wilkerson

-----------------------------------------------------------------------------

Date: Fri, 20 Apr 90 17:51 MET
From: Johannes Braams <JL_Braams%pttrnl.nl@UWAVM.ACS.WASHINGTON.EDU>
Subject: Problem with TeXtyl on VMS and lasy fonts
Keywords: TeXtyl V 1.30/VMS


    Hi all,

        We are using TeXtyl V 1.30/VMS. Today we discovered a problem with
        it. When a document uses one of the lasy fonts (lasy10 in this case)
        TeXtyl crashes with an access violation. The traceback information
        tells me it occurs in the routine INTFM, which reads the .tfm-file.

        Is there anybody who knows a solution to this problem? We have a
        workaround, because a symbol simalar to the one needed (\Box)
        exists in one of the ams-fonts (\square). Still we would very
        much appreciate a fix for this.

        Please respond to me directly, and I'll summarize for the list
        when I get a solution.

    Thanx in advance,

        Johannes Braams

PTT Research Neher Laboratorium,        P.O. box 421,
2260 AK Leidschendam,                   The Netherlands.
Phone    : +31 70 3325051               E-mail : JL_Braams@pttrnl.nl

E-mail was :
    EARN/BITnet : BRAAMS@HLSDNL5   UUCP        : hp4nl!dnlunx!johannes
    SURFnet     : DNLTS::BRAAMS    INTERnet    : BRAAMS%HLSDNL5@CUNYVM.cuny.edu
 PSS (DATAnet1) : +204 1170358::BRAAMS

-----------------------------------------------------------------------------

Date: Fri, 20 Apr 90 22:50:04 met
From: Piet van Oostrum <piet@cs.ruu.nl>
Subject: Bug in Latex not corrected
Keywords: LaTeX, bug

This is from latex.bug:

138.  A command like \index or \label could incorrectly suppress a
space after the next \end command.  (Reported by Johannes Braams.
Partially fixed on 30 Nov 88.  Problem can still occur if \index or
\label command comes inside the \end's environment.)

Well, it appears not to be fixed at all (Except in the comments).

This is from latex.tex as it appears on labrea: (Dated 7 Dec 1989)

% \begin{NAME} ==
%  BEGIN
%    IF \NAME undefined  THEN  \@tempa == BEGIN report error END
%                        ELSE  \@tempa ==  (\@currenvir :=L NAME) \NAME
%    FI
%    @ignore :=G F      %% Added 30 Nov 88
%    \begingroup            
%    \@currenvir :=L NAME
%    \NAME
%  END


\def\begin#1{\@ifundefined{#1}{\def\@tempa{\@latexerr{Environment #1
  undefined}\@eha}}{\def\@tempa{\def\@currenvir{#1}%
  \csname #1\endcsname}}\begingroup\@endpefalse\@tempa}

                       ^^ No \global\@ignoretrue

Now the bug report mentions that this will not solve all problems. There is
only one good solution for it:

Have two versions of the \@esphack macro, one with \@ignoretrue (to be used
in the end of an environment) and one without (to be used in simple
commands). As the second one is used the most, the first one should get
another name. It is currently used only in \end@float and its siblings.

Here is a suggested patch. I have tested it only with one (big) file.

Of course the fix is only official if endorsed by LL.

*** latex.tex.~1~	Mon Feb 12 18:02:12 1990
 --- latex.tex	Fri Apr 20 21:09:00 1990
***************
*** 2000,2005 ****
 --- 2000,2009 ----
      \ifhmode\@savsf\spacefactor\fi}
  
  \def\@esphack{\relax\ifhmode\spacefactor\@savsf
+      {}\ifdim \@savsk >\z@ \ignorespaces 
+   \fi \fi}
+ 
+ \def\@Esphack{\relax\ifhmode\spacefactor\@savsf
       {}\ifdim \@savsk >\z@ \global\@ignoretrue \ignorespaces 
    \fi \fi}
  
***************
*** 6494,6500 ****
  %               else \vadjust{\penalty -10004
  %                             \vbox{} 
  %                             \penalty \@floatpenalty}
! %                    \@esphack
  %      fi    fi
  %    END
  %
 --- 6498,6504 ----
  %               else \vadjust{\penalty -10004
  %                             \vbox{} 
  %                             \penalty \@floatpenalty}
! %                    \@Esphack
  %      fi    fi
  %    END
  %
***************
*** 6515,6521 ****
  %       if \@floatpenalty < 0
  %         then \@dbldeferlist :=G \@dbldeferlist * \@currbox
  %       fi
! %       if \@floatpenalty = -10002 then \@esphack  fi
  %     END
  %
  \newcount\@floatpenalty
 --- 6519,6525 ----
  %       if \@floatpenalty < 0
  %         then \@dbldeferlist :=G \@dbldeferlist * \@currbox
  %       fi
! %       if \@floatpenalty = -10002 then \@Esphack  fi
  %     END
  %
  \newcount\@floatpenalty
***************
*** 6560,6566 ****
          \vbox{}                 %% 26 May 87 to prevent extra vertical
          \prevdepth \@tempdima   %% space when used in vertical mode
          \penalty\@floatpenalty
!       \else \vadjust{\penalty -\@Miv \vbox{}\penalty\@floatpenalty}\@esphack
       \fi\fi}
  
  
 --- 6564,6570 ----
          \vbox{}                 %% 26 May 87 to prevent extra vertical
          \prevdepth \@tempdima   %% space when used in vertical mode
          \penalty\@floatpenalty
!       \else \vadjust{\penalty -\@Miv \vbox{}\penalty\@floatpenalty}\@Esphack
       \fi\fi}
  
  
***************
*** 6574,6580 ****
  \def\end@dblfloat{\if@twocolumn
       \par\vskip\z@\egroup %% \par\vskip\z@ added 15 Dec 87\egroup
       \ifnum\@floatpenalty <\z@ \@cons\@dbldeferlist\@currbox\fi
!       \ifnum \@floatpenalty =-\@Mii \@esphack\fi\else\end@float\fi}
  
  \newcount\c@topnumber
  \newcount\c@dbltopnumber
 --- 6578,6584 ----
  \def\end@dblfloat{\if@twocolumn
       \par\vskip\z@\egroup %% \par\vskip\z@ added 15 Dec 87\egroup
       \ifnum\@floatpenalty <\z@ \@cons\@dbldeferlist\@currbox\fi
!       \ifnum \@floatpenalty =-\@Mii \@Esphack\fi\else\end@float\fi}
  
  \newcount\c@topnumber
  \newcount\c@dbltopnumber
***************
*** 6689,6695 ****
  
  \def\@xympar{\ifnum\@floatpenalty <\z@\@cons\@currlist\@marbox\fi
       \setbox\@tempboxa\vbox   %% added 3 Jan 88
!      \bgroup\end@float\@esphack}
  
  \def\reversemarginpar{\global\@mparbottom\z@ \@reversemargintrue}
  \def\normalmarginpar{\global\@mparbottom\z@ \@reversemarginfalse}
 --- 6693,6699 ----
  
  \def\@xympar{\ifnum\@floatpenalty <\z@\@cons\@currlist\@marbox\fi
       \setbox\@tempboxa\vbox   %% added 3 Jan 88
!      \bgroup\end@float\@Esphack}
  
  \def\reversemarginpar{\global\@mparbottom\z@ \@reversemargintrue}
  \def\normalmarginpar{\global\@mparbottom\z@ \@reversemarginfalse}


Piet* van Oostrum, Dept of Computer Science, Utrecht University,
Padualaan 14, P.O. Box 80.089, 3508 TB Utrecht, The Netherlands.
Telephone: +31-30-531806   Uucp:   uunet!mcsun!ruuinf!piet
Telefax:   +31-30-513791   Internet:  piet@cs.ruu.nl   (*`Pete')

-----------------------------------------------------------------------

%%% Further information about the TeXhax Digest, the TeX
%%% Users Group, and the latest software versions is available
%%% in every tenth issue of the TeXhax Digest.
%%%
%%% Concerning subscriptions, address changes, unsubscribing:
%%%
%%%  BITNET: send a one-line mail message to LISTSERV@xxx
%%%         SUBSCRIBE TEX-L <your name>    % to subscribe
%%%      or UNSUBSCRIBE TEX-L
%%%
%%% Internet: send a similar one line mail message to
%%%           TeXhax-request@cs.washington.edu
%%% JANET users may choose to use
%%%           texhax-request@uk.ac.nsf
%%% All submissions to: TeXhax@cs.washington.edu
%%%
%%% Back issues available for FTPing as:
%%%          machine:              directory:  filename:
%%%   JUNE.CS.WASHINGTON.EDU         TeXhax/TeXhaxyy.nn
%%%              yy = last two digits of current year
%%%                       nn = issue number
%%%
%%%\bye
%%%

End of TeXhax Digest
**************************
-------