[comp.text] TeXhax Digest V88 #83

TeXhax@Score.Stanford.EDU (TeXhax Digest) (09/23/88)

TeXhax Digest   Friday, September 23, 1988   Volume 88 : Issue 83

Moderator: Malcolm Brown

Today's Topics:

                       \midspace and \midinsert
                     Re: \midspace and \midinsert
                           a test with key
        Access to TeX-related stuff by ANONYMOUS FTP from Yale
                        New version of DosTeX
                                BiBTeX
                          Sideways Printing
                      refer to BiBTeX conversion
                      Re: TeXhax Digest V88 #79
                         Index Help for Tex?
                  driver for Hewlett Packard Deskjet
                Re: The best way to do pictures in TeX
                     TeX and genealogy (2nd try)
               Re: Problem installing TeX on a 3B2/400
              Re: yaffb (yet another format for bibtex)
                    Re: DVI-to-PS drivers for VMS
             Previewers for IBM RT PC and for CGA screens

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

Date:         Thu, 15 Sep 1988 10:04 EDT
From:         Jim Walker <N410109%univscvm.BITNET@Forsythe.Stanford.EDU>
Subject:      \midspace and \midinsert

A colleague of mine, using AmS-TeX, ran into an instance in which a
\midspace that occurred at the end of a page foated to the top of the
same page. According to the manual, it's not supposed to do that; it's
supposed to float to the next page if it floats at all.  I haven't been
able to duplicate the effect with simpler data, but still, it's a
mystery.

\midspace expands into a \midinsert from plain TeX. According to the
TeXBook, if a \midinsert does not fit where it is, then it is converted
to a \topinsert. Doesn't that imply that it should always float to the
top of the current page? In other words, I don't understand why
\midspace EVER works as advertised.

%%% Barbara Beeton provides a quick initial reply:

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

Date: Thu 15 Sep 88 14:45:09-PDT
From: Barbara Beeton <BEETON@Score.Stanford.EDU>
Subject: Re: \midspace and \midinsert

we (at the math society) have seen the situation you describe, and also
haven't yet figured out how to fix it.

an insertion isn't going to be put on a page that would require the
place in text where it's invoked to be moved to the next page,  so
indeed all \midinsert's that are forced to move {\em should} go the
the following page, so that they come after the reference.  so why does
a \midinsert that doesn't fit at the bottom of the page fit at the top
of the same page?  this is what, in another field, would be known as a
boundary condition.  a \midinsert gets some vertical space put above and
below it; vertical space doesn't disappear at the bottom of a page like
it does at the top.  condition: the insert plus the space above it just
barely fits in what's left of the page, but the extra space following
it makes it too large, so it's changed to a \topinsert.  a \topinsert
doesn't get any space put above it, only below, and it's probably
about the same amount of space that was put above the \midinsert.
so now it (just barely) fits on the same page as the reference.
voil\`a!  you're just the victims of bad luck.
this could probably be fixed by tinkering with the macros in plain,
but that's not advised, and besides, as you said yourself, duplicating
the effect with simpler data for testing isn't all that easy.
					-- barbara beeton

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

Date:         Fri, 16 Sep 88 15:38:18 EDT
From:         Shashi Sathaye <SYSSHASH%ukcc.uky.edu@Forsythe.Stanford.EDU>
Subject:      a test with key

Some time in May (I think) some one sent a LaTeX file showing how to set up
a file such that one can make up a test with key in it, and conditionaly
print either a key or a test. I have texhax saved after 8858 and could not
find it in there. Will you please let me know who sent it or sent me the
copy of texhax that has the file in it. Thanks in advance. Shashi

%%% you could try looking at the file "<tex.texhax>texhax88.IN1", which is
%%% an index to the first 25 issues of 1988 texhax, prepared by Thiele and
%%% Beeton.  Otherwise you'll have to hunt around in the issues between
%%% 26 and 57.  Sorry, but I haven't the time to undertake searches of this
%%% sort.  Malcolm

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

Date: Mon, 19 Sep 88 14:50 EST
From: "Jerry Leichter (LEICHTER-JERRY@CS.YALE.EDU)"
Subject: Access to TeX-related stuff by ANONYMOUS FTP from Yale

I've been getting a number of messages from people who have been trying to get
several packages I announced on this list from Yale:  DVIDIS, the VWS DVI file
displayer; TEXTYL, John Renner's curve drawing program for TeX, as adapted for
VMS by yours truely; and TGRIND, my hacked at version of a program typesetter
originally written by Van Jacobson.

The problem is with Wollongong's latest versions the FTP server, which we run
here:  It will not let you access anything in any directory but the default
login directory for ANONYMOUS.  Wollongong claims this is a feature....  (At
one time, they would let you get at anything the operating system let you
read.  Then they allowed access to subdirectories of the login directory, but
no other directory trees.  I guess the next version will increase security
even more by not allowing you any access at all.)

The way I used to have stuff set up here was to have a subdirectory of the
ANONYMOUS accounts login directory for each distinct package.  That no longer
works.  So I've build backup savesets from the subdirectories, DVIDIS.A,
TEXTYL.A and TGRIND.A, and placed them in the ANONYMOUS login directory.  (For
TEXTYL.A, the saveset contains both the VMS and the original Unix versions, as
well as uuencoded and non-encoded versions of font and TFM files.  Trying to
keep the stuff separate was too much of a hassle.  Similarly, DVIDIS.A
contains a correct version of PXTOPK, because there was at one time a bogus
one floating around.)

Also in that directory are a pair of files, FIXREC.C and FIXREC.EXE, which fix
up the record size of the files after transfer.  Copy FIXREC.C; then, in
BINARY mode, copy FIXREC.EXE (optional, you can re-compile) and the .A file
you want.  Read the comment at the top of FIXREC.C to see how to use it.  The
savesets were produced using the default block size, which FIXREC knows
about....

I will be gone for a couple of weeks, so I will not be able to help you if
anything goes wrong, at least not right away.  If your write, I'll answer
when I get back - but please be prepared for a long wait.

							-- Jerry

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

Date: Mon, 19 Sep 88 09:44:54 CDT
From: beihl%cadillac.cad.mcc.com@mcc.com (Gary Beihl)
Subject: New version of DosTeX

The DosTeX distribution has been upgraded to version 2.93.  Excerpts
from the READ.ME file follow.  Ordering information from the original
TeXhax posting (issue #1, 1988) remains correct.

This version contains support for run-time sizing of the mem[] array
within INITEX.  In addition, a utility is provided to allow
construction of preloaded versions from INITEX dumps.  Preloaded
versions are therefore no longer distributed.  Instead, example .BAT
files are provided which may be used to produce preloaded versions of
"plain" and LaTeX.  A bug related to invisible rules has been fixed as
well as all previously reported bugs.
 
The Epson printer driver (based on Nelson Beebe's version 2.10) no longer 
swaps the bitmap to disk.  Instead, two passes are made over each page
with a half-size bitmap.  This gives approximately double the performance
of swapping. The bug related to closing log files is fixed.

The Hercules previewer is no longer distributed with screen resolution 
(120 dpi) fonts. The default is now to subsample the Epson (240 dpi)
.pk fonts.  Screen resolution fonts may still be selected, but the user
must provide these.

Gary Beihl (beihl@mcc.com)

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

Date:	20-SEP-1988 08:30:07 GMT -01:00
From:	THOWARD%graphics.computer-science.manchester.ac.uk@NSS.Cs.Ucl.AC.UK
Subject:        BiBTeX

I've ended up after a few years with a big BiBTeX database
of all my sources---which is great for making citations and source
lists in the usual LaTeX way.
Now I've got this big database, with a lot of effort invested in making it,
I need to process the database slightly differently.

I would like to have the command \getit{ref-key}, which expands to 
the formatted BiBTeX entry for the key just like it would appear in the 
source list.

(This is for an undergraduate `book list', with lists of books split into
subject areas, and comments on each. Listing [refs] under each heading 
pointing into a massive source list at the end is unhelpful to the reader.)

Any ideas, or help would be very welcome
Toby

Toby Howard 
Computer Graphics Unit, Department of Computer Science,
Manchester University, Oxford Road, Manchester, England, M13 9PL. 
Phone: +44 61 275 6224
Janet: toby@uk.ac.man.cs.cgu  
ARPA:  toby%cgu.cs.man.ac.uk@nss.cs.ucl.ac.uk                    

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

Date: Tue, 20 Sep 88 09:24:44 EDT
From: Eric Domeshek <domeshek-eric@YALE.ARPA>
Subject: Sideways Printing

I've been playing with the TeX version of Kernighan's PIC picture drawing
system in order to insert some charts into a LaTeX document.  It turns out
that the chart I want to construct is too wide for standard 8-1/2" paper, so
I wanted to try to print it sideways to get 11" of width to play with.  This
chart contains lots of text as well as boxes, arrows, etc.

Is there any way to get TeX to print sideways?  Is there a way to get it to
print just part of a document (a couple of LaTex figure environments)
sideways while the rest is normal?  Do I have to split the document into
seperate parts, leave gaps in the page numbering, and play with the print
drivers to get this to work?

Any help would be appreciated.  I don't normally read this group, so I'll
only see answers if you E-mail responses directly to me at:

ARPA:   Domeshek-Eric@cs.yale.edu
UUCP:   {harvard,cmcl2,decvax}!yale!Domeshek-Eric
BITNET: Domeshek@yalecs

Thanks for any tips,

-Eric

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

Date: Tue, 20 Sep 88 16:00:17 BST
From: bowen%prg.oxford.ac.uk@NSS.Cs.Ucl.AC.UK
Subject: refer to BiBTeX conversion

Further to Peter King's message in TeXhax Digest V88 #79, I have a
program which is also based on "awk" and "sed", called "ref2bib" which
will convert "refer" format to "BibTeX" format. The heuristics have
been developed by applying the program to a number of "refer" format
databases here at Oxford. Ref2bib can produce most of the types of
record which are allowed by BibTeX. Since it is a shell script, ref2bib
can easily be modified if required. Additionally, if a file called
".ref2bib" exists in the user's home directory, then this is used as a
"sed" script source for individual customization. Note that ref2bib 
was developed entirely separately from r2bib and Peter King's program.
It may or may not make a better job of conversion, depending on the
"refer" format input supplied. It also has options to determine the
naming of entries and the folding of output.

The program is accessible from the Programming Research Group archive
server at Oxford University by mailing a message containing:

	send prog ref2bib.shar

to <archive-server@prg.oxford.ac.uk> or <archive-server@ox-prg.uucp>.
A message containing "help" can also be sent for more information on
using the archive server. I am happy for this program to be placed
on any archive server, etc., in the US and elsewhere to reduce
international traffic.


Jonathan Bowen
Oxford University Computing Laboratory
Programming Research Group
England
JANET:	bowen@uk.ac.oxford.prg
ARPA:	bowen%prg.oxford.ac.uk@nss.cs.ucl.ac.uk
UUCP:	bowen@ox-prg.uucp (...!uunet!mcvax!ukc!ox-prg!bowen)
------------- snip snip ---------------
%%% Jonathan's file is too large for digest distribution.  As usual, it
%%% can be FTPed from the machine "score.stanford.edu" by GETting the
%%% file "<tex.texhax>bowen.txh".  I have sent a copy to TEX-L so that
%%% BITNET folks can fetch it using listserve commands.   Malcolm

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

Date:         Tue, 20 Sep 88 13:05:21 EDT
From:         "F. James Rohlf" <ROHLF%SBBIOVM.BITNET@Forsythe.Stanford.EDU>
Subject:      Re: TeXhax Digest V88 #79

The previewer DVIEW by Steve Ward was mentioned in Volume 88 - issue 79.  How
does one obtain a copy?


  . - - - - - - - - - - - F. James Rohlf  - - - - - - - - - - - - - - - .
  | Dept. of Ecology & Evolution,     State University of New York      |
  | Stony Brook, NY 11794-5245        Phone: 516 632-8580               |
  . - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - .

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

Date: Tue, 20 Sep 88 13:18:07 CDT
From: powers@skacsl.csc.ti.com
Subject: Index Help for Tex?

    I heard from a friend that there was a tool called IndxTex (or something
like that) that is in the public domain to help take the output of index 
commands and make an index that is useful.

Does anyone know about this?

Thanks,
Richard Powers

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

Date:     Tue, 20 Sep 88 11:58 PDT
From:     <COVENTRY%UWACHEM.BITNET@Forsythe.Stanford.EDU>
Subject:  driver for Hewlett Packard Deskjet

Is there a driver for the HP Deskjet printer yet?

regards,
Mary Coventry
coventry@uwachem (bitnet)

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

Date:     Tue, 20 Sep 88 14:51:44 CDT
From: William LeFebvre <phil@rice.edu>
Subject:  Re: The best way to do pictures in TeX

> I am normally working with pictures that should fit on an A4 page
> (approximately the same size as the US 11x8 inches format), and this is
> easily managed by MF.  

Wow!  I didn't know MF could deal with characters that large.  I was
assuming that it had some upper limit that was reasonable for
characters but not necessarily for whta I wanted to do.  Then again, a
8 by 11 inch character is not an unreasonable thing to request.....

You do point out the other reason I was concerned about tiling:
limitations of the output device.  I suspect that almost every DVI
driver in existence downloads every character as an individual
character "glyph" and that they don't try very hard to do anything
clever if the GF glyph is too large for the limitations of the
printer's font format.  Internal printer memory limitations are
important, too, but those are harder to detect.  

I'll have to try this out.  Thanks for the tips.

			William LeFebvre
			Department of Computer Science
			Rice University
			<phil@Rice.edu>

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

Date: Tue, 20 Sep 88 10:49:55 -0200
From: mcvax!eutrc3!eutrc4!rcpt@uunet.UU.NET (Piet Tutelaers)
Subject: TeX and genealogy (2nd try)

I am considering to typeset our genealogy with a hacked version of the
TeXbook-style. This style does not provide standard solutions to the
following typographic problems typical for genealogical publications:

   o the pagestyle -- Two-columns with the possibility to include photos
	either spanning one column or spanning both columns. The text
	should run  from top to bottom, also if the photo spans two
	columns.

   o kwartierstaat -- I couldn't find out what is the English word for our
   	Dutch `kwartierstaat'. It is the table you get if you denote the
	parents of the holder of the `kwartierstaat', their parents and so on. 
	Their are several ways to represent a `kwartierstaat'. I have seen
	a book published with TeX in which each left and right page
	contains a `kwartierstaat' consisting of 32 `kwartieren'.

   o family tree -- In the TeXbook, Donald Knuth gives (Excercise 22.14) an
   	example of a family tree. To typeset trees in this way would be a
	very painfull undertaking. I have seen a genealogy in which the
	main part of the tree was added as an appendix. While reading the
	book you could unfold this chart. Different locations in the family
	tree were indicated by different colours. It would be very nice
	to have a program which can generate such a chart for an
	electrostatic color plotter straight from the family tree.

   o coat of arms -- It would be very nice also to have a metafont program
  	which can generate the outline of a shield and the heraldic colors
	(the black and white standards for it) which are usually used here.

Experiences and suggestions in this area are very welcome!

___________			  Part I, hermine with two strokes of red
|    |\/\/|		   	<****>, part II silver with two triagular
|****|    |  Willem Tuteleers	 strokes at the top <\/\/> and a golden
|    |    |  Brustem 1562	 wheel <()>.
|****| () |  Belgium
 \___|___/

uucp:   mcvax!rcpt@eutrc3     | Piet Tutelaers        Room  RC 1.90
bitnet: rcpt@heithe5          | Eindhoven University of  Technology
phone:  +31 (0)40 474541      | P.O. Box 513, 5600 MB Eindhoven, NL

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

Date: Tue, 20 Sep 88 21:01:37 EDT
From: Chris Torek <chris@mimsy.umd.edu>
Subject: Re: Problem installing TeX on a 3B2/400

Michael Harpe writes:
>I am attempting to install TeX 2.93 and friends ... [with web2c or tex2c]
>onto a 3B2/400 with the floating point C compiler (fpcc).  All goes
>well until time to compile the last module of C code (tex8.c).  The compile
>fails with a compiler error indicating that the switch table has overflowed.
>This also happens when I try to compile DVItype for the trip test.

>I have looked everywhere for some kind of switch I can set to avoid this.

The switch you need is in the source code to your C compiler, which
AT&T does not provide.  Without source you have little hope of
increasing the size of the compiler's switch table.  What you can do is
to split up the switch statement that is overflowing the compiler's
tiny table.  For instance, if the switch table held four entries
and you needed six:

	switch (e) {
	case c1: code1; break;
	case c2: code2; break;
	case c3: code3; break;
	case c4: code4; break;
	case c5: code5; break;
	default: codedefault; break;
	}

becomes

	switch (e) {
	case c1: code1; goto done;
	case c2: code2; goto done;
	case c3: code3; goto done;
	case c4: code4; goto done;
	}
	switch (e) {
	case c5: code5; goto done;
	default: codedefault; goto done;
	}
done:

In other words, each `break' becomes a `goto' to skip successive
additional switch statements.  A different label must be used for
each split-up switch.

The best solution, albeit unlikely, would be to get AT&T to ship you
a more reasonably configured compiler.

Chris

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

Date: Tue, 20 Sep 88 23:34:59 EDT
From: toms@ncifcrf.gov
Subject: Re: yaffb (yet another format for bibtex)

TeXLaNdErS:

I ended up making my own format for Nucleic Acids Research.  I was able to
figure out that unsrt.bst contained instructions in a postfix language, and
I was able to modify the code to do some of what I needed.  Fortunately, I
contacted Stephen Gildea, who kindly sent documentation to me (it is called
btxdoc.tex), and so I could finish the task.

I would have been helped a lot if it had been more obvious
that all I needed was a nar.bst file and a nar.sty file.
For those of you wondering about making your own reference styles,
it turns out that there is a nicely defined language to do this.
The language is un-named, but is postfix and reasonably simple.

(FLAME ON: Yes, all you wizards, this is NEWS to most people!!!  If I had been
told this a few months ago, I would have been able to write what I need without
trouble!  Oren Patashnik, your long letters were useless, though your un-named
language is lovely.  If you had simply said
  'Well, I understand your desire for switches, but it is much more
  powerful to define a whole language!  Here is the document for the language.'
it would have kept me (and probably lots of other people) out of a bunch
of blind alleys.  FLAME OFF)

  I think that this language should be documented in the LaTeX book,
in a special appendix, along with other notes on how to change the .sty
macro to make the references come out the way a journal demands (examples
of how to switch from [] to (), examples of how to make the reference
be superscripted, how to make the bold face Reference lie down and be
what the journal asks.)

  It would also be immensely useful to have the file 00index in Rochester
refer to this documentation, and to provide it at Rochester.  (If it is
in our local documentation, I was unable to find it, when I looked after
Stephen sent it to me.)

I will submit nar.bst and nar.sty to Rochester once I've brushed them up
a bit more.

  Tom Schneider
  National Cancer Institute
  Laboratory of Mathematical Biology
  Frederick, Maryland
  toms@ncifcrf.gov

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

Date:		21-SEP-1988 08:32:23 GMT
From:		ABBOTTP%aston.ac.uk@NSS.Cs.Ucl.AC.UK
Subject: Re: DVI-to-PS drivers for VMS

John Manley enquires about DVI-to-PS on VAX/VMS.

Andrew Trevorrow has produced version 3 of PSPRINT for VAX/VMS systems. It
will drive Apple Laserwriters (and I assume other 300dpi postscript
printers but since we do not have any other brand of postcript printer it
has not been tried), LPS40 (tested elsewhere) and Linotronic (via a RIP,
tested here at Aston). Version 3 was developed here at Aston and includes
all the usual features you would expect from a DVI-to-PS program. Version
1.1 (and his DVItoVDU 1.7) are described on page 42 of Tugboat Vol 8 No 1
April 1987. A report on version 3 will appear in Vol 9  No 3 (due out end
of 1988). 

It is public domain software and versions are available from the usual 
sources. At present Version 3 is in the Aston archive but eventually will 
find its way across the `pond'.

Andrew read TeXhax so no doubt will respond himself with more details. All 
I will say is that it is very useful and well integrated into the TeX/LaTeX 
system here at Aston.

Peter
Computing Service     JANET	abbottp@uk.ac.aston
Aston University      ARPA	pabbott@nss.cs.ucl.ac.uk
Aston Triangle	            or  abbottp%uk.ac.aston@nss.cs.ucl.ac.uk
Birmingham B4 7ET     UUCP	...!ukc!aston!abbottp
U.K.		      BITNET	abbottp%uk.ac.aston@ac.uk
Tel (+44) 21 359 5492					     

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

Date:  Wed, 21 Sep 88 10:40:05 +0100
From: Stig Larsson <stig%math.chalmers.se@RELAY.CS.NET>
Subject: Previewers for IBM RT PC and for CGA screens

Is there a previewer for the IBM RT PC workstation running under UNIX
4.3 BSD?  

I am also hunting for the best previewer for a PC-XT with CGA.  In TeXhax
#79 Jacques Goldberg recommended Steve Ward's DVIEW.  Where can I get
that?

In TeXhax #80 Peter Flynn announced that Wayne Sullivan's CDVI
previewer is now available from the Rochester, Aston and Heidelberg
repositories.  Could somebody explain to me what to do to get the
stuff from there.

Thanks, 

Stig Larsson
Department of Mathematics
Chalmers University of Technology
S-412 96 G\"oteborg
Sweden

stig@math.chalmers.se

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

%%%
%%% Concerning subscriptions, address changes, unsubscribing:
%%%     BITNET: send a one-line mail message to LISTSERV@TAMVM1.BITNET:
%%%         SUBSCRIBE TEX-L <your name>    % to subscribe
%%%
%%%     All others: send mail to
%%%           texhax-request@score.stanford.edu
%%%     please send a valid arpanet address!!
%%%
%%%
%%% All submissions to: texhax@score.stanford.edu
%%%
%%% Back issues available for FTPing as:
%%%          machine:      directory:  filename:
%%%   [SCORE.STANFORD.EDU]<TEX.TEXHAX>TEXHAXnn.yy
%%%      nn = issue number
%%%      yy = last two digits of current year
%%%\bye
%%%

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

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