[net.text] TeX

nawyn@ittvax.ATC.ITT.UUCP (Ken Nawyn) (05/22/85)

	Hello fellow net people.  As I am new to this group I would
	appreciate some help.  I would like to obtain a copy of the
	text formatter "TeX" written by Donald Knuth at Stanford
	University.  I am interested in both commercial and public
	domain versions.  Many Thanks to all who answer.

	Please send responces by mail.  Just follow the Yellow Brick
	Road!
-- 

				Ken Nawyn
				ITT-ATC Systems Center,
				1 Research Drive Shelton, CT  06484
				Phone #: (203)-929-7341 x. 641
	
path = {allegra bunker ctcgrafx dcdvaxb dcdwest ucbvax!decvax duke eosp1
	ittral lbl-csam milford mit-eddie psuvax1 purdue qubix qumix 
	research sii supai tmmnet twg uf-cgrl wxlvax yale}!ittvax!nawyn

toni@erix.UUCP (Toni Roth) (01/20/86)

I have a problem with TeX that probably is easy to a TeXpert:
I get tex output from a filter that looks like this:

\srcvr {line 1 in column 1
line 2 in column 1
line 3 in column 1}
\rcvr {line 1 in column 2
line 2 in column 2}
\ctw {line 1 in column 3}
\receivers
some text after the tricky part

My problem is to design the macros \srcvr, \rcvr, \ctw and
\recivers to yield this:

line 1 in column 1	line 1 in column 2	line 1 in column 3
line 2 in column 1	line 2 in column 2
line 3 in column 1

some text after the tricky part

I've tried many different ways, but none seems to work.
If someone knows how to do this, please respond by mail.

		...!mcvax!enea!erix!toni

toni@erix.UUCP (Toni Roth) (02/09/86)

I have a problem with fast loading of macros:

I've written a macro package that uses many features from the plain
format. Now I want to create a .fmt-file for fast loading of these
macros.

The TeX-book says you should do "initex" and then type "&plain super"
in response to the "**" prompt (super is your extensions of plain).

If I do this, the command "virtex &super" takes more time to load the
macros than the command "tex" followed by "super".

The fastest way to load these macros must be to give the command "tex"
and then type "&super" in response to the "**" prompt. Is there a way
to create this "super.fmt" that doesn't include the plain format but
still uses it?

Thanks in advance:		...!mcvax!enea!erix!toni

chris@umcp-cs.UUCP (Chris Torek) (02/12/86)

In article <1058@erix.UUCP> toni@erix.UUCP writes:

>I've written a macro package that uses many features from the plain
>format. Now I want to create a .fmt-file for fast loading of these
>macros.

Since `super' builds on `plain', you can do this:

>The TeX-book says you should do "initex" and then type "&plain super"
>in response to the "**" prompt (super is your extensions of plain).

In fact you really should use `virtex &plain super', since you need
initex only to build hypenation tries and the like (I think---but
I am going to recommend something different).

But:

>If I do this, the command "virtex &super" takes more time to load the
>macros than the command "tex" followed by "super".

This is because `tex' has plain.fmt `preloaded'---in core (or more
precisely, in virtual space).  `virtex &plain' and `virtex &super'
would both take about the same amount of time.

What you really want to do is create a preloaded `supertex'.  The
way this is done is operating-system-dependent (and in some cases
compiler-dependent as well).  The distributed Unix TeX machinery
for doing this is as follows:

	% initex
	**\input plain \input super \dump
	[lots of output]
	[TeX writes `plain.fmt']
	% mv plain.fmt super.fmt
	% virtex
	**&super \read 0 to \blort
	\blort=
	[here you type your `quit' command key, normally ^\]
	Illegal instruction (core dumped)
	% undump supertex /usr/local/bin/virtex		[or whatever]
	Data segment size was 6144 now is 640000
	%

The `blort' hackery was required with the 4.1 TeX; I do not know
whether this was fixed in the 4.2 version.  The hack is not required
with Pastel TeX, and may even fail (I was told that it does not work
for LaTeX):

	% virtex
	**&super
	*^\

Note that undumped exectuables are quite large, and should thus be
used sparingly.
-- 
In-Real-Life: Chris Torek, Univ of MD Comp Sci Dept (+1 301 454 1415)
UUCP:	seismo!umcp-cs!chris
CSNet:	chris@umcp-cs		ARPA:	chris@mimsy.umd.edu