
This is words, a Latin-to English dictionary program, by William
Whitaker (whitaker@erols.com). It takes keyboard input or a file of
Latin text lines and provides an analysis of each word individually.
It uses the following files: INFLECT.SEC, UNIQUES., ADDONS.,
STEMFILE.GEN, INDXFILE.GEN, and DICTFILE.GEN.

This program was written originally in Ada, and designed for
DOS/Windows PCs.  It was first ported to Linux in early 1999 using
gnat 3.10p (Ada compiler) on a Pentium-II PC running RedHat Linux 5.2.
(Linux Kernel 2.0.36), and re-compiled in May 2000 using gnat 3.12p on a
similar PC running RedHat 6.0 (Linux Kernel 2.2.5).

In June 1999 it was found that the port of version 1.93 did not
run on a Slackware 3.6 distribution (libc5 rather than glibc).  The
`words' executable was re-compiled and run on the same Slackware 3.6 box,
under kernel 2.0.36.  That fix can be downloaded separately if you
need to run words 1.94 or previous on a libc5-based distribution.  Details
on this fix can be found at the end of this document.

However, all ports after version 1.95, which was completed in January 2000,
have the libraries compiled statically into the executable, and should therefore
run on any Linux installation.  Please let me know if this is untrue.

FreeBSD note:  Words will run on a FreeBSD 3.2 box under linux
emulation.  However, it is known not to work on FreeBSD-3.4 due to
the fact that the linux rt_sigaction() system call is either not supported
or ignored.

Please report to me any additional successes and failures running `words'
on another Linux distribution or platform (or even other Unices), and I
will post the information here.


====  

The latest release as of September 7, 2001 is words-1.97-linux.tar.gz.
(About 1.6 MB)

This and all older versions of words for Linux are available from:

    ftp://ftp1.thomasaquinas.edu/pub/linux/words/


Information about the program, current releases, and the words program
on other platforms can be found at:

    http://www.erols.com/whitaker/words.htm


The online interactive version of words can be found at:

    http://lysy2.archives.nd.edu


For complete documentation, read the included file wordsdoc.htm (in
HTML) or wordsdoc.txt (in ASCII); or find it on the web at:

    http://www.erols.com/whitaker/wordsdoc.htm


====

Untarring/ungzipping the file `words-1.xx-linux.tar.gz' will create a
subdirectory `words-1.xx' in your current directory. If you want to make
words available to all users on your system, you should do something
like this:  move the tar.gz file to a directory (e.g. /usr/local/) then
execute the following command:

    tar zxvf words-1.xx-linux.tar.gz

This results in a directory (/usr/local/words-1.xx) with about 9 MB of files
which contains the following: 

    README
    INFLECT.SEC
    UNIQUES.
    ADDONS.
    STEMFILE.GEN
    INDXFILE.GEN
    DICTFILE.GEN
    latin
    words
    wordsdoc.html
    wordsdoc.txt

After untarring, you may of course remove the file 
`words-1.xx-linux.tar.gz' from the directory /usr/local.

====

The simplest way to run the words program is to change to the directory
that has these zipped files (e.g. /usr/local/words-1.xx), and execute the
command:

    ./words

However, since you probably don't want to cd into this directory every
time you run words, I've included a simple shell script (suggested to me 
by Johannes Tevessen <johannes.tevessen@xlink.net>) that changes into the
words directory (e.g., /usr/local/words-1.xx), then runs the words
binary from there as a subtask.  On my system, I call the script `latin',
so that it's easy for my students to remember, but you can rename it
whatever you'd like.  It is as follows:

	#! /bin/bash
	WORDSPATH=/usr/local/words-1.xx
	( cd $WORDSPATH ; ./words $* )

You should put this shell script in one of your bin directories in
your path (mine is in /usr/local/bin). Then to run words, all you do is
type `latin' at any prompt, and it will execute words for you, and exit
gracefully.


I had to include this script because the program needs to be in its own
directory to know where to find its supporting files.  Creating a link
(ln) to the words program doesn't quite work.  Neither does putting
the whole `words' directory in your path. Remember, this program was
ported from DOS without any change in the source code, so it has some
DOS quirkyness.  Perhaps someone would like to write an install
script so that you don't have to do all of this stuff manually.  If so,
email it to me, and I will include it with the program.

====


For libc5 users only:  some users on libc5 systems had difficulties
executing the port of words 1.94, since it was compiled on a RedHat 5.2
glibc system.  The newer port of words 1.95, however was compiled
statically (i.e., library-independent). 

If you are still using words 1.94, please check your own system by
looking for the file /lib/ld-linux.so.2. If you have /lib/ld-linux.so.1
instead, you may need a different version of the `words' 1.94 executable.
You can get the libc5 verson of the `words' file by downloading the
following file from the same location: 

	words-1.94-libc5.tgz

This file will untar/ungzip to a single executable file `words'.  You
should use this file to replace the file `words' that came with the
original download.  This should allow you to run words on a libc5 system.

You should be able to circumvent this whole problem simply by upgrading to
any port after 1.95.  Try the latest version first, and let me know if it solves
the problem.

====


DISCLAIMER: I had nothing to do with the writing of this program.  I
only ported it from DOS to Linux.  Email me (bdragoo@thomasaquinas.edu)
with any Linux-related questions; email Whitaker (whitaker@erols.com) with
any program-related questions.


                       Pax Domini,

                       Brian Dragoo
                       Thomas Aquinas College
		       Santa Paula, CA
                       September 2001

		       bdragoo@thomasaquinas.edu






