
THE VOYNICH INTERLINEAR TRANSCRIPTION ARCHIVE TOOLS
VERSION 0.03 BETA
MICHAEL WINKELMANN, michael@weltretter.de

WHAT IS IT?

These are  some really  simple perl scripts  for handling  the Voynich
interlinear  transcription archive  edited  by Jorge  Stolfi. I  wrote
them, after I tried to compile and use the older VTT tool.

Because the tools are written  in a interpreted scripting language, it
is relatively easy to adapt them to fit your special needs.

This  little   file  is  always   out  of  date,  but   the  generated
documentation for the tools is up to date.


PURPOSE OF THE TOOLS

viat.pl

This tool  extracts lines  from the Voynich  interlinear transcription
archive, it can be used as a replacement for VTT.

vword.pl

This tool splits  a transcription file in separate  words or creates a
word  count statistic.  It can  be used  in a  pipe together  with the
viag.pl tool.

vhtml.pl

This tool creates a nice  formatted HTML document from a transcription
file. The manuscript  text is formatted in the EVA  Hand 1 font, which
eases the comparision with manuscript photos.

This is the most hackware-like script with messy source.

shuffle.pl

This tool  rearranges the input  lines in random  order. I used  it to
check the randomness of the single word repetitions.

linejoin.pl

This tool  joins it input  (assumed to be  a VMS transcription  in the
standard format)  to a  single, long string  of text. All  line ending
symbols and  inline comments  are removed, so  the weirdo  markup will
disappear  if there  is no  converting option  given to  viat.pl while
generating the input file for linejoin.pl

vindex.pl

Generate  an  index  file  (a  full concordance)  of  the  interlinear
archive. This index file can be used for analysis purposes or together
with the vfind.pl program. 

If  you want to  use the  vfind tool,  you have  to generate  an index
first. 

vfind.pl

Find all occurences of a voynichese word in the interlinear file. This
is done using the index file generated by vindex.pl

For a single  use, this program is probably much  slower than a simple
grep on  the whole archive, because  the whole index must  be read and
parsed. But  if you want  to query more  than one word  in interactive
mode, the program is really fast.

showeva.pl

This  is an experimental  program for  displaying transcriptions  in a
simple GUI  using the EVA font.  Some parts of the  description can be
made invisible by using parts of the GUI. 

If you want to use showeva in a  pipe, do not forget to call it with a
dash  as  file  specification.  Without arguments,  the  program  will
display nothing until you select a file by the menu.



WHERE IS THE DOCUMENTATION?

The documentation can be generated with the shipped mkdoc.pl tool. You
can create documentation for all programs just by typing 

	perl mkdoc.pl

This creates  a subdirectory "doc",  in which the  whole documentation
and an index file "index.html" is generated. 

All  documentation is  created from  the pod  section embedded  in the
script.

You can  use the perldoc tool,  if you want to  read the documentation
from  the command  line. Even  you  can create  documentation in  your
favorite format by  using the various pod2* tools  or the perl modules
in a simple script.


WHY PERL?

Okay, some  people say that  the syntax of  perl looks like  a erratic
file conversion,  and often they are  right. It is easy  to write, and
heavy to read.

There  is a  perl  interpreter  for nearly  every  platform, even  for
obsolete ones like the Commodore Amiga. So I do not expect any porting
problems, which arises often when C is used.

There is  no need to compile the  program to get a  binary. Many users
are not  un*x users and  do not know  anything about compiling  - they
hear or read some strange tech-speech and don't know what's to do.

There is no better language for text processing than perl. It is easy,
fast and practical.

Okay,  everything has it  drawbacks. To  use a  perl script,  the perl
interpreter has to be started,  which consumes time and slows down the
processing. But if you want to create a native binary for your system,
there is a perl compiler, which generates a C source you can compile.


NOTES TO MS/DOS and WINDOWS USERS

You  can make  the Perl  scripts directly  executable from  you MS/DOS
prompt by generating a batch file  using the pl2bat utility. It is not
like the Un*x mechanism of  making the file executable and providing a
hashbang at the first line.  The difference is, that an erratic return
value is not returned to the caller.

This problem is  not a huge one. But in a  pipe, the following process
can not notify the error and may handle empty or wrong input. If there
is a  lot of output  filling the console  or MS/DOS window,  the error
message may be unseen and the erratic output may cause confusion.

To  avoid this  problem,  you  can always  call  the perl  interpreter
directly, as in the following example:

perl viat.pl -t C -M | perl vhtml.pl > currier.htm


NOTES TO UN*X USERS

There is a hashbang line in the script, expecting the Perl interpreter
in  /usr/bin/perl. If this  path does  not reflect  your installation,
change this line or (if you  are root on the system) create a symbolic
link to the Perl interpreter in /usr/bin.

Some of my  scripts are written under MS/DOS and  did not have correct
unixoid end-of-line  markers. If this  causes problems, convert  it by
the tool of  your choice. A typical problem is  the error message "Bad
interpreter" when  you're executing the script direct  from your shell
prompt - but  no error at all when you call  the perl interpreter with
the script name  as parameter. It is caused by  the invisible ASCII 13
at the end of the hashbang line.


LICENSE

All tools are published under the  terms of the 2nd version of the GNU
General Public License. See COPYING for details.
