[mule-cgreek:00800] [rasmith@aristotle.tamu.edu: Re: Cgreek package]

[TAKAHASHI Naoto <ntakahas@xxxxxxxxx>]

Robin Smith $B;a$+$i$N%a!<%k$rE>Aw$7$^$9!#(B

$B9b66(B

------- Start of forwarded message -------
To: TAKAHASHI Naoto <ntakahas@xxxxxxxxx>
Subject: Re: Cgreek package 
In-Reply-To: Your message of "Mon, 03 Jul 2000 16:04:23 +0900."
             <200007030704.QAA15065@xxxxxxxxxxxxxxxx> 
Content-Type: text/plain; charset=US-ASCII
Date: Mon, 03 Jul 2000 06:25:24 -0500
From: Robin Smith <rasmith@xxxxxxxxxxxxxxxxxx>

>>>>> "TAKAHASHI" == TAKAHASHI Naoto <ntakahas@xxxxxxxxx> writes:

    >> Finally, it would be useful to have a search command that
    >> searched for a string of letters, ignoring breathings and
    >> accents.  Is your group interested in that?

    TAKAHASHI> Probably.  Have you already written such a program?

No, I haven't.  I've instead been continuing to modify your tlg2emacs
to parse references as well as text.  The version I have at present is
a little cleaner than what I sent you (lots of rules can be consolidated,
and a more careful reading of the documentation let me drop a few that
cover cases which aren't supposed to occur) and has these additions, 
among others:

 -A single function, 'setlevel', now both updates citation counters
and prints citations to the output stream.
 -Printing of citations is controlled by an array of format strings (one
element for each IDT level).  Different arrays can be selected by command
line argument, ranging from "print every value on a line by itself" at
the verbose end to "just emit a newline" (thus no citations at all are
printed: same behavior as the original tlg2emacs)
 -As this implies, I have built in a little argv parser and added a variety of
options:
    -c          choose a citation print level (various values possible)
    -f filename (obvious)
    -l string   value for lc_g2 (old functionality)
    -s string   seek for citation (still working on the details here)
    -o string   open file at block (atoi(string)*8192)

The -s function turms off text output until a match for the reference is
found (comparisons of values are done in accordance with the TLG's
definition of <=>).  The -o function is meant to go with the interpretation
of .idt files, which give 8K block offsets for works and references (I think
I told you that I have a utility that will parse these).

    TAKAHASHI> people?  I am an official TLG CD-ROM licensee, but have
    TAKAHASHI> not heard anything about the availability, even the
    TAKAHASHI> existence, of such documentations from the University
    TAKAHASHI> of California.

I just sent them a message (there's an indication on their web page that
you can ask for more information) saying I was working on a program, and they
sent me the documentation.  

    TAKAHASHI> By the way, I am planning to include classical Greek
    TAKAHASHI> support in Emacs-21.  The character set, however, will
    TAKAHASHI> be based on Unicode.  So far I have finished a variable
    TAKAHASHI> width font, file I/O routines using the UTF-8 format,
    TAKAHASHI> and an input method.  Now I am writing a TLG to Unicode
    TAKAHASHI> converter.  Support for Ibycus4 will be the next.  I
    TAKAHASHI> will keep you informed about the progress.

Wonderful.  I'll look forward to it.

Robin Smith              
Department of Philosophy            rasmith@xxxxxxxx
Texas A&M University                Voice (409) 845-5696
College Station, TX 77843-4237      FAX   (409) 845-0458
------- End of forwarded message -------