I wrote a gtk2-perl application, need advice

[Mitchell Laks <mlaks post harvard edu>]

Hi,

I wrote a gtk2-perl application (cybergaon) that I would like to distribute via sourceforge. It will be GPL 
(v3 I think).

I have a number of questions. This is my first application and it is a little messy. 

The application takes as input an archive of hebrew texts (such as the Bible and Mishna and Talmud and works 
of the 
medieval sage Maimonides) that are freely distributed on the web. That archive of texts, which store 
the texts with the  hebrew written in old IBM PC hebrew encoding CP862, also comes with an old Dos program to 
do searches for text. www.mechon-mamre.org/mtrpromo.htm.

1. I converted the directories of files of CP862 text to files of utf-8 (a simple perl script employing the 
iconv program).

2. I used gtk2-perl to essentially wrap grep. One can search for a REGEXP in the textentry 
and then there is a notebook with 2 tabs: 
in one tab I display the list of results in a treeview
and then I display full corresponding text in a textview in the other tab.
Thus when we click on a result in the list in the treeview, we display the full text of the corresponding 
file in the textview,  with the text prescrolled to the proper location in the file in the textview .

3. This is of course a general purpose grep display, and might be useful for others.

I am not sure how to set this up as an application for others to use. Ie: it is not structured right.
At present I just search through a list of subdirectories in my current directory.

I do my development on my debian sid box.


My problems:


1) A professional software engineer would know where to put things in a /usr/share and /usr/bin. I have all 
the code 
and data  in my current directory and subdirectories. What is the next step?

Perhaps I want to write my data to /usr/share/Cybergaon/MTR/*, and the application to /usr/bin/cybergaon.pl
or something like /usr/share/local/bin/cybergaon.pl. Suggestions?

I think that I can redistribute the raw data - ie the hebrew texts as they are free. However I sense
that sourceforge is not happy to have data rather than programs distributed via there site. 
On the other hand the texts are freely redistributable. Do I distribute a script that assumes the user
has downloaded the data file of texts from the mechon-mamre website?

The only modules I use are Gtk2 Gtk2::Pango Glib and Cwd.

2) I wanted to put up the application and code on sourceforge and I can't understand how to get it onto cvs. 
I am having
difficulty figuring out how to do the ssh-keygen and I really dont know what I am doing over there, and 
nothing
seems to be going up to sourceforge. Note that i have used ssh-keygen when I wanted to set up
a cron job to anonymous backup between machines in the past. I just cant  figure out what to do here. I would 
like
to zero my key and start over and upload the files for others to use.

Thanks,

Mitchell Laks