================
Scout version 1
================

This software is provided As-Is, with no warranty whatsoever.
If your computer dies, or your ISP excommunicates you... Don't blame me -
neither happen when I use it! (Not yet anyway)

===========
What is it?
===========
Scout is a basic Web browser, written in Tcl/Tk.
The aim was to have a 'no-frills' browser, which would allow me to:

	a: Scout around web sites, looking for things of interest.
	b: Perform an HTTP download of various archive types.
	c: Perform an FTP download of a file from a link on an HTTP site which
	   points at a file on an FTP server.

That's where it is now.

===================
Where it came from.
===================
It started when I downloaded the htmlwidget from Scriptics.After compiling it, I
found there were a couple of example apps produced, one of which was an Html
file viewer called HV.
This demonstrated the capabilities of the widget very well, and started me
thinking...

The http package is, of course, included with Tcl/Tk.

Including the Img1.2 package extended the range of images which could be
displayed - making it more practicable for browsing.

After a couple of absent-minded attempts to follow a link to an FTP server (not
recommended!), the FTP1.2 package was downloaded, and a very simple routine
written - mainly so I could pretend I was clicking on those links deliberately!

I used the ComboBox from the BWidgets package, but this could be replaced with
anything else that fits.

===========
The Basics.
===========
Entering a url causes a cache directory to be created, and the page downloaded
to an 'index.html' file. This is loaded into the htmlwidget, which parses the
file, and downloads the images from the site to the cache.The url is saved to a
file in the cache, and the page displayed.

Clicking on a link repeats the process - unless it points to an archive file,
or a file on an FTP server, in which case the file is brought down to the
download directory.

The 'Back/Forward' buttons navigate through the cache - redisplaying the data
stored there. They also reload the url from the cache into the htmlwidget,
which enables the built-in resolver to handle relative url's. This means you
sometimes have to back up through several pages, in a straight line, to get
back to the page you last jumped from. However, this is done so quickly, that
I don't find it too much of a hassle. ( Something to work on, maybe.)

Cache directories mount up, of course, but short scouting trips don't take up
too much disk space - and they can be deleted later. ( Add a cache delete/reset
button?)

At the moment, the 'Bookmarks' button loads a hand-written bookmark file kept
in an accessible 'localhost' location.

============
My Platform.
============

A home-brew P3, with 128 M.
SUSE 6.4 running Enlightenment 0.16.4.
The GUI was hacked about with tkBuilder 1.0.2.
Editing was done with ASED v2.0.
Tcl/Tk version is 8.3.17.

========
Licence.
========

Basically... Anyone is free to use, work on, change, or improve this script as
they like. The original intention was to keep it small, and reasonably simple.
Adding on things like mail and news readers, would probably be veering away
from this objective... I didn't set out to make another Netscape - just a
useful tool, that didn't take up too much space.

But then... Who am I to tell you how to spend your time? ( Look how I spend
mine!!)

If anyone does decide to make improvements, I would appreciate hearing about it.

mailto: harry.byers@ic24.net


Have fun....... Harry Byers.


