[Oberon] Re. Web usage

[easlab at absamail.co.za]

shark at gulfnet.sd64.bc.ca wrote:
 > Given a URI to a file, is there any direct way of 
> retrieving the file by http?  I know about 
> HyperDocTools.Fetch but it needs a link.

System.Directory HyperDoc*  ==
HyperDocTools.Obj
HyperDocs.Obj
HyperDocs.Panel
HyperDocs.Tool

All this stuff seems to be lacking documentation ?
There is a mistaken idea that Panels replace documentation/Tools.

Desktops.OpenDoc"http://URL" will fetch the file, if the
server serves it.
Apparently  a file named *.htm(l) will automagically be
converted to 'oberon rendering' and the original
'ascii string' will be lost. 
Perhaps n-o can also store the original 'ascii string' ?
For <non>.htm(l) the file is storable as received.

I think Oberon.Text :
MIME = {
...
	"application/GZip" = tar.gz
	"text/html" = html
	"text/plain" = txt
...
controls the actions ?

> Given a URI to an http accessible directory, is 
> there any way to get a list of the contents rather
> than a view of index.html?

Interesting:
http://www.complang.tuwien.ac.at/
is apparetly a directory, which probably contains  several files.
My guess is that if you address the server via the http protocol,
you cannot see the directory contents. 
Perhaps because http originated for use with htm(l) ie. hyperlinking,
where directory structure is 'redundant' ?
I'll ask on Newsgroups: comp.os.linux.networking if I get answers to
my other similar questions.

> How are cookies handled?  I see the command 
> HTTPDocs.ListCookies and wonder whether I should 
> be using it.

Apparently cookies is stuff written to YOUR file by the server,
and read-back by the server on later connections.

== Chris Glur.