Re: LYNX-DEV news: handling

[David Woolley]

> Is there a standard, or even common practice somewhere that implements
> retrieve by article number?  If so, exactly what notation does it use?

I assume we are talking URLs, as most newsreaders do it as a matter of 
course.

You don't need a standard, although, from memory, the nntp: URL
format uses //host/group/number, and Mosaic and Lynx, at least, use
article number ranges internally with news: to build the index (which
is not, I believe, in the old URL standards).  As the nntp: URL is being
merged with news: because of the popular misuse, it would probably be
worth checking the merged news: definition.  However, the obvious extension
of the internal Lynx processing would be for a single number after the
newgroup name to retrieve the article, rather than a one line index.

If it's not clear, you don't need a standard because we are talking about
internally generated URLs which are consumed by the same browser.

> And where is it useful?  Could you spell out what it is you are envisioning?
> With more detail, perhaps some student here on this list, in search of
> a fall project, might byte off implementing the feature you are describing.

I've not looked at the code, but I think it ought to be an easy change.  The
only obvious gotcha is /s in a message ID, but these ought not to be after the
@, and there ought to be no @s in the by number format.  The other thing, if
Lynx caches the connection, is to also cache the group selection - a good
idea even for indexes.

Note that links built from the contents of articles still need to use
message-IDs, so you still need to have a properly configured server.
;
; To UNSUBSCRIBE:  Send a mail message to address@hidden
;                  with "unsubscribe lynx-dev" (without the
;                  quotation marks) on a line by itself.
;