[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-core-devel
Subject:    Re: A new framework for embedding ... without CORBA
From:       Graham TerMarsch <gtermars () home ! com>
Date:       1999-10-01 22:30:18
[Download RAW message or body]

adawit@earthlink.net wrote:
[.....snip.....]
> > I'd looked at what'd really be involved in doing one, and didn't think that
> > it'd be too hard to do.  However, I did have question as to how NNTP URLs
> > would be represented, considering that NNTP itself has several other
> > extensions (e.g. 'xpat') that don't really map across into any sort of URL.
> > Other things such as "how do I just pull down the _headers_ for a single msg,
> > without the msg body" also left me as a bit puzzled as to how they'd map
> > properly.  I've read through the RFCs outlining NNTP URLs and for all that
> > they at least define how to get all of the subjects for a given group and a
> > message within the group, I didn't find anything else in there that'd be
> > useful for helping solve those issues.
> >
> > If someone out there has ideas on how these sorts of things could be mapped,
> > or that could point me in the direction of another kioslave that'd be a good
> > example to start with, I'd be willing to sit down and see what I can come up
> > with to create one for NNTP.
> 
> Well, I think I may be able to point you in the right direction with one of
> the things you mentioned up top.  The url stuff.  I am not familiar with the
> several url extensions that NNTP has, but I think some type of internal
> mapping is possible.  For example, the SMB protocol will probably use "smb://"
> or "smb:/" to denote a resource in a Windows server, however, a user will be
> allowed to type in \\HOSTNAME\SHARE.  A conversion is done as necessary from
> the \\HOSTNAME\SHARE format to the "smb:/" format internally or vise-versa.
> It won't matter which way you want to go as long as the resource identifier
> is unique.  Can you give some examples of the NNTP url extensions ?  I am only
> familiar with the NNTP url as it is defined in RFC 1738 ...

I'd been looking at both the NNTP url as defined in RFC1738, as well as the
proposed updated provided by RFC2396 (which as I understand it is still
considered draft, so do correct me if I'm barking up the wrong tree).  2396
and 1738 seemed to outline the basis of what the URL looks like, and I can see
that in the sense of wanting to either "get a list of all the articles in a
newsgroup" and "get me the full contents of a specific article in a newsgroup"
that the URL syntax they've provided was useful.

However, for the application that I'm working on I'm also going to be needing
to be able to do things like "get me all the 'x' header lines for the
newsgroup" (xhdr), as well as being able to do just a "head" request to get
the header information for a specific news article posting (e.g. Lines,
Subject, From, etc).  This is where I'm not quite sure how it fits together
right; presuming that a 'HEAD' request can be done (similarly to a 'HEAD'
request in HTTP), should that result in the value returned being the actual
header of the message itself?

I guess my whole wishy-washy-ness on this wasn't as to how the URL actually
gets turned into something that gets requested from the server, but in that I
already know something that I want to request from the server and haven't
quite figured out how (if at all) that maps back cleanly into a kioslave.  Did
that make any sense?

-- 
Graham TerMarsch

// -----------------------------------------------------------------
// God is real, unless declared integer.   
// -----------------------------------------------------------------

[prev in list] [next in list] [prev in thread] [next in thread]