Message-Id: <9403231646.AA01231@ulua.hal.com>
To: mitra@pandora.sf.ca.us (Mitra)
Subject: LISP for Complex URC Sytax [WAS: Re URC func spec ]
In-Reply-To: Your message of "Wed, 23 Mar 1994 03:41:40 GMT."
<Cn3M9H.2tn@pandora.sf.ca.us>
Date: Wed, 23 Mar 1994 10:46:52 -0600
From: "Daniel W. Connolly" <connolly@hal.com>
In message <Cn3M9H.2tn@pandora.sf.ca.us>, Mitra writes:
>
>For example a URC might be
>
>URN: path.net/mitra/1234
>Author: Mitra <mitra@path.net>
>OldVersion: URN:path.net/mitra/1233
>RelatedDoc: URC:URN: path.net/mitra/1232 Author: Fred
>
>I'm not sure how we do the escaping etc and syntax, but think this is a
>reasonable example of the content being conveyed.
I think all this parsing, syntax, escaping and quoting is getting a little
out of hand. How many parsers is it going to take to implement these
Internet Information Architecture applications?
So far, the players are:
SGML parsers
RFC822 header parsers
UR*-of-the-day parsers
Why don't we adopt the lisp printer syntax and be done with it? At
least as a canonical representation. The information in the
s-expressions can be translated to other formats (e.g. RFC822 headers)
as necessary. But lisp syntax would make the structure clear and delimit
the "visible" information from the "opaque" information.
So the above example might look like:
(URC :URN (URN "path.net/mitra/1234")
:Author "Mitra <mitra@path.net>"
:OldVersion (URN "path.net/mitra/1233")
:RelatedDoc (URC :URN (URN "path.net/mitra/1232") :Author "Fred"))
Or better yet, if we devised a LISP syntax for URNs, mail addresses, etc...
(URC :URN (URN "path.net" mitra 1234)
:Author (Address :friendly "Mitra" :mbox "mitra@path.net")
:OldVersion (URN "path.net" mitra 1233)
:RelatedDoc (URC :URN (URN "path.net" mitra 1232) :Author "Fred"))
Dan