ASID Working Group Martin Hamilton
INTERNET-DRAFT Loughborough University
July 1996
WHOIS++ URL Specification
Filename: draft-ietf-asid-whois-url-00.txt
Status of This Memo
This document is an Internet-Draft. Internet-Drafts are working
documents of the Internet Engineering Task Force (IETF), its
areas, and its working groups. Note that other groups may also
distribute working documents as Internet-Drafts.
Internet-Drafts are draft documents valid for a maximum of six
months and may be updated, replaced, or obsoleted by other
documents at any time. It is inappropriate to use Internet-Drafts
as reference material or to cite them other than as ``work in
progress.''
To learn the current status of any Internet-Draft, please check
the ``1id-abstracts.txt'' listing contained in the Internet-Drafts
Shadow Directories on ds.internic.net (US East Coast),
nic.nordu.net (Europe), ftp.isi.edu (US West Coast), or
munnari.oz.au (Pacific Rim).
Distribution of this memo is unlimited. Editorial comments should
be sent directly to the author. Technical discussion will take
place on the IETF ASID mailing list - ietf-asid@umich.edu.
This Internet Draft expires January 25th, 1997.
Abstract
This document defines a new Uniform Resource Locator (URL) scheme
"whois++", which provides a convention within the URL framework for
referring to WHOIS++ servers and the data held within them. It does
not specify a standard.
1. Overview of the WHOIS++ protocol
RFC 1835 [1] defines a simple Internet directory protocol known as
WHOIS++. In order that WHOIS++ may be used within the Uniform
Resource Locator (URL) framework defined by RFC 1738 [2], a URL
scheme definition for WHOIS++ is necessary. This document specifies
a URL scheme "whois++", for use with the WHOIS++ protocol.
[Page 1]
INTERNET-DRAFT July 1996
WHOIS++ is text based protocol after the fashion of many popular
Internet application protocols, such as SMTP [3] and FTP [4].
Although the protocol is TCP based, WHOIS++ is effectively stateless
- no state information is preserved across requests, there is no
concept of a session per se since each request/response pair is
self-contained, and there is no "login" phase.
WHOIS++ transactions normally consist of a single request from the
client and response from the server, followed by the TCP connection
between the two being torn down. Use of the "hold" constraint in the
WHOIS++ request makes it possible for the client to indicate that it
would like to keep the TCP connection open for more than one request/
response pair, but whether this is actually done is at the discretion
of the server.
2. WHOIS++ URL specification
The following information is necessary for a WHOIS++ client to
formulate and deliver a request:
o the domain name or IP address of the server to contact
o the port number of the server (63 by default)
o the request itself - normally a single line of text
This is a good match with the generic Uniform Resource Locator (URL)
scheme specified in RFC 1738. So, a URL of the following form would
seem to be appropriate:
whois++://host[:port][/<request-specification>]
Using the BNF grammar defined in RFC 1738, this could be written as:
whoisppurl = "whois++://" hostport [ "/" whoisppsrch ]
where
whoisppsrch = *uchar
The definitions for hostport and uchar are imported from RFC 1738:
hostport = host [ ":" port ]
uchar = unreserved | escape
These in turn depend upon the following:
unreserved = alpha | digit | safe | extra
alpha = lowalpha | hialpha
digit = "0" | "1" | "2" | "3" | "4" | "5" | "6" | "7" |
[Page 2]
INTERNET-DRAFT July 1996
"8" | "9"
safe = "$" | "-" | "_" | "." | "+"
extra = "!" | "*" | "'" | "(" | ")" | ","
lowalpha = "a" | "b" | "c" | "d" | "e" | "f" | "g" | "h" |
"i" | "j" | "k" | "l" | "m" | "n" | "o" | "p" |
"q" | "r" | "s" | "t" | "u" | "v" | "w" | "x" |
"y" | "z"
hialpha = "A" | "B" | "C" | "D" | "E" | "F" | "G" | "H" | "I" |
"J" | "K" | "L" | "M" | "N" | "O" | "P" | "Q" | "R" |
"S" | "T" | "U" | "V" | "W" | "X" | "Y" | "Z"
escape = "%" hex hex
hex = digit | "A" | "B" | "C" | "D" | "E" | "F" |
"a" | "b" | "c" | "d" | "e" | "f"
BNF for the WHOIS++ request format is defined in Appendix F of RFC
1835. This can contain characters which may confuse software which
deals with whois++ URLs, notably spaces and characters drawn from
non-ASCII character sets such as the UTF-8 variant of Unicode [5,6].
Hence, the usual rules about hex-escaping illegal and reserved
characters should apply - and the definiton of the WHOIS++ request as
"uchar". Note that the default WHOIS++ port of 63 should be used if
the port number component of the "hostport" construction is left out.
Global constraints such as authentication information, language and
character set preferences may be expressed as part of the WHOIS++
request. Consequently it is not thought necessary to specify them
separately in a mechanism such as the "user@host" construction
defined for the FTP URL.
Most WHOIS++ requests can be expected to consist of a single line of
text, followed by carriage return and line feed characters. It
should, however, be noted that it may be necessary to encode multi-
line requests within WHOIS++ URLs. Software which implements whois++
URLs should either be capable of handling this, or fail gracefully.
3. Examples
The whois++ URL scheme defined above should make it possible to write
URLs for any of the following:
(a) a reference particular WHOIS++ server, without implying
that a search should be done
(b) a "canned" search of a particular server
(c) individual objects within a server
Case (a) simply requires that the host and optionally the port number
be specified, e.g.
[Page 3]
INTERNET-DRAFT July 1996
whois++://acm.org/
or
whois++://acm.org:63/
When given a whois++ URL of this format, implementations may choose
to present the user with a search form or dialogue, contact the
server for information about which WHOIS++ options it supports, and
so on. The WHOIS++ default port 63 should be used if the port number
is not specified.
Case (b) requires a search specification to be present, e.g.
whois++://acm.org/name=phil%20and%20name=zimmerman
This may be sent verbatim to the server, once hex escaped chars in
the URL have been converted back to normal, e.g.
name=phil and name=zimmerman
Case, (c) is effectively an instance of (b). This may be implemented
as a search where the request consists of the WHOIS++ "handle" of the
requested object, e.g.
whois++://acm.org/handle=number6
4. Global constraints
Although there are no global constraints specified in these last two
URLs, the WHOIS++ client may choose to add global constraints of its
own, e.g. use of the "hold" constraint to request that the
connection be held open for a further request.
If in addition, global constraints are part of the URL, this can
easily be recognised by the presence of a colon ":" immediately after
the slash "/" which separates the host and port information from the
search specifier, e.g.
whois++://acm.org/:authenticate=password;name=foo;password=bar
At the implementor's discretion, the client may choose to pass these
global constraints on in any queries which are passed to this server,
e.g. if this URL was used in a search for "zimmerman", the request
passed to the server might be either of
zimmerman
[Page 4]
INTERNET-DRAFT July 1996
or
zimmerman:authenticate=password;name=foo;password=bar
or "zimmerman", followed by some combination of the global
constraints specified in the URL and other global constraints
introduced by the WHOIS++ client.
5. Compatibility with WHOIS and RWhois
The three protocols in the WHOIS family, NICNAME/WHOIS [7], WHOIS++,
and RWhois [8], are not particularly similar. WHOIS++ and RWhois use
different request and response formats, and have different well-known
port numbers. WHOIS responses are assumed to be plain text and human
readable.
Consequently, this document has not attempted to define a single URL
scheme for use with all three protocols.
6. World-Wide Web integration
These whois++ URLs may be used as hyperlinks in HTML [9] documents,
though it should be noted that the relative URL syntax defined in RFC
1808 [10] is not appropriate for use in these links. This is because
WHOIS++ requests do not map conveniently onto the generic resource
locator syntax used for relative URLs - the syntactic conventions
used in writing a WHOIS++ request are very different from those of
the generic resource locator.
The WHOIS++ protocol and the whois++ URL lend themselves to
implementation via a proxy HTTP [11] gateway, since the information
necessary to contact the server and deliver the request is embedded
within the URL itself. A simple proxy gateway has been implemented
which takes an HTTP "GET" request containing a whois++ URL, carries
out a WHOIS++ transaction and returns the results formatted as HTML.
This will probably be the preferred approach to providing WHOIS++
support by proxy for some time - there is no Internet Media Type
(aka. MIME content-type) registered for WHOIS++ responses as yet.
The proxy server implementation can be found at
<URL:http://www.roads.lut.ac.uk/pickup/>
It does not appear to be appropriate to use any HTTP methods other
than "GET" with whois++ URLs, and there does not appear to be any
value in using whois++ URLs in HTML forms.
The appearance of the "+" character in the protocol scheme component
of a URL is legal, according to RFC 1738. The author has lingering
[Page 5]
INTERNET-DRAFT July 1996
doubts about the ability of all software which processes URLs, for
example in parsing HTML documents, to cope with this character. No
evidence has been found to back these doubts up, however.
7. Security Considerations
Client software should check both the contents of the whois++ URL and
the results returned from WHOIS++ search requests for any unsafe
characters and character strings.
It is possible to embed requests for other protocols within this URL
format. This is an approach which may be used to defeat security
schemes, spoof protocols, and so on. Implementors should consider
requiring user confirmation when requests are directed to reserved
ports (i.e. those less than 1024) other than 63 and 43, or well-
known ports in the unreserved range.
Finally, implementations should take care not to cache authentication
information.
8. Acknowledgements
Thanks to Jeff Allen, Lorcan Dempsey, Patrik Faltstrom, Jon Knight,
William F. Maton, Larry Masinter, and Scott Williamson for their
comments on draft versions of this document.
This work was supported by grants from the UK Electronic Libraries
Programme (eLib) and the European Commission's Telematics for
Research Programme.
9. References
Request For Comments (RFC) and Internet Draft documents are available
from <URL:ftp://ftp.internic.net> and numerous mirror sites.
[1] P. Deutsch, R. Schoultz, P. Faltstrom and C.
Weider. "Architecture of the WHOIS++ service", RFC
1835. August 1995.
[2] T. Berners-Lee, L. Masinter and M. McCahill (eds).
"Uniform Resource Locators (URL)", RFC 1738.
December 1994.
[3] J. Postel. "Simple Mail Transfer Protocol", RFC
821. August 1982.
[Page 6]
INTERNET-DRAFT July 1996
[4] J. Postel, J. K. Reynolds. "File Transfer Proto-
col", RFC 959. October 1985.
[5] The Unicode Standard, Worldwide Character Encoding,
Version 1.0, Volume 1, Addison-Wesley, 1990. ISBN
0-201-56788-1.
[6] The Unicode Standard, Worldwide Character Encoding,
Version 1.0, Volume 2, Addison-Wesley, 1992. ISBN
0-201-60845-6.
[7] K. Harrenstien, M.K. Stahl, E.J. Feinler.
"NICNAME/WHOIS", RFC 954. October 1985.
[8] S. Williamson & M. Kosters. "Referral Whois Proto-
col (RWhois)", RFC 1714. November 1994.
[9] T. Berners-Lee, D. Connolly. "Hypertext Markup
Language - 2.0", RFC 1866. November 1995.
[10] R. Fielding. "Relative Uniform Resource Locators",
RFC 1808. June 1995.
[11] T. Berners-Lee, R. Fielding, H. Frystyk. "Hyper-
text Transfer Protocol -- HTTP/1.0", RFC 1945. May
1996.
10. Author's address
Martin Hamilton
Department of Computer Studies
Loughborough University of Technology
Leics. LE11 3TU, UK
Email: m.t.hamilton@lut.ac.uk
This Internet Draft expires January 25th, 1997.
[Page 7]
