Re: Client-Side Cookies

Dear Mr. Michael Zick,

  You are absolutely correct.  It appear that "tail matching" would be much 
simpler if the Jigsaw's ordering of the domain "parts" was reversed.  So, 
parts[0] would be "com", "org" "net" or whatever (instead of "www", "w3" or 
whatnot).  Then only parts[0] and parts[1] would be considered when sorting 
or looking up cookies (actually, according to the Netscape Cookie 
Specification, if the domain is something like the form "foo.bar.com.fr" or 
"jed.bush.gov.fl.us", the last three parts are supposed to used for "tail 
matching").

  The next time I start digging into the CookieFilter source code, I will 
try to make these changes (however, right now my Client-Side Jigsaw 
components are working good enough for my application and my attention lies 
elsewhere).

  Thank you very much for your input.

  Truly yours,
  John Philip Anderson


>From: "Michael S. Zick" <mszick@altavista.net>
>Reply-To: mszick@altavista.net
>To: John Philip Anderson <jpanderson_215@hotmail.com>
>CC: www-jigsaw@w3.org
>Subject: Client-Side Cookies
>Date: Tue, 19 Dec 2000 14:22:33 -0500 (EST)
>
>Dear Mr. Anderson,
>
>Having followed this discussion, I have the following to suggest:
>
>Wouldn't it be better to change "domainParts()" so that it returned
>the "parts" array in the order UR(L/I)'s are resolved?
>I.E...
>for <www.foo.bar.org>
>parts[0] == "org"
>parts[1] == "bar"
>parts[2] == "foo"
>parts[3] == "www"
>or for <anderson.net>
>parts[0] == "net"
>parts[1] == "anderson"
>
>This would be a more "intuitive" mapping between variable, multi-part
>strings and a variable length (elements) array.  This way, the array would
>never come up empty if the UR(L/I) has at least one part.
>
>Mike
>
>

_________________________________________________________________
Get your FREE download of MSN Explorer at http://explorer.msn.com

Received on Tuesday, 19 December 2000 17:18:25 UTC