To: Larry Masinter <masinter@parc.xerox.com>
Subject: Re: FTP URL mapping
In-Reply-To: Your message of Thu, 03 Mar 1994 18:19:50 -0800.
<94Mar3.181958pst.2732@golden.parc.xerox.com>
Date: Fri, 04 Mar 1994 00:35:25 -0500
From: John Curran <jcurran@nic.near.net>
Message-Id: <9403040536.aa17614@nic.near.net>
--------
] From: Larry Masinter <masinter@parc.xerox.com>
] Subject: Re: RE:FTP URL mapping
] Date: Thu, 3 Mar 1994 18:19:50 PST
]
] In an attempt to get some movement on this topic, I thought I would
] pursue a concrete proposal, that extends what I thought Keith Moore's
] proposed a while ago. Interpret:
]
] ftp://host.edu/a/b/c/d/e
]
] as
]
] ftp host.edu
] cwd a
] cwd b
] cwd c
] cwd d
] type binary
] retr e
]
] if error (no such file), then:
]
] nlist e
] ...
Interesting proposal. Salient features:
Provides for algorithmic mapping of parameter package
Allows for multiple "CWD" commands
Provides escape mechanism for retrieval optimization
I don't exactly love the proposal, but it's hard to judge since we still
lack the requirements. I asked earlier whether more than one CWD command
was necessary to retrieve any file. Obviously, the syntax could be made
friendlier if only one CWD command was necessary. For example:
ftp://cnri.reston.va.us/internet-drafts/draft-ietf-uri-url-01.txt
"OPEN cnri.reston.va.us, RETR internet-drafts/draft-ietf-uri-url-01.txt"
versus
ftp://cnri.reston.va.us:internet-drafts/draft-ietf-uri-url-01.txt
"OPEN cnri.reston.va.us, CWD internet-drafts, RETR draft-ietf-uri-url-01.txt"
On the other hand, there is no way to make a recommendation on the encoding
of the FTP URL until we know what the requirements are (i.e. one or multiple
CWD command support, types support, structure mode?, etc.)
/John