Ftp-WG: Telnet end-of-line
Pat LaVarre <[email protected]> Mon, 20 May 2002 14:24:56 -0500
| Newsgroups | gmane.ietf.ftpext |
|---|---|
| Message-ID | <[email protected]> |
Gregory L:
Yes please, I too am interested enough to read
something long.
My eye was first caught here by the mlst-nn change to:
pathname = utf-8-name / raw
from the rfc959:
<pathname> ::= <string>
of "any of the 128 ASCII characters except <CR> and
<LF>".
I took the lack of response to my last post, copied
below, to mean maybe the http://w3.hethmon.com/ftpext/
archives explain this change in full.
But in my newbie ignorance, I haven't yet found a
credible answer complete & precise enough to
implement.
Pat LaVarre
P.S. I have no idea what BCP is.
-----Original Message-----
From: Jeffrey Altman [mailto:[email protected]]
Sent: Mon 5/20/2002 12:42 PM
...
I would be interested in reading the doc to ensure its
accuracy.
... Thanks for writing it.
-----Original Message-----
From: Gregory A Lundberg [mailto:[email protected]]
Sent: Mon 5/20/2002 12:32 PM
To: FTPEXT Working Group
Cc:
Subject: Ftp-WG: Telnet end-of-line
I've got the beginnings of a BCP-style document which
explains the proper use of end-of-line, and leads to
the proper use of Telnet end-of-line, in the FTP.
Since it's a bit long, rather than post it, I thought
I'd first see if anyone is interested in seriously
discussing the issue. If so, I can make the document
available via anonymous FTP in both nroff and txt
formats.
-----Original Message-----
From: Pat LaVarre [mailto:[email protected]]
Sent: Mon 4/29/2002 5:42 PM
To: [email protected]
Cc:
Subject: Ftp-WG: FEAT for NUL CR LF etc.?
> > It just bugs me
> > that we can support any file, directory
> > or path name EXCEPT those
> > which include [Unicode x0..1F]
Bugs me too.
Do we see enough divergence in reasonable
interpretation here to say mlst-16 should include a
concrete example, one command of which has a
utf-8-name pathname that contains Ctrl+M CR, Ctrl+J
LF, Ctrl+@ NUL, and Ctrl+S XOFF?
> > unambiguous on paper ... no?
> >
> > For example, [an Ftp client]
> > could send the six byte
> > command "STOR \r" as the nine bytes
> > "STOR \r\0\r\n" to create a file
> > named by the one byte string
> > containing only x0D CR.
...
> From: Robert Elz [mailto:[email protected]]
> Sent: Wed 4/24/2002 11:33 PM
...
> If you want to put a CR in a file name,
> ... exactly as ... suggested.
> If you happen to be talking to a broken
> implementation, then you might not be able
> to put a CR in a file name
> - the answer is to fix the implementation,
> not to change the way FTP is defined.
Let's suppose in real life Ftp is normally Not robust
enough to handle x0A LF, x0D CR, x00 NUL, x13 Ctrl+S
XOFF, x11 Ctrl+Q XON, and the rest of UTF-8 codes
x00..1F ...
Should we then add protocol to help more compliant
clients and servers recognise each other?
We don't want servers and clients just volunteering
pathnames as arbitrary as mlst-15 allows in lines of
the LIST NLST MLST responses and as CWD RETR STOR MKD
RNTO arguments, do we?
Yours in breathtaking (refreshing?) ignorance,
Pat LaVarre
__________________________________________________
Do You Yahoo!?
LAUNCH - Your Yahoo! Music Experience
http://launch.yahoo.com