ietf-nntp Re: Folding

Charles Lindsey chl at clerew.man.ac.uk
Tue Jul 29 11:33:36 PDT 2003


In <20030729105805.GA62343 at finch-staff-1.thus.net> "Clive D.W. Feather" <clive at demon.net> writes:

>Combining Russ's and Charles's responses, here's what I propose doing.
>Scream now or forever hold your peace.

>> (1) Can a header line be folded between the colon and the first space?
>> (3) Can a header line be folded twice between two non-white space
>>     characters, so that a post-folding line contains only white space?

>Either side of the protocol MUST accept these from the other. They SHOULD
>NOT be sent, but a process receiving an article of this form from elsewhere
>is not required by this document to re-fold.

I would have made that SHOULD rather than MUST. I suspect many existing
implementations would not cope.

Also, I think NNTP servers should NEVER refold anything, certainly not
when relaying. Always pass on whatever was received, or drop it on the
floor (with 5xx maybe).

As a slight exception, the injecting agent hiding behind the POST command
has a little more leeway, according to Usefor. However, the NNTP protocol
as such should not be doing any folding (the injecting agent is best
regarded as a subcontractor for the POST command which takes its rules
from Usefor/wherever).

-- 
Charles H. Lindsey ---------At Home, doing my own thing------------------------
Tel: +44 161 436 6131 Fax: +44 161 436 6133   Web: http://www.cs.man.ac.uk/~chl
Email: chl at clerew.man.ac.uk      Snail: 5 Clerewood Ave, CHEADLE, SK8 3JU, U.K.
PGP: 2C15F1A9      Fingerprint: 73 6D C2 51 93 A0 01 E7 65 E8 64 7E 14 A4 AB A5



More information about the ietf-nntp mailing list