ietf-nntp OVER extension

Charles Lindsey chl at clw.cs.man.ac.uk
Thu Jul 4 03:54:47 PDT 2002


In <20020703111232.GC95156 at demon.net> "Clive D.W. Feather" <clive at demon.net> writes:

>We need to agree what is and isn't acceptable as a "byte count". For
>example, the above text could be changed to "... each line MAY be counted
>as one byte instead of two", which would allow the sensible thing while
>still permitting existing practice.

Either the client or the server or both may be storing line endings
as one or two bytes. I don't think the client wants to know what the server
is doing. What the client would like to know is "If I send an ARTICLE
command, how many bytes are you going to send me? I accept that stuffed
dots don't count in that.

If the client is going to set aside a buffer to store the article, it is
safer to send the larger number (i.e. CRLF=2). A client which stores only
NL internally might compute bytes:count-lines:count to get a more
accurate estimate.

But mostly this facility will be used by clients whose BOFH wants to drop
all articles greater than 100Kbytes, in which case it really does not
matter.

But I could live with the "MAY count as 1 or 2", though I would prefer 2.

-- 
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 clw.cs.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