Re: Lines count in USEFOR
"Charles Lindsey" <[email protected]> Mon, 16 Nov 2009 12:45:46 GMT
| Newsgroups | gmane.ietf.usenet.format |
|---|---|
| Message-ID | <[email protected]> |
In <[email protected]> Russ Allbery <[email protected]> writes: >Julien ÃLIE <[email protected]> writes: >> If I read RFC 5322: >> body = (*(*998text CRLF) *998text) / obs-body >> Isn't the final *998text counting for a line? >> I believe the line count is the number of CRLF separators in the <body> >> PLUS ONE. >> (Unless I'm mistaken by the definition of the body?) >NNTP as presently written cannot transfer an article that does not end in >a CRLF. In practice, clients will just add a CRLF to any article that >doesn't already end in one, since otherwise they can't POST. Then I suspect USEFOR ought to have said that the CRLF at the end was REQUIRED. Do we really want to fix that now? And what if you send it by UUCP? The definition in Son-of-1036 is The line count includes all body lines, including the signature if any, including empty lines (if any) at beginning or end of the body. (The single empty separator line between the headers and the body is not part of the body.) The "body" here is the body as found in the posted article, AFTER all transformations such as MIME encodings. which doew not really answer the question. But then a final <eol> was REQUIRED by Son-of-1036. Looking at the implementation of Cnews, if you contrive to inject a body with no CRLF (which most user agents would not permit), the injection process will add a CRLF. But I suspect that I have also hacked a bug into my cnews, because if you check the Lines header on this message, I have a suspicion you will find it has counted my signature twice :-( . -- Charles H. Lindsey ---------At Home, doing my own thing------------------------ Tel: +44 161 436 6131 Web: http://www.cs.man.ac.uk/~chl Email: [email protected] 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