Received: from localhost by CS.UTK.EDU with SMTP (cf v2.9s-UTK) id MAA26136; Mon, 11 Mar 1996 12:43:28 -0500 Received: by CS.UTK.EDU (bulk_mailer v1.4); Mon, 11 Mar 1996 12:42:34 -0500 Received: from leftbank.com by CS.UTK.EDU with SMTP (cf v2.9s-UTK) id MAA26072; Mon, 11 Mar 1996 12:42:32 -0500 Received: from elbereth.leftbank.com.leftbank by leftbank.com (4.1/Leftbank1.0) id AA28009; Mon, 11 Mar 96 12:43:30 EST From: cos@leftbank.com (Ofer Inbar) Message-Id: <9603111743.AA28009@leftbank.com> Subject: Re: Message format document outline To: drums@cs.utk.edu Date: Mon, 11 Mar 1996 12:44:00 -0500 (EST) In-Reply-To: <199603110654.BAA20559@wilma.cs.utk.edu> from "Keith Moore" at Mar 11, 96 01:54:40 am Organization: The Left Bank Operation - http://www.leftbank.com/ Reply-To: cos@leftbank.com X-Mailer: ELM [version 2.4 PL24] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Several years ago I began writing scripts to process mail for mailing lists, mail-news gateways, and the like. I'd read RFC822, and at first all my scripts were written to check for two consecutive line terminations as the end-of-headers marker. However, I soon discovered that this was failing for some messages, and rewrote the scripts to test for any line containing nothing but whitespace as end-of-headers. Some of these scripts are still in use today, and this aspect has yet to fail in any serious way. In other words: a) The number of messages which had whitespace in the end-of-headers marker was significant enough to force me to "fix" the scripts. b) There have been a few messages since then with blank continuation lines (or some other "bug" has caused a blank line in the middle of the headers), but they have been very rare. c) Failure to recognized end-of-headers, in the first case, would cause a serious failure - The entire message was interpreted as headers, resulting in many cases in the body of the message not being forwarded properly. d) Failure in the second case, seeing an end-of-headers marker where one was not intended, has never caused a serious failure. At worst, a Subject: or To: line ended up in the body of the message, resulting in a message with no subject, or an Apparently-To line, messing with people's filtering rules perhaps. But it has never caused a message body not to be forwarded to its destination. Any mail software that I write today will definitely interpret the first blank line, with or without whitespace, as end-of-headers, because that's what works. I think RFC822bis should say that too. -- Cos (Ofer Inbar) -- cos@leftbank.com cos@cs.brandeis.edu -- The Left Bank Operation -- lbo@leftbank.com http://www.leftbank.com "We all misuse the net for personal gain, one way or another." -- Larry Wall