On 2002-06-11, Rocco Rutte wrote: > Hi, > > * Christoph Bugel [02-06-11 22:21:30 +0200] wrote: > > My observation is that if someone with mutt-1.2.5 replies > > to a message by user1, it generates the following header: > > > In-Reply-To: <"from user1"@host1.org> > > The problem is that mutt cannot reliably distinct between a > message-id and a mail adress if both are given in angle > brackets. IIRC mutt assumes that a local part of a mail > address is at most 8 characters -- everything else is > considered to be a message-id. I don't have a better > solution.
hmmm... now that you mention it, yes, I did notice something that's connected to the string's length. but still, I thought that *anything* after the In-Reply-To: is supposed to be a message-id? Quote from RFC 2822: The "Message-ID:" field contains a single unique message identifier. The "References:" and "In-Reply-To:" field each contain one or more unique message identifiers, optionally separated by CFWS. So it seems that <"from user1"@host1.org> is not a valid thing to put after the In-Reply-To header, and since mutt-1.2.5?? does exactly that, I wonder if I'll have to live with broken threads until everyone will have stopped using mutt-1.2.5? > There's only one real solution (besides writing more robust > standards): only put those message-ids in the In-Reply-To > field you're really replying to. :-). Unfortunately, I don't have much control over what MUA other mailinglist participants are using. Anyway, I'm interested to know if this is considered to be a serious bug in mutt-1.2.5. BTW, I don't understand the duplication between 'In-Reply-To:' and 'References:', when mutt has to find the parent child relation between messages; seems like too much information can lead to ambiguity : In-Reply-To claims X but References claims Y. who do I believe?
