At 23:24 +0200 27 Dec 1999, Mikko Hänninen <[EMAIL PROTECTED]> wrote:
> I wonder if some sort of reply-hook would be needed in Mutt? Ie.
The idea has been brought up before.
> send-hooks would remain as they are, but when replying, the
> message-being-replied to is checked against reply-hooks first, and after
> that send-hooks are evaluated as currently. Something like that would
With that order there's really no way to undo any changes done by the
reply-hooks; you can't use a send-hook since that would undo the changes
before they had any affect, but you can't use a reply-hook since the
next message sent might not be a reply.
Of course executing reply-hooks after send-hooks could also cause
problems in some situations, so it's not ideal either.
> solve the current problem. But I'm not sure how complex adding that
> would be. I do know that I wouldn't want to try coding it. :-)
I don't think it would be all that hard to do, but I'm not sure if it's
the correct solution.
I think a better solution, although possibly much more difficult to do,
would be to have a pattern operator (~o perhaps) that modified the next
operator (or possibly the entire pattern) to match against the original
message for a reply, and never match for a new message. This would
allow (current) send-hooks and (pseudo-)reply-hooks to be mixed freely.
--
Aaron Schrab [EMAIL PROTECTED] http://www.execpc.com/~aarons/
Do not simplify the design of a program if a way can be found to make
it complex and wonderful.
