[Nmh-workers] $editalt and -noatfile

2012-12-03 Thread Oliver Kiddle
My script for handling replies relies on the $editalt variable. This broke recently and git bisect pointed me to commit 25581a94 which makes -noatfile the default. I don't care too much for the @ file but need $editalt. Does -atfile have wider effects than merely disabling the @ file or is there

Re: [Nmh-workers] $editalt and -noatfile

2012-12-03 Thread David Levine
Oliver wrote: My script for handling replies relies on the $editalt variable. This broke recently and git bisect pointed me to commit 25581a94 which makes -noatfile the default. I don't care too much for the @ file but need $editalt. Does -atfile have wider effects than merely disabling

Re: [Nmh-workers] $editalt and -noatfile

2012-12-03 Thread Ken Hornstein
In the meantime, can you use $mhaltmsg? I'm curious because maybe we should recommend using that instead of $editalt. David was nice enough to not mention that the original bug was mine; sorry about that! You know, I have no idea why there is both editalt and mhaltmsg; from looking at the

Re: [Nmh-workers] $editalt and -noatfile

2012-12-03 Thread David Levine
Ken wrote: You know, I have no idea why there is both editalt and mhaltmsg; from looking at the documentation, maybe mhaltmsg is just for whatnowproc? That makes sense. Hm. Looking at things further right now mhaltmsg is designed to be used by things like send (it uses it when doing

Re: [Nmh-workers] $editalt and -noatfile

2012-12-03 Thread Ken Hornstein
$editalt just seems out of place to me. It's the only lower-case environment variable that doesn't start with mh. I understand your issue, but we lack a time machine to go back and find the person responsible to complain about it :-) I see it predates the conversion of MH to RCS. Looking at