Re: [ANNOUNCE] mutt 1.5.16 released
On Jun 16 at 10:07, Brendan Cully wrote:
> On Saturday, 16 June 2007 at 12:46, Mike Hunter wrote:
> >
> > What do people think about allowing the user to specify in their muttrc
> > what values do and do not constitute a fatal error coming from their
> > editor? It seems to me mutt *should* react if it is told there's been a
> > fatal editor failure, and given that the standard means of communicating
> > such an error has occurred is inherently broken, we should allow for
> > another means.
>
> I don't think having some editor_error_code configuration list is
> substantially easier for users than
>
> #!/bin/sh
> vim "$@"; true
>
> The script approach, on the other hand, is a lot simpler for mutt.
As has been expressed by D. Martin (I went to high school with a kid
named Derek Martin, you aren't the same Derek Martin who throw crab
apples at passing 18 wheelers on I-44, are you?), even if you refine
this approach to allow for some exit codes to signify failure and some
to signify success, it still means that a user has to author a script in
a language present on a given machine to safely use an external editor
on many systems. I could dash this out in perl very quickly, but I (a
supposed developer!) don't know/remember csh (for example) and not all
systems have perl (FBSD, not to mention windows!), so I think the
expectations are actually broader than they first appear.
While I don't expect mutt to be as "easy to use" as a mainstream gui
mailer, I do think it should be a shared goal to have it be a "drop-in"
(or at least `configure && make && make install` :) ) tool on common
platforms. And since it *requires* the use of external tools to
function (again as expressed in this thread) we should expect it to
interface with those required tools in a sane way. One could argue that
solaris ex is inherently insane, which is true, but if we demand users
employ external editors with varying exit semantics (including
"insanity" semantics :) ), we should have a mechanism for translation.
My one concern is that I can't seem to find any documentation that ANY
common editor actually endorses a return value that signifies a fatal
error...but maybe we should encourage them :)
Mike
--
Mike Hunter
Confessed subject-line misspeller, author of "The Subject Spellcheck Patch"
http://marc.info/?l=mutt-dev&m=117268263816741&w=2