Re: rfc patch, abort rm instead of only removing the * from the cmdline

[Bart Schaefer <schaefer@xxxxxxxxxxxxxxxx>]

On Aug 21,  4:58pm, Peter Stephenson wrote:
} Subject: Re: rfc patch, abort rm instead of only removing the * from the c
}
} On Sun, 17 Aug 2008 13:42:15 -0700
} Bart Schaefer <schaefer@xxxxxxxxxxxxxxxx> wrote:
} > Ideally I suppose this prompt should work like the prompt for "correct",
} > giving you "nyae" as choices so you can choose to abort or to edit
} 
} That would certainly make things more consistent; I'm not sure how easy
} that is at this point in the code, however, which I think (without
} checking) is quite late into execution.  Does anyone want to look?

It can't be that late, as it has to happen before globbing in order to
find the Star token.  However, correction aborts by making hend() return
false, and relative to that in init.c:loop(), checkrmall() is down in
execode() which is after hend() has finished and even after preexec has
been run.  So the [yna] part would be easy (use Stephane's patch exept
only on a response of "a"), but it would have to unwind (or re-instate)
a fair bit of state to get back to the same command line for "e".

(This thread should probably move to zsh-workers at this point.)