Emacs crashing on C-x C-g

[Ryan Johnson]

On 25/10/2012 4:07 PM, Larry Hall (Cygwin) wrote:
> On 10/25/2012 1:06 PM, ping wrote:
>> On 10/25/2012 12:54 PM, Eli Zaretskii wrote:
>>>> Date: Thu, 25 Oct 2012 09:01:40 -0600
>>>> From: Eric Blake <eblake@redhat.com>
>>>>
>>>> On 10/25/2012 08:55 AM, Ryan Johnson wrote:
>>>>>>>> http://debbugs.gnu.org/cgi/bugreport.cgi?bug=12697
>>>>>>>>
>>>>>>>> I imagine it will be fixed before the release of Emacs 24.3.
>>>>>>> Agree the initial part is emacs. But I suspect the confused bash
>>>>>>> business is not. ^G should either send SIGINT or not, this core 
>>>>>>> dump
>>>>>>> thing is not cool.
>>>>>> Do you know whether this is Cygwin-specific?  Have you checked to 
>>>>>> see
>>>>>> what happens on Linux?
>>>>> Oops, you're right: I see the exact same behavior on linux. However,
>>>>> that's via ssh in a mintty window, so it could still be mintty. Can
>>>>> somebody running Linux directly verify, perhaps?
>>>> Linux (at least my setup on Fedora 17) has the same problem - when 
>>>> using
>>>> emacs as the editor under 'git commit', an ill-timed ctrl-G on my part
>>>> unceremoniously kills emacs as a result of sending the SIGINT to the
>>>> entire process group.  I hate the behavior, but it is definitely not
>>>> cygwin-specific.
>>> Why doesn't git block SIGINT when it invokes $EDITOR?  I think that's
>>> the prudent thing to do.
>> the simplest solution: change to vim.
>> I don't know of any extra benefit using emacs instead of vim...
>
> It's been 3 hours and not a single flame.  Have the Editor Wars
> actually died out?  Maybe there's hope for peace in the Middle East
> too. ;-)
At first I thought ping was suggesting to close all present and future 
vim bugs with "works for me, I use emacs."

But that would be completely absurd, so I concluded it must have been a 
joke.

Ryan


--
Problem reports:       http://cygwin.com/problems.html
FAQ:                   http://cygwin.com/faq/
Documentation:         http://cygwin.com/docs.html
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple