From: cgf AT kramden DOT cygnus DOT com (Christopher G. Faylor)
Subject: Re: BASH under B19
16 Apr 1998 22:14:49 GMT
Message-ID: <6h5vsp$n1d$1@cronkite.cygnus.com>
References: <8B40B8756FA1D111BCB900A02495E24F36B421 AT neptune DOT xstor DOT com> <3535BF6B DOT 539DB456 DOT cygnus DOT gnu-win32 AT tibco DOT com>
X-Newsreader: trn 4.0-test63 (15 March 1998)

In article <3535BF6B DOT 539DB456 DOT cygnus DOT gnu-win32 AT tibco DOT com>,
Matt Stupple <matts AT tibco DOT com> wrote:
>Bob McGowan wrote:
>> 
>> I am a bit confused by the second paragraph.  In the usual DOS
>> command.com and, I thought, in the cmd.exe of WinNT, a ^C is simply an
>> interrupt.  Unlike UNIX, it cannot be changed, that I know of, by any
>> simple command interface.  Regardless, it is just an interrupt and
>> should stop a running process.  If the Cygwin stuff is supposed to
>> maintain some compatability with the MS environment, then I do not see
>> why it would be necessary or desireable to have new threads spawned by
>> this key sequence.
>> 
>
>This is a straight copy from the MSVC help on the signal() function:

Actually, Cygwin32 does *not* start a new thread when ^C is hit.  A ^C
should behave just like UNIX.  It should either stop your process or,
in the case of a shell, it should intercept the signal and "do something".

MSVC does start a new thread when a ^C is hit.  Cygwin32 doesn't.
-- 
cgf AT cygnus DOT com             "Everything has a boolean value, if you stand
http://www.cygnus.com/      far enough away from it."  -- Galena Alyson Canada