I agree with your arguments, and if instead we put an option to disable this before compiling or a set in the psql cli?
On Sun, Oct 21, 2018, 20:20 Tom Lane <t...@sss.pgh.pa.us> wrote: > Renato dos Santos <shaz...@gmail.com> writes: > > I have been using psql for quite a few days. And I have accidentally > pressed the CTRL + \ keys that sends the signal QUIT+Coredump (CTRL+4 also > sends the same signal). > > I hope it's relevant to more people. (This has bothered me.) > > > this patch avoids the output of the CLI using ctrl + \ is the same as > ctrl + c > > I'm fairly confused about why this would be a good idea, for several > reasons: > > * If you have a tendency to make that typo, why would you want a fix that > only affects psql and not any of your other applications? (If you do > want the latter, there are already ways to do it, eg you could remap > SIGQUIT to some other key via stty, or disable core dumps via ulimit.) > > * If we put this in, what becomes of people who actually want a core dump, > eg for debugging? > > * SIGQUIT is a fairly well-known way to get out of an application when all > else fails. People who aren't familiar with psql's exit commands might > find it pretty unfriendly of us to block this off. > > regards, tom lane >
