Re: What's cooking in git.git (Nov 2012, #02; Fri, 9)

Paul Fox <pgf@foxharp.boston.ma.us> writes:

That's right. It is not a quirk of shell-mode in Emacs, because I get the same result with ^C in xterm too.

% EDITOR="$HOME/prefix/x86_64-unknown-linux-gnu/bin/emacsclient --current-frame" % export EDITOR % cvs commit BUGIT Waiting for Emacs...^Ccvs commit: warning: editor session failed

Log message unchanged or not specified
a)bort, c)ontinue, e)dit, !)reuse this message unchanged for remaining dirs
Action: (continue) a
cvs [commit aborted]: aborted by user
% 

While cvs was waiting from emacsclient:

% cat /proc/2030/stat 2030 (cvs) S 1849 2030 1849 34816 2030 4202496 598 0 0 0 0 0 0 0 20 0 1 0 94752537 34254848 410 18446744073709551615 140168182550528 140168183348316 140737407935424 140737407931680 140168163193950 0 0 6 20513 0 0 0 17 2 0 0 0 0 0 % grep 'Name\|Pid\|Sig' /proc/2030/status Name: cvs Pid: 2030 PPid: 1849 TracerPid: 0 SigQ: 0/28998 SigPnd: 0000000000000000 SigBlk: 0000000000000000 SigIgn: 0000000000000006 SigCgt: 0000000180005021 % cat /proc/2031/stat 2031 (emacsclient) S 2030 2030 1849 34816 2030 4202496 155 0 0 0 0 0 0 0 20 0 1 0 94752538 4169728 81 18446744073709551615 4194304 4210620 140735996104016 140735996095456 140664960886018 0 0 0 0 0 0 0 17 1 0 0 0 0 0 % grep 'Name\|Pid\|Sig' /proc/2031/status Name: emacsclient Pid: 2031 PPid: 2030 TracerPid: 0 SigQ: 0/28998 SigPnd: 0000000000000000 SigBlk: 0000000000000000 SigIgn: 0000000000000000 SigCgt: 0000000000000000 %

which I interpret to mean both processes were in process group 2030, the cvs process ignored SIGINT and SIGQUIT, the emacsclient process neither ignored nor handled any signals, and neither process blocked any signals (not even SIGCHLD as system(3) would). When ^C in the terminal sent SIGINT to the process group, it terminated the emacsclient process only.

If git did the same thing as cvs here, i.e. ignore the signals in the parent process only and check the exit status of the editor, I think that would be OK.