Return Code not set on SIGINT about launch

[audrow]

I'm going to mark this as backlog.

It's not 100% clear to me what the right behavior is here. What did we do in ROS 1? There also may be some funniness with Windows.

[mjcarroll]

So the longer discussion was that it should probably return 130 on ctrl-c, since that is the error code that represents that an interactive application was interrupted.

I don't think anybody would be opposed to launch having that behavior as well.

Adjacent to that, it would be interesting to know how roslaunch did it, and if not returning 130 was keeping in line with that precedent or just an oversight.

Additionally, all bets are off with Windows and how it handles interrupts and return codes, so it would be somewhat worthwhile to try to make the behaviors consistent.

[DLu]

Sad to report that roslaunch returns 0 in most cases except for exceptions (which sigint does not generate)