Greg Wooledge wrote:
>Imagine you run a command like "stty" which alters the terminal state
>(intentionally). You wouldn't want bash to undo this. So, if stty
>exits in a normal manner, the terminal changes are retained.
[...]
>A "crash" is indistinguishable from a normal exit, from the shell's point
>of view, unless the "crash" is the result of a signal.
As someone who's been bitten by this occasionally, though never badly
enough to look into the issue in detail, I'm wondering -- would it be
reasonable for bash to limit its interpretation of a "normal" exit to
"WIFEXITED(status) && WEXITSTATUS(status) == 0"? Conversely, is there
any case in which one would want to retain changes made to the terminal
state when stty or a similar program exits with a non-zero status?
(My "man stty" says nothing about the exit status, so it's entirely
possible I'm missing some subtleties.)
--Andrew Church
https://achurch.org/
