[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Bug-readline] Does _rl_interrupt_immediately still need to exist?
From: |
Pedro Alves |
Subject: |
[Bug-readline] Does _rl_interrupt_immediately still need to exist? |
Date: |
Thu, 21 Mar 2019 15:46:21 +0000 |
User-agent: |
Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101 Thunderbird/52.9.1 |
Hi,
I'm looking at readline's signal handling code, and I noticed that
in the current master, the _rl_interrupt_immediately variable is
never set anymore.
I see that _rl_interrupt_immediately used to be set around
rl_attempted_completion_function in complete.c in older releases,
but that signal handling was replaced with the safer
RL_SIG_RECEIVED check in the mainline code.
Is there a reason the _rl_interrupt_immediately variable still
exists, then? Could rl_signal_handler be simplified, like:
static RETSIGTYPE
rl_signal_handler (int sig)
{
- if (_rl_interrupt_immediately)
- {
- _rl_interrupt_immediately = 0;
- _rl_handle_signal (sig);
- }
- else
- _rl_caught_signal = sig;
+ _rl_caught_signal = sig;
SIGHANDLER_RETURN;
}
?
I was looking at signal handling on Windows, for gdb, assessing its
correctness (given Windows runs signals on a separate thread). Not
having the _rl_interrupt_immediately==true path simplifies things.
Thanks,
Pedro Alves
- [Bug-readline] Does _rl_interrupt_immediately still need to exist?,
Pedro Alves <=