Re: Signal handling bugaboo in command substitution

[Peter Stephenson <p.stephenson@xxxxxxxxxxx>]

On Tue, 8 Mar 2016 10:00:04 +0000
Peter Stephenson <p.stephenson@xxxxxxxxxxx> wrote:
> How about something like this?

Actually, probably ESUB_NOMONITOR should take precedence.  That's really
not a good place to allow this stuff, no matter who thought it might
be...

pws

diff --git a/ChangeLog b/ChangeLog
index 420eb26..ce05a3d 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,4 +1,4 @@
-2016-03-07  Peter Stephenson  <p.stephenson@xxxxxxxxxxx>
+-2016-03-07  Peter Stephenson  <p.stephenson@xxxxxxxxxxx>
 
 	* 38111: Src/parse.c: remove redundant return values from
 	par_list() and par_list1().
diff --git a/Src/exec.c b/Src/exec.c
index b60fc90..9fec16e 100644
--- a/Src/exec.c
+++ b/Src/exec.c
@@ -994,9 +994,19 @@ entersubsh(int flags)
     if ((flags & ESUB_REVERTPGRP) && getpid() == mypgrp)
 	release_pgrp();
     shout = NULL;
-    if (!job_control_ok) {
+    if (flags & ESUB_NOMONITOR)
+    {
+	/*
+	 * Allowing any form of interactive signalling here is
+	 * actively harmful as we are in a context where there is no
+	 * control over the process.
+	 */
+	signal_ignore(SIGTTOU);
+	signal_ignore(SIGTTIN);
+	signal_ignore(SIGTSTP);
+    } else if (!job_control_ok) {
 	/*
-	 * If this process is not goign to be doing job control,
+	 * If this process is not going to be doing job control,
 	 * we don't want to do special things with the corresponding
 	 * signals.  If it is, we need to keep the special behaviour:
 	 * see note about attachtty() above.