Date: Thu, 03 Feb 2005 15:42:59 +0000 From: Brian Candler <B.Candler@pobox.com> To: FreeBSD-gnats-submit@FreeBSD.org Subject: bin/77067: /bin/sh premature termination when 'set -e' is active Message-ID: <E1Cwj8F-0000MS-TP@billdog.local.linnet.org> Resent-Message-ID: <200502031550.j13FoGLK087346@freefall.freebsd.org>
next in thread | raw e-mail | index | archive | help
>Number: 77067
>Category: bin
>Synopsis: /bin/sh premature termination when 'set -e' is active
>Confidential: no
>Severity: serious
>Priority: low
>Responsible: freebsd-bugs
>State: open
>Quarter:
>Keywords:
>Date-Required:
>Class: sw-bug
>Submitter-Id: current-users
>Arrival-Date: Thu Feb 03 15:50:16 GMT 2005
>Closed-Date:
>Last-Modified:
>Originator: Brian Candler
>Release: FreeBSD 5.3-STABLE i386
>Organization:
>Environment:
System: FreeBSD billdog.local.linnet.org 5.3-STABLE FreeBSD 5.3-STABLE #0: Mon Dec 27 12:24:34 GMT 2004 root@billdog.local.linnet.org:/usr/obj/usr/src/sys/BILLDOG i386
>Description:
Under certain circumstances, 'set -e' causes /bin/sh to exit when it should
not. From the manpage:
-e errexit
Exit immediately if any untested command fails in non-interactive
mode. The exit status of a command is considered to be explic-
itly tested if the command is used to control an if, elif, while,
or until; or if the command is the left hand operand of an `&&''
or `||'' operator.
However, there are circumstances where even when a command status is
explicitly tested the script still terminates. I can replicate this using a
function call.
>How-To-Repeat:
This script demonstrates the problem:
------------------------------------------------------------
#!/bin/sh
set -e
foo ( ) {
if test "x$1" != "xyes"; then
return 1
fi
echo "OK!" # this line causes the problem!
}
if foo yes; then echo "yes"; fi
if foo no; then echo "no"; fi
echo "Completed!"
------------------------------------------------------------
It fails to complete. However, if you remove the 'echo "OK!"' line it *does*
complete. (This line can be anything - for example, replacing it with
"return 0" also causes the problem).
The script runs correctly under 'bash', so I don't think my understanding of
-e semantics is broken.
>Fix:
>Release-Note:
>Audit-Trail:
>Unformatted:
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?E1Cwj8F-0000MS-TP>