Re: chmod no longer clears sticky bit

2006-09-30 Thread Paul Eggert
Mike Frysinger <[EMAIL PROTECTED]> writes: > i guess this is allowed by spec huh ? Yes. The last sentence in the extract that you quoted clearly allows the current behavior. The area is not standardized that closely, so we tried to do the right thing. It was discussed in several threads, e.g.,

Re: chmod no longer clears sticky bit

2006-09-30 Thread Mike Frysinger
On Saturday 30 September 2006 19:20, Paul Eggert wrote: > Yes, that's mentioned in the NEWS file. It says: i really need to start reading that thing > chmod, install, and mkdir now preserve a directory's set-user-ID and > set-group-ID bits unless you explicitly request otherwise. E.g., >

Re: chmod no longer clears sticky bit

2006-09-30 Thread Paul Eggert
Mike Frysinger <[EMAIL PROTECTED]> writes: > looks like doing `chmod 0755` no longer clears sticky bits ... for example: Yes, that's mentioned in the NEWS file. It says: chmod, install, and mkdir now preserve a directory's set-user-ID and set-group-ID bits unless you explicitly request othe

Re: chmod no longer clears sticky bit

2006-09-30 Thread Mike Frysinger
On Saturday 30 September 2006 19:07, Mike Frysinger wrote: > looks like doing `chmod 0755` no longer clears sticky bits ... seems to arise from lib/modechange.c:mode_adjust() ... particularly: mode_t omit_change = (dir ? S_ISUID | S_ISGID : 0) & ~ changes->mentioned; omit_change here is set t

chmod no longer clears sticky bit

2006-09-30 Thread Mike Frysinger
looks like doing `chmod 0755` no longer clears sticky bits ... for example: $ tar xf coreutils-6.3.tar.bz2 $ cd coreutils-6.3 $ ./configure $ make -C lib $ make -C src chmod $ mkdir foo $ stat -c%a foo 755 $ ./src/chmod -v 2755 foo mode of `foo' changed to 2755 (rwxr-sr-x) $ stat -c%a foo 2755 $ ./

coreutils-6.3 released (stable)

2006-09-30 Thread Jim Meyering
Coreutils version 6.3 has been released. If you haven't heard about the GNU coreutils, the FAQ is a good place to start: . This is a stable release. Changes since 6.2 have been solely to fix bugs and to improve portability and robustness. A summary of

Re: pre-6.3 cygwin make check results

2006-09-30 Thread Jim Meyering
Eric Blake <[EMAIL PROTECTED]> wrote: > I am including the testsuite results from yesterday's CVS Thanks for all the testing and investigation. ___ Bug-coreutils mailing list Bug-coreutils@gnu.org http://lists.gnu.org/mailman/listinfo/bug-coreutils

pre-6.3 cygwin make check results

2006-09-30 Thread Eric Blake
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 I am including the testsuite results from yesterday's CVS; Paul's patch for Solaris openat emulation did indeed fix issues that cygwin was having. Some of the test failures are testsuite shortcomings for not recognizing cygwin limitations, such as the