Package: zsh-beta
Version: 4.3.6-dev-0+20080428-1
Severity: normal

lapse:~/debian|master|debian.ch%% echo $ZSH_VERSION
4.3.6-dev-0+0428
lapse:~/debian|master|debian.ch%% zmodload zsh/files
lapse:~/debian|master|debian.ch%% which mv
mv: shell built-in command
lapse:~/debian|master|debian.ch%% mv /tmp/debian.ch_T-Shirt_20080422-poll.png .
mv: /tmp/debian.ch_T-Shirt_20080422-poll.png: invalid cross-device link

-- System Information:
Debian Release: lenny/sid
APT prefers unstable
APT policy: (500, 'unstable'), (500, 'testing'), (500, 'stable'), (1, 'experimental')
Architecture: i386 (i686)

Kernel: Linux 2.6.24-1+scoflowctrl.1-686 (SMP w/1 CPU core)
Locale: LANG=en_GB, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

On Tue, May 06, 2008 at 03:42:08PM +0100, martin f krafft wrote:

Anyone up for changing this in a future version?

Note that this mv will not move files across devices. Historical
versions of mv, when actual renaming is impossible, fall back on copying
and removing files; if this behaviour is desired, use cp and rm manually.
This may change in a future version.

On May 6, 4:03pm, Clint Adams wrote:
}
} Anyone up for changing this in a future version?
}
} Note that this mv will not move files across devices. Historical
} versions of mv, when actual renaming is impossible, fall back on copying
} and removing files; if this behaviour is desired, use cp and rm manually.
} This may change in a future version.

I'd rather we didn't. rename(2) is atomic, but there's no reasonable way
to make cp+rm be so, and I think it's better if the user is forced to be
aware of the issue.

Also note that "cp" is (intentionally) NOT one of the commands supplied
by the zsh/files module, so changing mv means implementing cp.

also sprach Bart Schaefer brasslantern.com> [2008.05.06.1645 +0100]:

What, if I may ask, is the purpose of zsh/files?

--
.''`. martin f. krafft debian.org>
: :' : proud Debian developer, author, administrator, and user
`. `'` http://people.debian.org/~madduck - http://debiansystem.info
`- Debian - when you have better things to do than fixing systems

you can't assign IP address 127.0.0.1 to the loopback adapter,
because it is a reserved address for loopback devices.
-- micro$oft windoze xp professional

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)

On Tue, May 06, 2008 at 04:03:03PM +0100, Clint Adams wrote:

On 2008-05-06 at 17:55 +0100, martin f krafft wrote:

If I correctly recall the gist of a verbal conversation about 11 years
ago with the author, in a university environment with heavily multi-user
systems which were often abused by people writing My First Fork Bomb, it
was to have the availability to load in sufficient builtins to make it
easier to deal with systems which won't let you fork() anymore more; to
do basic recovery work and have the commands available; to avoid fork()
overheads when dealing with a loaded single-CPU Solaris box, etc.

But mostly, IIRC, to build a static version of the shell with the module
already loaded, to put on a boot/recovery diskette and so reduce what
was needed.

It pre-dates busybox, which has since become the common way of doing
this.

It wasn't intended to provide full versions of the commands, the
commands which are there don't implement all the POSIX options, etc.
It's intended to provide enough to be useful in emergencies.

Most people shouldn't need to use it.

On Tue, 6 May 2008 17:55:55 +0100
martin f krafft debian.org> wrote:

Its primary purpose has been as a fallback on a system where the shell
is working but nothing much else is. Hence it provides the limited,
easy-to-implement things you're likely to need to fix what needs
fixing. Moving things between file systems typically isn't something
you need (though in any given case it may be, as well as a whole load of
other stuff the shell doesn't provide).

On 2008-05-06 13:55:23 -0700, Phil Pennock wrote:

I use it *by default* just because of the Linux kernel bug^Wlimitation
concerning the argument list (I often exceed it). But perhaps I should
write my own shell function as a wrapper to coreutils and use xargs or
zargs when necessary.

--
Vincent Lefèvre vinc17.org> - Web: <http://www.vinc17.org/>
100%% accessible validated (X)HTML - Blog: <http://www.vinc17.org/blog/>
Work: CR INRIA - computer arithmetic / Arenaire project (LIP, ENS-Lyon)

On 2008-05-07 at 23:30 +0200, Vincent Lefevre wrote:

Would it make more sense for your usage if Zsh used the new "features"
capabilities to resolve this? Would this also help with the Debian
situation, in addition to documentation notes?

They were added by Peter after I sent a proposed patch renaming
zsh/stat's stat to zstat, with a stat deprecated backwards-compatible
name, since a stat(1) external command has come along in the years since
zsh's stat module was introduced with incompatible behaviour.

Peter just looked at the logical conclusion and put in a lot more work
to do the Right Thing.

On 2008-05-07 14:47:45 -0700, Phil Pennock wrote:

Yes. IIRC, the only features (standard or not) I really miss are
the -v option (verbose), the -n option for ln, the -u option for
mv, and mv across devices.

Names are not really important for me since it is easy to write a
shell function like:

rm() { command rm $@ }

or

rm() { builtin rm $@ }

--
Vincent Lefèvre vinc17.org> - Web: <http://www.vinc17.org/>
100%% accessible validated (X)HTML - Blog: <http://www.vinc17.org/blog/>
Work: CR INRIA - computer arithmetic / Arenaire project (LIP, ENS-Lyon)