On Tue, 2005-03-29 at 02:39 -0500, Paul Eggert wrote:
> Mike Hearn <[EMAIL PROTECTED]> writes:
> 
> > That sounds like the one, yes. Though 5.3.0 doesn't seem to show it.
> 
> Odd; it shows it for me.
> 
> $ _POSIX2_VERSION=199209 tail -c 123
> tail: cannot open `123' for reading: No such file or directory
> $ tail --version | head -n 1
> tail (GNU coreutils) 5.3.0

Right, but here you are setting _POSIX2_VERSION to something else. I
just used the default for my environment (FC3) which doesn't seem to
have _POSIX2_VERSION set.

> In coreutils' defense, the standard in question was the
> latest-and-greatest version until 2001.  (And the 2001 standard broke
> lots more shell scripts. :-)

Oh yipee :)

> Well, to some extent the horse is out of the barn door already.  The
> behavior you're objecting to is not just that of coreutils 5.3.0; it's
> also that of Solaris 9 /bin/tail, for example, and I assume for many
> other "tail" implementations.

Right, in this case the installers in question contain GNU/Linux
binaries so portability to Solaris isn't a large concern.

> If you're willing to assume just GNU/Linux, you can put
> "_POSIX2_VERSION=200112" into a README file, or into a wrapper around
> the nonportable installer.  

Yes I think our only solution is to update the templates and ask people
to rebuild their installers, then put a notice on our website informing
people about the _POSIX2_VERSION workaround.

thanks -mike



_______________________________________________
Bug-coreutils mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/bug-coreutils

Reply via email to