Bertram Franz wrote:
> Unsetting or setting to POSIX helped. So after all this is a bug in the
> en_US collation table, maintained by SuSE or Novell?
I think this is one of those cases where one person's bug is another
person's feature. To me it is a bug. But someone wanted everyone
else in the
Bernd Klein <[EMAIL PROTECTED]> writes:
> which is what I expect to happen, but if I use "sort -g +1 -2 example.txt"
> (-g according to man page: compare according to general numerical value)
> I get the confusing result:
> blabla 0.5 abcd
> xyz2 0.4 df
> xyz3 0.1 xyz5
> xyz4 0.002 bal
> xyz5 0.1
Hi,
I think, I may have found a bug in "sort".
If I take the following file:
sdf 4 trc
xya 3 trt
bal 1 trt
blabla 0.5 abcd
xyz 1 ggg
xyz2 0.4 df
xyz3 0.1 xyz5
xyz4 0.002 bal
xyz5 0.1 xyz2
"sort +1 -2 example.txt" gives me the following result:
xyz4 0.002 bal
xyz3 0.1 xyz5
xyz5 0.1 xyz2
xyz2 0.4
[EMAIL PROTECTED] wrote:
The POSIX specification requires that the prompt be issued to stderr:-
Fair enough. Thanks for the info.
What my patch essentially did was:
if (isatty(stderr)) { /* interactive */
human=open("/dev/tty");
fcntl(human,F_SETFD,FD_CLOEXEC);
} else {
human=stderr;
}
[comm 5.2.1]
Here are patches to add an option to `comm' so that it ignores case
while doing the comparisons, similar to the --ignore-case option in
`uniq'.
Most of the work is simply copy and paste from `uniq.c', so I doubt
that papers are necessary. In case they are, I've signed a kind of
`ge
Bob,
thanks a lot for your answer. You were right, LANG=en_US.UTF-8 on all
these systems, and that is propagated to all the other LC vars.
Unsetting or setting to POSIX helped. So after all this is a bug in the
en_US collation table, maintained by SuSE or Novell?
Anyway, I will now include LC_C
On Mon, Feb 28, 2005 at 10:06:23AM +, [EMAIL PROTECTED] wrote:
> Felipe Kellermann wrote:
> >On Sat, 26 Feb 2005 5:39am +0100, Alfred M. Szmidt wrote:
> >
> >
> >> > Did I mention ls should have a --no-total option
> >> > to remove those annoying
> >> > total 1120
> >> > without needing to
I've checked in this change:
2005-02-27 Jim Meyering <[EMAIL PROTECTED]>
* xnanosleep.c (xnanosleep): Work around bug in Linux-2.6.8.1's
nanosleep whereby it fails without setting errno upon being resumed
after being suspended.
Index: xnanosleep.c
==
Felipe Kellermann wrote:
On Sat, 26 Feb 2005 5:39am +0100, Alfred M. Szmidt wrote:
> Did I mention ls should have a --no-total option
> to remove those annoying
> total 1120
> without needing to pipe to a filter.
Another possibility would be to output the `total' to stderr.
The horror,
