[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: gawk: invalid read and write
From: |
Stephane Chazelas |
Subject: |
Re: gawk: invalid read and write |
Date: |
Sun, 28 Jan 2007 19:12:04 +0000 |
User-agent: |
Mutt/1.5.6i |
On Sun, Jan 28, 2007 at 07:50:57PM +0100, Jürgen Kahrs wrote:
> Andrew J. Schorr wrote:
>
> >> Wouldnt it be enough to just change it to
> >>
> >> awk '...' log.[0-9]* log.pid[0-9]*
> >
> > Or, even more simply, log.* (if you include
> > both wildcards, then one may not have any matches, and
> > that will cause an error, I think).
>
> Your reasoning is correct. But in reality, any
> user will have either valgrind 1 or valgrind 2,
> not both.
[...]
It should also be noted that "log.[0-9]*" also matches log.0foo
or log.1.gz.
So you could as well do
awk '
BEGIN {
for (i = 1; i < ARGC; i++)
if (ARGV[i] !~ /^log\.(pid)?[0-9]+$/)
delete ARGV[i]
}
...' log.*[0-9]*
Note that if there's no matching file, stdin will be considered.
--
Stéphane