2006/3/14, Anselm R. Garbe <garbeam_AT_wmii.de>:
>
> On Mon, Mar 13, 2006 at 12:27:01PM -0800, Suraj N. Kurapati wrote:
> > Suraj N. Kurapati wrote:
> > > echo "$PATH" | tr ':' '\n' | grep -v '^$' | sort -u | while read
> > > dir; do find "$dir" -perm -u+x -type f -print | sed 's,^.*/,,'; done
> > > | sort -u
> >
> > Whoops, a small correction (\! -type d):
> >
> > echo "$PATH" | tr ':' '\n' | grep -v '^$' | sort -u | while read
> > dir; do find "$dir" -perm -u+x \! -type d -print | sed 's,^.*/,,';
> > done | sort -u
>
> Can someone confirm if that works on BSD? Why are there two
> 'sort -u' filters?
First sort ensures, that there is no duplicates in PATH.
The find args and the sort -u arg smell like a gnu...
Solaris sort support this switch. Anyway - you can replace this with 'sort |
uniq'.
\! looks
> very strange. What is wrong with -type f?
Indeed. -type f should be sufficient.
hrr
Received on Tue Mar 14 2006 - 09:56:34 UTC
This archive was generated by hypermail 2.2.0 : Sun Jul 13 2008 - 16:01:09 UTC