Carlo Sogono wrote:
O Plameras wrote:
Carlo Sogono wrote:
True, actually not even man bash. If you execute the command you'll
most likely get an error complaining about the file 1, create a file
1 and re-execute it and by using the magic commands ls and cat he'd
figure it out without needing the man pages.
This is not going to produce errors because,
$cat 1>m1
means concatenate "std.input" into filename "m1".
1 is actually stdout as mentioned in my previous message. You're
redirecting to m1 what cat is outputting to the screen and not what
your entering to the keyboard so
Yes, 1 == std.out, but note that,
$cat 1>m1 is another way of writing
$cat >m1
which means
concatenate "std.input" into filename "m1".
In the context of $cat 1>m1, the output is specified
std.out == 1, but at the same time is re-directed to filename "m1'.
In the same context, the input is not specified, so the
default is used which is "std.input" == 0. That's why
$cat 1>m1
means concatenate "std.input" into filename "m1".
$ echo testme 1>m1
will put testme to m1, something that didn't come from stdin. Bash by
default sends to stdout whatever it reads from stdin so doing
$ echo testme >&0
just prints it to the screen anyway.
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
