Les Mikesell wrote:
On Fri, 2005-12-16 at 11:55, Mike McCarty wrote:
Normally, I'd suggest ./test and expect that to work reliably across
Unix-like operating systems.
Possibly. The word "test" is reserved in some shells.
It's a built-in, which means only that it is found ahead of a PATH
search for executables. You can still specify the path to a
real executable if you want.
Umm... read what I wrote. It is a built-in for bash. I did not
specify bash. I have used a shell in which "test" is a reserved
word, and could be used only under restricted contexts, and didn't
mean what you probably think it meant. It put the shell into
a "debug mode" of operation. So I'm simply exercising caution
in the "expect it to work reliably across Unix-like operating
systems" part of the quote.
Using "./test" on that shell would result in an error message
"/ is a directory"
after which further command entry would result in spitting out
displays of internal parse events during the execution of the
commands, things like "associating stdin with named pipe blah".
That is, if one had proper authority. Otherwise it would simply say
"/ is a directory"
[snip]
Yes, SCO UNIX is "more nearly" UNIX.
There have been very few variations in the bourne shell so I
wouldn't expect a difference in that respect. If you go back
far enough you might find one that didn't have test as a
built-in and would always have run /bin/test or it's '[' link
as an external program.
What does this have to do with SCO?
Mike
--
p="p=%c%s%c;main(){printf(p,34,p,34);}";main(){printf(p,34,p,34);}
This message made from 100% recycled bits.
You have found the bank of Larn.
I can explain it for you, but I can't understand it for you.
I speak only for myself, and I am unanimous in that!
