Lan Barnes wrote: > On Thu, March 22, 2007 2:08 pm, John Oliver wrote: >> On Thu, Mar 22, 2007 at 01:51:35PM -0700, James G. Sack (jim) wrote: >>> It appears that you didn't grasp the significance of what Greg said >>> about the '&&' separating your commands. >>> >>> If you want to: >>> try A, and then if that works try B, and then if that works try C >>> use >>> A && >>> B && >>> C >>> if you want to: >>> try A, and then (always) try B, and then (always) try C >>> use >>> A >>> B >>> C >> Ahhh! >> >> I thought the && would just move on to the next command after the >> completion, not necessarily the successful completion, of the first >> command. And that's what I want... I want it to try to install libgcc >> first, and only after that's complete, one way or the other, to move on >> to the next RPM command. >> >> I need to look up bash characters :-) >> >> > > OR (||) quits if the first is true, but checks the second if the first is > false. AND (&&) quits if the first is false but checks #2 if the first is > true. > > Their clever use to do conditionals in command lines is further obscured > by the fact that most posix programs return 0 for success (FALSE).
It's a cultural thing, I suppose. The association of FALSE with 0 is what appears reasonable to you, I suspect. In this context it is more consistent to say that 0 means NO-ERROR. Anyway, the bash conditionals were designed to avoid having to think about return codes. You don't (usually) need to examine return codes. If (operation) then <success branch> else <failure branch> fi and operation && successaction || failureaction Both seem _reasonable_ to me. ;-) > > I have always been ambivalent about this bit of cleverness, because I > distrust cleverness in programming in the first place, and because I fear > that if I have to move my lips when I puzzle it out as I write it, then > the next person to read it might just "correct" it into the third circle > of hell. But it does work if you get it right. > Regards, ..jim -- KPLUG-List@kernel-panic.org http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-list
