Public SourceMage Discussion List

[Paul <dufflebunk AT dufflebunk.homeip.net>]

We could make an invisiblize function...
function invisiblize() {
        local RET=$?
        local func=$1
        shift
        $func "$@"
        return $RET
}
which can wrap around any function. That's not a very nice thing though.
I believe that there's something wrong if we need to make nasty hacks
like that.
        

On Sun, 2004-04-18 at 04:37, Vladimír Marek wrote:
> > Hmm, also I had meant to point out that one of the reasons why sorcery is 
> > not detecting failures is due to structures like this:
> > if [ -x BUILD ] ; then
> > persistant_load
> > . BUILD
> > persistant_save
> > else
> > default_build
> > fi &&
> 
> I can see another danger there, in the
> 
> persistant_load
> . BUILD
> persistant_save
> 
> itself. If the BUILD fails, persistant_save is going to carry out the
> return also. There are two solutions I can see:
> 
> a)
> persistant_load
> . BUILD && persistant_save
> 
> b)
> persistant_save will maintain return value from last function:
> function persistant_save () {
>         local RET=$?
> 
>       ....
> 
>         return $RET
> }
> 
> Both solutions will work. But maybe a) is better, since if the spell
> failed, then no values should be saved ?
> 
> I will add b) anyway, since that does not stop us from implementing a).
> What do you think ?
> 
> --
>       Neuron
> _______________________________________________
> SM-Discuss mailing list
> SM-Discuss AT lists.ibiblio.org
> http://lists.ibiblio.org/mailman/listinfo/sm-discuss
-- 
De mortuis nihil nisi bonum.

Attachment: signature.asc
Description: This is a digitally signed message part