On Mon, Aug 23, 2004 at 09:54:05AM +0100 Nick Ing-Simmons wrote:
> Bjoern Hoehrmann <[EMAIL PROTECTED]> writes:
> >* Tassilo von Parseval wrote:
> >>On Mon, Aug 23, 2004 at 08:14:45AM +0200 Bjoern Hoehrmann wrote:
> >>
> >>> Should my XS code check whether an object 'can' a specific method
> >>> before calling call_method() for a reason other than performance? It
> >>> seems that such a check is not commonly performed so it is probably
> >>> not very important.
> >>
> >>Depends. When writing pure Perl code, do you usually check that a method
> >>exists before you call it? I'd say no. However, sometimes you may want
> >>to do it. The same is true for XS. If it's crucial not to trigger a
> >>non-existent method, then do a preliminary check.
> >
> >Right, I looked at it in the context of optional event handlers where
> >it is common that specific methods do not exist. In Perl, if $o->x is
> >called even though there is no method x, Perl would die; so my question
> >is rather whether call_method for a non-existant method has negative
> >side-effects.
>
> Calling the method and catching the die via
>
> call_method("Not_there",G_SCALARX|G_EVAL);
>
> is what I do in Tk. But Tk has the G_EVAL there anyway so
> that any die is reported in Tk context.
I was under the impression that G_EVAL catches errors that happen inside
"Not_there" (now provided this method exists). So G_EVAL would
eventually fill ERRSV with something like 'Can't locate object method
"Not_there" via package...' if "Not_there" does not exist?
Tassilo
--
$_=q#",}])!JAPH!qq(tsuJ[{@"tnirp}3..0}_$;//::niam/s~=)]3[))_$-3(rellac(=_$({
pam{rekcahbus})(rekcah{lrePbus})(lreP{rehtonabus})!JAPH!qq(rehtona{tsuJbus#;
$_=reverse,s+(?<=sub).+q#q!'"qq.\t$&."'!#+sexisexiixesixeseg;y~\n~~dddd;eval
