Front page | perl.perl5.porters |
Postings from July 2010
Quoth demerphq@gmail.com (demerphq): > On 12 July 2010 20:30, Dave Mitchell <davem@iabyn.com> wrote: > > On Mon, Jul 12, 2010 at 06:06:15PM +0200, Reini Urban wrote: > >> gvsv is just checking magic and doing the sideeffect > > > > Huh? gvsv doesn't call magic. > > > > Also, just as a data point, note that pp_concat *explicitly* calls get > > magic twice on $a . $a: > > > > � � � �if (left == right) > > � � � � � �/* $r.$r: do magic twice: tied might return different 2nd time */ > > � � � � � �SvGETMAGIC(right); > > Id like to argue that this was misguided. I dont think we guarantee > any particular order in this case for the fetch calls and thus the > statement is /still/ undefined even with this change. Perl doesn't have undefined behaviour. No matter what weasel words copied from stdc made it into the ++ docs, Perl's actual evaluation order has always been straightforward and well-defined. Changing this may be worth it, for a sufficiently beneficial optimisation, but it is definitely a backwards-incompatible change. BenThread Previous | Thread Next