Front page | perl.perl5.porters |
Postings from January 2014
On Tue, Jan 28, 2014 at 01:55:28AM -0000, Father Chrysostomos wrote:
> Tony Cook wrote:
> > On Mon Jan 27 16:02:45 2014, perl5-porters@perl.org wrote:
> > > What is wrong with that? It shows how perl understands the code.
> > > Evaluating it produces an op tree that behaves identically.
> >
> > The padrange optimization which this is built on, does preserve the deparse:
> >
> > $ ~/perl/blead/bin/perl5.19.8 -MO=Deparse -e 'sub foo { my ($x, $y); my $z; 1 }
> > '
> > sub foo {
> > my($x, $y);
> > my $z;
> > 1;
> > }
> > -e syntax OK
>
> I would consider that adventitious.
(is that a word?)
Yes, the padrange deparsing was based on keeping the old pre-optimisation
tree around because otherwise far too much stuff in Deparse would have
needed fixing up.
I also agree that it doesn't matter that the Deparse output is different
as long its logically equivalent; e.g.
$ p -MO=Deparse -e'$x && print $y'
print $y if $x
--
In my day, we used to edit the inodes by hand. With magnets.
Thread Previous
|
Thread Next