Perl subroutines may be called with or without leading & and with or without () at the end. Either one is strongly recommended – the command might end up as something else than a function call if both are missing.
I just discovered a nice Perl trick which bothered be years ago:
&foo
is different from
&foo()The first gets access to the global @_ (aka whatever function foo was called
from, its @_ is handed to foo), the latter does not.
Simply:
sub a { &b; }
sub b { print $_[0]; }
a("Hello World");
It works!
But this is unexpeced (usually) as few cases require silent argument passing. I switched to the b() form for every function call to be safe from unexpected parameter inheritance.
Maybe I’ll use the &b form (with a comment explaining why it’s used) when I need to pass a huge arguments on to the next called sub – it should be much faster to let the called sub also use the same @_ instead of creating a new copy for it.
Yor might even manipulate the parents @_ this way… no, better don’t think about it, the result would be un-maintainable source.
3 Kommentare. Schreib was dazu
hehe, gerade heute zufällig ein Script geschrieben, was das verhalten nutzt:
while () {
chomp; next unless $_;
@_ = split /s/;
push @_, ++$p;
&create_vhost
}
hehe, gerade heute zufällig ein Script geschrieben, was das verhalten nutzt:
while () {
chomp; next unless $_;
@_ = split /s/;
push @_, ++$p;
&create_vhost
}
hehe, gerade heute zufällig ein Script geschrieben, was das verhalten nutzt:
while () {
chomp; next unless $_;
@_ = split /s/;
push @_, ++$p;
&create_vhost
}