Re: Re: Re: Re: Never lock $0 inside of a BEGIN block

[demerphq]

Heh. Ive encountered a bunch of weirdness with using my and modifiers. (I've even argued that it should produce a warning.) I haven't seen the static variable trick. Thats scary. Cool Scary though. :-)

However I'm still not convinced that this error message couldn't be improved.

BTW, the static trick only works with if 0; and for (); and thus I still consider it something that should produce a warning.

sub umm()    { print "---\n"}
sub _caller(){(caller(1))[3]=~/^(?:main::)?(.*)$/}

sub for_blah
{
  my $foo for 1;
  printf "%-15s %d\n",_caller,$foo++;
}

sub not_for_blah
{
  my $foo for ();
  printf "%-15s %d\n",_caller,$foo++;
}

sub if_blah
{
  my $foo if 1;
  printf "%-15s %d\n",_caller,$foo++;
}

sub not_if_blah
{
  my $foo if 0;
  printf "%-15s %d\n",_caller,$foo++;
}

for_blah(),for_blah(),for_blah(),umm,
not_for_blah(),not_for_blah(),not_for_blah(),umm,
if_blah(),if_blah(),if_blah(),umm,
not_if_blah(),not_if_blah(),not_if_blah(),umm;
__END__

# No assignment        #With assignment 
for_blah        0      for_blah        0
for_blah        0      for_blah        0
for_blah        0      for_blah        0
---                    ---              
not_for_blah    0      not_for_blah    0
not_for_blah    1      not_for_blah    1
not_for_blah    2      not_for_blah    2
---                    ---              
if_blah         0      if_blah         1
if_blah         0      if_blah         1
if_blah         0      if_blah         1
---                    ---              
not_if_blah     0      not_if_blah     0
not_if_blah     1      not_if_blah     1
not_if_blah     2      not_if_blah     2
---                    ---

---
demerphq

_{<Elian> And I do take a kind of perverse pleasure in having an OO assembly language...}

[ChemBoy]

BTW, the static trick only works with if 0; and for ();

Not so. Observe:

#!/usr/local/bin/perl -l

sub horror { 
    my $foo if "reset" eq shift; 
    print ++$foo;
}

horror($_) for qw(foo bar baz reset bing bang boom);
__END__
1
2
3
4
1
2
3

(Not that the sub name is intended as an editorial comment, or anything... <grin>)

The nasty part is just that the cleanup (resetting $foo to undef) takes place at the end of the iteration in which my $foo is actually executed (which isn't exactly what you'd expect from a "reset" argument, but that's what you get with silly example code).




---

If God had meant us to fly, he would *never* have given us the railroads.
    --Michael Flanders

[demerphq]

Not so.

Yes so. The code you demonstrate illustrates this nicely too. Perhaps you thought I meant a literal 0? I didnt I meant if FALSE. This behaviour while nice is a bug IMO.


---
demerphq

_{<Elian> And I do take a kind of perverse pleasure in having an OO assembly language...}