You could try this.

#! perl -sw
use strict;
use Data::Dumper;

my $source = do {local $/; <DATA> }; # Slurp the source

# build an array of chunks that look like parms to Lique()
my (@parms) = $source =~ m/=\s+(?:new\s+)?CX::Lique(?:->new\s*?)?\((.*
+?)\);/gcs;

# Processs the array and build a hash from the chunks.
for my $parm (@parms) {
    my %info;
    $parm =~ s/'?(\w+)'?\s+=>\s+'?([^,')]+)'?,?\n/$info{$1} = $2;/seg;
    print Dumper(\%info);
}

__DATA__
my $fun = new CX::Lique(
  name   => 'Test Addon',
  'author'=> 'nutshell',
  email  => 'x@x.com',
  descrip => 'That',
  version => '1.1',
  build => 2
);

my $q = CX::Lique->new(

  name   => 'Test',
  'author'=> 'yeehaw',
  'email' => 'yeehaw@yahoo.com',
  descrip => 'Fun',
  version => '1.11',
  build  => 2
);
[download]

Which gives this. No guarentees it will work for all the possible ways this could be structured in the source though.

C:\test>202407
$VAR1 = {
          'descrip' => 'That',
          'build' => '2',
          'version' => '1.1',
          'email' => 'x@x.com',
          'name' => 'Test Addon'
        };
$VAR1 = {
          'descrip' => 'Fun',
          'build' => '2',
          'version' => '1.11',
          'email' => 'yeehaw@yahoo.com',
          'name' => 'Test'
        };

C:\test>
[download]

Thanks SOO much++

--nutshell

No regex needed. Just an extra set of curly braces takes care of the whole thing. Putting Lique() in a package left AAEFTR.

#! perl -sw
use strict;
use Data::Dumper;
my $debug = 1;

sub Lique  {
    my $inforef = shift;
    print Dumper( $inforef ) if $debug;
    # do stuff
    return $inforef;
}


my $fun = Lique(
    { # << Note!
        name   => 'Test Addon',
        'author'=> 'nutshell',
        email  => 'x@x.com',
        descrip => 'That',
        version => '1.1',
        build => 2
    } # << Note!
);

my $q = Lique(
    {  # << Note!
        name   => 'Test',
        'author'=> 'yeehaw',
        'email' => 'yeehaw@yahoo.com',
        descrip => 'Fun',
        version => '1.11',
        build  => 2
    } # << Note!
);

print Dumper( $fun, $q );

__DATA__
C:\test>202407
$VAR1 = {
          'author' => 'nutshell',
          'descrip' => 'That',
          'build' => 2,
          'version' => '1.1',
          'email' => 'x@x.com',
          'name' => 'Test Addon'
        };
$VAR1 = {
          'author' => 'yeehaw',
          'descrip' => 'Fun',
          'build' => 2,
          'version' => '1.11',
          'email' => 'yeehaw@yahoo.com',
          'name' => 'Test'
        };
$VAR1 = {
          'author' => 'nutshell',
          'descrip' => 'That',
          'build' => 2,
          'version' => '1.1',
          'email' => 'x@x.com',
          'name' => 'Test Addon'
        };
$VAR2 = {
          'author' => 'yeehaw',
          'descrip' => 'Fun',
          'build' => 2,
          'version' => '1.11',
          'email' => 'yeehaw@yahoo.com',
          'name' => 'Test'
        };

C:\test>
[download]

What's the origin of $code? Does this do what you want?

my $info = \%$fun;
# or
my $info = {};
@$info{ keys %$fun } = values %$fun;
[download]

You see - I cannot let it physically execute the code; the only way is messy regex I'm thinking :\

--nutshell

In that case, you going to have to explain your problem a lot more clearly.

What is "blow code"?

You mention a var, $code twice, but it doesn't appear anywhere in your examples.

What you appear to be saying is that you have a program source or sources, that makes calls to a function Lique() in a package CX, and that you want to parse these sources, locate each call to new CX::Lique() or CX::Lique->new() and the extract the parameters passes on these calls into a hashref %info using a regex?

Am I closer?

---

Cor! Like yer ring! ... HALO dammit! ... 'Ave it yer way! Hal-lo, Mister la-de-da. ... Like yer ring!

Yes!

That's exactly what I need to do :)

--nutshell