In Perl, on what basis should one choose between passing a reference to a hash/array and passing a hashref/arrayref scalar?

To be brief, I'm wondering if there are any best-practice reasons for deciding between:

my %hash = 开发者_StackOverflow中文版( foo => 1, bar => 2 );
# some in-between logic
some_func(\%hash);

and

my $hashref = { foo => 1, bar => 2 };
# some in-between logic
some_func($hashref);

Or is purely a style decision?

The two are equivalent and interchangeable. The decision should be made based on what is clearest for you.

You can also move back and forth:

my $hashref = {x => 1, y => 2};

our %hash; *hash = $hashref;

some_func($hashref);
some_func(\%hash);    # \%hash == $hashref

In general I prefer to work with the plural forms %name and @name since it results in less line noise due to dereferencing. That and some_func(\%var) is clearer with regard to var's type than some_func($var)

These two examples do exactly the same thing. So it mainly depends on which is more convenient for what else, if anything, you do with the %list variable and/or $listref variable.

Or maybe you'd like to skip the extra variable entirely:

some_func( { foo => 1, bar => 2 } );

(Less likely now that you've added those "in-between logic" comments.)

As indicated above, they're the same though the first may give you better context to what type of variable (i.e. a hash) you're using in your "some in-between logic" code.

In most cases it won't make a difference, but I have had some nasty bugs that occurred from trying to change where my reference points if I pass in a reference-of list.

Unless I really want to change the contents of a pre-existing array or hash, I personally favor using anonymous lists/hashes (like your second example).

This is just personal experience, though. I will admit that maybe if I understood Perl better I would know the objective best practices (if any exist).