For any specific Javascript, you can implement the same function in Perl and then get the result that way.

If you want a generic solution for getting the result of arbitrary Javascript code, you will have to run the Javascript. Example environments for Javascript in a web context could be WWW::Mechanize::Firefox and WWW::Mechanize::PhantomJS.

For your specific example (not tested):

/replace_image\(([^,]+)/
[download]

/replace_image\(["']([^"']+)/
[download]

should fetch the URL into $1 for you.

The url in the JS is fixed per page, so one per page. I parse lots of pages and want to extract the URL. I tried the code you gave

my $wanted = $page =~ /push_delayed_image\(([^,]+)/;

print Dumper $wanted;
[download]

But this prints $VAR - 1; not the URL

You need list context to get the capture groups back.

my ($wanted) = $page =~ /push_delayed_image\(([^,]+)/;
#  ^       ^
[download]

($q=q:Sq=~/;[c](.)(.)/;chr(-||-|5+lengthSq)`"S|oS2"`map{chr |+ord
}map{substrSq`S_+|`|}3E|-|`7**2-3:)=~y+S|`+$1,++print+eval$q,q,a,
[download]

Alternately to what Choroba said, you could:

$page =~ /push_delayed_image\(([^,]+)/;
my $wanted = $1;
[download]

Often, I'm using a conditional, so will do this:

if ($page =~ /push_delayed_image\(([^,]+)/)
{
    my $wanted = $1;
    ...
}
[download]

That depends on what the part of the page source where the URLs are stored looks like, show us that?