Perl for Web Site Management

Scoping via Anonymous Blocks

And there is. The solution is to stick something called an anonymous block around the subroutine (which just means a pair of curly braces with no associated subroutine name), with the my %date_seconds declaration inside that block but outside the subroutine proper. That is, something like:

{   # start of anonymous block
    my %date_seconds;
    sub get_seconds {
       # subroutine definition goes here
    }
}   # end of anonymous block

This way, the same %date_seconds hash is visible to every invocation of the &get_seconds subroutine, and thus is available for caching timelocal lookup results, but the hash remains invisible for the rest of the script.

Remember that %month_num hash we initialized up at the top of the script? Since the &get_seconds subroutine is the only thing in the script that uses it, it would be nice to limit its scope, too, by putting it inside the anonymous block. (Putting it inside the subroutine itself would work, but it would be inefficient because the initialization of the hash would have to be repeated each time the subroutine was invoked.)

There is a problem with putting the %month_num initialization inside the anonymous block, though. You will recall that Perl goes through two phases when it runs a script. The first phase, called the compilation phase , is the time when things like subroutine definitions are processed. After that comes the runtime phase, when the sequential commands in the script are actually executed. This is why we can successfully ...