[Purdue-pm] Can do better with map somehow?

[Dave Jacoby]

There is something to that. I don't speak "map" as fluently as I should yet.

On 11/28/2010 03:36 PM, Mark Senn wrote:
> Dave Jacoby<jacoby at purdue.edu>  wrote:
>      My new solution:
>
>      my $first = {
>          'a' =>  { 'foo' =>  1, 'bar' =>  'Cheers', },
>          'b' =>  { 'foo' =>  3, 'bar' =>  'Crow', },
>          'c' =>  { 'foo' =>  1, 'bar' =>  'Neon Cactus', } } ;
>      my $second ;
>      %$second =
>          map { $_, $first->{ $_ } }
>          grep { $first->{ $_ }->{ foo } == 1 } #<-- WOW!
>          keys %$first ;
>      say Dumper $second ;
>
> I think (but don't know for sure) Perl 5 will do the grep and
> save all those results and then do the map.  For this small
> example it probably won't matter much but if %$first was very
> large the above may run out of memory.
>
> I prefer this:
>
> my $third;
> map  { ($first->{$_}->{foo} == 1)  and  $third->{$_} = $first->{$_} }
>      keys %$first;
> say Dumper $third;
>
> -mark


-- 
Dave Jacoby                         Address: WSLR S049
Code Maker                          Mail:    jacoby at purdue.edu
Purdue University                   Phone:   765.49.67368