in reply to Schwartzian Transform vs. Building Hash of Function Results, Then Sorting on Function Results
in thread Schwartzian Transform

davebaker,
In both cases, you are calculating the expensive_func() only once per item and sorting the list based off the result of the expensive_func(). The biggest difference is that in the standard ST, you do it in a single step and have no left over variable.

Cheers - L~R

  • Comment on Re: Schwartzian Transform vs. Building Hash of Function Results, Then Sorting on Function Results

Replies are listed 'Best First'.
Left over variable
by davebaker (Pilgrim) on Oct 16, 2006 at 19:35 UTC
    Thanks! I hadn't considered the left over variable. I'd want to add this line before my sorted-hash code snippet: 
    my %result_for;

    [download]
    (That doesn't eliminate any drawbacks to having a left over variable as compared to the ST, but it would be safer coding I think.)
[reply]
[d/l]

In Section Cool Uses for Perl