minor correction, you need a '->' in the $a in the sort:
@sorted = map {$_->[1]}
sort { $a->[0] <=> $b->[0] }
map { [$_[4],$_] } @unsorted;
and correct me if I'm mistaken (please), but isn't the Schwartzian
Transform used in cases where it is necessary to modify
the sort key? ie. dictonary sorting mixed case words.
/\/\averick
| [reply] [Watch: Dir/Any] [d/l] |
Thanks for the correction :)
Naw, the transform doesn't have to modify the sort key... its more general purpose than that (though that is an excellent use). It is just a good algorithm for sorting (array based) data that is more complicated (and or a lot larger) than a plain array. Recently, for instance, I had an array of hashes, and used the transform to sort the array by the contents of one of the hash items.
For instance, if we had a phone book or something, consisting of an array of hashes, each hash being an entry ($phonebook[0]{firstname} would access the firstname, for example) we could do the transform to sort by say... or something.
@sorted_phonebook = map {$_->[1]}
sort {$a->[0] cmp $b->[0]}
map {[$_{lastname},$_]} @phonebook;
Now, we could indeed do a more complex grabbing of the field we are sorting by and/or a more complex sort algorithm, one in which we needed to modify the key even, but the transform is useful for this case as well.
--awwaiid
| [reply] [Watch: Dir/Any] [d/l] [select] |