The reason for the sort is that it starts dropping elements from the end of the array. If you start from the beginning, you won't drop the elements that you want to drop. Say you want to drop the elements at indices 0 and 2; you drop the element at index 0, but then the other elements squish towards the front of the array. So element 2 is now element 1. So you'll drop the wrong elements. That's why you should start from the end.splice @array, $_, 1 for sort { $b <=> $a } @indices;
Update: tilly pointed out a bug where duplicate indices could lead to dropping incorrect elements in the array. Here's a fix:
This is just getting silly in terms of the number of times we're looping over the data; but it does still seem to be quite fast. Faster than I would have expected, in fact.my %u; splice @array, $_, 1 for grep !$u{$_}++, sort { $b <=> $a } @indices;
In reply to Re: Removing certain elements from an array
by btrott
in thread Removing certain elements from an array
by Anonymous Monk
| For: | Use: | ||
| & | & | ||
| < | < | ||
| > | > | ||
| [ | [ | ||
| ] | ] |