A ranger station labels every checkpoint along a trail with a lowercase codename. For a safety audit, the station only cares about one contiguous stretch of the checkpoint list, identified by two indices, and wants to know how many codenames in that stretch both begin and end with a vowel ('a', 'e', 'i', 'o', 'u') -- a pattern the rangers use to flag checkpoints near a water source.

Given the full list of codenames and the stretch's boundary indices (inclusive, 0-indexed), count how many codenames in that stretch satisfy the vowel-bracket pattern. A codename of length 1 satisfies the pattern exactly when its single letter is a vowel.

Input format

• Line 1: a single integer n, the number of checkpoints.
• Line 2: n space-separated lowercase strings, the checkpoint codenames, in trail order.
• Line 3: two space-separated integers left and right, the inclusive 0-indexed boundaries of the audited stretch.

Output format

A single integer: the number of codenames among codenames[left..right] whose first and last letters are both vowels.

Constraints

• 1 <= n <= 100000
• each codename consists only of lowercase English letters ('a'-'z')
• 1 <= length of each codename <= 10
• 0 <= left <= right <= n - 1

4
area book ultra look
0 2

2

1
e
0 0

1