Given a string s
. You should re-order the string using the following algorithm:
- Pick the smallest character from
s
and append it to the result. - Pick the smallest character from
s
which is greater than the last appended character to the result and append it. - Repeat step 2 until you cannot pick more characters.
- Pick the largest character from
s
and append it to the result. - Pick the largest character from
s
which is smaller than the last appended character to the result and append it. - Repeat step 5 until you cannot pick more characters.
- Repeat the steps from 1 to 6 until you pick all characters from
s
.
In each step, If the smallest or the largest character appears more than once you can choose any occurrence and append it to the result.
Return the result string after sorting s
with this algorithm.
Input: s = "aaaabbbbcccc" Output: "abccbaabccba" Explanation: After steps 1, 2 and 3 of the first iteration, result = "abc" After steps 4, 5 and 6 of the first iteration, result = "abccba" First iteration is done. Now s = "aabbcc" and we go back to step 1 After steps 1, 2 and 3 of the second iteration, result = "abccbaabc" After steps 4, 5 and 6 of the second iteration, result = "abccbaabccba"
Input: s = "rat" Output: "art" Explanation: The word "rat" becomes "art" after re-ordering it with the mentioned algorithm.
Input: s = "leetcode" Output: "cdelotee"
Input: s = "ggggggg" Output: "ggggggg"
Input: s = "spo" Output: "ops"
1 <= s.length <= 500
s
contains only lower-case English letters.
impl Solution {
pub fn sort_string(s: String) -> String {
let mut cnt = [0; 26];
let mut ret = Vec::new();
for ch in s.bytes() {
cnt[(ch - b'a') as usize] += 1;
}
while ret.len() < s.len() {
for i in 0..26 {
if cnt[i] > 0 {
ret.push(i as u8 + b'a');
cnt[i] -= 1;
}
}
for i in (0..26).rev() {
if cnt[i] > 0 {
ret.push(i as u8 + b'a');
cnt[i] -= 1;
}
}
}
String::from_utf8(ret).unwrap()
}
}