CF41C · Email address

• 难度：1300
• 标签：expression parsing、implementation
• 链接：https://codeforces.com/problemset/problem/41/C
• 时间限制：2 seconds　内存限制：256 megabytes
• 出现位置：Day02-表达式解析-栈-括号序列

英文原题面

Statement

Sometimes one has to spell email addresses over the phone. Then one usually pronounces a dot as dot, an at sign as at. As a result, we get something like vasyaatgmaildotcom. Your task is to transform it into a proper email address ([email protected]). It is known that a proper email address contains only such symbols as . @ and lower-case Latin letters, doesn't start with and doesn't end with a dot. Also, a proper email address doesn't start with and doesn't end with an at sign. Moreover, an email address contains exactly one such symbol as @, yet may contain any number (possible, zero) of dots. You have to carry out a series of replacements so that the length of the result was as short as possible and it was a proper email address. If the lengths are equal, you should print the lexicographically minimal result. Overall, two variants of replacement are possible: dot can be replaced by a dot, at can be replaced by an at.

Input

The first line contains the email address description. It is guaranteed that that is a proper email address with all the dots replaced by dot an the at signs replaced by at. The line is not empty and its length does not exceed 100 symbols.

Output

Print the shortest email address, from which the given line could be made by the described above replacements. If there are several solutions to that problem, print the lexicographically minimal one (the lexicographical comparison of the lines are implemented with an operator < in modern programming languages). In the ASCII table the symbols go in this order: . @ ab...z

样例

样例 1

输入：

vasyaatgmaildotcom

输出：

[email protected]

样例 2

输入：

dotdotdotatdotdotat

输出：

[email protected]

样例 3

输入：

aatt

输出：

a@t