Let us call a non-empty sequence of lowercase English letters a word. Prefix of a word x is a word y that can be obtained from x by removing zero or more last letters of x.

Let us call two words similar, if one of them can be obtained from the other by removing its first letter.

You are given a set S of words. Find the maximal possible size of set of non-empty words X such that they satisfy the following:

• each word of X is prefix of some word from S;
• X has no similar words.

Input

Input data contains multiple test cases. The first line of the input data contains an integer t — the number of test cases. The descriptions of test cases follow.

The first line of each description contains an integer n — the number of words in the set S (1 ≤ n ≤ 106). Each of the following n lines contains one non-empty word — elements of S. All words in S are different.

It is guaranteed that the total length of all words in one input data doesn't exceed 106.

Output

For each test case print one line that contains one integer m — the maximal number of words that X can contain.

见题面 Input 部分

见题面 Output 部分