| You, a perfect speller, have a vocabulary of \(N\) distinct words, \(V_1, ..., V_N\), each consisting of exactly \(L\) lowercase letters from the alphabet \(\{\)`'m'`, `'e'`, `'t'`, `'a'`\(\}\). | |
| Your friend, a truely terrable speler, has attempted to write \(Q\) of these words as \(W_1, ..., W_Q\), each also consisting of \(L\) lowercase letters from the same alphabet. | |
| Let \(S_i\) be the number of words in your vocabulary that differ from \(W_i\) at exactly two indices. Please determine the sum \(S_1 + ... + S_Q\). | |
| # Constraints | |
| \(1 \le T \le 95\) | |
| \(1 \le N, Q \le 750{,}000\) | |
| \(1 \le L = |V_i| = |W_i| \le 20{,}000\) | |
| \((N+Q)*L \le 15{,}000{,}000\) | |
| \(V_{ij} \in \{\)`'m'`, `'e'`, `'t'`, `'a'`\(\}\) | |
| \(W_{ij} \in \{\)`'m'`, `'e'`, `'t'`, `'a'`\(\}\) | |
| All \(V_i\) in a given test case are distinct. | |
| The sum of lengths of all strings across all cases is at most \(18{,}000{,}000\). | |
| # Input Format | |
| Input begins with a single integer \(T\), the number of test cases. For each test case, there is first a line containing a single integer \(N\). Then, \(N\) lines follow, the \(i\)th of which contains the string \(V_i\). Then, there is a line containing a single integer \(Q\). Then, \(Q\) lines follow, the \(i\)th of which contains the string \(W_i\). | |
| # Output Format | |
| For the \(i\)th test case, output a single line containing `"Case #i: "` followed by a single integer, the sum \(S_1 + ... + S_Q\). | |
| # Sample Explanation | |
| The first case is depicted below: | |
| {{PHOTO_ID:1332823754208247|WIDTH:500}} | |
| The answer is \(4\), since: | |
| - \(W_1\) = "`teammate`" differs from "`metamate`" at three indices, so \(S_1 = 0\). | |
| - \(W_2\) = "`meatmate`" differs from "`metamate`" at exactly two indices, so \(S_2 = 1\). | |
| - \(W_3\) = "`metatame`" differs from "`metamate`" at exactly two indices, so \(S_3 = 1\). | |
| - \(W_4\) = "`mememate`" differs from "`metamate`" at exactly two indices, so \(S_4 = 1\). | |
| - \(W_5\) = "`metameme`" differs from "`metamate`" at exactly two indices, so \(S_5 = 1\). | |
| In the second case, the answer is \(0\), since: | |
| - \(W_1\) = "`tata`" differs from \(V_1\) = "`meet`" at four indices, \(V_2\) = "`emma`" at three indices, and \(V_3\) = "`tate`" at only one index, so \(S_1 = 0\). | |
| - \(W_2\) = "`maam`" differs from \(V_1\) = "`meet`" at three indices, \(V_2\) = "`emma`" at four indices, and \(V_3\) = "`tate`" at three indices, so \(S_2 = 0\). | |
| In the third case, the answer is \(5\), since: | |
| - \(W_1\) = "`tam`" differs from both \(V_1\) = "`mem`" and \(V_3\) = "`mat`" at exactly two indices, so \(S_1 = 2\). | |
| - \(W_2\) = "`mat`" differs from \(V_1\) = "`mem`" at exactly two indices, so \(S_2 = 1\). | |
| - \(W_3\) = "`tea`" differs from both \(V_1\) = "`mem`" and \(V_2\) = "met" at exactly two indices, so \(S_3 = 2\). | |