PROGRAMMING:Praise the crazy devil
Microblog has a "like" function, you can like for your favorite blog to show support. Every blog post has some labels that describe its characteristics, and the type of blog post you like indirectly describes your characteristics. However, there is such a kind of person, they will brush the sense of existence by praising everything they see. This kind of person is called "praise maniac". The tags they like are too scattered to show obvious characteristics. This problem requires you to write a program, by counting the number of different tags that each person likes, to find out the top three likes.
###Input format:
Enter a positive integer in the first line, $$n $$($$Le 100 $$), which is the number of users to be counted. This is followed by $$n $$lines, each listing a user's like tag. The format is "` name '$$k $$$f"_ 1 \cdots F_ K $$, where 'name' is a non empty user name with no more than 8 English lowercase letters, $$1 / Le K / Le 1000 $$, $$f_ I $$($$I = 1, cdots, K $$) is the number of the feature tag. We will number all the feature tags from 1 to $$10 ^ 7 $. Numbers are separated by spaces.
###Output format:
Count the number of different tags that each person likes, find out the top three with the largest number, and output their user names in order in a line, separated by a space, and there should be no extra space at the end of the line. If there are parallels, the output label with the smallest average number of times will be output, and the title will ensure that such users are not paralleled. If there are less than three people, use '-' to make up for the missing. For example, 'Mike Jenny -' means there are only two people.
###Input example:
```in
five
bob 11 101 102 103 104 105 106 107 108 108 107 107
peter 8 1 2 3 4 3 2 5 1
chris 12 1 2 3 4 5 6 7 8 9 1 2 3
john 10 8 7 6 5 4 3 2 1 7 5
jack 9 6 7 8 9 10 11 12 13 14
```
###Output example:
```out
jack chris john
```
answer：If there is no answer, please comment
###Input format:
Enter a positive integer in the first line, $$n $$($$Le 100 $$), which is the number of users to be counted. This is followed by $$n $$lines, each listing a user's like tag. The format is "` name '$$k $$$f"_ 1 \cdots F_ K $$, where 'name' is a non empty user name with no more than 8 English lowercase letters, $$1 / Le K / Le 1000 $$, $$f_ I $$($$I = 1, cdots, K $$) is the number of the feature tag. We will number all the feature tags from 1 to $$10 ^ 7 $. Numbers are separated by spaces.
###Output format:
Count the number of different tags that each person likes, find out the top three with the largest number, and output their user names in order in a line, separated by a space, and there should be no extra space at the end of the line. If there are parallels, the output label with the smallest average number of times will be output, and the title will ensure that such users are not paralleled. If there are less than three people, use '-' to make up for the missing. For example, 'Mike Jenny -' means there are only two people.
###Input example:
```in
five
bob 11 101 102 103 104 105 106 107 108 108 107 107
peter 8 1 2 3 4 3 2 5 1
chris 12 1 2 3 4 5 6 7 8 9 1 2 3
john 10 8 7 6 5 4 3 2 1 7 5
jack 9 6 7 8 9 10 11 12 13 14
```
###Output example:
```out
jack chris john
```
answer：If there is no answer, please comment