In this challenge, we're using regular expressions to detect the various tags used in an HTML document.
- Tags come in pairs. Some tag name, , will have an opening tag, , followed by some intermediate text, followed by a closing tag, . The forward slash in a closing tag will always come before the tag name.
- The exception to this is self-closing tags, which consist of a single tag (not a pair) with a forward slash after the tag name:
Here are a few examples of tags:
- The tag is for paragraphs:
- There may be or more spaces before or after a tag name:
- A void or empty tag involves an opening and closing tag with no intermediate characters:
Some tags can also have attributes, such as the tag, which is used to add a hyperlink to another document:
In the above case, is the tag name and is an attribute having the value .
Task
Given lines of HTML, find the tag names (ignore any attributes) and print them as a single line of lexicographically ordered semicolon-separated values (e.g.: ).
Input Format
The first line contains an integer, , the number of HTML fragments.
Each of the subsequent lines contains a fragment of an HTML document.
Constraints
- Each fragment contains ASCII characters.
- The fragments are chosen from Wikipedia, so analyzing and observing their markup structure may help.
- Leading and trailing spaces/indentation have been trimmed from the HTML fragments.
Output Format
Print a single line containing all of the unique tag names found in the input. Your output tags should be semicolon-separated and ordered lexicographically (i.e.: alphabetically). Do not print the same tag name more than once.
Sample Input
2
<p><a href="http://www.quackit.com/html/tutorial/html_links.cfm">Example Link</a></p>
<div class="more-info"><a href="http://www.quackit.com/html/examples/html_links_examples.cfm">More Link Examples...</a></div>
Sample Output
a;div;p
Explanation
The first line contains tag names: .
The second line contains tag names: .
Our set of unique tag names is .
When we order these alphabetically and print them as semicolon-separated values, we get "".