extract_html method removes some wanted characters

[jasoncorum]

In the extract_html method there's are some expath methods called to remove a large set of characters:

doc.xpath("//h2[translate(text(),concat('~!@#\{$%^&*()_+-=}-";:<>,.?/|[]',"'"),'')='#{title}']")`

We recently wanted to use parens and a dash in an OpenVault Special Collection tab, but their removal in CMLess made it unable to associate the records with the tab.