Lithuanian gives you good practice with regex

I'm parsing Lithuanian verbs, and in tidying up the raw HTML, I'm compelled to come up with little beauties like this:

<a href="\/[a-z]*\/[a-z]*\W*[a-z]*\W*[a-z]*">

...in order to to clear away the HTML formatting.

The problem lies in the fact that Lithuanian diacritics (č, ž, ė, ę and so on) are not recognised as letters [a-z] but rather as non-word characters \W.