Write script to filter some xml code (regex) like the one attached.

I believe the steps to do this are: 1. Remove <commentary> tags and their parameters, but leave their contents. 2. Remove all <a tref= … > tags. 3. Remove all tag parameters apart from href=. 4. Change <h1 … > tags to <p class="cleaned">. 5. Remove <h4 … > tags AND their contents. 6. Change <p> tags into <p class="cleaned"> 7. Close up repeated new-lines. The result should be just a list of <p class="cleaned"> tags with text and simple <a href=> links. This will be run on a Mac but I suspect standard Linux/Unix text processing tools will work.