When using wget -rk to locally mirror a website, wget misses to do some things which these two scripts fix:

• if a url contains a GET query string like http://mysite.com/foo?bar=baz&blub=bla then a file with name foo?bar=baz&blub=bla gets saved. The links from other documents to this file do not get urlencoded. This is fixed by urlencode.py.
• if the copy is put online then the webserver will most likely determine the content type of the static content by the filename extension. But files that are saved under a name like foo?bar=baz&blub=bla do not have a recognized extension. This is fixed by filenameext.sh.

Both scripts together allowed to fix a mirror of a phpbb forum for me.