find $DOCUMENTROOT -type f -name '*.html' | wc -l
[download]

You might want to tweak this, depending on what you consider a webpage.

Nothing fancy, not even perl, needed.

Abigail