> but it doesn't download pdf files from:

Are the links converted as absolute (http://....) links ?
Did you try to use the follwing scan rule:
+*.pdf
?

Roderick Lengkeek

Xavier Roche