GoogleBot constantly crawls non-existant PDFs with ‘viagra’ or ‘cialis’ in the name?

My access log is full of requests for non-existent pdfs relating to ‘viagra’ and ‘cialis’ or other similar drugs from GoogleBot (user-agent is: Googlebot/2.1 (+http://www.google.com/bot.html) ip range is 66.249.64.*)

examples:

/WdjUZ/LXWKZ/cialis-hearing-loss.pdf /WKYnZ/viagra-questions.pdf /ZWohZ/LhfkZ/canadian-viagra-and-healthcare.pdf /YSXaZ/XnoZZ/buy-propecia-no-prescription.pdf /MRWQZ/MeWXZ/TZlWZ/UlaMZ/drug-manufacturers-buy-softtabs-viagra.pdf /PnddZ/NKdZZ/generic-viagra-no-prescription-australia.pdf /QQWVZ/RoRbZ/URObZ/LdNgZ/levitra-10-mg-order.pdf 

Why would google think these are urls that need to be crawled? If someone else is hosting a page with links like this, what purpose would that serve?

Is there any harm in creating a robots.txt rule to tell google to stop like so:

User-agent * Disallow /*viagra*.pdf$   Disallow /*cialis*.pdf$   Disallow /*propecia*.pdf$   Disallow /*levitra*.pdf$