Squid and Phishy filtering
A month or two ago I had a random thought, “Wouldn’t it be cool to integrate phishtank.com’s database of online “phishy” URLs into Squid somehow?”
My next target is to extend it to include Google’s SafeBrowsing API and downloadable hash database, which will provide enterprise organisations, ISPs, educational instutions and the like more of a reason to run Squid to provide extra safety against all the evil crap thats out there.
September 11, 2007 at 8:56 pm
Hey,
I’m also quite interested in integrating Phistank’s database into Squid. Did you actually manage to do so? And if you did, would you mind to write a little how-to?
September 14, 2007 at 7:41 am
It wasn’t that difficult really. The trouble is that they’ve asked me to not release it in its current form - which sucks down the whole database once an hour - since, and quite understandably, they never really intended the full download to be used by potentially thousands of end-users.
In the end I just took the xml, cut out the URLs, populated a hash table on disk and used that table in an external_acl helper. Perl made this all very simple.
October 10, 2007 at 1:01 pm
I just released a simple version of redirector for squid, that use Google Safe Browsing API to check URLs. You can find it at http://xtalk.msk.su/~ott/en/squid-gsb.html
If you will interested, i could change some parts to decrease dependencies on Boost C++ Libraries