There are a few examples given in my further posts; most URL structures are SEO friendly, offering an easy way to match differences between them.
Those subtle differences don't need big code to detect; rather strict rules before the routine is satisfied with it's own results and creativity/knowledge where to start.

I've checked a few approaches including but not limited to heuristics, fuzzy matching, Regexp::Assemble etc.. but found no real good results yet.

Currently my best approach is keeping a hash with known sites and their structure, but like to get that hash automatic instead of by hand because the Internet is rather big ;)