Slightly off-topic: Does HN fail to parse an URL with no slashes?
(def parse-site (url)
(rev (tokens (cadr (tokens url [in _ #\/ #\?])) #\.)))
It basically tokenizes the URL "http://example.com/foo/bar?quux#woah" into a list "http:", "example.com", "foo", etc. (no "" since tokens uses consif internally) and takes the second element as a domain name. (It has a prior knowledge of url starting with either "http://" or "https://".) Obviously "#" is missing in the list of separators.Edit: I filed an issue to the HN issue tracker. https://github.com/HackerNews/HN/issues/35