url-spec.clj

(require
  '[cemerick.url :as url]
  '[clojure.spec.alpha :as s]
  '[clojure.spec.gen.alpha :as sgen])

(defn non-empty-string-alphanumeric
  []
  (sgen/such-that #(not= "" %)
    (sgen/string-alphanumeric)))

(defn url-gen
  "Generator for generating URLs; note that it may generate 
  http URLs on port 443 and https URLs on port 80, and only 
  uses alphanumerics"
  []
  (sgen/fmap
    (partial apply (comp str url/->URL))
    (sgen/tuple
      ;; protocol
      (sgen/elements #{"http" "https"})
      ;; username
      (sgen/string-alphanumeric)
      ;; password
      (sgen/string-alphanumeric)
      ;; host
      (sgen/string-alphanumeric)
      ;; port
      (sgen/choose 1 65535)
      ;; path
      (sgen/fmap #(->> %
                       (interleave (repeat "/"))
                       (apply str))
        (sgen/not-empty
          (sgen/vector
            (non-empty-string-alphanumeric))))
      ;; query
      (sgen/map
        (non-empty-string-alphanumeric)
        (non-empty-string-alphanumeric)
        {:max-elements 2})
      ;; anchor
      (sgen/string-alphanumeric))))

(s/def ::url (s/with-gen
               (s/and string?
                      #(try
                         (url/url %)
                         (catch Throwable t false)))
               url-gen))
               
(sgen/generate (url-gen))

(s/valid? ::url "http://conan.is")