2 files changed, 29 insertions(+), 3 deletions(-)

M crawler.lisp
M packages.lisp

@@ 1,4 1,7 @@
-(defvar *wikipedia-api* "https://en.wikipedia.org/w/api.php")
+(in-package wikisophy.crawler)
+
+(defvar *wikipedia-api-url* "https://en.wikipedia.org/w/api.php")
+(defvar *wikipedia-url* "https://en.wikipedia.org")
 
 (defun query-wikipedia (input)
   "Given a string input return the url of the most relevant wikipedia


@@ 11,7 14,7 @@
      (cadddr
       (cl-json:decode-json-from-string
        (map 'string #'code-char
-	    (drakma:http-request *wikipedia-api* :parameters parameters)))))))
+	    (drakma:http-request *wikipedia-api-url* :parameters parameters)))))))
 
 (defun parse-html (url)
   "Given a wikipedia url parse the html into a PLUMP-DOM element"


@@ 52,3 55,25 @@
 		   do (if (validate-link a p)
 			  (return-from outer
 			    (aref (lquery:$ a (attr :href)) 0)))))))
+
+(defun title-from-url (url)
+  "Given a wikipedia url return the articles title."
+  (let ((mark (search "/wiki/" url)))
+    (subseq url (+ 6 mark)))) ; 6 denotes the length of "/wiki/"
+
+(defun path-to-philosophy (input)
+  "Given a string as an input return a list that shows the path from
+  the first wikipedia article to Philosophy. Endings include detecting
+  a cycle or a page without a first link."
+  (let ((url (query-wikipedia input))
+	(end (concatenate 'string *wikipedia-url* "/wiki/Philosophy")))
+    (if url
+	(loop while (string-not-equal url end)
+	      when (member (title-from-url url) titles-list :test #'string-equal)
+		do (return (append titles-list '("Cycle detected!")))
+	      when (not url)
+		do (return (append titles-list '("No links page detected!")))
+	      collect (title-from-url url) into titles-list
+	      do (setq url (concatenate 'string *wikipedia-url* (grab-link url)))
+	      finally (return (append titles-list '("Philosophy"))))
+	'("There is no article to start crawling from."))))

@@ 1,2 1,3 @@
 (defpackage wikisophy.crawler
-  (:use :cl))
+  (:use :cl)
+  (:export :path-to-philosophy))