Slim style HTML Parser

slim.clj

(ns slim.core)
(use '[clojure.contrib.duck-streams :only (read-lines spit)])

(defstruct html-node :depth :leading_whitespace :interpreted :selector :attributes :content)

(defn process-line [acc line]
  (let [[full-text 
         leading_whitespace 
         interpreted
         selector 
         full_attributes 
         attributes 
         content] (re-find #"^([\s]*)([-=;|]?)([\w]*)([(](.*?)[)])?(.*)$" line)]
    (if (or (= (.trim full-text) "")  ; empty line
            (= interpreted ";"))  ;  comment
      acc
      (conj acc (struct-map html-node :depth (count leading_whitespace), 
                                    :leading_whitespace leading_whitespace,
                                    :interpreted_type interpreted,
                                    :interpreted (or (= interpreted "=") (= interpreted "-")),
                                    :selector selector, 
                                    :attributes attributes, 
                                    :content (.trim content))))))

(defn process-file [file-name line-accum]
  (reduce process-line line-accum (read-lines file-name)))

(def self-closing-tags
  #{"meta" "br" "link"})

(defn attributes [attr]
  (if (nil? attr) "" (str " " attr)))

(defn opening-tag [node]
  (if (contains? self-closing-tags (node :selector))
    (str "\n" (node :leading_whitespace) "<" (node :selector) (attributes (node :attributes)) " />")
    (str "\n" (node :leading_whitespace) "<" (node :selector) (attributes (node :attributes)) ">")))

(defn closing-tag [node]
  (if (contains? self-closing-tags (node :selector)) "" (str "\n" (node :leading_whitespace) "</" (node :selector) ">")))

(defn interpreted-clojure [node]
  (load-string (node :content)))
  
(defn content [node]
  (if (= "" (node :content)) "" (str "\n" (node :leading_whitespace) "  " (node :content))))

(defn write-html [data stack]
  (if (empty? data)
    (reduce #(str %1 (closing-tag %2)) "" stack)
    (let [current (first data)]
    (if (<= (get current :depth -1) (get (first stack) :depth -2))
      (str (closing-tag (first stack)) (write-html data (rest stack)))
      (if (current :interpreted)
        (if (= "=" (current :interpreted_type)) 
          (str "\n" (current :leading_whitespace) (interpreted-clojure current) (write-html (rest data) stack))
          (do (interpreted-clojure current) (str (write-html (rest data) stack))))
        (if (= (current :interpreted_type) "|")
          (str (content current) (write-html (rest data) stack))
          (str (opening-tag current) (content current) (write-html (rest data) (conj stack current)))))))))
    
(time (write-html (process-file (str "test.slim.html") []) '()))
(spit "test.html" (write-html (process-file (str "test.slim.html") []) '()))

;;  TODO
;;  Add Doctype Directive

test.slim.html

- (def things ["cars" "trucks" "automobiles" "kittens" "clothes"])
- (def other_things ["losers" "winners" "fighters"])
- (def and_more_things ["coders" "kangaroos" "guns"])

- (defn greeting [name] (str "Hello, " name))

- (defn fib-step [ab] [(last ab) (+ (first ab) (last ab))])
- (defn fib-seq [] (map first (iterate fib-step [0 1])))

; This is a comment

html(xmlns="http://www.w3.org/1999/xhtml")
  head
    title Page Title
    meta(name='keywords' content="template language")
    link(rel="alternate" href="/styles.css")
    
  body
    h1 Hello World
    div
      p(class='my classes' id='my_id') This is a paragraph. Lorem ipsum dolor ( ) sit amet, 
        | consectetur adipisicing elit, sed do eiusmod tempor incididunt ut labore et 
        | dolore magna aliqua. & Ut enim ad minim veniam, quis nostrud exercitation 
        | ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor 
        | in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur. 
        | Excepteur sint occaecat cupidatat non proident, sunt in culpa qui officia deserunt 
        | mollit anim id est laborum.
      ul
        = (reduce #(str %1 "<li>" %2 "</li>") "" things)
        = (reduce #(str %1 "<li>" %2 "</li>") "" other_things)
        = (reduce #(str %1 "<li>" %2 "</li>") "" and_more_things)
        li
          a(class="hello" href='http://google.com') Go To Google
      p
        = (greeting "Kathleen")
        | You are awesome
        | and fun
    = (str "test")
    h3(style="color:green") Fibonacci Numbers    
    ul
      = (reduce #(str %1 "<li>" %2 "</li>") "" (take 10 (fib-seq)))
    
    div(id="footer") Copyright &copy; 2010 Chas Lemley
    script(type="text/javascript")
      | $(content).do_something();