use markdown parser to identify code blocks in literate coffeescript

package.json

@@ -23,7 +23,7 @@
   },
   "preferGlobal": true,
   "scripts": {
-    "test":         "node ./bin/cake test",
+    "test": "node ./bin/cake test",
     "test-harmony": "node --harmony ./bin/cake test"
   },
   "homepage": "http://coffeescript.org",
@@ -38,5 +38,8 @@
     "highlight.js": "~8.0.0",
     "underscore": "~1.5.2",
     "docco": "~0.7.0"
+  },
+  "dependencies": {
+    "marked": "~0.3.3"
   }
 }

src/helpers.coffee

@@ -2,6 +2,17 @@
 # the **Lexer**, **Rewriter**, and the **Nodes**. Merge objects, flatten
 # arrays, count characters, that sort of thing.
 
+marked = require 'marked'
+# marked.setOptions
+#   renderer: new marked.Renderer()
+#   gfm: true
+#   tables: true
+#   breaks: false
+#   pedantic: false
+#   sanitize: true
+#   smartLists: true
+#   smartypants: false
+
 # Peek at the beginning of a given string to see if it matches a sequence.
 exports.starts = (string, literal, start) ->
   literal is string.substr start, literal.length
@@ -67,19 +78,22 @@ exports.some = Array::some ? (fn) ->
   return true for e in this when fn e
   false
 
-# Simple function for inverting Literate CoffeeScript code by putting the
-# documentation in comments, producing a string of CoffeeScript code that
-# can be compiled "normally".
+# Simple function for extracting code from Literate CoffeeScript by stripping
+# out all non-code blocks,  producing a string of CoffeeScript code that can
+# be compiled "normally".
 exports.invertLiterate = (code) ->
-  maybe_code = true
-  lines = for line in code.split('\n')
-    if maybe_code and /^([ ]{4}|[ ]{0,3}\t)/.test line
-      line
-    else if maybe_code = /^\s*$/.test line
-      line
-    else
-      '# ' + line
-  lines.join '\n'
+  # don't know how to avoid this hack using token as placeholder for tabs, then
+  # re-inserting tabs after code extraction. The token has been split in two
+  # so that it does not end up getting parsed in this src code
+  token = '9ddb1d26184bdaf8'+'d4de55835d82eb56'
+  code = code.replace "\t", token
+  # parse as markdown, discard everything except code blocks
+  out = ""
+  for item in marked.lexer code, {}
+    out += "#{item.text}\n" if item.type == 'code'
+  # put the tabs back in
+  out.replace token, "\t"
+  out
 
 # Merge two jison-style location data objects together.
 # If `last` is not provided, this will simply return `first`.

test/literate.litcoffee

@@ -55,3 +55,111 @@ Tabs work too:
 
 				test "tabbed code", ->
 					ok yes
+
+---
+
+    # keep track of whether code blocks are executed or not
+    executed = false
+
+<p>
+
+    executed = true # should not execute, this is just HTML para, not code!
+
+</p>
+
+    test "should ignore indented sections inside HTML", ->
+      eq executed, false
+
+---
+
+*   A list item with a code block:
+
+        test "basic literate CoffeeScript parsing", ->
+          ok yes
+
+---
+
+*   Lorem ipsum dolor sit amet, consectetuer adipiscing elit.
+    Aliquam hendrerit mi posuere lectus. Vestibulum enim wisi,
+    viverra nec, fringilla in, laoreet vitae, risus.
+
+*   Donec sit amet nisl. Aliquam semper ipsum sit amet velit.
+    Suspendisse id sem consectetuer libero luctus adipiscing.
+
+---
+
+1.  This is a list item with two paragraphs. Lorem ipsum dolor
+    sit amet, consectetuer adipiscing elit. Aliquam hendrerit
+    mi posuere lectus.
+
+    Vestibulum enim wisi, viverra nec, fringilla in, laoreet
+    vitae, risus. Donec sit amet nisl. Aliquam semper ipsum
+    sit amet velit.
+
+2.  Suspendisse id sem consectetuer libero luctus adipiscing.
+
+---
+
+1.  This is a list item with two paragraphs. Lorem ipsum dolor
+    sit amet, consectetuer adipiscing elit. Aliquam hendrerit
+    mi posuere lectus.
+
+    Vestibulum enim wisi, viverra nec, fringilla in, laoreet
+    vitae, risus. Donec sit amet nisl. Aliquam semper ipsum
+    sit amet velit.
+
+2.  Suspendisse id sem consectetuer libero luctus adipiscing.
+
+---
+
+*   A list item with a blockquote:
+
+    > This is a blockquote
+    > inside a list item.
+
+---
+
+This next one probably passes because a string is inoffensive in compiled js, also, can't get `marked` to parse it correctly, and not sure if empty line is permitted between title and reference
+
+This is [an example][id] reference-style link.
+[id]: http://example.com/
+
+    "Optional Title Here"
+
+---
+
+    executed = no
+
+1986. What a great season.
+          executed = yes
+
+and test...
+
+    test "should recognise indented code blocks in lists", ->
+      ok executed
+
+---
+
+    executed = no
+
+1986. What a great season.
+
+          executed = yes
+
+and test...
+
+    test "should recognise indented code blocks in lists with empty line as separator", ->
+      ok executed
+
+---
+
+    executed = no
+
+1986\. What a great season.
+           executed = yes
+
+and test...
+
+    test "should ignore indented code in escaped list like number", ->
+      eq executed, no
+