1 files changed, 1021 insertions, 0 deletions
diff --git a/webapps/qooxdoo-0.6.3-sdk/frontend/framework/tool/modules/treegenerator.py b/webapps/qooxdoo-0.6.3-sdk/frontend/framework/tool/modules/treegenerator.py
new file mode 100755
index 0000000000..d473520e80
--- /dev/null
+++ b/webapps/qooxdoo-0.6.3-sdk/frontend/framework/tool/modules/treegenerator.py
@@ -0,0 +1,1021 @@
+#!/usr/bin/env python
+
+import sys, optparse
+import tokenizer, tree, filetool, variableoptimizer
+
+
+SINGLE_LEFT_OPERATORS = [ "NOT", "BITNOT", "ADD", "SUB", "INC", "DEC" ]
+
+SINGLE_RIGHT_OPERATORS = [ "INC", "DEC" ]
+
+MULTI_TOKEN_OPERATORS = [ "HOOK", "ADD", "SUB", "MUL", "DIV", "MOD", \
+  "LT", "LE", "GT", "GE", "EQ", "NE", "SHEQ", "SHNE", \
+  "AND", "OR", "BITOR", "BITXOR", "BITAND", "POWEROF", \
+  "LSH", "RSH", "URSH" ]
+
+MULTI_PROTECTED_OPERATORS = [ "INSTANCEOF", "IN" ]
+
+ASSIGN_OPERATORS = [ "ASSIGN", "ASSIGN_ADD", "ASSIGN_SUB", "ASSIGN_MUL", \
+  "ASSIGN_DIV", "ASSIGN_MOD", "ASSIGN_BITOR", "ASSIGN_BITXOR", "ASSIGN_BITAND", \
+  "ASSIGN_LSH", "ASSIGN_RSH", "ASSIGN_URSH" ]
+
+LOOP_KEYWORDS = [ "WHILE", "IF", "FOR", "WITH" ]
+
+
+class TokenStream:
+  def __init__ (self, tokens):
+    self.tokens = tokens
+    self.commentsBefore = None
+    self.parsepos = -1
+    self.eolBefore = False
+
+  def curr (self):
+    """Returns the current token."""
+    return self.tokens[self.parsepos]
+
+  def currType (self):
+    return self.curr()["type"]
+
+  def currDetail (self):
+    return self.curr()["detail"]
+
+  def currSource (self):
+    return self.curr()["source"]
+
+  def currLine (self):
+    return self.curr()["line"]
+
+  def currColumn (self):
+    return self.curr()["column"]
+
+  def currMultiline (self):
+    return self.curr()["multiline"]
+
+  def currConnection (self):
+    return self.curr()["connection"]
+
+  def currIsType (self, tokenType, tokenDetail = None):
+    if self.currType() != tokenType:
+      return False
+    else:
+      if tokenDetail == None:
+        return True
+      elif type(tokenDetail) == list:
+        return self.currDetail() in tokenDetail
+      else:
+        return self.currDetail() == tokenDetail
+
+  def expectCurrType (self, tokenType, tokenDetail = None):
+    if not self.currIsType(tokenType, tokenDetail):
+      expectedDesc = tokenType
+      if type(tokenDetail) == str:
+        expectedDesc += "/" + tokenDetail
+      raiseSyntaxException(self.curr(), expectedDesc)
+
+  def finished (self):
+    # NOTE: the last token is end of file
+    return self.parsepos >= len(self.tokens) - 1
+
+  def next (self, item=None, after=False):
+    length = len(self.tokens)
+    self.eolBefore = False
+    self.breakBefore = False
+
+    token = None
+    while self.parsepos < length - 1:
+      self.parsepos += 1
+
+      token = self.tokens[self.parsepos]
+
+      if token["type"] == "eol":
+        if self.eolBefore:
+          self.breakBefore = True
+
+        self.eolBefore = True
+        # ignore end of line
+        pass
+
+      elif token["type"] == "comment":
+        # After current item
+        if token["connection"] == "after":
+          if not token.has_key("inserted") or not token["inserted"]:
+            if item:
+              commentNode = tree.Node("comment")
+              commentNode.set("line", token["line"])
+              commentNode.set("column", token["column"])
+              commentNode.set("text", token["source"])
+              commentNode.set("detail", token["detail"])
+              commentNode.set("multiline", token["multiline"])
+              commentNode.set("connection", token["connection"])
+              commentNode.set("begin", token["begin"])
+              commentNode.set("end", token["end"])
+
+              if after:
+                item.addListChild("commentsAfter", commentNode)
+              else:
+                item.addChild(commentNode)
+
+              self.eolBefore = False
+              self.breakBefore = False
+
+            else:
+              print "Found unresolved after comment in line %s, column %s" % (token["line"], token["column"])
+              print token["source"]
+              pass
+
+        # Documentation and Block comments of next item
+        else:
+          if not self.commentsBefore:
+            self.commentsBefore = []
+
+          commentNode = tree.Node("comment")
+          commentNode.set("line", token["line"])
+          commentNode.set("column", token["column"])
+          commentNode.set("text", token["source"])
+          commentNode.set("detail", token["detail"])
+          commentNode.set("multiline", token["multiline"])
+          commentNode.set("connection", token["connection"])
+          commentNode.set("begin", token["begin"])
+          commentNode.set("end", token["end"])
+
+          self.commentsBefore.append(commentNode)
+
+          self.eolBefore = False
+          self.breakBefore = False
+
+      else:
+        break
+
+    #print "next token: " + str(token)
+
+    if token == None:
+      # return end of file token
+      return self.tokens[length - 1]
+    else:
+      return token
+
+  # alternative to use, when we want to check if the next token
+  # is a comment, but are not able to use next() because if there is
+  # no comment we want to leave in our position
+  def comment (self, item, after=False):
+    length = len(self.tokens)
+
+    token = None
+    pos = self.parsepos
+
+    while pos < length - 1:
+      pos += 1
+      token = self.tokens[pos]
+
+      if token["type"] == "comment" and token["connection"] == "after" and (not token.has_key("inserted") or not token["inserted"]):
+        commentNode = tree.Node("comment")
+        commentNode.set("line", token["line"])
+        commentNode.set("column", token["column"])
+        commentNode.set("text", token["source"])
+        commentNode.set("detail", token["detail"])
+        commentNode.set("multiline", token["multiline"])
+        commentNode.set("connection", token["connection"])
+        commentNode.set("begin", token["begin"])
+        commentNode.set("end", token["end"])
+
+        token["inserted"] = True
+
+        if after:
+          item.addListChild("commentsAfter", commentNode)
+        else:
+          item.addChild(commentNode)
+
+      else:
+        break
+
+  def hadEolBefore(self):
+    return self.eolBefore
+
+  def hadBreakBefore(self):
+    return self.breakBefore
+
+  def clearCommentsBefore(self):
+    commentsBefore = self.commentsBefore
+    self.commentsBefore = None
+    return commentsBefore
+
+
+
+class SyntaxException (Exception):
+  pass
+
+
+
+def createItemNode(type, stream):
+  # print "CREATE %s" % type
+
+  node = tree.Node(type)
+  node.set("line", stream.currLine())
+  node.set("column", stream.currColumn())
+
+  commentsBefore = stream.clearCommentsBefore()
+  if commentsBefore:
+    for comment in commentsBefore:
+      node.addListChild("commentsBefore", comment)
+
+  return node
+
+
+
+def raiseSyntaxException (token, expectedDesc = None):
+  if expectedDesc:
+    msg = "Expected " + expectedDesc + " but found "
+  else:
+    msg = "Unexpected "
+
+  msg += token["type"]
+
+  if token["detail"]:
+    msg += "/" + token["detail"]
+
+  msg += ": '" + token["source"] + "'. file:" + \
+    token["id"] + ", line:" + str(token["line"]) + \
+    ", column:" + str(token["column"])
+
+  raise SyntaxException(msg)
+
+
+
+def createSyntaxTree (tokenArr):
+  """Creates a syntax tree from a token stream.
+
+  tokens: the token stream."""
+
+  stream = TokenStream(tokenArr)
+  stream.next()
+
+  rootBlock = tree.Node("file")
+  rootBlock.set("file", stream.curr()["id"])
+
+  while not stream.finished():
+    rootBlock.addChild(readStatement(stream))
+
+  return rootBlock
+
+
+
+def readExpression (stream):
+  return readStatement(stream, True)
+
+
+
+def readStatement (stream, expressionMode = False, overrunSemicolon = True, inStatementList = False):
+  item = None
+
+  eolBefore = stream.hadEolBefore()
+  breakBefore = stream.hadBreakBefore()
+
+  # print "PROGRESS: %s - %s (%s) [expr=%s]" % (stream.currType(), stream.currDetail(), stream.currLine(), expressionMode)
+
+  if currIsIdentifier(stream, True):
+    # statement starts with an identifier
+    variable = readVariable(stream, True)
+    variable = readObjectOperation(stream, variable)
+
+    if stream.currIsType("token", ASSIGN_OPERATORS):
+      # This is an assignment
+      item = createItemNode("assignment", stream)
+      item.set("operator", stream.currDetail())
+      stream.next(item)
+
+      item.addListChild("left", variable)
+      item.addListChild("right", readExpression(stream))
+    elif stream.currIsType("token", "COLON") and not expressionMode:
+      # This is a label
+      item = variable
+      item.type = "label"
+      stream.next(variable)
+    else:
+      # Something else comes after the variable -> It's a sole variable
+      item = variable
+
+    # Any comments found for the variable belong to the extracted item
+    commentsChild = variable.getChild("commentsBefore", False)
+    if item and commentsChild != None:
+      variable.removeChild(commentsChild)
+      item.addChild(commentsChild, 0)
+  elif stream.currIsType("protected", "FUNCTION"):
+    item = createItemNode("function", stream)
+    stream.next(item)
+
+    # Read optional function name
+    if stream.currIsType("name"):
+      item.set("name", stream.currSource())
+      stream.next(item)
+
+    readParamList(item, stream)
+    item.addListChild("body", readBlock(stream))
+
+    # Check for direct execution: function() {}()
+    if stream.currIsType("token", "LP"):
+      # The function is executed directly
+      functionItem = item
+      item = createItemNode("call", stream)
+      item.addListChild("operand", functionItem)
+      readParamList(item, stream)
+      item = readObjectOperation(stream, item)
+  elif stream.currIsType("protected", "VOID"):
+    stream.next(item)
+    item = createItemNode("void", stream)
+    stream.next(item)
+    item.addChild(readStatement(stream, expressionMode))
+    stream.expectCurrType("token", "RP")
+    stream.next(item, True)
+    item = readObjectOperation(stream, item)
+  elif stream.currIsType("token", "LP"):
+    igroup = createItemNode("group", stream)
+    stream.next(igroup)
+    igroup.addChild(readStatement(stream, expressionMode))
+    stream.expectCurrType("token", "RP")
+    stream.next(igroup, True)
+    oper = readObjectOperation(stream, igroup)
+
+    # supports e.g. (this.editor.object || this.editor.iframe).style.marginTop = null;
+    if stream.currIsType("token", ASSIGN_OPERATORS):
+      # This is an assignment
+      item = createItemNode("assignment", stream)
+      item.set("operator", stream.currDetail())
+      stream.next(item)
+
+      item.addListChild("left", oper)
+      item.addListChild("right", readExpression(stream))
+    else:
+      # Something else comes after the variable -> It's a sole variable
+      item = oper
+
+  elif stream.currIsType("string"):
+    item = createItemNode("constant", stream)
+    item.set("constantType", "string")
+    item.set("value", stream.currSource())
+    item.set("detail", stream.currDetail())
+    stream.next(item, True)
+    # This is a member accessor (E.g. "bla.blubb")
+    item = readObjectOperation(stream, item)
+  elif stream.currIsType("number"):
+    item = createItemNode("constant", stream)
+    item.set("constantType", "number")
+    item.set("value", stream.currSource())
+    item.set("detail", stream.currDetail())
+    stream.next(item, True)
+    # This is a member accessor (E.g. "bla.blubb")
+    item = readObjectOperation(stream, item)
+  elif stream.currIsType("regexp"):
+    item = createItemNode("constant", stream)
+    item.set("constantType", "regexp")
+    item.set("value", stream.currSource())
+    stream.next(item, True)
+    # This is a member accessor (E.g. "bla.blubb")
+    item = readObjectOperation(stream, item)
+  elif expressionMode and (stream.currIsType("protected", "TRUE") or stream.currIsType("protected", "FALSE")):
+    item = createItemNode("constant", stream)
+    item.set("constantType", "boolean")
+    item.set("value", stream.currSource())
+    stream.next(item, True)
+  elif expressionMode and stream.currIsType("protected", "NULL"):
+    item = createItemNode("constant", stream)
+    item.set("constantType", "null")
+    item.set("value", stream.currSource())
+    stream.next(item, True)
+  elif expressionMode and stream.currIsType("token", "LC"):
+    item = readMap(stream)
+  elif expressionMode and stream.currIsType("token", "LB"):
+    item = readArray(stream)
+  elif stream.currIsType("token", SINGLE_LEFT_OPERATORS):
+    item = createItemNode("operation", stream)
+    item.set("operator", stream.currDetail())
+    item.set("left", True)
+    stream.next(item)
+    item.addListChild("first", readExpression(stream))
+  elif stream.currIsType("protected", "TYPEOF"):
+    item = createItemNode("operation", stream)
+    item.set("operator", "TYPEOF")
+    item.set("left", True)
+    stream.next(item)
+    item.addListChild("first", readExpression(stream))
+  elif stream.currIsType("protected", "NEW"):
+    item = readInstantiation(stream)
+    item = readObjectOperation(stream, item)
+  elif not expressionMode and stream.currIsType("protected", "VAR"):
+    item = createItemNode("definitionList", stream)
+    stream.next(item)
+    finished = False
+    while not finished:
+      if not currIsIdentifier(stream, False):
+        raiseSyntaxException(stream.curr(), "identifier")
+
+      childitem = createItemNode("definition", stream)
+      childitem.set("identifier", stream.currSource())
+      stream.next(childitem)
+      if stream.currIsType("token", "ASSIGN"):
+        assign = createItemNode("assignment", stream)
+        childitem.addChild(assign)
+        stream.next(assign)
+        assign.addChild(readExpression(stream))
+
+      item.addChild(childitem)
+
+      # Check whether anothe definition follows, e.g. "var a, b=1, c=4"
+      if stream.currIsType("token", "COMMA"):
+        stream.next(item)
+      else:
+        finished = True
+
+    stream.comment(item, True)
+
+  elif not expressionMode and stream.currIsType("protected", LOOP_KEYWORDS):
+    item = readLoop(stream)
+  elif not expressionMode and stream.currIsType("protected", "DO"):
+    item = readDoWhile(stream)
+  elif not expressionMode and stream.currIsType("protected", "SWITCH"):
+    item = readSwitch(stream)
+  elif not expressionMode and stream.currIsType("protected", "TRY"):
+    item = readTryCatch(stream)
+  elif not expressionMode and stream.currIsType("token", "LC"):
+    item = readBlock(stream)
+  elif not expressionMode and stream.currIsType("protected", "RETURN"):
+    item = createItemNode("return", stream)
+    stream.next(item)
+    # NOTE: The expression after the return keyword is optional
+    if not stream.currIsType("token", "SEMICOLON") and not stream.currIsType("token", "RC"):
+      item.addListChild("expression", readExpression(stream))
+      stream.comment(item, True)
+  elif not expressionMode and stream.currIsType("protected", "THROW"):
+    item = createItemNode("throw", stream)
+    stream.next(item)
+    item.addListChild("expression", readExpression(stream))
+    stream.comment(item, True)
+  elif not expressionMode and stream.currIsType("protected", "DELETE"):
+    item = createItemNode("delete", stream)
+    stream.next(item)
+    item.addListChild("expression", readExpression(stream))
+    stream.comment(item, True)
+  elif not expressionMode and stream.currIsType("protected", "BREAK"):
+    item = createItemNode("break", stream)
+    stream.next(item)
+    # NOTE: The label after the break keyword is optional
+    if not stream.hadEolBefore() and stream.currIsType("name"):
+      item.set("label", stream.currSource())
+      # As the label is an attribute, we need to put following comments into after
+      # to differenciate between comments before and after the label
+      stream.next(item, True)
+  elif not expressionMode and stream.currIsType("protected", "CONTINUE"):
+    item = createItemNode("continue", stream)
+    stream.next(item)
+    # NOTE: The label after the continue keyword is optional
+    if not stream.hadEolBefore() and stream.currIsType("name"):
+      item.set("label", stream.currSource())
+      stream.next(item, True)
+
+  if not item:
+    if stream.currIsType("token", "SEMICOLON") and not expressionMode:
+      # This is an empty statement
+      item = createItemNode("emptyStatement", stream)
+      stream.next(item)
+    else:
+      if expressionMode:
+        expectedDesc = "expression"
+      else:
+        expectedDesc = "statement"
+      raiseSyntaxException(stream.curr(), expectedDesc)
+
+  # check whether this is an operation
+  if stream.currIsType("token", MULTI_TOKEN_OPERATORS) or stream.currIsType("protected", MULTI_PROTECTED_OPERATORS) or (stream.currIsType("token", SINGLE_RIGHT_OPERATORS) and not stream.hadEolBefore()):
+    # its an operation -> We've already parsed the first operand (in item)
+    parsedItem = item
+
+    oper = stream.currDetail()
+
+    item = createItemNode("operation", stream)
+    item.addListChild("first", parsedItem)
+    item.set("operator", oper)
+    stream.next(item)
+
+    if oper in MULTI_TOKEN_OPERATORS or oper in MULTI_PROTECTED_OPERATORS:
+      # It's a multi operator -> There must be a second argument
+      item.addListChild("second", readExpression(stream))
+      if oper == "HOOK":
+        # It's a "? :" operation -> There must be a third argument
+        stream.expectCurrType("token", "COLON")
+        stream.next(item)
+        item.addListChild("third", readExpression(stream))
+
+    # Deep scan on single right operators e.g. if(i-- > 4)
+    if oper in SINGLE_RIGHT_OPERATORS and stream.currIsType("token", MULTI_TOKEN_OPERATORS) and expressionMode:
+      paroper = stream.currDetail()
+
+      paritem = createItemNode("operation", stream)
+      paritem.addListChild("first", item)
+      paritem.set("operator", paroper)
+      stream.next(item)
+
+      if paroper in MULTI_TOKEN_OPERATORS or paroper in MULTI_PROTECTED_OPERATORS:
+        # It's a multi operator -> There must be a second argument
+        paritem.addListChild("second", readExpression(stream))
+        if paroper == "HOOK":
+          # It's a "? :" operation -> There must be a third argument
+          stream.expectCurrType("token", "COLON")
+          stream.next(item)
+          paritem.addListChild("third", readExpression(stream))
+
+      # return parent item
+      item = paritem
+
+
+
+  # check whether this is a combined statement, e.g. "bla(), i++"
+  if not expressionMode and not inStatementList and stream.currIsType("token", "COMMA"):
+    statementList = createItemNode("statementList", stream)
+    statementList.addChild(item)
+    while stream.currIsType("token", "COMMA"):
+      stream.next(statementList)
+      statementList.addChild(readStatement(stream, False, False, True))
+    item = statementList
+
+  # go over the optional semicolon
+  if not expressionMode and overrunSemicolon and stream.currIsType("token", "SEMICOLON"):
+    stream.next(item, True)
+
+
+  item.set("eolBefore", eolBefore)
+  item.set("breakBefore", breakBefore)
+
+  return item
+
+
+
+def currIsIdentifier (stream, allowThis):
+  det = stream.currDetail()
+  return stream.currIsType("name") or stream.currIsType("builtin") \
+    or (stream.currIsType("protected") and \
+       (det == "INFINITY" or det == "PROTOTYPE" or det == "CALL" or \
+        det == "APPLY" or (allowThis and det == "THIS")))
+
+
+
+def readVariable (stream, allowArrays):
+  # Note: keywords may be used as identifiers, too
+  item = createItemNode("variable", stream)
+
+  done = False
+  firstIdentifier = True
+  while not done:
+    if not currIsIdentifier(stream, firstIdentifier):
+      raiseSyntaxException(stream.curr(), "identifier")
+
+    identifier = createItemNode("identifier", stream)
+    identifier.set("name", stream.currSource())
+    stream.next(identifier)
+
+    if allowArrays:
+      while stream.currIsType("token", "LB"):
+        accessor = createItemNode("accessor", stream)
+        stream.next(accessor)
+        accessor.addListChild("identifier", identifier)
+        accessor.addListChild("key", readExpression(stream))
+
+        stream.expectCurrType("token", "RB")
+        stream.next(accessor, True)
+
+        identifier = accessor
+
+    item.addChild(identifier)
+
+    firstIdentifier = False
+
+    if stream.currIsType("token", "DOT"):
+      stream.next(item)
+    else:
+      done = True
+
+  return item
+
+
+
+def readObjectOperation(stream, operand, onlyAllowMemberAccess = False):
+  if stream.currIsType("token", "DOT"):
+    # This is a member accessor (E.g. "bla.blubb")
+    item = createItemNode("accessor", stream)
+    stream.next(item)
+    item.addListChild("left", operand)
+
+    # special mode for constants which should be assigned to an accessor first
+    if operand.type == "constant":
+      item.addListChild("right", readVariable(stream, False))
+      item = readObjectOperation(stream, item)
+    else:
+      item.addListChild("right", readObjectOperation(stream, readVariable(stream, False)))
+
+  elif stream.currIsType("token", "LP"):
+    # This is a function call (E.g. "bla(...)")
+    item = createItemNode("call", stream)
+    item.addListChild("operand", operand)
+    readParamList(item, stream)
+    item = readObjectOperation(stream, item)
+  elif stream.currIsType("token", "LB"):
+    # This is an array access (E.g. "bla[...]")
+    item = createItemNode("accessor", stream)
+    stream.next(item)
+    item.addListChild("identifier", operand)
+    item.addListChild("key", readExpression(stream))
+
+    stream.expectCurrType("token", "RB")
+    stream.next(item, True)
+    item = readObjectOperation(stream, item)
+  else:
+    item = operand
+
+  # Any comments found for the operand belong to the item
+  if operand != item:
+    commentsChild = operand.getChild("commentsBefore", False)
+    if commentsChild != None:
+      operand.removeChild(commentsChild)
+      item.addChild(commentsChild, 0)
+
+  return item
+
+
+
+def readParamList (node, stream):
+  stream.expectCurrType("token", "LP")
+
+  params = createItemNode("params", stream)
+  node.addChild(params)
+
+  stream.next(params)
+
+  firstParam = True
+  lastExpr = None
+  while not stream.currIsType("token", "RP"):
+    if firstParam:
+      firstParam = False
+    else:
+      stream.expectCurrType("token", "COMMA")
+      stream.next(lastExpr, True)
+
+    lastExpr = readExpression(stream)
+    params.addChild(lastExpr)
+
+  # Has an end defined by the loop above
+  # This means that all comments following are after item
+  stream.next(params, True)
+
+
+def readBlock(stream):
+  stream.expectCurrType("token", "LC")
+  item = createItemNode("block", stream)
+
+  # Iterate through children
+  stream.next(item)
+  while not stream.currIsType("token", "RC"):
+    item.addChild(readStatement(stream))
+
+  # Has an end defined by the loop above
+  # This means that all comments following are after item
+  stream.next(item, True)
+
+  return item
+
+
+def readMap(stream):
+  stream.expectCurrType("token", "LC")
+
+  item = createItemNode("map", stream)
+  stream.next(item)
+
+  # NOTE: We use our own flag for checking whether the array already has entries
+  #       and not item.hasChildren(), because item.hasChildren() is also true
+  #       when there are comments before the array
+  hasEntries = False
+
+  while not stream.currIsType("token", "RC"):
+    if hasEntries:
+      stream.expectCurrType("token", "COMMA")
+      stream.next(item)
+
+    if not currIsIdentifier(stream, True) and not stream.currIsType("string") and not stream.currIsType("number"):
+      raiseSyntaxException(stream.curr(), "map key (identifier, string or number)")
+
+    keyvalue = createItemNode("keyvalue", stream)
+    keyvalue.set("key", stream.currSource())
+
+    if stream.currIsType("string"):
+      keyvalue.set("quote", stream.currDetail())
+
+    stream.next(keyvalue)
+    stream.expectCurrType("token", "COLON")
+    stream.next(keyvalue, True)
+    keyvalue.addListChild("value", readExpression(stream))
+
+    item.addChild(keyvalue)
+
+    hasEntries = True
+
+  # Has an end defined by the loop above
+  # This means that all comments following are after item
+  stream.next(item, True)
+
+  return item
+
+
+
+def readArray(stream):
+  stream.expectCurrType("token", "LB")
+
+  item = createItemNode("array", stream)
+  stream.next(item)
+
+  # NOTE: We use our own flag for checking whether the array already has entries
+  #       and not item.hasChildren(), because item.hasChildren() is also true
+  #       when there are comments before the array
+  hasEntries = False
+  while not stream.currIsType("token", "RB"):
+    if hasEntries:
+      stream.expectCurrType("token", "COMMA")
+      stream.next(item)
+
+    item.addChild(readExpression(stream))
+    hasEntries = True
+
+  # Has an end defined by the loop above
+  # This means that all comments following are after item
+  stream.next(item, True)
+
+  # Support constructs like [ "foo", "bar" ].join("")
+  item = readObjectOperation(stream, item)
+
+  return item
+
+
+
+def readInstantiation(stream):
+  stream.expectCurrType("protected", "NEW")
+
+  item = createItemNode("instantiation", stream)
+  stream.next(item)
+
+  # Could be a simple variable or a just-in-time function declaration (closure)
+  # Read this as expression
+  stmnt = readStatement(stream, True, False)
+  item.addListChild("expression", stmnt)
+
+  return item
+
+
+
+def readLoop(stream):
+  stream.expectCurrType("protected", LOOP_KEYWORDS)
+
+  loopType = stream.currDetail()
+
+  item = createItemNode("loop", stream)
+  item.set("loopType", loopType)
+
+  stream.next(item)
+  stream.expectCurrType("token", "LP")
+
+  if loopType == "FOR":
+    stream.next(item)
+
+    if not stream.currIsType("token", "SEMICOLON"):
+      # Read the optional first statement
+      first = createItemNode("first", stream)
+      item.addChild(first)
+      first.addChild(readStatement(stream, False, False))
+      stream.comment(first, True)
+
+    if stream.currIsType("token", "SEMICOLON"):
+      # It's a for (;;) loop
+      item.set("forVariant", "iter")
+
+      stream.next(item)
+      if not stream.currIsType("token", "SEMICOLON"):
+        # Read the optional second expression
+        second = createItemNode("second", stream)
+        item.addChild(second)
+        second.addChild(readExpression(stream))
+        stream.comment(second, True)
+
+      stream.expectCurrType("token", "SEMICOLON")
+      stream.next(item)
+
+      if not stream.currIsType("token", "RP"):
+        # Read the optional third statement
+        third = createItemNode("third", stream)
+        item.addChild(third)
+        third.addChild(readStatement(stream, False, False))
+        stream.comment(third, True)
+
+    elif stream.currIsType("token", "RP"):
+      # It's a for ( in ) loop
+      item.set("forVariant", "in")
+      pass
+
+    else:
+      raiseSyntaxException(stream.curr(), "semicolon or in")
+
+    stream.expectCurrType("token", "RP")
+
+  else:
+    expr = createItemNode("expression", stream)
+    stream.next(expr)
+    expr.addChild(readExpression(stream))
+    item.addChild(expr)
+    stream.comment(expr, True)
+    stream.expectCurrType("token", "RP")
+
+  # comments should be already completed from the above code
+  stmnt = createItemNode("statement", stream)
+  item.addChild(stmnt)
+  stream.next()
+  stmnt.addChild(readStatement(stream))
+
+  if loopType == "IF" and stream.currIsType("protected", "ELSE"):
+    elseStmnt = createItemNode("elseStatement", stream)
+    item.addChild(elseStmnt)
+    stream.next(elseStmnt)
+    elseStmnt.addChild(readStatement(stream))
+
+  return item
+
+
+
+def readDoWhile(stream):
+  stream.expectCurrType("protected", "DO")
+
+  item = createItemNode("loop", stream)
+  item.set("loopType", "DO")
+  stream.next(item)
+
+  stmnt = createItemNode("statement", stream)
+  item.addChild(stmnt)
+  stmnt.addChild(readStatement(stream))
+
+  stream.expectCurrType("protected", "WHILE")
+  stream.next(item)
+
+  stream.expectCurrType("token", "LP")
+
+  expr = createItemNode("expression", stream)
+  item.addChild(expr)
+  stream.next(expr)
+
+  expr.addChild(readExpression(stream))
+
+  stream.expectCurrType("token", "RP")
+  stream.next(item, True)
+
+  return item
+
+
+def readSwitch(stream):
+  stream.expectCurrType("protected", "SWITCH")
+
+  item = createItemNode("switch", stream)
+  item.set("switchType", "case")
+
+  stream.next(item)
+  stream.expectCurrType("token", "LP")
+
+  expr = createItemNode("expression", stream)
+  stream.next(expr)
+  item.addChild(expr)
+  expr.addChild(readExpression(stream))
+
+  stream.expectCurrType("token", "RP")
+  stream.next(expr, True)
+
+  stream.expectCurrType("token", "LC")
+  stmnt = createItemNode("statement", stream)
+  item.addChild(stmnt)
+  stream.next(stmnt)
+
+  while not stream.currIsType("token", "RC"):
+    if stream.currIsType("protected", "CASE"):
+      caseItem = createItemNode("case", stream)
+      stream.next(caseItem)
+      caseItem.addListChild("expression", readExpression(stream))
+      stmnt.addChild(caseItem)
+
+      stream.expectCurrType("token", "COLON")
+      stream.next(caseItem, True)
+
+    elif stream.currIsType("protected", "DEFAULT"):
+      defaultItem = createItemNode("default", stream)
+      stmnt.addChild(defaultItem)
+      stream.next(defaultItem)
+
+      stream.expectCurrType("token", "COLON")
+      stream.next(defaultItem, True)
+
+    else:
+      raiseSyntaxException(stream.curr(), "case or default")
+
+    while not stream.currIsType("token", "RC") and not stream.currIsType("protected", "CASE") and not stream.currIsType("protected", "DEFAULT"):
+      stmnt.addChild(readStatement(stream))
+
+  stream.next(stmnt, True)
+
+  return item
+
+
+def readTryCatch(stream):
+  stream.expectCurrType("protected", "TRY")
+
+  item = createItemNode("switch", stream)
+  item.set("switchType", "catch")
+  stream.next(item)
+
+  item.addListChild("statement", readStatement(stream))
+
+  while stream.currIsType("protected", "CATCH"):
+    catchItem = createItemNode("catch", stream)
+    stream.next(catchItem)
+
+    stream.expectCurrType("token", "LP")
+
+    exprItem = createItemNode("expression", stream)
+    catchItem.addChild(exprItem)
+    stream.next(exprItem)
+    exprItem.addChild(readExpression(stream))
+
+    stream.expectCurrType("token", "RP")
+    stream.next(exprItem, True)
+
+    stmnt = createItemNode("statement", stream)
+    catchItem.addChild(stmnt)
+    stmnt.addChild(readStatement(stream))
+
+    item.addChild(catchItem)
+
+  if stream.currIsType("protected", "FINALLY"):
+    finallyItem = createItemNode("finally", stream)
+    stream.next(finallyItem)
+
+    stmnt = createItemNode("statement", stream)
+    finallyItem.addChild(stmnt)
+    stmnt.addChild(readStatement(stream))
+
+    item.addChild(finallyItem)
+
+  return item
+
+
+
+
+
+
+
+
+
+def main():
+  parser = optparse.OptionParser()
+
+  parser.add_option("-w", "--write", action="store_true", dest="write", default=False, help="Writes file to incoming fileName + EXTENSION.")
+  parser.add_option("-e", "--extension", dest="extension", metavar="EXTENSION", help="The EXTENSION to use", default=".compiled")
+  parser.add_option("--optimize-variables", action="store_true", dest="optimizeVariables", default=False, help="Optimize variables. Reducing size.")
+  parser.add_option("--encoding", dest="encoding", default="utf-8", metavar="ENCODING", help="Defines the encoding expected for input files.")
+
+  (options, args) = parser.parse_args()
+
+  if len(args) == 0:
+    print "Needs one or more arguments (files) to compile!"
+    sys.exit(1)
+
+  for fileName in args:
+    if options.write:
+      print "Generating tree of %s => %s%s" % (fileName, fileName, options.extension)
+    else:
+      print "Generating tree of %s => stdout" % fileName
+
+    restree = createSyntaxTree(tokenizer.parseFile(fileName, "", options.encoding))
+
+    if options.optimizeVariables:
+      variableoptimizer.search(restree, [], 0, "$")
+
+    compiledString = tree.nodeToXmlString(restree)
+    if options.write:
+      filetool.save(fileName + options.extension, compiledString)
+
+    else:
+      try:
+        print compiledString
+
+      except UnicodeEncodeError:
+        print "  * Could not encode result to ascii. Use '-w' instead."
+        sys.exit(1)
+
+
+
+if __name__ == '__main__':
+  try:
+    main()
+
+  except KeyboardInterrupt:
+    print
+    print "  * Keyboard Interrupt"
+    sys.exit(1)