W3C DOM code refactoring

src/DOM/DFT.py

@@ -17,7 +17,7 @@
 # MA  02111-1307  USA
 
 import os
-import w3c
+import W3C.w3c as w3c
 import hashlib
 import logging
 import Window
@@ -63,16 +63,17 @@ def fix(self, script):
     def handle_onload(self):
         try:
             body = self.window.doc.body
-            if body and body.tag.has_key('onload'):
-                self.window.evalScript(self.fix(body.tag['onload']), tag = body.tag.contents[-1])
         except:
-            pass
+            body = self.window.doc.getElementsByTagName('body')[0]
+
+        if body and body.tag.has_key('onload'):
+            self.window.evalScript(self.fix(body.tag['onload']), tag = body.tag.contents[-1])
 
         if hasattr(self.window, 'onload'):
             self.window.evalScript(self.fix(self.window.onload))
 
     def handle_onclick(self):
-        inputs = self.window._findAll('input')
+        inputs = self.window._findAll(('input', 'a'))
         for input in inputs:
             for k, v in input.attrs:
                 if k in ('onclick', ):
@@ -163,7 +164,11 @@ def handle_applet(self, applet):
         if not archive:
             return
 
-        response, content = self.window._navigator.fetch(archive)
+        try:
+            response, content = self.window._navigator.fetch(archive)
+        except:
+            return
+
         self.log.warning('Saving applet %s' % (archive, ))
         with open(archive, 'wb') as fd:
             fd.write(content)
@@ -219,17 +224,26 @@ def handle_frame(self, frame):
     def handle_iframe(self, iframe):
         self.handle_frame(iframe)
 
+    def handle_body(self, body):
+        pass
+
     def run(self):
-        self.log.warning(self.window.doc)
+        self.log.debug(self.window.doc)
 
         soup = self.window.doc.doc
 
+        # Dirty hack
+        for p in soup.findAll('object'):
+            self.handle_object(p)
+
         for child in soup.recursiveChildGenerator():
             name = getattr(child, "name", None)
-            if name is not None:
-                handler = getattr(self, "handle_%s" % (name, ), None)
-                if handler: 
-                    handler(child)
+            if name is None or name in ('object', ):
+                continue
+
+            handler = getattr(self, "handle_%s" % (name, ), None)
+            if handler:
+                handler(child)
 
         self.handle_onload()
         self.handle_onclick()

src/DOM/W3C/Attr.py

@@ -0,0 +1,88 @@
+#!/usr/bin/env python
+from __future__ import with_statement
+
+import sys, re, string
+
+from HTML import BeautifulSoup
+import PyV8
+
+from DOMException import DOMException
+from Node import Node
+
+class Attr(Node):
+    _value = ""
+
+    def __init__(self, parent, attr):
+        self.parent = parent
+        self.attr   = attr
+
+        self._value = self.getValue()
+
+    def __repr__(self):
+        return "<Attr object %s%s at 0x%08X>" % ("%s." % self.parent.tagName if self.parent else "", self.attr, id(self))
+
+    def __eq__(self, other):
+        return hasattr(other, "parent") and self.parent == other.parent and \
+               hasattr(other, "attr") and self.attr == other.attr
+
+    @property
+    def nodeType(self):
+        return Node.ATTRIBUTE_NODE
+
+    @property        
+    def nodeName(self):
+        return self.attr
+
+    def getNodeValue(self):
+        return self.getValue()
+
+    def setNodeValue(self, value):
+        return self.setValue(value)
+
+    nodeValue = property(getNodeValue, setNodeValue)
+
+    @property
+    def childNodes(self):
+        from NodeList import NodeList
+
+        return NodeList(self.parent.doc, [])
+
+    @property
+    def parentNode(self):
+        return self.parent
+
+    # Introduced in DOM Level 2
+    @property
+    def ownerElement(self):
+        if self.parent: 
+            if self.parent.nodeType == Node.ELEMENT_NODE:
+                return self.parent
+
+        return None
+
+    @property
+    def ownerDocument(self):
+        return self.parent.doc
+
+    @property
+    def name(self):
+        return self.attr
+
+    def specified(self):
+        return self.parent.has_key(self.attr)
+
+    def getValue(self):
+        if self.parent:
+            if self.parent.tag.has_key(self.attr):
+                return self.parent.tag[self.attr]
+
+        return self._value 
+
+    def setValue(self, value):
+        self._value = value
+
+        if self.parent:
+            self.parent.tag[self.attr] = value
+
+    value = property(getValue, setValue)
+

src/DOM/W3C/CDATASection.py

@@ -0,0 +1,16 @@
+from __future__ import with_statement
+
+import sys, re, string
+
+from DOMException import DOMException
+from Text import Text
+
+
+class CDATASection(Text):
+    def __repr__(self):
+        return "<CDATA '%s' at 0x%08X>" % (self.tag, id(self))
+
+    @property
+    def nodeName(self):
+        return "#cdata-section"
+

src/DOM/W3C/CharacterData.py

@@ -0,0 +1,45 @@
+#!/usr/bin/env python
+from __future__ import with_statement
+
+import sys, re, string
+import PyV8
+
+from DOMException import DOMException
+from Node import Node
+
+class CharacterData(Node):
+    def __init__(self, doc, tag):
+        Node.__init__(self, doc)
+
+        self.tag = tag
+
+    def __str__(self):
+        return str(self.tag)
+
+    def getData(self):
+        return unicode(self.tag)
+
+    def setData(self, data):
+        raise DOMException(DOMException.NO_MODIFICATION_ALLOWED_ERR)
+
+    data = property(getData, setData)
+
+    @property
+    def length(self):
+        return len(self.tag)
+
+    def substringData(self, offset, count):
+        return self.tag[offset:offset+count]
+
+    def appendData(self, arg):
+        raise DOMException(DOMException.NO_MODIFICATION_ALLOWED_ERR)
+
+    def insertData(self, offset, arg):
+        raise DOMException(DOMException.NO_MODIFICATION_ALLOWED_ERR)
+
+    def deleteData(self, offset, count):
+        raise DOMException(DOMException.NO_MODIFICATION_ALLOWED_ERR)
+
+    def replaceData(self, offset, count, arg):
+        raise DOMException(DOMException.NO_MODIFICATION_ALLOWED_ERR)
+

src/DOM/W3C/Comment.py

@@ -0,0 +1,12 @@
+#!/usr/bin/env python
+from __future__ import with_statement
+
+from DOMException import DOMException
+from CharacterData import CharacterData
+
+
+class Comment(CharacterData):
+    @property
+    def nodeName(self):
+        return "#comment"
+

src/DOM/W3C/DOMException.py

@@ -0,0 +1,29 @@
+#!/usr/bin/env python
+from __future__ import with_statement
+
+import sys
+import PyV8
+
+class DOMException(RuntimeError, PyV8.JSClass):
+    def __init__(self, code):
+        self.code = code
+
+    # ExceptionCode
+    INDEX_SIZE_ERR                 = 1  # If index or size is negative, or greater than the allowed value
+    DOMSTRING_SIZE_ERR             = 2  # If the specified range of text does not fit into a DOMString
+    HIERARCHY_REQUEST_ERR          = 3  # If any node is inserted somewhere it doesn't belong
+    WRONG_DOCUMENT_ERR             = 4  # If a node is used in a different document than the one that created it (that doesn't support it)
+    INVALID_CHARACTER_ERR          = 5  # If an invalid or illegal character is specified, such as in a name. 
+    NO_DATA_ALLOWED_ERR            = 6  # If data is specified for a node which does not support data
+    NO_MODIFICATION_ALLOWED_ERR    = 7  # If an attempt is made to modify an object where modifications are not allowed
+    NOT_FOUND_ERR                  = 8  # If an attempt is made to reference a node in a context where it does not exist
+    NOT_SUPPORTED_ERR              = 9  # If the implementation does not support the type of object requested
+    INUSE_ATTRIBUTE_ERR            = 10 # If an attempt is made to add an attribute that is already in use elsewhere    
+
+    # Introduced in Level 2
+    INVALID_STATE_ERR              = 11 # If an attempt is made to use an object that is not, or is no longer, usable
+    SYNTAX_ERR                     = 12 # If an invalid or illegal string is specified
+    INVALID_MODIFICATION_ERR       = 13 # If an attempt is made to modify the type of the underlying object
+    NAMESPACE_ERR                  = 14 # If an attempt is made to create or change an object in a way which is incorrect with regards to namespaces
+    INVALID_ACCESS_ERR             = 15 # If a parameter or an operation is not supported by the underlying object
+