Patch #102492, fixing bug #116677:

    give minidom.py behaviour that complies with the DOM Level 1 REC,
    which says that when a node newChild is added to the tree, "if the
    newChild is already in the tree, it is first removed."

    pulldom.py is patched to use the public minidom interface instead
    of setting .parentNode itself.  Possibly this reduces pulldom's
    efficiency; someone else will have to pronounce on that.

2 files changed, 42 insertions, 10 deletions

diff --git a/Lib/xml/dom/minidom.py b/Lib/xml/dom/minidom.py
index 2d3755c..d078f39 100644
--- a/Lib/xml/dom/minidom.py
+++ b/Lib/xml/dom/minidom.py
@@ -40,6 +40,7 @@ class Node(_Node):
 
     def __init__(self):
         self.childNodes = []
+        self.parentNode = None
         if Node._debug:
             index = repr(id(self)) + repr(self.__class__)
             Node.allnodes[index] = repr(self.__dict__)
@@ -98,6 +99,8 @@ class Node(_Node):
             return self.childNodes[-1]
 
     def insertBefore(self, newChild, refChild):
+        if newChild.parentNode is not None:
+            newChild.parentNode.removeChild(newChild)
         if refChild is None:
             self.appendChild(newChild)
         else:
@@ -116,6 +119,8 @@ class Node(_Node):
         return newChild
 
     def appendChild(self, node):
+        if node.parentNode is not None:
+            node.parentNode.removeChild(node)
         if self.childNodes:
             last = self.lastChild
             node.previousSibling = last
@@ -129,6 +134,8 @@ class Node(_Node):
         return node
 
     def replaceChild(self, newChild, oldChild):
+        if newChild.parentNode is not None:
+            newChild.parentNode.removeChild(newChild)
         if newChild is oldChild:
             return
         index = self.childNodes.index(oldChild)
@@ -144,6 +151,12 @@ class Node(_Node):
 
     def removeChild(self, oldChild):
         self.childNodes.remove(oldChild)
+        if oldChild.nextSibling is not None:
+            oldChild.nextSibling.previousSibling = oldChild.previousSibling
+        if oldChild.previousSibling is not None:
+            oldChild.previousSibling.nextSibling = oldChild.nextSibling            
+        oldChild.nextSibling = oldChild.previousSibling = None
+        
         if self._makeParentNodes:
             oldChild.parentNode = None
         return oldChild
@@ -606,11 +619,23 @@ class Document(Node):
     implementation = DOMImplementation()
 
     def appendChild(self, node):
+        if node.parentNode is not None:
+            node.parentNode.removeChild(node)
+
         if node.nodeType == Node.ELEMENT_NODE \
            and self._get_documentElement():
             raise TypeError, "two document elements disallowed"
         return Node.appendChild(self, node)
 
+    def removeChild(self, oldChild):
+        self.childNodes.remove(oldChild)
+        oldChild.nextSibling = oldChild.previousSibling = None
+        oldChild.parentNode = None
+        if self.documentElement is oldChild:
+            self.documentElement = None
+             
+        return oldChild
+
     def _get_documentElement(self):
         for node in self.childNodes:
             if node.nodeType == Node.ELEMENT_NODE:
diff --git a/Lib/xml/dom/pulldom.py b/Lib/xml/dom/pulldom.py
index e674385..7f5ef79 100644
--- a/Lib/xml/dom/pulldom.py
+++ b/Lib/xml/dom/pulldom.py
@@ -55,7 +55,9 @@ class PullDOM(xml.sax.ContentHandler):
             attr.value = value
             node.setAttributeNode(attr)
 
-        node.parentNode = self.curNode
+##        print self.curNode, self.curNode.childNodes, node, node.parentNode
+        self.curNode.appendChild(node)
+#        node.parentNode = self.curNode
         self.curNode = node
 
         self.lastEvent[1] = [(START_ELEMENT, node), None]
@@ -77,7 +79,8 @@ class PullDOM(xml.sax.ContentHandler):
             attr.value = value
             node.setAttributeNode(attr)
 
-        node.parentNode = self.curNode
+        #node.parentNode = self.curNode
+        self.curNode.appendChild(node)
         self.curNode = node
 
         self.lastEvent[1] = [(START_ELEMENT, node), None]
@@ -93,8 +96,9 @@ class PullDOM(xml.sax.ContentHandler):
 
     def comment(self, s):
         node = self.document.createComment(s)
-        parent = self.curNode
-        node.parentNode = parent
+        self.curNode.appendChild(node)
+#        parent = self.curNode
+#        node.parentNode = parent
         self.lastEvent[1] = [(COMMENT, node), None]
         self.lastEvent = self.lastEvent[1]
         #self.events.append((COMMENT, node))
@@ -102,24 +106,27 @@ class PullDOM(xml.sax.ContentHandler):
     def processingInstruction(self, target, data):
         node = self.document.createProcessingInstruction(target, data)
 
-        parent = self.curNode
-        node.parentNode = parent
+        self.curNode.appendChild(node)
+#        parent = self.curNode
+#        node.parentNode = parent
         self.lastEvent[1] = [(PROCESSING_INSTRUCTION, node), None]
         self.lastEvent = self.lastEvent[1]
         #self.events.append((PROCESSING_INSTRUCTION, node))
 
     def ignorableWhitespace(self, chars):
         node = self.document.createTextNode(chars)
-        parent = self.curNode
-        node.parentNode = parent
+        self.curNode.appendChild(node)
+#        parent = self.curNode
+#        node.parentNode = parent
         self.lastEvent[1] = [(IGNORABLE_WHITESPACE, node), None]
         self.lastEvent = self.lastEvent[1]
         #self.events.append((IGNORABLE_WHITESPACE, node))
 
     def characters(self, chars):
         node = self.document.createTextNode(chars)
-        parent = self.curNode
-        node.parentNode = parent
+        self.curNode.appendChild(node)
+#        parent = self.curNode
+#        node.parentNode = parent
         self.lastEvent[1] = [(CHARACTERS, node), None]
         self.lastEvent = self.lastEvent[1]