[XML-SIG] TreeWalker: NextNode

Jeremy Kloth jeremy.kloth@fourthought.com
19 Dec 2002 16:19:46 -0700

On Wed, 2002-12-18 at 23:48, Jag wrote:
> Hi...
> I have found a bug in the TreeWalker code in PyXML 0.8.1, and previous 
> versions.
> Using the following snippet of code:
>     mywalker = TreeWalker(mydom.documentElement, NodeFilter.SHOW_ALL, None, 1)

A thing to note, the official way of creating a TreeWalker is to use the
createTreeWalker method of the Document object.  The only DOM
implementation in PyXML that offers this method is 4DOM.  And this is
why the code is written the way it is, it was designed to only work with


> The NodeIterator does not have this problem. Infact, it seems that it has the 
> solution but I cant find what other dependencies are present that prevent the 
> modification of code from the NodeIterator to allow the TreeWalker to work.
> The NodeIterator just uses node.nodeType but this does not want to work for 
> the TreeWalker??
> Have others had this problem? I would appreciate some feedback on this issue.

Now with that said, there is no longer a compelling reason to not update
TreeWalker to use attribute access (it will make its use in 4DOM slower

Jeremy Kloth