- From: <keshlam@us.ibm.com>
- Date: Fri, 16 Oct 1998 15:34:32 -0400
- To: www-dom@w3.org
>1) Where can I find the DOM FAQ? I don't know of an official FAQ, though Don Park has just started assembling a list clarifying some of the less obvious points ( http://www.docuverse.com/domspec/consensus.html) which might evolve into one. >2) Is the root node "<TABLE>" or is "<TABLE>" the first child node of >the document? The example is only a portion of a document, out of context ("taken from an HTML document"). But if it was the whole document, the <TABLE> Element would be a child of a Document, and would in fact be the only Element child of that Document, retrievable either as a child or via the Document.getDocumentElement method. >3) Finally, the spec says NodeList "contains all children of this >node". That would seemingly include grandchildren, right? Children means immediate children only. There are two methods that return NodeList. One, getChildren, returns the immediate kids as per your quote. The other, getElementsByTagName (which can only be issued against Documents and Elements), returns all descendents which are Elements and have the specified name. I haven't gone through your dumpNode() method in detail, but the concept's valid: Recursive tree-walk each node outputs its immediate contents (in whatever format is appropriate for that node type -- obviously you haven't dealt with them all, and you have to watch out for the fact that CDATASection isa Text!), followed by outputting each of the children if any, in turn (via either getFirstChild/getNextSibling, or getChildNodes/item), followed by the end-tag if this node type has one. ______________________________________ Joe Kesselman / IBM Research Unless stated otherwise, all opinions are solely those of the author.
Received on Friday, 16 October 1998 15:20:02 UTC