There doesn't seem to be any way of traversing from one top level node
(e.g. doctype, html, #text, etc) to any of its siblings, once you no
longer have the original NodeList obtained from the Parser class. This
means, for example, that you cannot do a get previous/next sibling
method currently on these nodes.
A few quick ideas about how to do this (none of which are particularly idea=
l):
- Create a node named something like DocumentNode (we have HTMLPage
but that doesn't really seem to do what we want) that whose children
are the NodeList obtained by the Parser class. Make the Parser class
create one of these instead of the NodeList as it does now.
- Have a method named something like getDocument for each tag, which
simply returns a reference to the initial NodeList obtained by the
Parser class.
Any other ideas are welcome.
Ian
|
