XOM API for Processing XML with Java

[Wolfgang Hoschek <whoschek AT lbl.gov>]

Perhaps normalizeText(ParentNode) should take an extra parameter to 
tell whether an entire subtree should be normalized or just a part of 
it. That way a user could call it after all insert() and remove() calls 
on an given element, saving unnecessarily deep tree traversals. As in

public static void normalizeText(ParentNode node, int maxDepth)

For example normalizeText(node, 0) would just normalize the Texts that 
are direct children of a given node, whereas
normalizeText(node, Integer.MAX_VALUE) would recurse over the entire 
subtree.

Perhaps it's a gadget and not worth the extra API complexity. Perhaps 
it's useful. If you have an opinion either way, drop me a line.

Wolfgang.

> "Michael Kay" writes:
>
> > Rather than having Saxon create virtual text nodes to represent a 
> > sequence
> > of adjacent XOM text nodes, I think it would be better if XOM had an 
> > option
> > to merge adjacent text nodes as soon as they are added to the tree.
>
> JDOM use to do this and it was dropped because it caused confusion 
> with the
> index methods. Someone would add or remove a node and the index of the
> remaining nodes wouldn't be what they expected. Examples of methods in 
> XOM
> that would have this problem are getChild(int index), insert(Node node,
> int index), indexOf(), and especially getChildCount() to name a few. In
> hindsight, should of keep the merge behavior and dump the methods.  If 
> you
> want the 2nd child element of a parent node, and unless you take extra
> measures to build a index certain nodes or find some other way to keep 
> track
> of their positions, you're going to have to search from the first 
> child anyway.
> May as well just use a iterator.

On Jan 10, 2005, at 1:59 AM, Hallvard Trætteberg wrote:

> My point is that the programmer will have to understand this problem 
> and
> that this cannot be handled automatically by a XOM XPath api in a 
> clean way
> (not by auto-merging, not by including virtual nodes). The most helpful
> support is Wolfgang's suggestion as shown below: a normalizeTexts 
> method.