> I really want to walk the object-tree internal implementation objects that get built
> when the xpath string gets parsed by the engine. I just want to have a sense of what
> is going on. Well, i have some more questions for you.
> Currently, i'm working on a project that selects nodes on an encoded XML file using
> XPath to query. However, i just want to implement only a small subset of XPath. I'm
> thinking of two strategies at the moment.
> 1. I use the full Jaxen and only implement small and necessary methods in the
> DocumentNavigator interface (am i right here?)
Ayup, that's the intention, typically. There are several axes that are
optional, in case you never have a need for following-sibling:: or whatnot.
Some you get for free (ie, you get ancestor:: when you implement parent::).
> 2. OR implement a small subset of XPath based on Jaxen and integrate tightly with
> methods to traverse axes on an encoded XML stream.
I guess you could do that.
If you want to drop down lower, you could also work with SAXPath directly,
and build your own expression trees. I haven't really thought about
subsetting the Jaxen expression trees.
-bob
_______________________________________________
Jaxen-interest mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/jaxen-interest
