While fixing up my recent 'lenient' sample...
I noticed that leniency is not dealth with in a consistent manner throughout the binding implementations:
one binding in the pack 'ValueJXpathBinding' catches the possible JXPathException and just logs them as a warning...
I take it this try/catch attitude stems from a time when I didn't know of the existance of setLenient() in JXPath :-)
So, since 1/ we hapilly do use that now 2/ have it available as an attribute on all bindings 3/ and by default set leniency to true (and since 4/ I hopefully smashed the last bug in that area some minutes ago)
I am removing the try-catch-warn approach: if the binding is set to non-lenient then the consequences of that should be as obvious as possible and thus thrown up in the face of the end-user (which should be the testing web-app developer at that time)
I don't think I overlooked something here, but am wide open for other views and insights...
Sounds good!
-marc=
PS: by the way: I noticed that setting the leniency doesn't seem to have any effect on Javascript objects, they seem to be lenient by themselves?
That's something I once started to investigate, but quickly got lost in the guts of JXPath DynamicPropertyPointer and JS Scriptable properties...
It would be nice to have a consistent behaviour in this area also.
Sylvain
-- Sylvain Wallez Anyware Technologies http://www.apache.org/~sylvain http://www.anyware-tech.com { XML, Java, Cocoon, OpenSource }*{ Training, Consulting, Projects } Orixo, the opensource XML business alliance - http://www.orixo.com
