In some of the standard packages from Micro Focus, there is code to generate
and update an XML object (often at startup). Think PCRS,
EntrilementConfiguration etc.

However, an optimisation is added that tries to detect if the resulting data is
identical to that in the vault and then skip the expensive write.
This is done via string comparison after some replace-all to normalise the
data, which is expensive if the XML in question is huge.

Does anyone know of a better way in IDM to determine that two XML documents are
essentially identical (ignore whitespace)? One that scales to 20MB mapping
tables and the like?
Essentially a way to compare without needing to serialize the data back to
strings. Simply a true/false.

I guess an ECMAScript that calls some java classes might be the best option
here. Anyone been down this path before, got some code to share?