Question

So I’ve been looking at a lot of rule engine software, and there is this one thing that I want to do that is either really inconvenient, or else I’m just not understanding documentation (quite likely).

Basically, I want an instance of an xml schema to be accessible to a rule engine either as a fact base, rule base (not likely), or at least as an object of some sort that can have its values replaced for variables within the rule base for sensible comparisons. I’m specifically looking at NxBRE, but I’m not married to any one BRE, so if you’ve achieved this with a different piece of software, I’m open to that, too.

My question is, do you think this is possible? Most of what I’ve looked at has stated that I have to load information in whatever rule/fact format that is used by the engine (often bragging about how this information can be loaded from any uri, which is irrelevant to me), but is there a software that can convert xml files always based on a unique schema to a series of corresponding facts in a format read by the engine? I can try to clarify if I’m not being clear in what I’m asking.

Answer 1

(Cross posted from SF.net)

To load Flow Engine’s context objects from a custom XML document, the only option consists in loading the XML yourself and programmatically add objects one by one in the context after the rule file (xBRE) has been loaded.

With the Inference Engine, you have more options:

• adding facts one by one from your main app,
• creating a binder that does the same at the boundary between your app and NxBRE,
• XSLT-ing your custom XML to RuleML facts and load these facts in the engine after loading the rulebase.