|Understanding the BPEL Validator [message #559427]
||Thu, 16 September 2010 11:10
Registered: September 2010
For several reasons I'm working on a (OCL-based) implementation of the BPEL validator. Currently I'm hanging on SA00010 and want to find out how the "original" validator checks the rule.|
The validation rule seems to be implemented in org.eclipse.bpel.validator.rules.VariableValidator. Actually there are three methods that implement the rule for MessageType, Element and Type nodes. But I don't get the point how these methods actually validate the rule. Basically, all methods do the same: They check whether the attribute node is not null (e.g. checkAttributeNode (mNode, fElementNode, AT_ELEMENT , KIND_NODE) ), but as far as I see the result of this check is not used.
So I have two questions:
1. What is the idea behind validating SA00010 as it has been done here (The description "make sure that * is visible from the imports" does not help me much)?
2. Why is it sufficient to do it the way as in the VariableValidator?
Powered by FUDForum
. Page generated in 0.02842 seconds