Skip to main content
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [qvtd-dev] [Dev] QVTi Evaluator
Hi

There are various things that may make things better.

a) Incorporation of QVTi.ocl with good content as part of the editor validation.

This is getting quite close to the top of my stack.

b) Debugging the compiled rather than interpreted execution.

Needs me to complete the CG for the new MiddleProperty etc.

this would then allow use of the Java debugger with better context/breakpoints.

but will this actually help

c) Better trace print-outs

your activity

d) Direct QVTi debugger

On my stack, but probably a couple of months away.

e) Reflection

Go and find a nice quiet place in the sun and think about what it is that is making it so difficult to work out what is going wrong. This may shed some insight into debugging aids that we need.

-----

Validation needs doing anyway, so I suggest you go through every single property of every QVTi class and decide what could be invalid about it, and then express it in Complete OCL.

e.g. something like

context MappingCall
inv ConsistentNumberOfBindings: bindings.size() = referredMapping.boundVariable->size()

No need to check basic multiplicity of type conformance that EMF does anyway.

You should now be able to load this into the editor and just use it.

This should give you some good OCL practice. If you're stuich trying writing it in form,al words, the it may be clear, else forward to me.

(No using allInstances().)

-----

If this isn't enough to make progress and you really need the QVTi debugger, then it may be best to postpone this activity for a couple of months and concentrate on the sub-ETL transformations.

    Regards

        Ed



On 12/07/2013 14:29, Horacio Hoyos Rodriguez wrote:

Hi Ed,

 

I have found a wrong binding in a mapping call in the UML2RDBMS transformation which was causing the exception during the test. However, the transformation still fails and I think is somehow due to smaller errors in the transformation or in the population of the trace model.

 

I need to come up with a better debugger because tracking them at the moment is quite hard.

 

Regards,

 

Horacio Hoyos Rodríguez

EngD Student

University of York

 

http://www.york.ac.uk/docs/disclaimer/email.htm



_______________________________________________
qvtd-dev mailing list
qvtd-dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/qvtd-dev


No virus found in this message.
Checked by AVG - www.avg.com
Version: 2013.0.3349 / Virus Database: 3204/6484 - Release Date: 07/11/13


Back to the top