I have a corresponding generic function for which I have described methods for each of these classes in a signature with one other variable. There are also some ad hoc variable outside the signature which have different default values for each class.
Each method calls callNextMethod() to populate the slots which it has in common with the previous class and then some additional code to populate the new slots in the class.
Before writing each new method the selectMethod() function correctly identifies the method which I would like to call from the callNextMethod() function. But after 3-4 levels of nesting the method seem unable to identify the correct next method and returns nothing but continues without error.
I'm sorry, but I couldn't reproduce the problem with a simple example. But after countless hours of debug, a thorough reading of the Green Book and various attempts with the trace() function - this just drops the trace within callNextMethod(), again without error - , I have exhausted all my ideas for what is wrong.
All I can think is there might be a problem in callNextMethod(), making it inconsistent with selectMethod(). It also not clear to me how default values are passed through to the next method.
Is there a recommended debug process for this sort of problem? Is there a way call a specific method with a defined signature from within the new method definition?
Regards,
John Marsland
______________________________________________ R-devel@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
