[Ecls-list] Two bugs in ECL's CLOS implementation

[Juan Jose Garcia-Ripoll]

I think both errors are unrelated. One of them is probably due to a
fix submitted recently, which might not be working properly.

The other one might be due to the form of the lambda function. I am
not sure that I designed set-funcallable-instance-function to take a
function with the same signature as the generic function, but rather a
function that works as a dispatch (something like taking the
funcallable instance and a list of arguments). I have to look it up.

> How do I get better error messages here?
> I would like to know what operation caused
> this error / what memory or slot was actually accessed...

Debugging errors may be clearer compiling with DEBUG = 3 but in this
case the error happens inside the core, not in the compiled code
itself. Besides this, the C compiler does not allow such a low level
access and even knowing the address that caused the error is in
general impossible.

I will look into both problems. There are other major stoppers in the
Windows port that may delay the next release anyway.

Juanjo

-- 
Instituto de Física Fundamental, CSIC
c/ Serrano, 113b, Madrid 28006 (Spain)
http://juanjose.garciaripoll.googlepages.com