Re: SmartEiffel shortcomings?

Philippe Ribet wrote:

> Wolfgang Jansen wrote:
>
>> When provided a STORABLE class and its supporting framewark
>> (namely some additions to the SE compiler) I added also print command
>> to the SE debugger. (Both, class STORABLE and the print command,
>> rely on object and class introspection, so it was straightforward
>> to apply to the debugger what had been developed for the class.)
>> Being on the way, I added more to the debugger, e.g. that it stops
>> at an exception before the stack gets unwound. The last version goes
>> still further: even if the system has not been compiled with -sedb
>> (and not with -boost) then a rudimentary debugger is invoked in case
>> of an exception. This way, you can (besides seeing the stack)
>> evaluate the heap.
>
>
> Do you know that almost all of this is included in the compiler 
> nowadays? (introspection, storable, recursive attributes printing from 
> debugger) Of course the mini-debugger is not, but it might be 
> interesting. Would you like to contribute it?
>
Do you know that I published introspection, storable, and
evaluation and printing of expressions from the debugger
more than two years ago? So, the unnessecary repeated work
has been done by the SE developers.
By the way, my old version of STORABLE works very well
while yours produces terrible errors.

> May be stopping at an exception is not already done. If so, I think 
> you could easily provide some small patch. Thanks in advance!
>
Forcing the debugger to stop at an exception before the stack
gets unwound is easy (whether in a debugger session or not).
How can I provide a patch (sending plain text, editing SE's
snapshot version, ...)?
All the other things rely on the introspection developed
originally for use by STORABLE. The question is, are you
willing to drop your late and insufficient work in this field
and instead including (mutadis mutandis) what I've published? 

WJ