[Ecls-list] exception question

[Juan Jose Garcia Ripoll]

Michael O'Connor wrote:

>I manged to fix the C++ compile errors in the latest
>cvs code by making the following edits:[...]
>I don't know if I've done the right thing here but it
>seems to work ok
>
It is the right thing and I should've done it myself a couple of days 
ago, but I am now making the last steps towards integrating a real 
pretty-printer and the whole of CMUCL's format into ECL, and did not 
manage to even answer your e-mail.

> (except that instead of getting a
>system error when throwing an exception I now get
>nothing at all - i.e. exception is ignored, which
>makes me wonder if I'm missing something else).
>
Really, from the ECL part there is nothing fancy. We only use 
setjmp/longjmp, and tweak no registers or whatever. This should not 
interfere with C++'s exception mechanism. And actually the fact that ECL 
does not interfere or detect exceptions is a major concern, because it 
cannot recover from them: if C++ calls lisp and lisps calls C++ which 
emits an exception that is captured by the outermost C++ environment, 
there are chances that the frames stack, binding stacks, etc, get corrupted.

Regards,

Juanjo

-- 
Max-Planck-Institut fuer Quantenoptik	+49/(0)89/32905-345
Hans-Kopfermann-Str. 1, D-85748		www.mpq.mpg.de/Theorygroup/CIRAC/
Garching b. Muenchen, Germany		Juan.Ripoll at mpq.mpg.de