[LLVMdev] RFC: Exception Handling Proposal II

[Duncan Sands]

Hi Renato,

>> OK :)  Essentially what happens is as follows: when running destructors
>> when exiting a scope (which may be a nested scope), if a destructor throws
>> an exception then any remaining destructors are first run, then the
>> Program_Error exception is thrown at the point of the scope exit.  This
>> may be caught by an enclosing handler.
>
> That's a neat model.
>
> What happens if two (or more) exceptions are thrown while you're
> cleaning up another? For instance, if A's destructor also throws an
> exception?

in that case, any remaining destructors are run and Program_Error is thrown at
the point of scope exit :)  In short, all destructors are run, and if any throw
exceptions then that fact is noted somewhere (and the exception is not allowed
to propagate) and once all destructors have had a chance to run then one
instance of Program_Error is thrown at the point of scope exit.  Note that the
fact that the exception is thrown not at the place where a particular
destructor went wrong but at the end of the scope means that exactly the same
exception is thrown no matter which (or how many) destructors barfed with an
exception.  If you had to throw the exception at the particular point in the
destructor's code where the failure happened then you would be in trouble if
multiple destructors failed (at which spot would you throw the exception?).

> A's exception was not thrown while processing B's exception (but while
> cleaning up *before* processing it), so it's not a nested exception,
> but an "aside" exception...
>
> It makes my head hurt... I think Stroustroup didn't want to think
> about that, too... :D

Actually it's pretty simple, see above.

Ciao,

Duncan.