2a43a256   
   From: bouncer@dev.null   
      
   Daryle Walker wrote:   
      
   > If an exception happens while the return value is transferred from   
   > my_function to whatever called it, does my_function have to worry about   
   > its noexcept specification being violated?  Or has my_function already   
   > ended, and any exception emitted gets compared against the noexcept   
   > setting of the caller function?  As I said, I hope it's the latter,   
   > because the transfer happens at a point where the called function can't   
   > use try-catch or any other way of keeping the exception from leaking.   
      
   At the point where a function's return value is constructed, its local   
   context is still very much alive.  Local variables and temporaries can   
   only be destroyed later - otherwise we would not be able to access them   
   from the function's return expression.  To me, this suggests that an   
   exception escaping from the function's return expression, or from any   
   of the destructors triggered when returning from the function, would   
   violate a promise to not throw.   
      
   I can't think of any technical reason why a return expression could not   
   be wrapped in a try-block; logically, however, I wouldn't know what to   
   do in the corresponding catch-block to remedy the situation.  A function   
   should return a meaningful value; a promise to not throw implies it is   
   either able to do so or not return at all - which is exactly what   
   noexcept brings us.   
      
   - Wil   
      
      
   --   
         [ See http://www.gotw.ca/resources/clcm.htm for info about ]   
         [ comp.lang.c++.moderated.    First time posters: Do this! ]   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)