«

For vector and deque, the 'at( )' member function throws out_of_range
exception if the argument to the 'at( )' function is not in range. But
the subscript operator [ ] does not throw this exception for the same
situation.

My question:
Why can't the subscript operator [ ] itself throw the out_of_range
exception in which case the 'at ( )' member function may not be
needed ?

Kindly explain

Thanks
V.Subramanian

On 2007-08-12 15:47, , India wrote:
> For vector and deque, the 'at( )' member function throws out_of_range
> exception if the argument to the 'at( )' function is not in range. But
> the subscript operator [ ] does not throw this exception for the same
> situation.
>
> My question:
> Why can't the subscript operator [ ] itself throw the out_of_range
> exception in which case the 'at ( )' member function may not be
> needed ?

Because sometimes you can be sure that the index provided will be in
range and don't want the extra overhead associated with the check, the
operator[] is meant for occasions where you know that the code is safe
and you want as much performance as possible.

--
Erik Wikström

, India wrote:
> For vector and deque, the 'at( )' member function throws out_of_range
> exception if the argument to the 'at( )' function is not in range. But
> the subscript operator [ ] does not throw this exception for the same
> situation.

> My question:
> Why can't the subscript operator [ ] itself throw the out_of_range
> exception in which case the 'at ( )' member function may not be
> needed ?

Because that's usually not what you want. You want some sort of
implementation defined crash, the equivalent of an assertion
failure. Undefined behavior allows the implementation to
provide this (and also allows removing the checks, if you find
that they create too much runtime overhead).

--
James Kanze (GABI Software) email:james.ka...@gmail.com
Conseils en informatique orientée objet/
Beratung in objektorientierter Datenverarbeitung
9 place Sémard, 78210 St.-Cyr-l'École, France, +33 (0)1 30 23 00 34