Comments on: Differnece between oprator[] and "at" function of vector

By: bobobobo

bobobobo — Wed, 24 Jun 2009 16:40:06 +0000

sweet post! I like your writing style.

By: out of bound vector [] access. « Computing Life

out of bound vector [] access. « Computing Life — Wed, 10 Oct 2007 14:33:46 +0000

[...] what should happen when out of bound access is made? Once you have made up your mind, check out the rational behind the way STL [...]

By: S a r a t h

S a r a t h — Thu, 14 Jun 2007 12:03:35 +0000

Thanks.
KD,
it’s an exception on accessing invalid memory. That’s why it’s getting inside the common catch block. It could be any error, not the bug of vector, suppose if the copy contructor or operator= of the type we are passing to the vector could also throw the exception, or any other function. It can be anything which causes the exception. but it’s not possible for us to catch anything related out of range access of vector.
As I said in the post, If you are using Visual Studio 2005 and you disabled the checked iterators, the exception will not be caught in the common block. It will simply crash.

The behavior of the same may differ vendor to vendor and as per the standard there’s no exception specification for operator[]
HTH

By: kd

kd — Thu, 14 Jun 2007 08:19:12 +0000

nice post.
what happen if we use like this
try{ v[1000] = 10; }
catch(…){}

v is vector of zero size.
I still get the exception in this case.

try{ v[1000] ; } //In this case no exception, i think since we are not accessing thing

catch(…){}