Re: [boost] BOOST_CHECK_EQUAL() dangers

From: David Abrahams (dave_at_[hidden])
Date: 2002-11-20 10:51:53

Markus Schöpflin <markus.schoepflin_at_[hidden]> writes:

> 1. Just document it. Not so good, IMHO.
>
> 2. Document the special case and add a check for NULL pointers before calling strcmp().
>
> 3. Remove the special case alltogether. After all, I might want to
> check that the pointers are equal and not the string they point
> to. This might be the best solution, but breaks backward
> compatibility.

I vote for #3, strongly. It's too capricious and inconsistent
otherwise, and might make the check useless in generic code. The user
can always convert to std::string if he wants the other check.

-- 
                       David Abrahams
   dave_at_[hidden] * http://www.boost-consulting.com
Boost support, enhancements, training, and commercial distribution