Re: [Boost-testing] Re: [testing] Possible changes to testing..

From: John Maddock (john_at_[hidden])
Date: 2005-03-09 08:08:38

>> This situation is likely a reaction to reduce test times to
>> something manageable.
>
> I'm not 100% positive about Boost.Config (John?), but for other
> libraries it's just an atavism.

The "full" tests in libs/config/test used to take a lot longer to run than 
they do currently, but mainly they aren't 100% reliable: some common sense 
is required in interpreting some of the results, we also don't have the bjam 
rules needed to make the tests more reliable.  Whatever refactoring is not a 
problem, I'm sure this is true for all the libraries.

John.