Re: [ros-dev] Strategy for invalid parameters

[Thomas Weidenmueller]

M Bealby wrote:
> So what you're saying is that the functions in the modules of code
> which actually do the work should perform the checking.  What about
> this: If a parameter is passed straight through a function to another
> no checking should take place, but if any manipulation of the
> parameters occurs then they should be checked?  I think this might be
> slightly different to your idea Alex.  What does everyone think?  I
> personally think this is the best approach as it requires little, if
> any duplicated checking code.

I agree with Alex because the only place the handle can be checked is in 
ob. It doesn't make any sense to add pseudo checks.

- Thomas