Re: [boost] Re: Boost implementation of the Library TR?

From: Peter Dimov (pdimov_at_[hidden])
Date: 2003-05-03 08:33:54

From: "Beman Dawes" <bdawes_at_[hidden]>
> At 09:44 AM 5/2/2003, David Abrahams wrote:
>  >"Joel de Guzman" <djowel_at_[hidden]> writes:
>  >
>  >> Beman Dawes wrote:
>  >>
>  >>> * Do TR components need a formal review?
>  >>>
>  >>>     Some kind of review is needed to ensure quality, but the process
>  >>>     should be simplified compared to a normal formal review. Details
>  >>>     need to be worked out.
>  >>
>  >> As alluded to by Beman, there was a previous discussion between me,
>  >> Doug, Dave and Jaakko. There's an implementation of the TR compliant
>  >> version of the tuple library written by Jaakko and me a few months
ago.
>  >> Right now, it is still in the Spirit CVS. Some of us would like to
have
>
>  >it
>  >> in boost's CVS in the tuple directory. I asked for a go signal. So
far,
>  >> Dave, Jaakko and Doug gave it a go. If a formal review is required, I
>  >> guess I can't proceed yet at this point?
>  >>
>  >> I await your comments regarding this matter.
>  >
>  >I think a careful review of the implementation is required, but maybe
>  >it doesn't rise to the level of a formal review.
>
> Yes, that was what I was thinking. At the very minimum, someone not part
of
> the development effort should look over the code.

Doesn't seem a good plan to me. You can't find bugs in TR libraries by
looking. It's a TR library if it passes a TR test suite, not a code review.