Re: [boost] [iostreams] fix for vc-6_5

From: Joaquín Mª López Muñoz (joaquin_at_[hidden])
Date: 2006-07-20 13:54:47

Gennaro Prota ha escrito:

> On Thu, 20 Jul 2006 08:03:08 +0200, Joaquín Mª López Muñoz
> <joaquin_at_[hidden]> wrote:
>
> >Unless we all begin chasing red cells down we will never get 1.34
> >out of the door.
>
> Yes, we have admittedly some problems with that. I asked myself why
> and I think it is in part due to the number of interesting proposals
> which are regularly submitted, in part to the high commitment of many
> boost members to the WG21 committee, which has set a pretty aggressive
> schedule both for TR2 and C++0x.

I'm not so sure about those reasons you propose. Having a cursory inspection
of the regression tests landscape, I'd say the main sources of red and yellow
cells are, roughly ordered in decreasing importance measured as no of
failures:

1. More or less exotic compilers for which nobody cares to try to fix things or,
at least, mark whole libs as unusable on them.
2. Toolsets with some kind of configuration problem.
3. Abandoned libs.
4. Libs which are strictly not abandoned (the author hangs around the lists)
but need some more attention: even if the codebase does not change,
problems in 3rd party libs can result in regressions here.

Probably the most worrysome areas are #3 and #4. I don't think the reasons
you brought up can be used to explain these, at least not #3. In the
short term, the only thing we can probably do is devote some
spare time to greenify regressions, which is what I try to do when I can
spend an hour or two on this: many of the problems are relatively simple
to fix if you have access to the problematic compiler. In the longer term,
declaring some libs as abandoned and eventually getting someone to
maintain them would be IMHO the best course of action.

> Another facet of the problem is coupling between libraries. I think
> once 1.34.0 is out it will be worth to devote some thoughts to this.

Beman Dawes proposed a very attractive maintenance and release
scheme designed to avoid this kind of situations.

http://lists.boost.org/Archives/boost/2006/05/104654.php

Joaquín M López Muñoz
Telefónica, Investigación y Desarrollo