Re: [boost] Re: Re: Re: Attempting resolution of Threads &ExceptionsIssue

From: David Abrahams (dave_at_[hidden])
Date: 2002-08-13 15:54:50

I think your statement is too sweeping. It's a good guideline if:

1. You have to cope with systems that integrate "crash" exceptions with C++
EH
2. They don't provide a workaround, like the ability to handle the "crash"
at the site of detection
3. You hope to recover from these crashes somehow or debugging requirements
mean you have no alternative

My rationale is:

1. We shouldn't advise every programmer to pick a coding style based on
what some misguided implementations do
2. Single-base-class hierarchies are old-fashioned ;-)
3. We shouldn't encourage misguided implementations to propagate
4. Your guideline doesn't always help, and it's too restrictive. The real
problem is interaction with catch(...). Lots of stdlib implementations and
existing code uses catch(...)

-----------------------------------------------------------
           David Abrahams * Boost Consulting
dave_at_[hidden] * http://www.boost-consulting.com

----- Original Message -----
From: "Hillel Y. Sims" <hsims_at_[hidden]>
Newsgroups: gmane.comp.lib.boost.devel
To: <boost_at_[hidden]>
Sent: Tuesday, August 13, 2002 5:05 PM
Subject: [boost] Re: Re: Re: Attempting resolution of Threads
&ExceptionsIssue

> Does this mean you agree or disagree with my statement? If disagree,
please
> provide rationale.
>
> thanks,
> hys
>
> "David Abrahams" <dave_at_[hidden]> wrote in message
> news:0b5b01c24306$254e52f0$6501a8c0_at_boostconsulting.com...
> >
> > From: "Hillel Y. Sims" <hsims_at_[hidden]>
> >
> > > I really hope that is just a faulty recollection; std::exception
should
> > > always be the ultimate base class for all user-defined exceptions in
any
> > > good design, especially libraries (except maybe under very
specialized
> > > circumstances).
> >
> > I'm fairly certain that Bjarne is not intimately familiar with the
> > non-standard extensions to C++ EH which drive you to your conclusions
> about
> > what should and should not be done.
> >
>
> --
> Hillel Y. Sims
> FactSet Research Systems
> hsims AT factset.com
>
>
>
>
> _______________________________________________
> Unsubscribe & other changes:
http://lists.boost.org/mailman/listinfo.cgi/boost
>