|
Boost : |
From: David Abrahams (dave_at_[hidden])
Date: 2003-10-08 10:18:00
"Peter Dimov" <pdimov_at_[hidden]> writes:
> David Abrahams wrote:
>> "Pavel Kuznetsov" <pavel_at_[hidden]> writes:
>>
>>> Peter Dimov (pdimov_at_[hidden]) wrote:
>>>
>>>> You can't disable the g++ warning with a pragma
>>>
>>> Then I suppose the only solution is to check whether given symbol is
>>> defined before using it in BOOST_WORKAROUND. Does everybody agrees
>>> to this strategy? I mean, if nobody is against it, I'll make
>>> appropriate changes and send the diffs. Or rather some voting
>>> procedure should be followed?
>>
>> I guess I'm against it, on the grounds that it's both verbose and
>> error-prone.
>
> But it works, and the alternatives do not. "Works" as in "makes Boost usable
> with -Wundef".
I know :(.
Before accepting the inevitable, I'd like some comments on this:
If we *were* going to do that, though, maybe we should simplify
BOOST_WORKAROUND to not deal with the !defined(SYMBOL) case, or even
cause an error if SYMBOL == 0.
-- Dave Abrahams Boost Consulting www.boost-consulting.com
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk