From: Samuel Krempp (krempp_at_[hidden])
Date: 2003-03-19 11:43:54
Le mer 19/03/2003 à 15:19, Beman Dawes a écrit :
> >I am currently doing a search for other places where borland v 0x0561 is
> >assumed, as I don't think the latest patch fixed any issues that would
> >affect boost and it would be a shame to have to choose between boost and
> >the patch.
> >(especially as Samuel put so much effort into getting this version to be
> >Borland compatible! I have mailed him separately off-list in case he is
> >not actively monitoring right now, but I know release is close)
>
> OK, I've made the second set of suggested changes.
thanks a lot.
I hadn't followed the ML for a week. (And I didn't really notice the new
__BORLANDC__ version)
Sorry for not responding sooner.
Indeed, updating 0x561 to 0x564 is fine. and has really zero possible
secondary effects on any other part of boost.
About the
|| BOOST_WORKAROUND(__BORLANDC__, BOOST_TESTED_AT( 0x570 ) ) )
versus
&& [etc..], it does not really matter.
with ||, all __BORLANDC__ are caught.
While, hopefully, there could be a future 0x565 version which does not
need the workaround.
so && was in fact a bit more precise. (but requires updating at each new
faulty __BORLANDC__ version).
since this workaround doesnt have much impact anyway, enabling for
future version is fine, so the || is fine.
-- Samuel