|
Boost : |
From: Reid Sweatman (borderland_at_[hidden])
Date: 2000-08-25 05:32:08
> -----Original Message-----
> From: Gary Powell [mailto:Gary.Powell_at_[hidden]]
> Sent: Tuesday, August 22, 2000 11:51 AM
> To: 'boost_at_[hidden]'
> Subject: RE: [boost] VC++ 7.0 Challenge
> The MSDeveloper magazine, listed that MS has fixed the error messages from
> using templates, and updated the library to be fully standard
> compliant. I'm
> not sure how they managed it without partial specialization, but that was
> the claim. They also claim to have fixed some compiler bugs. The inside
> rumor is that the code base is very brittle and needs to be
> re-written, and
> reworked.
So they just made the suppression pragmas internal to the compiler. As for
fully compliant, I suggest you go have a look at the announcements on their
site about the compiler. They explicitly list several template features
that they *still* won't support, all of which are necessary for Boost to
compile correctly, without workarounds.
> I too am stuck using MS for reasons partly political and partly platform
> related. Therefore I have used gcc to do my C++ library development work.
> While I can't benefit from this code at work, I am at least
> helping the C++
> world, and advancing my own skills.
Same here. I switched compilers personally after joining this list and
finding out who really supported the spec, but generally, when I work for
someone else, which I'm shortly going to be doing again, they usually
require VC++.
> My personal annoyance with MS is that they released C#. I would have
> preferred the effort had gone to fixing C++.
Exactly what I posted to their web page on C#, which has a cutesy DHTML
goodie at the bottom to let you post messages and read what others have
posted. I think I might have seen two positive messages there. And yeah, I
personally suspect that C# is why VC++ 7 isn't much of an improvement.
Reid Sweatman
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk