Prev: Module definition file and C++ mangled names
Next: Пояса массажеры с термо-эффектом
From: Hendrik Schober on 29 Apr 2008 04:38
Alex Blekhman <tkfx.REMOVE(a)yahoo.com> wrote:
> "Ben Voigt [C++ MVP]" wrote:
>
> [...]
>
> I'm using VC++ 2005/2008. I agree with you that some of C++0x
> features are already there. (For example, notable ">>" sequence is
> accepted when declaring a template.) However, "cool" features like
> lambdas, variadic templates, auto type inference etc. aren't
> available with VC++ yet.
>
>> I'd put a lot more stock in that and the timeline given at
>> http://www.open-std.org/JTC1/SC22/WG21/docs/papers/2008/n2565.html
>
> Esentially it's the same timetable that I posted. The commettee
> strives to finish the final draft in 2009. It means that it will
> take at least a year or two until final international
> ratification. I don't expect compiler vendors to run ahead of the
> pack and implement the draft before it becomes the standard.

Doesn't that contradict what you wrote further up?

Even MS seems to have realized that the C++ standard is
important and has put a lot of effort into implementing
it. While some important bits are still missing (namely
separate compilation of templates and a much stricter
rejecting of bad code), in the last 7 years VC has become
a pretty good C++ compiler and has caught up to be among
the leaders of the pack.
Would they wait until C++0x is officially released by ISO
before they start implementing it, they'd throw all this
away and would be back in the dreaded VC6 era, where they
were about a decade behind (old for-scoping, anyone?) and
would spoil all the good work they have done in the last
decade. I seriously doubt they are going to do that.

> Alex

Schobi

--
SpamTrap(a)gmx.de is never read
I'm HSchober at gmx dot de
"I guess at some point idealism meets human nature and
explodes." Daniel Orner


From: Alex Blekhman on 29 Apr 2008 05:55
"Hendrik Schober" wrote:
> Would they wait until C++0x is officially released by ISO
> before they start implementing it, they'd throw all this away
> and would be back in the dreaded VC6 era, where they were about
> a decade behind (old for-scoping, anyone?) and would spoil all
> the good work they have done in the last decade. I seriously
> doubt they are going to do that.

Most likely they already started to implement C++0x. However,
C++0x features are pretty heavy, so it will take time to shape
them up. I have doubts about whether we will see any of
significant C++0x features in VC++10.

VC++6 era was not so dreaded after all. VC++6 was a good compiler
when it came out. It still is the one of most widely used
compilers today. It definetely faster than newer VC++ compilers.

Alex


From: Ben Voigt [C++ MVP] on 29 Apr 2008 11:43
Alex Blekhman wrote:
> "Hendrik Schober" wrote:
>> Would they wait until C++0x is officially released by ISO
>> before they start implementing it, they'd throw all this away
>> and would be back in the dreaded VC6 era, where they were about
>> a decade behind (old for-scoping, anyone?) and would spoil all
>> the good work they have done in the last decade. I seriously
>> doubt they are going to do that.
>
> Most likely they already started to implement C++0x. However,
> C++0x features are pretty heavy, so it will take time to shape
> them up. I have doubts about whether we will see any of
> significant C++0x features in VC++10.

I take it you missed this statement from Herb Sutter's blog post I linked
to:

(Usability note: The lambda version was the only one I wrote correctly the
first time as I tried these examples on compilers to check them. 'Nuff said.
<tease type="shameless"> Yes, that means I tried it on a compiler. No, I'm
not making any product feature announcements about VC++ version 10. At least
not right now. </tease>)

Not only will compiler vendors not wait for ISO to ratify the standard, they
aren't even waiting for the final committee recommendation.

http://www.comeaucomputing.com/439features.html
http://gcc.gnu.org/projects/cxx0x.html

>
> VC++6 era was not so dreaded after all. VC++6 was a good compiler
> when it came out. It still is the one of most widely used
> compilers today. It definetely faster than newer VC++ compilers.
>
> Alex


From: Alex Blekhman on 29 Apr 2008 12:14
"Ben Voigt [C++ MVP]" wrote:
> Not only will compiler vendors not wait for ISO to ratify the
> standard, they aren't even waiting for the final committee
> recommendation.

I hope that you're right and we'll see the full C++0x
implementation rather sooner than later. However, currently, as
Herb Sutter said himself, there is no feature announcements about
VC++10. You tend to interpret his teaser as if these features were
already there. I prefer to be more careful in my prognoses. Given
the huge user base of MS products the company is utterly
conservative when introducing new features.

Alex


From: Bo Persson on 29 Apr 2008 12:38
Alex Blekhman wrote:
> "Ben Voigt [C++ MVP]" wrote:
>> Not only will compiler vendors not wait for ISO to ratify the
>> standard, they aren't even waiting for the final committee
>> recommendation.
>
> I hope that you're right and we'll see the full C++0x
> implementation rather sooner than later. However, currently, as
> Herb Sutter said himself, there is no feature announcements about
> VC++10. You tend to interpret his teaser as if these features were
> already there. I prefer to be more careful in my prognoses. Given
> the huge user base of MS products the company is utterly
> conservative when introducing new features.
>

On the other hand, there seems to have been an awakening late last
year, when they realized that they have to catch up with this:

http://gcc.gnu.org/projects/cxx0x.html


Otherwise, there will be applications running on Linux or OS X Server,
that just cannot be ported to Windows Server.

Horror! :-)


Bo Persson


 |  Next  |  Last
Pages: 1 2 3
Prev: Module definition file and C++ mangled names
Next: Пояса массажеры с термо-эффектом