|
This isn't just about "C++ 'Buzzwords' ", this is about truly usable libraries, had one access to a C++ compiler that is, and a list of violations against the C++ standard Microsofts VC7 performs (making me think some companies should be forbidden to call their compilers "C++" compilers until they truly are - being forced to use the more correct "C++ wannabe compiler", and possibly adding the word "maybe" when dealing with Micros~1).
Since none of these things work with any publicly available Microsoft compiler (and I seriously dubt they "got it" even in the "secret club release" 7.1 either), why even bother with a poll?
Since I've NEVER BEEN ABLE TO USE any of this with the MS POS so-called C++ compiler, there is NO WAY I (if I only used MS tools and only targetted Windows, using MS tools) even could be "familiar enough with to use" it.
What kind of troll wrote this poll? It's, literally, like someone asking "Without ever have been presented by your monopoly vendor the tools to parse or use conforming ISO C++ code, what of the following libraries and tools, of which all requires following the international ISO C++ standard ratified half a decade ago, are you familiar enough with to use if your vendor got its thumb out of its arse to really create something even resembling the international standard C++ parser requirements?".
Would "Duh?" be a good answer. I (and You) CAN'T USE IT - therefore the only way to have gotten enough experience with this would have to be by using better tools from some other vendor!
If I had indeed done that, used better tools (e.g. como, or even GCC C++), why in the name of Goat(se.cx) would I ever want to pay several more thousands of dollars to get access to another beta-quality release that "maybe" and "in the future" will still not compile conforming code?!
I think someone is either trolling, or stuck so full with MS X-mas snort they don't know heads from tails. I'm sorry if this is offensive, but this "poll" just don't compute.
Merry christmas and piss off. If you are so dumb you're offended by this post, don't blame me - blame your parents for being stupid enough to not use appropriate protection
|
|
|
|
|
Damn you're bitching. Settle down beavies.
--
Only in a world this sh*tty could you even try to say these were innocent people and keep a straight face.
|
|
|
|
|
Me? Bitchin'? Seen any MS compiler around tellin' me not to? :->
I only ask for small things like PTS, correct Koenig lookup and ... just following the international C++ standard. Instead, like holding on to inches and feet, they go for what is not accepted in the rest of the world.
Hubris? Must be...
|
|
|
|
|
Mike Nordell wrote:
Me? Bitchin'? Seen any MS compiler around tellin' me not to? :->
I was referring to the "Poll writer is a troll" part
But I don't see how you can be so negative about the new C++ compiler. It's not released yet, so there's a chance that it may actually support PTS, König et al a 100%.
Ropa inte hej förrän...
--
Only in a world this sh*tty could you even try to say these were innocent people and keep a straight face.
|
|
|
|
|
Jörgen Sigvardsson wrote:
I was referring to the "Poll writer is a troll" part
Ahh, yes, maybe I was a bit too critical. It could also have been some kind of wicked x-mas joke "Look here at all the goodies you can't use! :-P". Hmmm, maybe there is some hidden message in this, like <voice=ghostly>"time to switch... time to switch..."</voice>.
Then, maybe I'm just dreaming.
But I don't see how you can be so negative about the new C++ compiler.
They paired it with an offensive IDE? I know Microsoft? They already told us it won't work?
It's not released yet, so there's a chance that it may actually support PTS, König et al a 100%.
When pigs fly...
Seriously, not if you follow this NG. Maybe this link could also be of use. Microsoft would NEVER fix all known bugs in a product for the next release, it goes against everything they stand for and their business practice. They ALWAYS leave known bugs, just for the hell of it some say, but the truth is they do it to be able to tease people into shelling out even more money for the next version, and the next version and the next version...
It has also been said by at least one Microsoft spokesperson in the mentioned NG that some vital C++ conformance won't be added until (at the earliest) version 8.
Ropa inte hej förrän...
du behöver ett bäcken.
--
-There are 10 kinds of people in the world;
those who understand binary, and those who don't.
|
|
|
|
|
I followed the microsoft-link on compiler limitations. Apparently it was about Visual Studio .NET current release. I thought the poll was about Everett, the next version. I read somewhere that they'll be very close to 100% compliance in Everett. To reach 100% they'd have a bit more time. Not reaching 100% fully doesn't really bother me (until I'm suffering from it )
I thought the jump in compliance from VC6 to VC7 was a great leap. It really simplified a lot of programming for me, so I'm at least happier. Sure, I would like to see better tools, but I'm afraid I'm trapped in the Microsoft world. It's hard to find packages which gives more value per buck than MSDN subscriptions. Buying an extra compiler for shitloads of cash to gain the last 2% of compliance isn't very tempting after having shelled out $2000+
Mike Nordell wrote:
Ropa inte hej förrän...
du behöver ett bäcken.
--
Only in a world this sh*tty could you even try to say these were innocent people and keep a straight face.
|
|
|
|
|
I don´t know what the hell are these things! (excepting two of them), but now I know what should I do when connected to the internet and having nothing to do
Regards.
"When I look into your eyes, there´s nothing there to see, nothing but my own mistakes staring back at me"
|
|
|
|
|
|
Nishant S wrote:
Next time I demand two votes
And that's your right!!!
Regards,
Brian Dela
"There should be an amendment to the constitution, that every president must be examined for paranoia before moving into office." - peterchen
|
|
|
|
|
Unless something significant has changed between the final beta and the release version, unexpected() will still be essentially useless, because exception specifications are still essentially ignored.
VC++ only understands two exception specifications; throw() (throws nothing) and throw(...) (throws anything); any other exception specifications are treated as throw(...).
This is not a bad thing, mind you. Exception specifications are dangerous and can result in resource leaks and other nasties. Calling unexpected() by default calls terminate(). terminate() doesn't unwind the stack -- so if you were relying on destructors to release resources, that's not going to happen.
You can't even solve this by replacing unexpected(). unexpected() has few choices about what it can do:
1) terminate the program with terminate(), exit(), or abort(). This is bad as it doesn't unwind the stack.
2) throw an object of type bad_exception. IFF the exception specification includes bad_exception exception handling will continue as usual (but most exception handlers don't, and unexpected() can't tell anyway). If it doesn't, the program will call terminate() -- no stack unwinding.
3) throw an object of a type specified in the exception specification. But since unexpected() doesn't know what was in the exception specification, this in general impossible. If it manages it, stack unwinding continues as normal. If it doesn't manage it, terminate() gets called.
Exception specifications (except for throw(), which has notable optimization benefits) are broken and dangerous. They shouldn't even be in the C++ Standard; they were an experiment and a bad one at that. They should not be supported by compilers as they reduce the fault-tolerance of programs.
|
|
|
|
|
This works like a charm:
template <typename T, int ARR_SIZE>
class array
{
T data[ARR_SIZE];
};
Did you mean template template paremeters?
|
|
|
|
|
I thought we did, but I was not sure.
Christian
No offense, but I don't really want to encourage the creation of another VB developer.
- Larry Antram 22 Oct 2002
C# will attract all comers, where VB is for IT Journalists and managers - Michael
P Butler 05-12-2002
Again, you can screw up a C/C++ program just as easily as a VB program. OK, maybe not
as easily, but it's certainly doable. - Jamie Nordmeyer - 15-Nov-2002
|
|
|
|
|
Yes. There are 3 kinds of template parameters:
1. Type parameters (class, union, built-in type)
2. Non-type parameters (constant integral values)
3. template template parameters which are templates themself.
Of those, VC does not support only 3. I guess that VC 7.1 will.
|
|
|
|
|
Seems to me you're right. VC++ 6.0 supports non-template parameters, and I'm not aware of any bug relating to the compiler support in this area.
Joaquín M López Muñoz
Telefónica, Investigación y Desarrollo
|
|
|
|
|
There actually is a flaw in MSVC 6.0 (don't know about 7)
template<unsigned N><br />
int Fib(void)<br />
{<br />
return Fib(N-1)+Fib(N-2);<br />
}<br />
template<><br />
int Fib<1>(void)<br />
{<br />
return 1;<br />
}<br />
template<><br />
int Fib<2>(void)<br />
{<br />
return 1;<br />
}<br />
This doesn't work.
Writing something like
int i=Fib<5>();
Will instead of the correct (recursive) instantiation just instantiate Fib<1>()
|
|
|
|
|
Oh yeah. That bug. I have run into that before. Happens only with functions and not classes.
Tim Smith
I'm going to patent thought. I have yet to see any prior art.
|
|
|
|
|
just to make me learn C++. The last 2 or 3 polls have been C++ related and I haven't a clue what any of these features are.
I already said I would do it over intersession...or rather, next week(I have a class scheduled for intersession ). I promise! I'll learn C++! Really...so could we please return to the polls that apply to everybody?
I don't know whether it's just the light but I swear the database server gives me dirty looks everytime I wander past.
-Chris Maunder
Microsoft has reinvented the wheel, this time they made it round.
-Peterchen on VS.NET
|
|
|
|
|
Actually, most of those have nothing to do with C++. It is just a big joke we are all in on to make the non-C++ people feel stupid.
Tim Smith
I'm going to patent thought. I have yet to see any prior art.
|
|
|
|
|
|
|
Since the poll said something to the effect of, without searching, blah blah, of course, I went and Googled several of these just to see what they were. As they came up, a couple looked familiar. Some others I had never heard of.
Even a broken clock is right twice a day.
|
|
|
|
|
I never heard of any of these C++ Buzzwords and would love to read a few words explaining very shortly each of these buzzwords.
Thanks !
Jerome
|
|
|
|
|
I wrote an article on Koenig lookup a day or so ago, and explained empty aggregate initialisers below somewhere. I hope to do an article on PTS sometime very soon. The top ones are all libraries, Boost contains stuff that generally has a good chance of making the next standard, Loki is Andrei Alexandrescu's policy template library, I don't really know the others.
Christian
No offense, but I don't really want to encourage the creation of another VB developer.
- Larry Antram 22 Oct 2002
C# will attract all comers, where VB is for IT Journalists and managers - Michael
P Butler 05-12-2002
Again, you can screw up a C/C++ program just as easily as a VB program. OK, maybe not
as easily, but it's certainly doable. - Jamie Nordmeyer - 15-Nov-2002
|
|
|
|
|
Thanks !!!
Jerome
|
|
|
|
|
Blitz is a numerical library to add fortan level support to c++ for scientific apps.
unexpected is called by the code when a type not specified as being thrown is thrown, the standard has the best example of course.
non type templatee params are integrals passed straight to the template, it makes constants that change possible and allow for a lot of compile time optimization
PTS is probably Partial Template specialization, its a complex subject since I hope by that he means all the other template features missing from the MS compiler (minus export which i don't care for).
|
|
|
|