Comments by AndreFratelli (Top 40 by date)

Yeah, it kind of loses its purpose...

That would catch the exceptions not listed in the catch blocks, not the one not listed in the function declaration.

What I make of this: it can't be done. If VS doesn't support exception specification there is no way to know if an exception is not specified. Hell, none is.

I agree.

And I can confirm it. I tested it.

Yes, VC doesn't even support exception specification (http://msdn.microsoft.com/en-us/library/sa28fef8.aspx), so it doesn't even make sense to talk about unexpected exceptions there.

Anyway, how would you suggest that to be done at the compiler level?

that is what I call an awesome answer! =)

I didn't take exactly the same approach, though. But the idea to pass an encapsulated array to the constructor gave me an idea. So I did this:

<pre>
// "const_array" and "array" are inner classes, size_type and T are defined in the parent class
template < size_type _Capacity > class const_array
{
public:
const T& operator[](size_type index) const { ... }
protected:
T items[_Capacity];
};

template < size_type _Capacity > class array
: public const_array<_Capacity>
{
public:
T& operator[](size_type index) { /* not const */ }
};
</pre>

Now, to initialize first construct an "array" and pass it to "const_array". Worked like a charm =)

What do you think?

Thank you!

Appreciated ;)

Cheers.

My later experiments seem to suggest that the only problem will be with DLLs. It seems the compiler actually tries to compile the templates when they are exported in a DLL. Unfortunately, this really is to export in a DLL.

Anyway, the conditions have changed and I can't even use STL at all! It seems some target platforms do not support it, don't really know which. That said, I wrote my own implementation of red-black trees and I'm using them in the slots of an hash table (it's even more efficient). But I'll take a look at those compilers ;)

Actually the class being mapped was written by me. I usually insist in designing my apps as well as I can and I believe that a default constructor should not be there when a class shouldn't support default construction. This is the case. It has a mandatory argument. Of course I could work around this and allow default construction and then allow setting the value with a setter. But then again, it would allow default construction =) it's just a design issue.

Thank you for your reply!
Regards

I was kind of fearing that :P

std::map::operator [] does one of two things:

1) Return a reference to an existing value with the given key; or
2) Instantiate a new value using the default constructor and associate it with the key.

So what to do when a default constructor is not available? I inherited from std::map (assoc_map) and allow the passage of a default value at construction time.

Then I overload operator[] and copy the default instance instead of propagating to std::map::operator[], when the value does not exist.

This works because it's a template class and, therefore, its methods only get compiled when a specialization exists for them. What I would like to know is if this is indeed true. If an attempt exists to try to compile std::map::operator[] it will complain that the default constructor is not available, even though that does not happen in VS08.

Thank you

I'm sorry, didn't notice this before. andre.fratelli@hotmail.com

Thanks!

You are? Why?
Where do I send it to?

Thx!

Maybe a bug report, then?

Indeed, I'm not inlining it anymore. I posted because I think it's odd and couldn't understand the reason. Right know I just wan't to understand it =)

regards

Sorry, didn't notice that last line. Yes, I did. Please note that not inlining the operator works, so everything else works.

Yes, I know all that. Writing the body with the declaration (in the header file) gives other unresolved externals, oddly enough:

Error 1 error LNK2019: unresolved external symbol "public: class std::basic_string<char,struct std::char_traits<char="">,class std::allocator<char> > const & __thiscall SGEEmotion::id(void)const " (?id@SGEEmotion@@QBEABV?$basic_string@DU?$char_traits@D@std@@V?$allocator@D@2@@std@@XZ) referenced in function "public: bool __thiscall SGEEmotion::Comparator::operator()(class SGEEmotion const &,class SGEEmotion const &)const " (??RComparator@SGEEmotion@@QBE_NABV1@0@Z) SGEMood.obj

I've read it, and I know what inlining is. I don't get the point you're trying to make..

Hi, please see my reply above. Thx!

It's SGEEmotion::Comparator::operator ()


Error 1 error LNK2019: unresolved external symbol "public: bool __thiscall SGEEmotion::Comparator::operator()(class SGEEmotion const &,class SGEEmotion const &)const " (??RComparator@SGEEmotion@@QBE_NABV1@0@Z) referenced in function "public: float & __thiscall std::map<class sgeemotion,float,class="" sgeemotion::comparator,class="" std::allocator<struct="" std::pair<class="" sgeemotion="" const="" ,float=""> > >::operator[](class SGEEmotion const &)" (??A?$map@VSGEEmotion@@MVComparator@1@V?$allocator@U?$pair@$$CBVSGEEmotion@@M@std@@@std@@@std@@QAEAAMABVSGEEmotion@@@Z) SGEMood.obj

This:

freeglut_static.lib(freeglut_main.obj) : error LNK2001: unresolved external symbol __iob

I posted in the C# forum? =\ didn't even realise..

It's neither, actually. The problem arose while developing a Qt application, but still, I'm interested in the more abstract idea of the object handling the thread dying before the thread does.

Ah, so without that call it wouldn't be call at all?

QWidget and QGLWidget are Qt classes, so I really can't do anything.

1) The only thing they have in common is QWidget. Doesn't it just need to be virtual somewhere in the inheritance tree?
2) Nice =)
3) What do you mean the most specified? The one that's supposed to be called more times?
4) Oh my...

I was exploring the virtual inheritance approach, but I see I have more to lose then I have to gain here... I think I'll just QWidget as the base class and drop the Widget class at all.

Thank you for your patience!
Best.

I did read about it. That's how I got this questions in the first place =)

What do you mean no matter what? Isn't that what virtual inheritance is supposed to do? =\

Thanks!

Oh.. =\ I get it.

I don't think so. You have to fetch the data and apply the algorithm to it outside the database. I don't recommend it though.. Instead, you should save paths and store clustered information about the files (using IDs).

Oh, Ok. But why strike?

(did I just double post? I'm still figuring out some details on this interface =P)

I'm not "cheers"... See here

Sorry, but can you be a bit clearer? I don't understand what you'r saying...

Sorry, what? =\

Actually, no, I don't quite get your questions...

In the first one are you trying to know how the {n} marks associate with the arguments? If so, take a look at varargs, you could write something similar with it.

As to the second one, empirically I would say you are trying to insert something at the beginning of a dictionary. Is that it? If so, then it makes no sense because dictionaries are not ordered.

I don't really know D3D, only OpenGL. But still, why would you need to access the texture's data to detect collisions? I recently wrote a collision detection system, and I think you are taking the wrong approach here. Have you considered other options to "pixel" collision detection? Bound-boxes, Octrees, Triangle-triangle collisions..

Anyway, whenever you want something transformed, you better transform it in CPU (assuming you are not transforming ** everything ** in a mesh, instead of just whatever needs to be checked).

regards

Since you are already using structs, you could also go old school and implement a linked list yourself. STL is probably a better option, though.

Moved this comment so OP gets it.