C / C++ / MFC

Re: about change array length

14-Dec-11 11:43

Very good link. Smile | :)

If the Lord God Almighty had consulted me before embarking upon the Creation, I would have recommended something simpler.
-- Alfonso the Wise, 13th Century King of Castile.

This is going on my arrogant assumptions. You may have a superb reason why I'm completely wrong.
-- Iain Clarke

[My articles]

szpillka15-Dec-11 5:32

szpillka

15-Dec-11 5:32

Unfortunaltely C/C++ does not allow programmer to use something similar to Delphi's setlength procedure, so that array once set length cannot be resized. When u want to resize, u must declare new array with bigger size, copy elements from old to new, delete old array. So therefore STL has some classes-containers, most often used by me is a list (from #include (list))

Read about list, vector, map, check Google

A quastion in programming language desing' scope

desatir731613-Dec-11 20:43

desatir7316

13-Dec-11 20:43

hello
what is difference between these assigning values for "a" and "b" ?

C++

class c1{
int a;
int b;
c1():b(0){
a=3;
}
}

thanks a lot

Re: A quastion in programming language desing' scope

Orjan Westin14-Dec-11 0:05

Orjan Westin

14-Dec-11 0:05

b is assigned first, since class initialising is executed before the constructor body. This is also the only place where you can assign a value to a non-static const or reference member variable.

Re: A quastion in programming language desing' scope

David Crow14-Dec-11 3:32

David Crow

14-Dec-11 3:32

See here.

what is the result(union & structure)?

desatir731613-Dec-11 20:32

desatir7316

13-Dec-11 20:32

hi there
can someone say what is the answer of this piece of code
by the way it is so important that why this answer made by the code in Programming language's scope
tnx

C++

union U{
struct S{
shortint x;
shortint y;
}
float f;
long l;
}u1;
u1.S.x=3;
u1.s.y=4;
cout<<f;
cout<<l;

Richard MacCutchan13-Dec-11 22:13

13-Dec-11 22:13

The answers will most likely be garbage as the code refers to three different types in the same piece of memory. The float and long values will not have been initialised properly.

Unrequited desire is character building. OriginalGriff
I'm sitting here giving you a standing ovation - Len Goodman

CPallini14-Dec-11 2:33

14-Dec-11 2:33

Your code is utterly broken. A more compiler-friendly one could be:

C++

#include <iostream>
using namespace std;

union U
{
  struct S
  {
    short int x;
    short int y;
  }s;
  float f;
  long l;
} u1;

int main()
{
  u1.s.x=3;
  u1.s.y=4;

  cout << u1.f << endl;
  cout << u1.l << endl; 

}

Now let's have a look at the u1 memory layout (at least on my 32 bit machine):

ADDR      00 01 02 03
VALUE     03 00 04 00

where addresses are relative to &u1.
Since the (my) machine is little endian, we have the following bit configuration

BYTES   0    0    0    4    0    0    0    3
BITS  0000 0000 0000 0100 0000 0000 0000 0011

That is, for the float member,

SIGN     0
EXP      0
MANTISSA 00001000000000000000011

That is a denormalized float (see here[^]) with value

(2^-5 + 2^-22 + 2^-23) * 2^-126 = 3.6735039242275079514265571025945 * 10^-40

that looks similar to the program output: 3.67346E-040

Richard MacCutchan14-Dec-11 2:40

14-Dec-11 2:40

Excellent answer.

Unrequited desire is character building. OriginalGriff
I'm sitting here giving you a standing ovation - Len Goodman

CPallini14-Dec-11 2:42

14-Dec-11 2:42

Thank you, Richard. Smile | :)

Stefan_Lang14-Dec-11 5:34

Stefan_Lang

14-Dec-11 5:34

While C.Pallini's response may be correct, there is no guarantee, as the internal format of float is not defined by the standard and therefore compiler-specific! There may even be compiler options to use a specific representation, so the result may change depending on those options.

Also the standard does not specify that long is a 32 bit type, only that it must be at least 32 bit. While I know of no compiler that does this*, any compiler is free to use more than 32 bits, especially when the OS is not 32 bit! If for instance long were a 64 bit type, the value of l would be 3*2^48+4*2^32, which is a lot bigger than 3*2^16+4 in case of a 32 bit long type.

*compilers that support bigger than 32 bit integer types usually provide them by different names, such as 'long long' or '__int64', rather than just 'long'

CPallini15-Dec-11 10:10

How to Convert vector::reference to vector::iterator ?

15-Dec-11 10:10

As you correctly note, the particular memory layout I considered is not assured (this is why I specified 'on my 32 bit', 'little endian' machine and so on). My aim was to give him a method for investigating such matter.
Smile | :)

002comp13-Dec-11 20:21

002comp

13-Dec-11 20:21

Hello Friends

I am using vector and I used vector::iterator to iterate over the elements.
And after that I am using vector::reference = vector.back() to access last element.

Now,i want to assign all these refernce value to iterator one.

Any Ideas ?

Thanks & Regards
Yogesh

Re: How to Convert vector::reference to vector::iterator ?

C^hris L^osinger14-Dec-11 2:22

C^hris L^osinger

14-Dec-11 2:22

C++

for (int i=1;i<vec.size();i++) vec[i] = vec[0];

image processing toolkits | batch image processing

Re: How to Convert vector::reference to vector::iterator ?

Stefan_Lang14-Dec-11 5:40

Stefan_Lang

14-Dec-11 5:40

You can't. The type of vector<T>::reference is T&, and has no reference to the original vector object, only to one element of type T.

The only way to assign an iterator is either through one of the methods of vector, or by assigning another iterator.

If you want an iterator to the last element, you can do this:

C++

std::vector<int> myvector(5); // 5 elements
std::vector::iterator last = myvector.end(); // points past the last element
--last; // now points at the last element

P.S.: after rereading your question I think I may have misinterpreted your question. If this isn't what you wanted and need more help, please clarify.

Re: How to Convert vector::reference to vector::iterator ?

002comp14-Dec-11 17:30

002comp

14-Dec-11 17:30

Thanks For your Answers.
I also proceed with the Stefan Method.

Thanks A Lot.
Regards
Yogesh

how to test a compressed file in MFC?

Le@rner13-Dec-11 20:20

13-Dec-11 20:20

hi all,

i want to test comressed file or Archive in my MFC application.

i want its a valid zip file or not.

i have no idea from where i have to start.

please guide me and help me for this.

thanks in advance.

Richard MacCutchan13-Dec-11 22:09

13-Dec-11 22:09

Use Google to find details on the different types of compressed files and their signatures.

Unrequited desire is character building. OriginalGriff
I'm sitting here giving you a standing ovation - Len Goodman

Le@rner13-Dec-11 23:52

13-Dec-11 23:52

now how to open these files and decide these are valid archive or not.

Richard MacCutchan14-Dec-11 0:08

14-Dec-11 0:08

I just explained what you need to do; what more information do you need?

Unrequited desire is character building. OriginalGriff
I'm sitting here giving you a standing ovation - Len Goodman

Le@rner14-Dec-11 0:35

14-Dec-11 0:35

on signature basis i just found file is which type not that is valid or not.

so i want any class or function using the i can find the file is valid or not.

like test method in 7zip application.

Richard MacCutchan14-Dec-11 0:38

14-Dec-11 0:38

Given the problem you are trying to resolve I would think you will have to write the function yourself. There are lots of resources on the internet concerned with compressed files that will help you.

Unrequited desire is character building. OriginalGriff
I'm sitting here giving you a standing ovation - Len Goodman

Le@rner14-Dec-11 0:46

14-Dec-11 0:46

can provide any one out of this

thanks.