Adding two matrix objects and assigning the result to third object using overloaded operators ? - C / C++ / MFC Discussion Boards

Re: Hackerrank:Down to Zero problem

David Crow25-Mar-18 15:41

David Crow

25-Mar-18 15:41

See here.

"One man's wage rise is another man's price increase." - Harold Wilson

"Fireproof doesn't mean the fire will never come. It means when the fire comes that you will be able to withstand it." - Michael Simmons

"You can easily judge the character of a man by how he treats those who can do nothing for him." - James D. Miles

In type conversion of 2 different classes can i convert both ways ?

Tarun Jha25-Mar-18 5:14

Tarun Jha

25-Mar-18 5:14

i want to make a program that can convert from class a type to class b & vice versa..
i already made a program that does one type of conversion..

//define 2 classes polar & rectangle to represent points int the polar & rectanglular system.
#include <iostream>
#include <cmath>
using namespace std;

class Polar;

//====================================================
class Rectangle{
    float length, bredth;
public:
    Rectangle(){}
    Rectangle(float ,float);

    /*
      operator Polar(){
        Polar temp;
        float a = atan(length/bredth), r = sqrt(x*x + y*y);
        temp.getAngle(a);
        temp.getRadius(r);

        return(temp);
    }
    */

    int getLength(){return(length);}
    int getBredth(){return(bredth);}
    void putLength(float x){length = x;}
    void putBredth(float y){bredth = y;}

    void putData(){cout<<"Length : "<<length<<"\tBreadth : "<<bredth<<endl;}
    ~Rectangle(){}
};

Rectangle::Rectangle(float x, float y){
    length = x;
    bredth = y;
}

//====================================================

class Polar{
    float angle, rad;
public:

    Polar(){}
    Polar(float , float);

    operator Rectangle(){
        Rectangle temp;
        float l = rad*cos(angle), b =rad*sin(angle) ;
        temp.putLength(l);
        temp.putBredth(b);

        return(temp);
    }

    int getAngle(){return(angle);}
    int getRadius(){return(rad);}
    void putData(){cout<<"Angle : "<<angle<<"\tRadius : "<<rad<<endl;}
    ~Polar(){}
};

Polar::Polar(float x, float y){
    angle = x;
    rad = y;
}

// Rectangle::operator Polar(){
//     Polar temp;
//     float a = atan(length/bredth), r = sqrt(x*x + y*y);
//     temp.getAngle(a);
//     temp.getRadius(r);
//
//     return(temp);
// }
//===========================================================

int main(){
    Rectangle r1(40, 20.23);
    Polar p1(120, 25);

    r1 = p1;
    r1.putData();

    //p1 = r1;
    //p1.putData();

    return 0;

}

as in the above code i have converted class Polar to class Rectangle, but i also want to have a conversion of class Rectangle to class Polar.

Thank you

Re: In type conversion of 2 different classes can i convert both ways ?

Victor Nijegorodov25-Mar-18 8:07

Victor Nijegorodov

25-Mar-18 8:07

Tarun Jha wrote:
as in the above code i have converted class Polar to class Rectangle,

Did you test it?
Does it work?

I ask because I see neither the assignment operator

Rectangle& Rectangle::operator =(const Polar& polar)

nor the ctor to create Rectangle instance from the Polar one... Confused | :confused:

Re: In type conversion of 2 different classes can i convert both ways ?

Tarun Jha25-Mar-18 23:26

Tarun Jha

25-Mar-18 23:26

but i am not using any variables with reference to heap, so i thought it shouldn't be needed for ex:- new, *variable etc.

Re: In type conversion of 2 different classes can i convert both ways ?

CPallini25-Mar-18 10:50

CPallini

25-Mar-18 10:50

C++

#include <iostream>
#include <cmath>
using namespace std;

class Polar;

//====================================================
class Rectangle
{
  float length, bredth;
public:
  Rectangle(){}
  Rectangle(float ,float);

  operator Polar();

  int getLength(){return(length);}
  int getBredth(){return(bredth);}
  void putLength(float x){length = x;}
  void putBredth(float y){bredth = y;}

  void putData(){cout<<"Length : "<<length<<"\tBreadth : "<<bredth<<endl;}
  ~Rectangle(){}
};


//====================================================

class Polar{
  float angle, rad;
public:

  Polar(){}
  Polar(float , float);

  operator Rectangle();

  int getAngle(){return(angle);}
  int getRadius(){return(rad);}
  void putAngle(float a){angle=a;}
  void putRadius(float r){rad = r;}

  void putData(){cout<<"Angle : "<<angle<<"\tRadius : "<<rad<<endl;}
  ~Polar(){}
};

//===========================================================
Rectangle::Rectangle(float x, float y)
{
  length = x;
  bredth = y;
}

Rectangle::operator Polar()
{
  Polar temp;
  float a = atan2(bredth, length), r = sqrt(length*length + bredth*bredth);
  temp.putAngle(a);
  temp.putRadius(r);

  return(temp);
}

//===========================================================
Polar::Polar(float x, float y)
{
  angle = x;
  rad = y;
}
Polar::operator Rectangle()
{
  Rectangle temp;
  float l = rad*cos(angle), b =rad*sin(angle);
  temp.putLength(l);
  temp.putBredth(b);
  return(temp);
}

//===========================================================
int main()
{
  float radians = 120 * M_PI / 180; // convert to radians

  // roundtrip polar->rectangle->polar
  Polar p1(radians, 25);
  p1.putData();

  Rectangle r1 = p1;
  r1.putData();

  Polar p2 = r1;
  p2.putData();

  return 0;
}

Re: In type conversion of 2 different classes can i convert both ways ?

Tarun Jha25-Mar-18 23:25

Tarun Jha

25-Mar-18 23:25

thank you

Re: In type conversion of 2 different classes can i convert both ways ?

CPallini26-Mar-18 0:18

CPallini

26-Mar-18 0:18

You are welcome.

Adding two matrix objects and assigning the result to third object using overloaded operators ?

Tarun Jha24-Mar-18 8:19

Tarun Jha

24-Mar-18 8:19

here ih the code below i have tried to add two matrix objects and assign it to the third object but it doesn't give expected result and the rather some garbage value.

include <iostream>
using namespace std;

class matrix{
    int **p;    //pointer to Matrix
    int d1, d2;

public:
    //matrix(){}                  //default constructor
    matrix(int x=20, int y=20);       //parameterised

    void get_element();
    void put_element();
    int checkAdd(matrix &);                                     //to check the feasibility of the operation

    //Rule of three
    matrix operator + (matrix &);
    void operator = (matrix );

    ~matrix(){
        for (int i = 0; i < d1; i++)
            delete p[i];
        delete p;
    }
};

matrix temp;                    //global matrix temp

//=============================================

void matrix::operator = (matrix obj){

    this->d1 = obj.d1;
    this->d2 = obj.d2;
    for(int i=0; i<d1; i++)
        for(int j=0; j<d2; j++)
            this->p[i][j] = obj.p[i][j];

    //put_element();
    //return(*this);
}

matrix matrix::operator+(matrix &obj){
    if(checkAdd(obj)==0)
    {
        //matrix temp;
        for(int i=0; i<obj.d1; i++)
            for(int j=0; j<obj.d2; j++)
                temp.p[i][j] = this->p[i][j] + obj.p[i][j];


        //temp.put_element();
        return(temp) ;
    }

    cout<<"coloumn and rows of both matrix are not equal !"<<endl;
    return (*this);
}

matrix ::matrix(int x, int y)
{
    d1 = x;
    d2 = y;
    p = new int *[d1];                  //creates an array of pointers

    for(int i=0; i<d1; i++)
        p[i] = new int [d2];            //creates row for each row
}

int matrix::checkAdd(matrix &obj){
    if((this->d1 == obj.d1) && (this->d2 == obj.d2))
        return 0;
    else return 1;
}

void matrix::get_element(){
    for(int i=0; i<d1; i++)
        for(int j=0; j<d2; j++){
            cout<<"m["<<i<<"]["<<j<<"] = ";
            cin>>p[i][j];
        }
    //cout<<"Following is your Matrix : \n\n";
    //put_element();
}

void matrix::put_element(){
    cout<<"\n\n";
    for(int i=0; i<d1; i++){
        for(int j=0; j<d2; j++){
            cout<<p[i][j]<<"\t";
        }
        cout<<endl;
    }

}

//=============================================

int main(){
    matrix A(3, 2), B(3, 2), C;                     //matrix objects a constucted

    cout<<"Enter matrix elements row by row \n\n";
    A.get_element();
    A.put_element();

    B.get_element();
    B.put_element();

    C = A + B;
    cout<<"\n\n";
    C.put_element();

    return 0;
}

thank you

Re: Adding two matrix objects and assigning the result to third object using overloaded operators ?

Victor Nijegorodov24-Mar-18 8:35

Victor Nijegorodov

24-Mar-18 8:35

What result did you expect and what did you get?

Re: Adding two matrix objects and assigning the result to third object using overloaded operators ?

Tarun Jha24-Mar-18 10:06

Tarun Jha

24-Mar-18 10:06

Quote:
So basically i want to add 2 matrix object and assign the result to the third non-initiated matrix object, without changing the values of first 2 original matrix.

this is the result i am getting.

Quote:
Enter matrix elements row by row

m[0][0] = 1
m[0][1] = 2
m[1][0] = 3
m[1][1] = 4
m[2][0] = 5
m[2][1] = 6

1 2
3 4
5 6
m[0][0] = 7
m[0][1] = 8
m[1][0] = 9
m[1][1] = 10
m[2][0] = 11
m[2][1] = 12

7 8
9 10
11 12

8 10 47120720 0 0 0 0 0 0 0 50331651 41272 47127056 0 47120720 0 0 2396 136 0
12 14 47120720 0 1550742898 1148219457 1549890657 1633906508 1700027500
1426092141 1146242387 1229016399 1162100046 1330924371 1429482832 844253517 1398079566 1329877573 1313423693 1095717471
16 18 47120720 0 1162690894 1918981181 1426091637 1347568979 1229344594
1128088908 1934974010 1551069797 1970430292 1398145134 1127232561 1414417743 1397509967 1547322173 1735357008 544039282
47120720 0 47127424 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0
47120720 0 47127424 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0
47120720 0 47127424 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0
47120720 0 47127424 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0
47120720 0 47127424 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0
47120720 0 47127424 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0
47120720 0 47127424 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0
47120720 0 47127424 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0
47120720 0 47127424 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0
47120720 0 47127424 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0
47120720 0 47127424 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0
47120720 0 47127424 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0
47120720 0 47127424 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0
47120720 0 47127424 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

i want a result:
8 10
12 14
16 18

Re: Adding two matrix objects and assigning the result to third object using overloaded operators ?

Richard MacCutchan24-Mar-18 21:13

Richard MacCutchan

24-Mar-18 21:13

You already posted this exact question in the ATL/STL forum, and I gave you a suggestion. If you google for "operator overload C++" you will find plenty of samples.

Re: Adding two matrix objects and assigning the result to third object using overloaded operators ?

Tarun Jha25-Mar-18 5:08

Tarun Jha

25-Mar-18 5:08

i was told that these type of questions are to be posted here.
and i did the program..

#include <iostream>
using namespace std;

class matrix{
    int **p;                        //pointer to Matrix
    int d1, d2;
    int id=0;                        //identif which operator is used

public:
    //matrix(){}                      //default constructor
    matrix(int x=20, int y=20);       //parameterised

    //to pass private data-members to other objects
    int getD_1(){return (d1);}
    int getD_2(){return (d2);}
    //int getId(const int x){id =x ;}
    //void showId(){cout<<"ID :\t"<<id<<endl;}

    //for taking input nad printing output
    void get_element();
    void put_element();

    //to check feasibility
    friend int checkAdd(matrix &, matrix &);                                     //to check the feasibility of the operation
    friend int checkMul(matrix &, matrix &);

    //Rule of three
    friend matrix operator+(matrix &, matrix &);
    friend matrix operator*(matrix &, matrix &);
    //friend matrix operator-(matrix &, matrix &);

    void operator = (matrix );

    ~matrix(){
        for (int i = 0; i < d1; i++)
            delete p[i];
        delete p;
    }
};

//=================================================================================================================

//defining overloaded assingment operator
void matrix::operator = (matrix obj){
    this->d1 = obj.d1;
    this->d2 = obj.d2;
    this->id = obj.id;
    for(int i=0; i<d1; i++)
        for(int j=0; j<d2; j++)
            this->p[i][j] = obj.p[i][j];
}

//defining overloaded addition operator
matrix operator+(matrix &a, matrix &b){
    matrix temp;
    temp.d1 = a.d1;
    temp.d2 = a.d2;
    temp.id = 1;

    if(checkAdd(a, b) == 0){                //checks the viability if it's feasible or not.
        for(int i=0; i<temp.d1; i++)
            for(int j=0; j<temp.d2; j++)
                temp.p[i][j] = a.p[i][j] + b.p[i][j] ;

        return(temp);
    }
    else {
        cout<<"Matrix addition not possible as rows and coloumn's of given 2 matrices are not equal"<<endl;
        return(a);
    }
}

//defining overloaded Multiplication operator
matrix operator*(matrix &a, matrix &b){
    matrix temp;
    temp.d1 = a.d1;
    temp.d2 = b.d2;
    temp.id = 2;

    if(checkMul(a, b) == 0){              //checks the viability if it's feasible or not.
        int sum=0;
        for (int c = 0; c < a.d1; c++) {
            for (int d = 0; d < b.d2; d++) {
                for (int k = 0; k <b.d1; k++) {
                    sum = sum + a.p[c][k]*b.p[k][d];
                }
                temp.p[c][d] = sum;
                sum = 0;
            }
        }
        return (temp);
    }
    else {
        cout<<"Matrix Multiplication is not possible as coloumn of 1st Matrix != Row of 2nd Matrix "<<endl;
        return(a);
    }
}

//defining parameterised constucter
matrix ::matrix(int x, int y){
    d1 = x;
    d2 = y;
    p = new int *[d1];                  //creates an array of pointers

    for(int i=0; i<d1; i++)
        p[i] = new int [d2];            //creates row for each row
}

//defining checkAdd f^n to check if given 2 matrices obj can be added or not
int checkAdd(matrix &a, matrix &b){
    if(a.getD_1() == b.getD_1() && a.getD_2() == b.getD_2())
        return 0;
    else return 1;
}

//defining checkMul f^n to check if 2 matrices obj can be Multiplied.
int checkMul(matrix &a, matrix &b){
    if(a.getD_2() == b.getD_1())
        return 0;
    else return 1;
}

//gets the element from the user.
void matrix::get_element(){
    for(int i=0; i<d1; i++)
        for(int j=0; j<d2; j++){
            cout<<"m["<<(i+1)<<"]["<<(j+1)<<"] = ";
            cin>>p[i][j];
        }
}


//displays the current matrix
void matrix::put_element(){
    if(id == 1) cout<<"\nThe result of addition of 2 matrices is :"<<endl;
    else if(id==2)  cout<<"\nThe result of product of 2 matrices is:"<<endl;

    for(int i=0; i<d1; i++){
        for(int j=0; j<d2; j++){
            cout<<p[i][j]<<"\t";
        }
        cout<<endl;
    }
    cout<<"\n";
}

//=====================================================================================================================

int main(){
    matrix A(3, 2), B(2, 4), C;                     //matrix objects a constucted

    cout<<"Enter matrix elements row by row \n\n";
    A.get_element();
    A.put_element();

    B.get_element();
    B.put_element();

    C = A * B;
    C.showId();
    C.put_element();

    C = A + B;
    C.showId();
    C.put_element();

    return 0;
}

and thank you for the advice

Re: Adding two matrix objects and assigning the result to third object using overloaded operators ?

Richard MacCutchan25-Mar-18 5:32

Richard MacCutchan

25-Mar-18 5:32

Forget about matrix calculations. You need to go and study operator overloading and get that working on a basic class first. Once you understand how to do it correctly so you get no errors, then you can move on to the matrix functions.

Re: Adding two matrix objects and assigning the result to third object using overloaded operators ?

Tarun Jha25-Mar-18 6:37

Tarun Jha

25-Mar-18 6:37

is there anything wrong with the program i have posted before ?

Re: Adding two matrix objects and assigning the result to third object using overloaded operators ?

Richard MacCutchan25-Mar-18 7:16

Richard MacCutchan

25-Mar-18 7:16

I don't know; you have to do your own testing. But I find one of the easiest way to debug issues is to create a small test program that just focused on the area that appears not to work. So a small program with a very basic class, that just tests the operator+ overload, would be much easier to understand.

Re: Adding two matrix objects and assigning the result to third object using overloaded operators ?

Tarun Jha25-Mar-18 8:06

Tarun Jha

25-Mar-18 8:06

ok...
thanks for the advice

Linux C++ implementation of ioctl SOLVED

Vaclav_23-Mar-18 8:50

Vaclav_

23-Mar-18 8:50

Broken 5V supply to LCD!

I should have know better since the LCD did not react to power-up - reset and show 5/7 characters - but the power LED was on on the i2c module itself !

Learned few Linux tools on how to troubleshoot i2c, so no big loss.

Now I can proceed with implementing ioctl.

I would like to discuss implementation of Linux ioctl function.
There are few reasons I am posting my question here and not on Linux forum. However, if you feel it is inappropriate you have two options - ignore my request ( preferred to sermon ) or flag it to be removed.
I am basically aware WHAT ioctl does, and have been thru few documents ( Linux Device drivers, CodeProject driver tutorials etc.)
I am trying to implement ioctl "driver" to output (Raspberry Pi) to I2C LCD.
The problems I cannot get straight is how to configure I2C speed (using C++, not Linux commands ) and implement ACK read from slave I2C device.

I understand I2C protocol, but all of the sample codes I have read so far are "limited" to open ioctl file and "write " to I2C slave address.

I like to see or make my own code to implement I2C "start /stop " and read ACK/ NACK from slave.

Any to the point help will as always be appreciated.

Thanks

-- modified 7-Apr-18 15:35pm.

Re: Linux C++ implementation of ioctl

Jochen Arndt24-Mar-18 23:39

Jochen Arndt

24-Mar-18 23:39

You have two options:

Use the existing I2C support on the Raspberry Pi with the I2C GPIO pins
Implement your own I2C support on general use GPIO pins

The I2C speed is defined by the master by generating the clock. A device might use "clock stretching" to slow down (Clock, Stretching, Arbitration - I2C Bus[^]).

So implementations are not time critical. They can use a predefined (max.) clock and have not to care about being slowed down by other tasks on the system.

I have implemented I2C in assembler on micro controllers. Implementing it in C is not a problem but be aware that it might be a frustrating tasks because it can't be simply debugged. I highly recommend to use a logic analyser or at least a storage oscilloscope to check the signals on the bus.

Quote:
all of the sample codes I have read so far are "limited" to open ioctl file and "write " to I2C slave address.

With device drivers use i2c_transfer(). Have a look at the source of any Linux driver for a I2C device like rtc-pcf8563.c
[^].

When using ioctl() on the I2C device just use read():

C++

int fd = open("/dev/i2c-1", O_RDWR);
// Use I2C_SLAVE_FORCE if the device is in use (e.g. when a driver is installed)
ioctl(fd, I2C_SLAVE, addr);
read(fd, buffer, length);

Or use ioctl(I2C_RDWR) or ioctl(I2C_SMBUS) for reading which both does not require the I2C_SLAVE[_FORCE]. In any case check first if the used ioctl() call is supported.

Re: Linux C++ implementation of ioctl

leon de boer26-Mar-18 16:53

leon de boer

26-Mar-18 16:53

The speed is configured outside the program or on the library you are using on the Pi

Edit /boot/config.txt if it doesn't exist create it
add the line or edit the line with your baudrate

dtoverlay=i2c-speed,i2c1_baudrate=400000

i2c1 may also be disabled by default I can't remember if so you need

dtparam=i2c_vc=on

In the userland repository there is also a direct header and library
bcm2835: C library for Broadcom BCM 2835 as used in Raspberry Pi[^]
and a forum
bcm2835 : Google Groups[^]
Standard use

#include <bcm2835.h>  

int main(int argc, char **argv)  
{  
    char buf[1];

    if (!bcm2835_init())return 1;  
    bcm2835_i2c_begin();                //Start I2C operations.
    bcm2835_i2c_setSlaveAddress(0x20);  //I2C address
    bcm2835_i2c_set_baudrate(10000);    //1M baudrate

    while(1)  
    {   
        buf[0] = 0xEF;    //LED ON
        bcm2835_i2c_write(buf,1);
        bcm2835_delay(500);
        buf[0] = 0xFF;     //LED OFF
        bcm2835_i2c_write(buf,1);
        bcm2835_delay(500); 
    }    
    bcm2835_i2c_end();  
    bcm2835_close();  
    return 0;  
}

In vino veritas

modified 26-Mar-18 23:06pm.

Re: Linux C++ implementation of ioctl

Vaclav_4-Apr-18 5:08

Vaclav_

4-Apr-18 5:08

Thanks for replies.
I am currently "on the road" and will reply when I get home.

In short
I did try "just GPIO" , it worked to the point, but it got too convoluted.
Hence I am after using ioctl.
I'll take a look at the bcm2835.h, hopefully I can use it for bcm2837 chip used in RPi 3B.
I think my main concern is still to be able to "see" the code following / running the I2C spec - mainly changing hardware pins directions and reading START /STOP and ACK/NACK back. Of course I am verifying the actual LCD and sometime even using scope.

Thanks for all your comments, appreciate that.

-- modified 5-Apr-18 12:42pm.

VS2017: Cannot debug Dll's

indrekm23-Mar-18 1:13

indrekm

23-Mar-18 1:13

I converted VS2008 solution (sln) to VS2017. Visual Studio is up-to-date - version 15.6.4.
My solution contains tens of C++ projects - executables, dll's and lib's. Now I have a weird issue - dll's cannot be debugged in Debug configuration.
In Release configuration it works - breakpoints can be used.
But in Debug configuration all the breakpoints are disabled and there is a hint - "The breakpoint will not currently be hit. No symbols have been loaded for this document."
When I try to load symbols (pdb) manually I'm getting a message "A matching symbol file was not found in this folder."
In DLL project settings (vcxproj file) GenerateDebugInformation is true.
I'm a bit confused why generated pdb file cannot be used by Visual Studio.

Re: VS2017: Cannot debug Dll's

Richard MacCutchan23-Mar-18 1:21

Richard MacCutchan

23-Mar-18 1:21

The chances are that your DLL needs to be rebuilt.

Re: VS2017: Cannot debug Dll's

Jochen Arndt23-Mar-18 1:25

Jochen Arndt

23-Mar-18 1:25

Have you converted the solution in-place (with existing output files from old VS 2008 builds)?

Then try to delete all output files (especially old ones which can be identified by the time stamp).

Re: VS2017: Cannot debug Dll's

Randor 23-Mar-18 8:09

Randor

23-Mar-18 8:09

Hi,

The 'Debug' build is just a text label... in other words it sounds like your DLL is being compiled with the symbols stripped out. There is nothing stopping someone from configuring 'Debug' exactly as a 'Release' build.

Try using DUMPBIN /SYMBOLS[^] on your DLL to check if the symbol tables exists.

Don't forget that you can PIPE the output into a text file:

DUMPBIN /SYMBOLS YourDLL.DLL > symbols.txt

Best Wishes,
-David Delaune

Select and Pick lines in OpenGL

Gopi Nath22-Mar-18 19:40

Gopi Nath

22-Mar-18 19:40

Hello Everybody,

I am using VC++ and OpenGL for one application.

I am new to OpenGL. I can able to load different objects and also able to create vertices. After that, I was able to select and pick those using glReadPixels and gluUnProject.

Now, I need to select and pick lines (edges of the triangles). I searched in google and not able to find a proper result.

Can anybody give me ideas how to implement this?

Thanks in Advance.

Regards,
Gopi.

Use Ctrl+Left/Right to switch messages, Ctrl+Up/Down to switch threads, Ctrl+Shift+Left/Right to switch pages.