remove() not working ?

I've included <algorithm>...

Code:

---

void operator-(double number){
remove(doublevector.begin(), doublevector.end(), number);
}

---

normally it should remove all elements that are equal to the number from the vector ... I don^t get any compile errors but it seemply doesn't remove them.

http://cboard.cprogramming.com/showthread.php?t=78649
floats and doubles are approximations.
Getting actual equality is much harder than you might imagine.

yeah but replace()

with double works...

Post a complete compilable example.

Ok obviously remove() works but my problem is that the vector seems to occupy the same memory as before after remove ..
for some weird reason doublevector.end() doesn't change after remove()...

Code:

---

#include <iostream>
#include <algorithm>
#include <vector>


int main(){

  vector <double> doublevector;
  vector <double> :: iterator theiterator;
 


  doublevector.push_back(5.67);
  doublevector.push_back(5.67);
 
  doublevector.push_back(-0.45);
  doublevector.push_back(-0.45);

  doublevector.push_back(15);

  //Prints the vector
  for  (theiterator=doublevector.begin(); theiterator != doublevector.end();theiterator++)
    cout<<"\t"<<*theiterator;
  cout<<endl;

  //should remove all numbers that equal 5.67 from vector
  remove(doublevector.begin(),doublevector.end(), 5.67);

  //replaces all numbers that equal -0.45 with -0.55
  replace(doublevector.begin(),doublevector.end(),-0.45,-0.55);

  //Prints the vector
  for  (theiterator=doublevector.begin(); theiterator != doublevector.end();theiterator++)
    cout<<"\t"<<*theiterator;
  cout<<endl;


  return 0;
}

---

You have to use the erase() member function to actually erase the elements removed.

yeah but then remove() is useless ?

Quote:

---

yeah but then remove() is useless ?

---

Of course not. Read SGI's Standard Template Library Programmer's Guide. Still as Salem mentioned, simple comparison with floating point numbers can be problematic.

why of course not... if i remove an element and i can access it later what is the use of it... ?

and you get kinda lost on the site you gave me..

maybe someone else has used remove() efficiently ?

Quote:

---

if i remove an element and i can access it later what is the use of it

---

That's where the erase() member function comes into play. remove() works on a range denoted by an iterator pair, hence the container itself is left unchanged (only the container elements are swapped around).

You could consider something like:

Code:

---

#include <algorithm>
#include <functional>
#include <cmath>
#include <vector>
#include <iostream>

struct IsApprox : std::binary_function<double, double, bool> {
        bool operator()(double lhs, double rhs) const {
                return fabs(lhs - rhs) < 0.001;
        }
};

int main() {
        std::vector<double> doublevector;
        doublevector.push_back(5.67);
        doublevector.push_back(5.67);
        doublevector.push_back(-0.45);
        doublevector.push_back(-0.45);
        doublevector.push_back(15.0);

        for (std::vector<double>::iterator i = doublevector.begin();
                i != doublevector.end(); ++i) {
                        std::cout << *i << " ";
        }
        std::cout << "\n";

        doublevector.erase(
                std::remove_if(
                        doublevector.begin(),
                        doublevector.end(),
                        std::bind2nd(IsApprox(), 5.67)),
                doublevector.end());

        for (std::vector<double>::iterator i = doublevector.begin();
                i != doublevector.end(); ++i) {
                        std::cout << *i << " ";
        }
        std::cout << "\n";
}

---

Quote:

---

why of course not... if i remove an element and i can access it later what is the use of it... ?

---

Speed. Erasing elements isn't necessary.

Code:

---

vector <double> :: iterator theNewEnd;
theNewEnd = remove(doublevector.begin(),doublevector.end(), 5.67);

//Prints the vector
  for  (theiterator=doublevector.begin(); theiterator != theNewEnd; ++theiterator)
    cout<<"\t"<<*theiterator;
  cout<<endl;

---

wow i am stunned i didn^t know you could do that in c++ ... !

Code:

---

doublevector.erase(
                std::remove_if(
                        doublevector.begin(),
                        doublevector.end(),
                        std::bind2nd(IsApprox(), 5.67)),
                doublevector.end());

---

so if erase() didn't find anything to erase how do i check that ?

let's say i have

Code:

---

while (doublevector.erase(remove(doublevector.begin(), doublevector.end(),9)) != ??? );

---

with what do i replace the ???

thanks a ton guys.. this has been really helpful so far :)

Quote:

---

so if erase() didn't find anything to erase how do i check that ?

---

The size of the vector would be unchanged. Still, erase() doesnt actually 'find' anything. The comparison is done in remove() or remove_if(), which returns an iterator pointing to one past the end of the non-removed portion of the sequence.

This is why you may not even need to use erase(), you could just use this new endpoint iterator and ignore the elements in the removed portion of the sequence, as per 7stud's example.

Quote:

---

with what do i replace the ???

---

You really have to read the STL Guide on remove(), remove_if() and vector.

you were right ! that STL Guide is very useful, i will consult it from time to time. I also find out that i can do

Code:

---

doublevector.erase(newend, doublevector.end());

---

which is actually combined with what 7stud said and is exactly what i needed ! :)