and why doesnt the standard vector have such conversion available?

<bluekite2000@gmail.com> wrote in message
news:1125526444.578739.319490@o13g2000cwo.googlegr oups.com...[color=blue]
> and why doesnt the standard vector have such conversion available?[/color]

You can use the vector constructor that takes two iterators:

#include <vector>
#include <assert.h>

using namespace std;

int main()
{
vector<float> a;
a.push_back(1.2f);

vector<double> b(a.begin(), a.end());

assert(!b.empty());
cout << b.front() << '\n';
}

Ali

> Re: how do u convert a vector<float> A to a vector<double> B

std::vector<float> A;
/* etc */
std::vector<double> B(A.begin(), A.end())

<bluekite2000@gmail.com> wrote in message
news:1125526444.578739.319490@o13g2000cwo.googlegr oups.com...[color=blue]
> and why doesnt the standard vector have such conversion available?[/color]

It does. See above.

-Mike

What if b is already constructed?
Another question
Say you have
vector<float> a(3);
//init a
vector<std::complex<float> > c(3);
//init c
Now I want something like
real(c)=a;

Currently I have
for(int i=0;i<c.size();i++)
c[i].real()=a[i];

Which I dont really like.

<bluekite2000@gmail.com> wrote in message
news:1125527793.917748.118130@o13g2000cwo.googlegr oups.com...[color=blue]
> What if b is already constructed?[/color]

Then it's assignment:

b = vector<double>(a.begin(), a.end());
[color=blue]
> Another question
> Say you have
> vector<float> a(3);
> //init a
> vector<std::complex<float> > c(3);
> //init c
> Now I want something like
> real(c)=a;
>
> Currently I have
> for(int i=0;i<c.size();i++)
> c[i].real()=a[i];
>
> Which I dont really like.[/color]

There is nothing wrong with it. Just wrap it in a function and you are done:

typedef vector<float> Reals;
typedef vector<complex<float> > Complexes;

void set_reals(Complexes & complexes, Reals const & reals)
{
// some checks
if (complexes.size() != reals.size())
{
throw SomeError;
}

/* the logic here */
}

Very neat :)

More cool (and possibly more obscure) things can probably be done, but I
really don't think that it's worth it.

Ali

"Ali Çehreli" <acehreli@yahoo.com> wrote in message
news:df5dem$88c$1@domitilla.aioe.org...[color=blue]
> <bluekite2000@gmail.com> wrote in message
> news:1125527793.917748.118130@o13g2000cwo.googlegr oups.com...[color=green]
>> What if b is already constructed?[/color]
>
> Then it's assignment:
>
> b = vector<double>(a.begin(), a.end());[/color]

Actually, a probably better expression (for clarity,
maintainability, and performance) is:

b.assign( a.begin(), a.end() );


hth, Ivan
--
http://ivan.vecerina.com/contact/?subject=NG_POST <- email contact form
Brainbench MVP for C++ <> http://www.brainbench.com

On 2005-08-31 18:14:04 -0400, bluekite2000@gmail.com said:
[color=blue]
> and why doesnt the standard vector have such conversion available?[/color]

#include <vector>
#include <algorithm> //for std::copy

....

std::vector<float> A = ...;

//Construct B with the contents of A
std::vector<double> B(A.begin(), A.end());

//Or if B is already constructed:
B.clear();
B.insert(B.begin(), A.begin(), A.end());

//Or:
B.resize(A.size());
std::copy(A.begin(), A.end(), B.begin());


--
Clark S. Cox, III
clarkcox3@gmail.com

On 2005-09-01 00:09:55 -0400, "Ivan Vecerina"
<INVALID_use_webform_instead@vecerina.com> said:
[color=blue]
> "Ali Çehreli" <acehreli@yahoo.com> wrote in message
> news:df5dem$88c$1@domitilla.aioe.org...[color=green]
>> <bluekite2000@gmail.com> wrote in message
>> news:1125527793.917748.118130@o13g2000cwo.googlegr oups.com...[color=darkred]
>>> What if b is already constructed?[/color]
>>
>> Then it's assignment:
>>
>> b = vector<double>(a.begin(), a.end());[/color]
>
> Actually, a probably better expression (for clarity,
> maintainability, and performance) is:
>
> b.assign( a.begin(), a.end() );[/color]

There is no "assign" function on std::vector.


--
Clark S. Cox, III
clarkcox3@gmail.com

Clark S. Cox III wrote:
<snip>[color=blue][color=green]
>> Actually, a probably better expression (for clarity,
>> maintainability, and performance) is:
>>
>> b.assign( a.begin(), a.end() );[/color]
>
> There is no "assign" function on std::vector.
>
>[/color]

23.2.4.1 vectors constructors, copy, and assignment
<snip>
template <class InputIterator>
void assign(InputIterator first, InputIterator last);

Effects:
erase(begin(), end());
insert(begin(), first, last);

template <class Size, class U> void assign(Size n, const U& u = U());

Effects:
erase(begin(), end());
insert(begin(), n, t);
<snip>

Maybe your compiler lacks member template support?

Marc

Clark S. Cox III wrote:
<snip>[color=blue]
> //Or:
> B.resize(A.size());
> std::copy(A.begin(), A.end(), B.begin());[/color]

b.clear();
b.reserve( a.size() );
std::copy( a.begin(), a.end(), std::back_inserter( b ) );

Marc

On 2005-09-01 08:52:49 -0400, Marc Mutz
<marc@klaralvdalens-datakonsult.se> said:
[color=blue]
> Clark S. Cox III wrote:
> <snip>[color=green][color=darkred]
>>> Actually, a probably better expression (for clarity,
>>> maintainability, and performance) is:
>>>
>>> b.assign( a.begin(), a.end() );[/color]
>>
>> There is no "assign" function on std::vector.
>>
>>[/color]
>
> 23.2.4.1 vectors constructors, copy, and assignment
> <snip>
> template <class InputIterator>
> void assign(InputIterator first, InputIterator last);
>
> Effects:
> erase(begin(), end());
> insert(begin(), first, last);
>
> template <class Size, class U> void assign(Size n, const U& u = U());
>
> Effects:
> erase(begin(), end());
> insert(begin(), n, t);
> <snip>
>
> Maybe your compiler lacks member template support?[/color]

Well, I'll be. I can't believe I missed that.

I think I was looking at the first listing in 23.2.4 without
remembering the "Descriptions are provided here only for operations on
vector that are not described in one of these tables or for operations
where there is additional semantic information." part.


--
Clark S. Cox, III
clarkcox3@gmail.com