It is critical to be able to instantiate a class with different template argumetns to be able to do a cast function (i.e. a double vector to a single vector) or in this case to be able to make a smaller vector from a bigger one. Doing this with static functions outside the class is painful and the syntax is clunky (no operators, no implicit members that actually call get/set functions). test program that fails:

struct vec(int size) {
float v[size];
vec!(1) first() {vec!(1) t; t.v[0]=v[0];return t;}
}
int main () {
vec!(2) v;
vec!(1) g =v.first();

}

<hellcatv@hotmail.com> wrote in message news:c7ctkt$1all$1@digitaldaemon.com...
> It is critical to be able to instantiate a class with different template argumetns to be able to do a cast function (i.e. a double vector to a
single
> vector) or in this case to be able to make a smaller vector from a bigger
one.
> Doing this with static functions outside the class is painful and the
syntax is
> clunky (no operators, no implicit members that actually call get/set
functions).
> test program that fails:
>
> struct vec(int size) {
> float v[size];
> vec!(1) first() {vec!(1) t; t.v[0]=v[0];return t;}
> }
> int main () {
> vec!(2) v;
> vec!(1) g =v.first();
>
> }
>


There is no bug here, you missed something that isn't documented
(at least i didn't find it)
It should be:

struct vec(int size) {
float v[size];
.vec!(1) first() {.vec!(1) t; t.v[0]=v[0];return t;}
}
int main () {
vec!(2) v;
vec!(1) g =v.first();
return 1;
}

template vec belongs to global scope so . gives acces to global scope.

D is my new favorite language :-) thanks

In article <c7d1kc$1gr1$1@digitaldaemon.com>, Ivan Senji says...
>
><hellcatv@hotmail.com> wrote in message news:c7ctkt$1all$1@digitaldaemon.com...
>> It is critical to be able to instantiate a class with different template argumetns to be able to do a cast function (i.e. a double vector to a
>single
>> vector) or in this case to be able to make a smaller vector from a bigger
>one.
>> Doing this with static functions outside the class is painful and the
>syntax is
>> clunky (no operators, no implicit members that actually call get/set
>functions).
>> test program that fails:
>>
>> struct vec(int size) {
>> float v[size];
>> vec!(1) first() {vec!(1) t; t.v[0]=v[0];return t;}
>> }
>> int main () {
>> vec!(2) v;
>> vec!(1) g =v.first();
>>
>> }
>>
>
>
>There is no bug here, you missed something that isn't documented
>(at least i didn't find it)
>It should be:
>
>struct vec(int size) {
>float v[size];
>.vec!(1) first() {.vec!(1) t; t.v[0]=v[0];return t;}
>}
>int main () {
>vec!(2) v;
>vec!(1) g =v.first();
>return 1;
>}
>
>template vec belongs to global scope so . gives acces to global scope.
>
>