PureBasic Forums - English

I need this Operator!

Example:

But you just showed yourself how to use the pow() command, why do you need ** ?

For jokes please go to offtopic area

Three Reasons that support this request:
1) It's much more comfortable to use.
2) It looks more cool.
3) This would also work correctly with Longs (however, Pow accepts Float only).

For the case, that this will be inplemented, please don't use **. In my opinion ^ is better.

Well, I support zis idea too.

"**" is no good idea

That's why I prefer ^

about ^, yes as many other languages.

but the rules about types conversion should be clearly described before.

type? = var_long ^ var_long
type? = var_long ^ var_float
type? = var_long ^ var_double
...

type? = var_float ^ var_long
type? = var_float ^ var_float
type? = var_float ^ var_double
...

and so on...

AND51 wrote:That's why I prefer ^

That's why i told '**' was a joke.

And it was already requested:
http://www.purebasic.fr/english/viewtopic.php?t=7730

So then this thread seems a double joke.

Flype wrote:about ^, yes as many other languages.

but the rules about types conversion should be clearly described before.

type? = var_long ^ var_long
type? = var_long ^ var_float
type? = var_long ^ var_double
...

type? = var_float ^ var_long
type? = var_float ^ var_float
type? = var_float ^ var_double
...

and so on...

long ^ long -> long
long ^ float -> float
long ^ double -> double
float ^ long -> float
float ^ float -> float
float ^ double -> double
double ^ * -> double
quad ^ long -> quad
long ^ quad -> quad
quad ^ double -> double
double ^ quad -> double
quad ^ float -> float
float ^ quad -> float

Using ^ would be ok.

Trond wrote: long ^ long -> long
long ^ float -> float
long ^ double -> double
float ^ long -> float
float ^ float -> float
float ^ double -> double
double ^ * -> double
quad ^ long -> quad
long ^ quad -> quad
quad ^ double -> double =quad
double ^ quad -> double =quad
quad ^ float -> float =quad
float ^ quad -> float =quad

I would of thought that if either argument was a quad then the answer should be quad, same with doubles etc

Derek wrote:Using ^ would be ok.

Trond wrote: long ^ long -> long
long ^ float -> float
long ^ double -> double
float ^ long -> float
float ^ float -> float
float ^ double -> double
double ^ * -> double
quad ^ long -> quad
long ^ quad -> quad
quad ^ double -> double =quad
double ^ quad -> double =quad
quad ^ float -> float =quad
float ^ quad -> float =quad

I would of thought that if either argument was a quad then the answer should be quad, same with doubles etc

no. Quads do not support decimals while doubles do, hence result has to be a double.

I just assumed that if you were using quads then you would be going for a higher number rather than a more accurate number including decimals but I guess it could be argued both ways.

Derek wrote:I just assumed that if you were using quads then you would be going for a higher number rather than a more accurate number including decimals but I guess it could be argued both ways.

I second the request for ^ as "to the power of"

and I also request an optimized implementation.

if I use only integers and get an integer result, the returned type should be integer, too,
not forced float like the Pow()-Function returns.

additionally, it would be really great if it would be translated into a bitshift (in the asm-code) if the base is a power of two.