Hi guys,

i'm pretty new to D (wrote a simple Hello World), but ... shouldn't the out
assert be more like (x * x) == result?

long square_root(long x)
in
{
assert(x >= 0);
}
out (result)
{
assert((result * result) == x);
}
body
{
return math.sqrt(x);
}

Serge Ratke wrote:
> Hi guys,
> 
> i'm pretty new to D (wrote a simple Hello World), but ... shouldn't the out
> assert be more like (x * x) == result?
> 
> long square_root(long x)
> in
> {
> assert(x >= 0);
> }
> out (result)
> {
> assert((result * result) == x);
> }
> body
> {
> return math.sqrt(x);
> }
> 
> 

Yup, else it does something completely insane.

Regards,
Sjoerd

Sjoerd van Leent schrieb am 2004-11-15:
>> i'm pretty new to D (wrote a simple Hello World), but ... shouldn't the out
>> assert be more like (x * x) == result?
>> 
>> long square_root(long x)
>> in
>> {
>> assert(x >= 0);
>> }
>> out (result)
>> {
>> assert((result * result) == x);
>> }
>> body
>> {
>> return math.sqrt(x);
>> }
>
> Yup, else it does something completely insane.

so you are saying:

1) result := sqrt(x)
2) result == x * x

mhhhh... the only non-complex solution would be 0?

Thomas

On Mon, 15 Nov 2004 21:43:41 +0000 (UTC), Serge Ratke <Serge_member@pathlink.com> wrote:
> Hi guys,
>
> i'm pretty new to D (wrote a simple Hello World), but ... shouldn't the out
> assert be more like (x * x) == result?
>
> long square_root(long x)
> in
> {
> assert(x >= 0);
> }
> out (result)
> {
> assert((result * result) == x);
> }
> body
> {
> return math.sqrt(x);
> }

No, I think it's correct, using an example:

square_root(25);

if x == 25;
then result == 5;
therefore 5*5 == 25
or result*result == x

Regan

-- 
Using M2, Opera's revolutionary e-mail client: http://www.opera.com/m2/

Serge Ratke wrote:

> Hi guys,
> 
> i'm pretty new to D (wrote a simple Hello World), but ... shouldn't the
> out assert be more like (x * x) == result?
> 
> long square_root(long x)
> in
> {
> assert(x >= 0);
> }
> out (result)
> {
> assert((result * result) == x);
> }
> body
> {
> return math.sqrt(x);
> }

try plugging in some example values. For example take x = 4. then result = 2 and 2*2 == 4.

In article <cnbeqf$13bg$1@digitaldaemon.com>, Ben Hinkle says...

Of course you are right,

sorry for that (should have read the function name closer, missed the *root*
part).

Sorry for that, again

Regan Heath wrote:
> On Mon, 15 Nov 2004 21:43:41 +0000 (UTC), Serge Ratke <Serge_member@pathlink.com> wrote:
> 
>> Hi guys,
>>
>> i'm pretty new to D (wrote a simple Hello World), but ... shouldn't the out
>> assert be more like (x * x) == result?
>>
>> long square_root(long x)
>> in
>> {
>> assert(x >= 0);
>> }
>> out (result)
>> {
>> assert((result * result) == x);
>> }
>> body
>> {
>> return math.sqrt(x);
>> }
> 
> 
> No, I think it's correct, using an example:
> 
> square_root(25);
> 
> if x == 25;
> then result == 5;
> therefore 5*5 == 25
> or result*result == x
> 
> Regan
> 

Ah I did mistake sqrt (A square root) with a power.

My mistake,
Sjoerd

Serge Ratke wrote:
> Hi guys,
> 
> i'm pretty new to D (wrote a simple Hello World), but ... shouldn't the out
> assert be more like (x * x) == result?
> 
> long square_root(long x)
> in
> {
> assert(x >= 0);
> }
> out (result)
> {
> assert((result * result) == x);
> }
> body
> {
> return math.sqrt(x);
> }

As has been noted, the general idea of this contract is correct. However, the out contract is incorrect because it assumes that x is a square number.  This contract will obviously fail if x is, for instance, 2.

WALTER: This needs to be changed on the website...

On Tue, 16 Nov 2004 06:53:44 -0700, Russ Lewis <spamhole-2001-07-16@deming-os.org> wrote:

> Serge Ratke wrote:
>> Hi guys,
>>  i'm pretty new to D (wrote a simple Hello World), but ... shouldn't the out
>> assert be more like (x * x) == result?
>>  long square_root(long x)
>> in
>> {
>> assert(x >= 0);
>> }
>> out (result)
>> {
>> assert((result * result) == x);
>> }
>> body
>> {
>> return math.sqrt(x);
>> }
>
> As has been noted, the general idea of this contract is correct. However, the out contract is incorrect because it assumes that x is a square number.  This contract will obviously fail if x is, for instance, 2.
>
> WALTER: This needs to be changed on the website...
>

I'm assuming he meant double/real instead of long. (possibly from C's long double?)
I don't think i've seen (m)any root functions taking integrals.

-- 
Using Opera's revolutionary e-mail client: http://www.opera.com/m2/

Simon Buchan wrote:
> On Tue, 16 Nov 2004 06:53:44 -0700, Russ Lewis  <spamhole-2001-07-16@deming-os.org> wrote:
>> As has been noted, the general idea of this contract is correct.  However, the out contract is incorrect because it assumes that x is a  square number.  This contract will obviously fail if x is, for instance,  2.
>>
>> WALTER: This needs to be changed on the website...
>>
> 
> I'm assuming he meant double/real instead of long. (possibly from C's long  double?)
> I don't think i've seen (m)any root functions taking integrals.

Well, if they were floats, then it would stil be broken, because of float rounding issues.  You'd have to do something like
  assert(abs((result*result) - x) < SOME_SMALL_VALUE);