Smalltalk › Squeak › Squeak - Beginners

Rouding Floats up to a number of decimals - Method Finder

_‹ Previous Topic Next Topic _›

Classic

List

Threaded

14 messages Options

cedreek

Rouding Floats up to a number of decimals - Method Finder

I couldn't find such a function. I tryed in the method finder but find
nothing: MethodFinder methodFor: #(#(1.2288 3) 1.228).

Do somebody know if such a method exist ?

Meanwhile I did that:

Float>>trimAfter: aNumberOfDecimal "roundAfter:"
| number |
number := 10 raisedTo: aNumberOfDecimal.
^(self * 10^aNumber) truncated "rounded" / 10^aNumber asFloat

and still MethodFinder methodFor: #(#(1.2288 3) 1.228) returns '(no
single method does that function) '
whereas 1.2288 roundedAfter: 3 returns 1.229

Does anybody know what was wrong in my method finder use ?

Thanks

Cédrick

_______________________________________________
Beginners mailing list
[hidden email]
http://lists.squeakfoundation.org/mailman/listinfo/beginners

Herbert König

Re: Rouding Floats up to a number of decimals - Method Finder

Hello Cédrick,

c> I couldn't find such a function. I tryed in the method finder but find
c> nothing: MethodFinder methodFor: #(#(1.2288 3) 1.228).

afaik method finder doesn't find *every* method but tries a a fixed
set of messages. Methods that would do the job but are not in the list
will not be found.

MethodFinder>>initialize and friends suggest this.

Cheers,

Herbert

_______________________________________________
Beginners mailing list
[hidden email]
http://lists.squeakfoundation.org/mailman/listinfo/beginners

cedreek

Re: Rouding Floats up to a number of decimals - Method Finder

> afaik method finder doesn't find *every* method but tries a a fixed
> set of messages. Methods that would do the job but are not in the list
> will not be found.
>
> MethodFinder>>initialize and friends suggest this.

stupid me :)

but.. I think I'll read these methods another day...

Thanks Herbert

So anybody has an idea of methods that truncate/round Floats ?
_______________________________________________
Beginners mailing list
[hidden email]
http://lists.squeakfoundation.org/mailman/listinfo/beginners

Herbert König

Re[2]: Rouding Floats up to a number of decimals - Method Finder

Hello Cédrick,

c> So anybody has an idea of methods that truncate/round Floats ?
sorry for ignoring your main question.

Float has a truncate and round off protocal with metods that truncate
and round to integer.

So it looks you have to do something like

(aFloat * (10^desiredDecimals) + 0.5) truncated / (10^desiredDecimals)

or without the + 0.5 for truncation.

Cheers,

Herbert

_______________________________________________
Beginners mailing list
[hidden email]
http://lists.squeakfoundation.org/mailman/listinfo/beginners

cedreek

Re: Re[2]: Rouding Floats up to a number of decimals - Method Finder

>
> Float has a truncate and round off protocal with metods that truncate
> and round to integer.
>
> So it looks you have to do something like
>
> (aFloat * (10^desiredDecimals) + 0.5) truncated / (10^desiredDecimals)
>
> or without the + 0.5 for truncation.
>

close to what I proposed earlier ;-) Just need a conversion to Float
as it returns a fraction.

so no existing method... Isn't it something common ?
_______________________________________________
Beginners mailing list
[hidden email]
http://lists.squeakfoundation.org/mailman/listinfo/beginners

Marcin Tustin

Re: Re[2]: Rouding Floats up to a number of decimals - Method Finder

In reply to this post by Herbert König

Why is it important to you to have a float? Once you've truncated it, you have a rational number, by definition.
Are you sure you want to truncate the number, rather than performing the truncation in the presentation layer?

On Wed, Jul 23, 2008 at 10:34 AM, cdrick <[hidden email]> wrote:

>
> Float has a truncate and round off protocal with metods that truncate
> and round to integer.
>
> So it looks you have to do something like
>
> (aFloat * (10^desiredDecimals) + 0.5) truncated / (10^desiredDecimals)
>
> or without the + 0.5 for truncation.
>

close to what I proposed earlier ;-) Just need a conversion to Float
as it returns a fraction.

so no existing method... Isn't it something common ?

_______________________________________________
Beginners mailing list
[hidden email]
http://lists.squeakfoundation.org/mailman/listinfo/beginners

_______________________________________________
Beginners mailing list
[hidden email]
http://lists.squeakfoundation.org/mailman/listinfo/beginners

cedreek

Re: Re[2]: Rouding Floats up to a number of decimals - Method Finder

> Why is it important to you to have a float? Once you've truncated it, you
> have a rational number, by definition.
> Are you sure you want to truncate the number, rather than performing the
> truncation in the presentation layer?

In my case, this is only for presentation. The real float doesn't
change. And anyway rounded and truncated doesn't change the number
value as they are immutable.

a := 3.4456.
a rounded.
a "-> 3.4456"

You meant maybe I use ScaledDecimal instead of Float?
_______________________________________________
Beginners mailing list
[hidden email]
http://lists.squeakfoundation.org/mailman/listinfo/beginners

Herbert König

Re[4]: Rouding Floats up to a number of decimals - Method Finder

In reply to this post by Marcin Tustin

Hello Marcin,

agreed with the the beginning of your post.

Floats (their finite precision representation) often create as many
problems as they solve, if it weren't for trigonometric functions I'd
try to keep away from them as far as I could :-))

MT> Are you sure you want to truncate the number, rather than
MT> performing the truncation in the presentation layer?

And then Cédrick will go searching the printing protocol of float and
find the same problem in a different disguise.

Float's private protocol might contain a start with absPrint.....

Cheers,

Herbert

_______________________________________________
Beginners mailing list
[hidden email]
http://lists.squeakfoundation.org/mailman/listinfo/beginners

Herbert König

Re[4]: Rouding Floats up to a number of decimals - Method Finder

In reply to this post by cedreek

Hello Cédrick,

search the "truncating and rounding" and "printing" protocols of Number
there you'll find some help.

--
Cheers,

Herbert

_______________________________________________
Beginners mailing list
[hidden email]
http://lists.squeakfoundation.org/mailman/listinfo/beginners

cedreek

Re: Re[4]: Rouding Floats up to a number of decimals - Method Finder

arffff thanks all ;)

1.23456 printShowingDecimalPlaces: 3

I knew that had to exist...Since I know the method finder I became
lazy to search far ;)

See you

Cédrick

2008/7/23 Herbert König <[hidden email]>:

> Hello Cédrick,
>
> search the "truncating and rounding" and "printing" protocols of Number
> there you'll find some help.
>
> --
> Cheers,
>
> Herbert
>
> _______________________________________________
> Beginners mailing list
> [hidden email]
> http://lists.squeakfoundation.org/mailman/listinfo/beginners
>

_______________________________________________
Beginners mailing list
[hidden email]
http://lists.squeakfoundation.org/mailman/listinfo/beginners

Zulq Alam-2

Re: Rouding Floats up to a number of decimals - Method Finder

In reply to this post by cedreek

Are #truncateTo: and #roundTo: what you want?

1.2288 roundTo: 1 "1"
1.2288 roundTo: 0.1 "1.2"
1.2288 roundTo: 0.01 "1.23"
1.2288 roundTo: 0.001 "1.229"
1.2288 roundTo: 0.0001 "1.2288"
1.2288 roundTo: 0.05 "1.25"
1.2288 roundTo: 0.002 "1.228"

1.2288 truncateTo: 1 "1"
1.2288 truncateTo: 0.1 "1.2"
1.2288 truncateTo: 0.01 "1.22"
1.2288 truncateTo: 0.001 "1.228"
1.2288 truncateTo: 0.0001 "1.2288"
1.2288 truncateTo: 0.05 "1.2"
1.2288 truncateTo: 0.002 "1.228"

Z.

cdrick wrote:
>
> So anybody has an idea of methods that truncate/round Floats ?

_______________________________________________
Beginners mailing list
[hidden email]
http://lists.squeakfoundation.org/mailman/listinfo/beginners

cedreek

Re: Re: Rouding Floats up to a number of decimals - Method Finder

> 1.2288 roundTo: 1 "1"
> 1.2288 roundTo: 0.1 "1.2"
> 1.2288 roundTo: 0.01 "1.23"
> 1.2288 roundTo: 0.001 "1.229"
> 1.2288 roundTo: 0.0001 "1.2288"
> 1.2288 roundTo: 0.05 "1.25"
> 1.2288 roundTo: 0.002 "1.228"

Stupid me again ;)

I only tried #roundTo: with Integer.

But #printShowingDecimalPlaces: is alright ;)

Thanks...

Cédrick

_______________________________________________
Beginners mailing list
[hidden email]
http://lists.squeakfoundation.org/mailman/listinfo/beginners

Nicolas Cellier-3

Re: Rouding Floats up to a number of decimals - Method Finder

cdrick a écrit :

>> 1.2288 roundTo: 1 "1"
>> 1.2288 roundTo: 0.1 "1.2"
>> 1.2288 roundTo: 0.01 "1.23"
>> 1.2288 roundTo: 0.001 "1.229"
>> 1.2288 roundTo: 0.0001 "1.2288"
>> 1.2288 roundTo: 0.05 "1.25"
>> 1.2288 roundTo: 0.002 "1.228"
>
> Stupid me again ;)
>
> I only tried #roundTo: with Integer.
>
> But #printShowingDecimalPlaces: is alright ;)
>
> Thanks...
>
> Cédrick
>

You might want to check for http://bugs.squeak.org/view.php?id=5640

Installer mantis ensureFix: 5640.

_______________________________________________
Beginners mailing list
[hidden email]
http://lists.squeakfoundation.org/mailman/listinfo/beginners

Nicolas Cellier-3

Re: Rouding Floats up to a number of decimals - Method Finder

In reply to this post by cedreek

cdrick a écrit :

> I couldn't find such a function. I tryed in the method finder but find
> nothing: MethodFinder methodFor: #(#(1.2288 3) 1.228).
>
> Do somebody know if such a method exist ?
>
> Meanwhile I did that:
>
> Float>>trimAfter: aNumberOfDecimal "roundAfter:"
> | number |
> number := 10 raisedTo: aNumberOfDecimal.
> ^(self * 10^aNumber) truncated "rounded" / 10^aNumber asFloat
>
> and still MethodFinder methodFor: #(#(1.2288 3) 1.228) returns '(no
> single method does that function) '
> whereas 1.2288 roundedAfter: 3 returns 1.229
>
> Does anybody know what was wrong in my method finder use ?
>
> Thanks
>
> Cédrick
>
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Beginners mailing list
> [hidden email]
> http://lists.squeakfoundation.org/mailman/listinfo/beginners

Try with

MethodFinder methodFor: #(#(1001.2288 0.001) 1001.228).

It works even if (1001.2288 truncateTo: 0.001) ~= 1001.228, because of
inexact arithmetic (in fact, it may depends which library is reading the
Float; results are different if above decimal representation are rounded
to nearest Float as SqNumberParser does).

The magic behind is that MethodFinder uses closeTo: rather than =, and
this is right because no one should rely on Float equality after
performing some inexact operations.

Nicolas

_______________________________________________
Beginners mailing list
[hidden email]
http://lists.squeakfoundation.org/mailman/listinfo/beginners