math round woes

math round woes

[Ivan Kanis]

Hello,

I am having problem this week with org-drill failing due to a rounding
error.

(round 586514472.354) -> eval: Arithmetic range error: "round",
586514472.354

I thought I'd patch things with calc:

(require 'calc-arith)
(math-round 586514472.354) -> 586514472.354

I looked at math-round and it calls math-trunk:

(math-trunk 1.1) -> 1.1

What's going on?

I am on the emacs-24 branch.
-- 
Static from nylon underwear.
    -- BOFH excuse #57

Re: math round woes

[David Kastrup]

Ivan Kanis <ivan@kanis.fr> writes:

> Hello,
>
> I am having problem this week with org-drill failing due to a rounding
> error.
>
> (round 586514472.354) -> eval: Arithmetic range error: "round",
> 586514472.354
>
> I thought I'd patch things with calc:
>
> (require 'calc-arith)
> (math-round 586514472.354) -> 586514472.354
>
> I looked at math-round and it calls math-trunk:
>
> (math-trunk 1.1) -> 1.1
>
> What's going on?

Uh, calc does calculations with its own number representation rather
than Lisp floating point numbers.  But calc cannot actually help you in
any way since the basic problem is that the result _will_ _not_ _fit_
into a Lisp integer.

If you are fine with putting it in a floating point number, use fround
rather than round.

-- 
David Kastrup

Re: math round woes

[Andreas Schwab]

Ivan Kanis <ivan@kanis.fr> writes:

> (require 'calc-arith)
> (math-round 586514472.354) -> 586514472.354
>
> I looked at math-round and it calls math-trunk:
>
> (math-trunk 1.1) -> 1.1
>
> What's going on?

Since you are calling an internal calc function you need to use its
internal number representation (and s/k/c/).

Andreas.

-- 
Andreas Schwab, schwab@linux-m68k.org
GPG Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

Re: math round woes

[Rüdiger Sonderfeld]

On Saturday 30 August 2014 11:42:46 Ivan Kanis wrote:
> (round 586514472.354) -> eval: Arithmetic range error: "round",
> 586514472.354

It seems you are using GNU Emacs on a 32 bit system without wide-int support.  
Try building GNU Emacs using `--with-wide-int`.

Regards,
Rüdiger

Re: math round woes

[Ivan Kanis]

August, 30 at 11:57 Andreas wrote:

> Ivan Kanis <ivan@kanis.fr> writes:
>
>> (require 'calc-arith)
>> (math-round 586514472.354) -> 586514472.354
>>
>> I looked at math-round and it calls math-trunk:
>>
>> (math-trunk 1.1) -> 1.1
>>
>> What's going on?
>
> Since you are calling an internal calc function you need to use its
> internal number representation (and s/k/c/).

What's s/k/c?

Re: math round woes

[Andreas Schwab]

Ivan Kanis <ivan@kanis.fr> writes:

> August, 30 at 11:57 Andreas wrote:
>
>> Ivan Kanis <ivan@kanis.fr> writes:
>>
>>> (require 'calc-arith)
>>> (math-round 586514472.354) -> 586514472.354
>>>
>>> I looked at math-round and it calls math-trunk:
>>>
>>> (math-trunk 1.1) -> 1.1
>>>
>>> What's going on?
>>
>> Since you are calling an internal calc function you need to use its
>> internal number representation (and s/k/c/).
>
> What's s/k/c?

See sed(1).

Andreas.

-- 
Andreas Schwab, schwab@linux-m68k.org
GPG Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

Re: math round woes

[Ivan Kanis]

August, 30 at 12:29 Rüdiger Sonderfeld wrote:

> On Saturday 30 August 2014 11:42:46 Ivan Kanis wrote:
>> (round 586514472.354) -> eval: Arithmetic range error: "round",
>> 586514472.354
>
> It seems you are using GNU Emacs on a 32 bit system without wide-int support.  
> Try building GNU Emacs using `--with-wide-int`.

It works, thank you! I am sure I have seen the error on a 64bit Linux,
though... 
-- 
In science the credit goes to the man who convinces the world, not to
the man to whom the idea first occurs.
    -- Francis Darwin