Charles Greathouse on Wed, 09 May 2012 20:21:07 +0200


[Date Prev] [Date Next] [Thread Prev] [Thread Next] [Date Index] [Thread Index]

Re: ispolygonal


I'm not sure if the explosion of error types would be useful.  Maybe
it would be best to stick to just one or two (e_INEQUALITY, e_BADINPUT
by whatever name).

Charles Greathouse
Analyst/Programmer
Case Western Reserve University

On Wed, May 9, 2012 at 10:35 AM, Karim Belabas
<Karim.Belabas@math.u-bordeaux1.fr> wrote:
> * Bill Allombert [2012-05-09 16:15]:
>> On Wed, May 09, 2012 at 02:20:52PM +0200, Karim Belabas wrote:
>>> The rationale was to avoid the e_MISC error type [ which is essentially
>>> impossible to trap cleanly ], and consider that the expected input
>>> "type" was 'a t_INT in Z_{> 2}'.
>>>
>>> This is a bit far-fetched and potentially confusing, I temporarily changed
>>> it to e_MISC.
>>>
>>> Presumably we need a general error type for "violated inequality" ?
>>>
>>>   pari_err(e_INEQ, "ispolygonal", "s > 2", s);
>>
>> We could define
>> pari_err(e_TOOSMALL,"ispolygonal",s,gen_2)
>>     --> "%s: %Ps must be > %Ps"
>> In practice I expect this would cover almost all case and allow the user
>> to retrieve the inequality.
>
> We might need
>
>  e_TOOSMALL (must have s > ...)
>  e_TOOLARGE (must have s < ...)
>  e_BADVALUE (must have s != ...)
>
> In fact, I'd expect the last one (BADVALUE) to be the most common (0 in
> arithmetic functions, poles of transcendental functions, etc.).
>
> The second one might not be needed at all since e_OVERFLOW might cover
> all existing cases.
>
> Cheers,
>
>    K.B.
> --
> Karim Belabas, IMB (UMR 5251)  Tel: (+33) (0)5 40 00 26 17
> Universite Bordeaux 1          Fax: (+33) (0)5 40 00 69 50
> 351, cours de la Liberation    http://www.math.u-bordeaux1.fr/~belabas/
> F-33405 Talence (France)       http://pari.math.u-bordeaux1.fr/  [PARI/GP]
> `