Re : Re: "Unknown function" message for compiled function

["Ewan Delanoy" <ewan.delanoy@gmx.fr>]

Thanks Karim for your answer.

> why the 'local_' prefix in your (indeed, local) variable names; (to my
 > taste) it makes the script a little harder to read.

Because (as far as I could understand) it seems that even though I declare a variable
"toto" to be local in some function, the GP interpreter will complain if I try to re-use the name "toto" later as a function name.

  Generally speaking, my impression is that reusability of names is limited in GP somehow.
  Is that another gross misunderstanding of mine, or is there really a limitation of Pari-GP on this ?

> any particular reason to stick to pari-2.3.2 ?

  Besides this user's natural laziness, no.


Ewan

Â

----- Message d'origine -----

De : Karim Belabas

EnvoyÃs : 17.05.12 11:46

à : Ewan Delanoy

Objet : Re: "Unknown function" message for compiled function

 
* Ewan Delanoy [2012-05-17 11:31]: 
>  I have a function below (called "polforcedshrink") that does not work (whenever I call it I get the error message " *** unknown function or error in formal parameters: "), although the interpreter remains silent at compile time. 
> 
>  Can anyone tell me what's wrong with the code ? 
> 
>  Thanks in advance, 
> 
>  Ewan 
> 
>  The code : 
> 
>  polforcedshrink(paul,var,a,b,l)= 
>  { 
>  local(local_runner,local_c,local_adj,answer); 
>  answer=[]; 
>  local_runner=paul; 
>  for(k=1,length(l),\ 
>  local_c=l[k];\ 
>  answer=concat([subst(local_runner,var,local_c)],answer);\ 
>  local_adj=1-2*(local_c==b);\ 
>  local_runner=local_adj*((local_runner-subst(local_runner,var,local_c))/(var-local_c));\ 
>  ); 
>  return(answer); 
>  } 
>  ? test1=paulforcedshrink(g1,x,0,1,[1,0,0,0,0]) 
           ^^^^ replace by pol, it will work :-) 

Cheers, 

    K.B. 

P.S: Some minor comments: 
- you don't need the trailing \ (since the multi-line text is already enclosed 
  within braces { }) 

- why the 'local_' prefix in your (indeed, local) variable names; (to my 
  taste) it makes the script a little harder to read. 

- I am not too fond of (valid) cryptic constructs like 
      local_adj=1-2*(local_c==b); 
  (ab)using boolean truth values as integers. Why not 
      local_adj = if (local_c==b, -1, 1); 
  ? 

- any particular reason to stick to pari-2.3.2 ? 
-- 
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] 
`