Re: Some GP misbehaviours

[Karim BELABAS <Karim.Belabas@math.u-psud.fr>]

On Wed, 28 Aug 2002, Bill Allombert wrote:
> During my holidays, I found the following bugs:
> 1) the protoype of trap0 is DI,DI which is invalid, it should be DIDI
> 2) ??numdiv state that the result is a long. In fact gnumbdiv and numbdiv
> return a GEN.

Both fixed.

> 3) ? M=mathilbert(5);M[,1][2]=3;M
>   ***   incorrect type or length in matrix assignment: M=mathilbert(5);M[,1][2]
> 4) ? M=mathilbert(2);M[1,][2]=3;M
> %2 =
> [1 1/2]
> [1/2 1/3]
> This is wrong...

Fixed. It's easy to collapse [r,][c] into [r,c].  [ matcell() ]

> 5)? 1 /**/ 3
> %1 = 13
> This is in accordance with the doc
> ? 1 /*
> %3 = 1
> comment> */ 3
> %4 = 3
>
> It look like newline inside comments are not ignored

Fixed.  [ filtre0() ]

> 6) ? printtex("a"(x+1)2)
> a{x + 1}{2}
> Maybe it is ``smart'' but in accordance to the doc, printtex do
> automatic concatenation, not automatic splitting, so this should be identical
> to
> ? printtex(Str("a"(x+1)2))
> ax + 12

This one is not fixed yet. A brute force fix is obvious, but detrimental to
efficiency (for all routines using a 's*' prototype). I'm still thinking
about it.

    Karim.
-- 
Karim Belabas                    Tel: (+33) (0)1 69 15 57 48
Dép. de Mathematiques, Bat. 425  Fax: (+33) (0)1 69 15 60 19
Université Paris-Sud             Email: Karim.Belabas@math.u-psud.fr
F-91405 Orsay (France)           http://www.math.u-psud.fr/~belabas/
--
PARI/GP Home Page: http://www.parigp-home.de/